Nitrogen removal from wastewater through microbial electrolysis cells and cation exchange membrane
© Haddadi et al.; licensee BioMed Central Ltd. 2014
Received: 16 May 2013
Accepted: 10 February 2014
Published: 17 February 2014
Vulnerability of water resources to nutrients led to progressively stricter standards for wastewater effluents. Modification of the conventional procedures to meet the new standards is inevitable. New technologies should give a priority to nitrogen removal. In this paper, ammonium chloride and urine as nitrogen sources were used to investigate the capacity of a microbial electrolysis cell (MEC) configured by cation exchange membrane (CEM) for electrochemical removal of nitrogen over open-and closed-circuit potentials (OCP and CCP) during biodegradation of organic matter. Results obtained from this study indicated that CEM was permeable to both organic and ammonium nitrogen over OCP. Power substantially mediated ammonium migration from anodic wastewater to the cathode, as well. With a urine rich wastewater in the anode, the maximum rate of ammonium intake into the cathode varied from 34.2 to 40.6 mg/L.h over CCP compared to 10.5-14.9 mg/L.h over OCP. Ammonium separation over CCP was directly related to current. For 1.46-2.12 mmol electron produced, 20.5-29.7 mg-N ammonium was removed. Current also increased cathodic pH up to 12, a desirable pH for changing ammonium ion to ammonia gas. Results emphasized the potential for MEC in control of ammonium through ammonium separation and ammonia volatilization provided that membrane characteristic is considered in their development.
KeywordsMicrobial electrolysis cell Ammonium Urine Diffusion Closed-circuit potential Open-circuit potential
Wastewater treatment technologies are energy demanding processes. As an example, 3-4% of the total energy demand in the United Kingdom comes from wastewater processes. Also, the Unites States consumes 21 billion kwh for this purpose . Nitrogen removal in wastewater treatment is a main concern. The majority of nitrogen in municipal wastewater originates from urine. An adult person typically produce an average of 11.5 g-N/d urine . Nitrogen release into environment could lead to autrification in water reservoirs. Most treatment methods have mainly been established for organic removal. Recent standards have forced treatment processes to be enhanced for nitrogen elimination. On the other hand, extra considerations should be taken into account when nutrient removal is desirable beside organic removal; this complicates the treatment method  and poses extra costs. Hence, urine separation at the source can hold the promise of sustainability of wastewater management .
Bioelectrochemical systems convert organic matter to protons and recoverable electrons within an anode compartment. Electrons flow through an external circuit to cathode and conduct reduction reactions. Usually hydroxide ions are produced from oxygen reduction in the cathode . Simply, two configurations of BESs are used for wastewater treatment purposes: microbial fuel cell (MFC) or microbial electrolysis cell (MEC). Anodic reactions are almost the same in both configurations. Anode and cathode are connected to each other by a load in MFC; whereas, in MEC, anode potential adjusted to a specific amount or external voltage is invested to prepare enough energy for accomplishment of reactions in the cathode to produce hydrogen gas . These systems have been tested for purification of synthetic and real wastewaters, organic and inorganic pollutants [7, 8]. They were able to capture energy from treatment of domestic wastewater , food-processing wastewater , landfill leachate , cellulose , and even alcohol [13, 14]. They have shown recycling energy from waste and it is time to evaluate their capability for removal of pollutants that might not have any energy content but are also environmentally important such as nitrogen compounds.
Once removing nitrogen from wastewater using bioelectrochemical systems comes to mind, the bioremediation aspect of BESs might be highlighted. During bioremediation, nitrate is used in the biocathode, acts as the final electron acceptor and is reduced to nitrite or nitrogen gas [15, 16]. In fact, bacteria take electrons from electrodes and participate in nitrate reduction. Biological standard potential (at pH = 7) of NO3-/NO2- and NO3-/N2 are +0.43 V and +0.74 V respectively; while, this potential for the commonly oxidant in the MFC, O2/H2O, is +0.82 V [17, 18]. In BES systems anode compartment is operated in anaerobic condition . Oxygen or oxygen compounds interfere with energy production process . Consequently, biological removal of nitrate is not suggested in the bioanode. Moreover, lack of oxygen in the anode prevents nitrification. In theory, bioanode seems to have the least chance for dealing with this nutrient since it makes nitrification or denitrification impossible. However, oxygen leakage from air-cathode might contribute to nitrification . Nitrification and denitrification is the widely used concept for nitrogen removal in biological technologies. However, it should be noted that this process has some limitations for nitrogen rich wastewaters. Physicochemical processes also can remove nitrogen from wastewater but they are much more expensive than biological methods. Bioelectrocemical systems as their name implies are not sole biological methods. They also involve physical and chemical reactions to clean wastewater and produce energy.
Beside electrons, organic biodegradation in BES is also associated with proton production. As electrons leave the anode, charge would be imbalanced in the system. Then protons might also immigrate to the cathode; otherwise, some negative charges from the cathode should come to the anode to account for charge neutrality. In most designs, a membrane separates anode and cathode cells and so charges must cross the permeable membrane. Hydroxyl ions or protons are responsible for this charge neutrality, but other cations or anions may also be involved. Some studies have reported ions traveling other than protons or hydroxide . Hence, the use of membrane in BES is sometimes controversial .
Despite the fact that bioelectrochemical systems have potential for removing organic matters and producing energy, simultaneous biotransformation of organic waste and nutrient from wastewater in one bio-cell seems to be impractical. In addition, nutrient removal and scaling up remain unaddressed. Nevertheless, bio-mediated electrochemical removal of nitrogen in the bioanode, if feasible, can overcome those limitations and enables simultaneous nitrogen and organic matter removal in the anode. Ion transfer instead of charge transfer can provide a chance for taking off some unpleasant ions from wastewater. Here, in this study, MEC system configured by cation exchange membrane (CEM) was tested to remove nitrogen as ammonium ion from ammonium and urine-rich wastewaters.
Cation exchange membranes are supposed to exchange ions but this study would have a look at its permeability to organic nitrogen when faced with urine, a rich source of organic nitrogen.
Material and methods
Urine and acetate
Concentration g/L Ammonium and acetate (20 mM)
Ammonium and acetate (25 mM)
MEC configuration and operation
Measurements and calculations
dQ states cumulative current (mA.s) and F is Farady’s constant (96485 A.s/mol).
pH was measured by a pH meter. For urine experiments, ammonium and TKN were traced every day almost over a one-week period. Ammonium in ammonium chloride fed reactors was traced over 24 hours with 4 hour-intervals.
Total Kjeldahl Nitrogen (TKN) and ammonium nitrogen (NH4-N) were determined via colorimetry method by an ammonia analyzer (Bran + Luebbe AutoAnalyzer 3). Prior to quantifying TKN, samples were digested at 200°C for 1.5 hours and at 350°C for 3.5 hours in sulfuric acid and left overnight to be cooled. Organic nitrogen was calculated as:
Where, parameters are: W ammonium mass (mmol), Ct ammonium concentration at time t (mg-N/L), Vt water volume in the cathode at time t (L), Mw molecular weight of nitrogen (14 g/mol).
Results and discussion
Nitrogen removal from mimic urine medium by CEM
Following appearance of ammonium in the anode, it was transferred to the cathode. Figure 2b shows catholyte ammonium as a function of time and anolyte ammonium. At the end of operation, ammonium in the anode was found to be 2580 mg-N/L that resulted in a 782 mg-N/L cumulative ammonium in the cathode.
The test was replicated to ensure the repeatability of the test results. Some differences were observed between two series runs with similar urine concentration owing to urine dissociation rate. In the second run, however, urine decomposition took place faster which might be related to bacteria activity. More than 30% of the TKN was as ammonium after 45 h of operation and it approached 80% over 62 h. Such as previous run, most of the organic nitrogen was transferred to the cathode within the first two days. The maximum organic nitrogen in the cathode was 1500 mg/L and took place over 138 h (Figure 2c); parallelly, NH4-N was also migrated and reached to 1120 mg/L (Figure 2d).
Reactors fed by mimic urine showed that nitrogen in organic and ammonium forms could cross cation exchange membrane. Ammonium transfer is consistent with the characteristic of this type of membrane; but organic nitrogen is not supposed to be able to go through the membrane. Over 100 hours, the cathode TKN concentration was around 20% of the anode TKN. However, it does not mean that 20% of the TKN of anolyte went to the cathode, as the cathode was smaller than the anode. In both runs, organic nitrogen was transferred to the cathode exceeded ammonium nitrogen. The reason behind this might be its concentration. The driving force for nitrogen transfer was concentration gradient. The highest gradient between two chambers was at the beginning or in the first two days of the operation over which organic nitrogen was the prominent species. It constitutes 70-90% of the TKN during the first two days. As nitrogen left the anode toward the cathode, gradient concentration decreased and slowed further nitrogen transfer rate.
In addition to TKN, water could also pass through the membrane. Water, as expected in osmosis phenomenon, moved in the opposite direction of ions through the membrane. As the cathode was filled with deionized water, high gradient concentration was created between catholyte and anolyte. Then, the anode pulled water from the cathode. Nitrogen transfer might not have been significant, but when water left the cathode, cathodic ammonium seemed more concentrated than it really was. Unfortunately, we did not measure water transfer rate, but in two other experiments running with 11 g-N/L we had to add 15 mL water to the cathode, after 5 days of operation, to compensate water loss. Water loss would mislead in showing concentration based on C1V1 = C2V2 relationship in which parameters are: C1 real concentration, V1 initial volume, C2 the measured concentration, and V2 remained volume after water loss or intake.
Water and even acetate passed through the nafion 117 membrane [25, 26], a proton exchange membrane which typically is used in BES studies. Rozendal et al. (2007) observed 33.5 ml water loss through nafion membrane having surface area of 256 cm2 over 48 h.
Maximum rates of ammonium increase in the cathode compartment
Maximum rates (mg/L.h)
Over 24 h
Over 48 h
Following the urine dissociation, pH of the solutions changed. Figure 2e shows pH changes of anolyte due to ammonium production. pH increased up to 9.7 and 9.5 in the anode and cathode, respectively.
Current was unattainable with this urine medium under the stated operation condition. The medium composition to attain current was also changed. To trace nitrogen over current generation process, some pretests were conducted and they revealed that MEC was able to produce current from a medium containing 4000 mg-N/L urine and acetate as an electron donor.
Nitrogen transfer during closed-circuit voltage from urine rich substrate
Power density curves for MEC reactors are shown in Figures 3b and 3c. The reactors began energy production at different times. As shown in Figure 3b, the power generation and urine hydrolysis started after 50 hours of operation. During this delay, nitrogen was transferred as organic form and diminished concentration gradient between anolyte and catholyte. A response in ammonium transmission to current was observed quickly after power production. NH4-N transport rates peaked when power increased, but this peak was also combined with ammonium peak in the anode. The ammonium transfer as a function of power density and ammonium content in the anode chamber are also presented in Figures 3b and 3c.
Before power generation, nitrogen had a chance to transfer as organic form. In the previous section, it was realized that the TKN transfer was independent of nitrogen type. Whatever the dominant nitrogen type was in the anode, the final TKN within cathode was similar in both series runs. As presented in Figures 3(a-c), during the current generation cycle, more than 45% of the all TKN transferred over a 150 h period occurred within 24 hours during which the current and urine dissociation were in maximum level. Urine dissociation into ammonium increased chance of nitrogen to pass through the membrane. Nitrogen should have been as ion form to take the responsibility of charge neutralization. Hence, as it was in organic form, the MEC could not have expedited its removal via transferring to the other chamber.
It was not very clear whether the transmission rate peak was due to current or ammonium concentration gradient due to coinciding of power production and urine dissociation peaks. To find whether the nitrogen equilibrium following current generation was a response to charge balance or it was a result of anolyte ammonium increase, the maximum ammonium concentrations in components for OCP and CCP were compared. Maximum ammonium concentration in the anode chamber with initial urine of 4 g-N/L reached 2680 mg-N/L; while, for this anolyte ammonium, 1580 mg-N/L ammonium was observed in the cathode (Figure 3d). In the second CCP run, anolyte and catholyte ammonium reached 2620 and 1460, respectively. Whereas upon OCP at 11 g-N/L initial urine, the ammonium in two different runs ranged from 2575 to 8345 mg-N/L in the anodes and 782 and 1260 mg-N/L in the cathodes, respectively. Although the initial concentrations of the ammonium in the CCP tests seemed to be lower than those shown by OCP tests, the transfer rates to the cathode were much higher. Based on anodic concentration, ammonium transferring upon CCP and OCP resulted in catholyte ammonium of 56-59% and 15-30% of anolyte ammonium, respectively. To understand the effect of current more clearly, the maximum catholyte ammonium increase rates occurred over 24-hour and 48-hour periods are provided in Table 2. Samples were taken every 24 hours.
pH was also changed during operations. pH increases in the anode compartment was deduced from the ammonium content and in the cathode from the ammonium transfer. pH changes in the cathode under CCP mode was mainly a result of current. Ammonium concentration in the anolyte always exceeded from the catolyte which raised pH only up to 9 (Figure 3e); However, catholyte pH in CCP increased up to 12. Following current drop, catholyte pH was decreased as much as 1–2 magnitudes.
Comparison between two modes disclosed the role of current on ammonium separation but dependency of ammonium upon ammonium dissociation made it difficult to distinguish the effect of current from diffusion and to define an absolute relationship for nitrogen transport through current within a urine rich wastewater. Controlling bacteria to hydrolyze urine in a constant or desirable rate is not easy. Hence, a series of short-term experiments were carried out with substrate containing ammonium (ammonium chloride) instead of urine. The results will be explained in the next section.
Ammonium removal from ammonium rich substrate over OCP and CCP
To balance the charge in bioelectrochemical systems, as electrons transfer to the cathode, a movement of ions begins from or toward the anode based on the applied membrane. CEM is designed to transfer cation or to inhibit anion transportation. In this study, reactors used CEM; hence, protons or cations should have moved from the anode to the cathode.
Considering that for each electron produced one proton should leave the anode, total ammonium was brought to mmol and was compared to cumulative current. The ammonium transport to the cathode during closed circuit potential is a function of current. Results can be seen in Figures 4b and 4c. Water loss and water intake during sampling were considered when calculating ammonium mass transfer. Over 24 hours, for 3.13 and 1.90 mmol electrons produced 2.12 and 1.46 mmol ammonium (29.7 and 20.5 mg-N) took the responsibility of charge neutralization.
As shown in Figures 4b and 4c, the share of current dedicated to ammonium removal decreased over time. This is perhaps due to proton concentration. In fact, the reason behind ion transferring instead of charge transferring is their higher concentration in the reactor other than protons . In the current study, the reactors were operated in the batch modes; hence, anolyte ammonium decreased over time because of ammonium transmission to the cathode. Moreover, in the first day, current had increasing trend, which can be attributed to increase in proton concentration. Increase in the current and decrease in the ammonium concentration over time would diminish the ratio of ammonium to protons. Then, over time, protons were most probable to compete with ammonium and to slow down ammonium removal. Consequently, it can be concluded here that the fraction of current dedicated to ammonium separation decreased over time since the distance between ammonium removal curve and current production curve increased. This means α, the share of ammonium for charge balancing [mmol NH4(transferred) = ∝ (mmol electron) produced ], is a function of the current, ammonium concentration in the anode or even concentration of other ions. Comparison between Figures 4b and 4c supports this statement. These runs used the same ammonium concentration, but different acetate concentrations, lower acetate concentration produced lower current. Acetate was 20 mM in Figure 4b and 25 mM in Figure 4c; Clearly, the current curve and ammonium-transferred curve in Figure 4b are closer to each other than in Figure 4c. In both figures, the distance between current curve and ammonium curve increased over time. Current production rate was higher than ammonium removal rate. Although increase in the current decreased the contribution of the ammonium to charge neutralization, it improved the total ammonium removal as mg/L.
The nature of driving forces for ammonium transfer over OCP and CCP are different. Transfer of ammonium due to diffusion depends on concentration gradient between anolyte and catholyte. However, the ammonium transfer due to charge neutrality depends on the ratio of protons to the ammonium in the anode and is not related to ammonium concentration in the cathode. Hence, regardless of the limiting factors, physical ammonium removal is preferred for wastewaters with high ammonium concentration.
Ammonia stripping is a well-known process for ammonia removal from some industrial wastewaters over which pH is raised by addition of chemicals and the ammonium ion changes to ammonia and then air stripping brings out ammonia gas from wastewater. Reactions in MXC usually tend to increase pH in the cathode. The difference between catholyte pH in CCP mode and OCP mode was significant as shown in Figure 3e. Ammonium ion carries a hydrogen ion, although it is a basic compound and its transmission to the cathode causes more pH increase. In some cases, the cathodic pH did even reach 12. This pH enhances ammonium transformation to ammonia gas. Depending on temperature, more than 99% of the total ammonium is as ammonia form in pH 12 . Blowing air in the air-cathode microbial fuel cells can also ease its removal afterward. Then ammonia stripping might happen in MFC.
The role of current generation in bioelectrochemical systems on nitrogen fate and the potential of cation exchange membrane (CEM) for nitrogen transmission were addressed in this paper through an experimental work. Based on the results, urine and water could pass through CEM and transported between anode and cathode due to diffusion. The most portion of organic nitrogen transfer took place over the first two days. Ammonium was produced from urine and crossed the CEM; but, rates varied between OCP and CCP. Upon OCP while anolyte ammonium was 2575 and 8345 mg-N/L, catholyte ammonium was measured as 782 and 1260 mg-N/L. The catholyte NH4-N increased remarkably over current generation and terminated between 1580 and 1460 mg-N/L with the anolyte ammonium of 2680 and 2620 mg-N/L. Using ammonium chloride as a nitrogen source, CEM achieved a majority of charge balance by ammonium instead of protons and enabled ammonium removal. Difference between proton and ammonium concentrations was determining factor for ammonium separation. Of 3.13 and 1.90 mmol charges produced over a 24-hour period 2.12 and 1.46 mmol were balanced by ammonium. MEC systems made physical ammonium removal feasible during organic biodegradation and energy production. For a given reactor, if there is no limiting factor, physical ammonium removal by MEC systems would be better for higher nitrogen concentrations.
Cation exchange membrane
Microbial electrolysis cell
Microbial fuel cell
Total Kjeldahl nitrogen.
The authors would like to thank United Nations Educational Scientific and Cultural Organization, for funding this research through UNESCO/Keizo Obuchi Research Fellowships Programme, cycle 2011.
- Oh ST, Kim JR, Premier GC, Lee TH, Kim C, Sloan WT: Sustainable wastewater treatment: how might microbial fuel cells contribute. Biotechnol Adv 2010, 28(6):871–881. 10.1016/j.biotechadv.2010.07.008View ArticleGoogle Scholar
- Larsen TA, Gujer W: Separate management of anthropogenic nutrient solutions (human urine). Water Sci Technol 1996, 34(3–4):87–94.View ArticleGoogle Scholar
- Khin T, Annachhatre AP: Novel microbial nitrogen removal processes. Biotechnol Adv 2004, 22(7):519–532. 10.1016/j.biotechadv.2004.04.003View ArticleGoogle Scholar
- Escher BI, Pronk W, Suter MJF, Maurer M: Monitoring the removal efficiency of pharmaceuticals and hormones in different treatment processes of source-separated urine with bioassays. Environ Sci Technol 2006, 40(16):5095–5101. 10.1021/es060598wView ArticleGoogle Scholar
- Fornero JJ, Rosenbaum M, Cotta MA, Angenent LT: Carbon Dioxide addition to microbial fuel cell cathodes maintains sustainable catholyte pH and improves anolyte pH, alkalinity, and conductivity. Environ Sci Technol 2010, 44(7):2728–2734. 10.1021/es9031985View ArticleGoogle Scholar
- Nam J-Y, Tokash JC, Logan BE: Comparison of microbial electrolysis cells operated with added voltage or by setting the anode potential. Int J Hydrog Energy 2011, 36(17):10550–10556. 10.1016/j.ijhydene.2011.05.148View ArticleGoogle Scholar
- Logan BE, Hamelers B, Rozendal R, Schroder U, Keller J, Freguia S, Aelterman P, Verstraete W, Rabaey K: Microbial fuel cells: methodology and technology. Environ Sci Technol 2006, 40(17):5181–5192. 10.1021/es0605016View ArticleGoogle Scholar
- Heilmann J, Logan BE: Production of electricity from proteins using a microbial fuel cell. Water environment research : a research publication of the Water Environment Federation 2006, 78(5):531–537. 10.2175/106143005X73046View ArticleGoogle Scholar
- Liu G, Yates MD, Cheng S, Call DF, Sun D, Logan BE: Examination of microbial fuel cell start-up times with domestic wastewater and additional amendments. Bioresour Technol 2011, 102(15):7301–7306. 10.1016/j.biortech.2011.04.087View ArticleGoogle Scholar
- Oh S, Logan BE: Hydrogen and electricity production from a food processing wastewater using fermentation and microbial fuel cell technologies. Water Res 2005, 39(19):4673–4682. 10.1016/j.watres.2005.09.019View ArticleGoogle Scholar
- Greenman J, Galvez A, Giusti L, Ieropoulos I: Electricity from landfill leachate using microbial fuel cells: comparison with a biological aerated filter. Enzym Microb Technol 2009, 44(2):112–119. 10.1016/j.enzmictec.2008.09.012View ArticleGoogle Scholar
- Rismani-Yazdi H, Christy AD, Dehority BA, Morrison M, Yu Z, Tuovinen OH: Electricity generation from cellulose by rumen microorganisms in microbial fuel cells. Biotechnol Bioeng 2007, 97(6):1398–1407. 10.1002/bit.21366View ArticleGoogle Scholar
- Kim JR, Jung SH, Regan JM, Logan BE: Electricity generation and microbial community analysis of alcohol powered microbial fuel cells. Bioresour Technol 2007, 98(13):2568–2577. 10.1016/j.biortech.2006.09.036View ArticleGoogle Scholar
- Luo H, Liu G, Zhang R, Jin S: Phenol degradation in microbial fuel cells. Chem Eng J 2009, 147(2–3):259–264.View ArticleGoogle Scholar
- Park HI, Kim D, Choi Y-J, Pak D: Nitrate reduction using an electrode as direct electron donor in a biofilm-electrode reactor. Process Biochem 2005, 40(10):3383–3388. 10.1016/j.procbio.2005.03.017View ArticleGoogle Scholar
- Gregory KB, Bond DR, Lovley DR: Graphite electrodes as electron donors for anaerobic respiration. Environ Microbiol 2004, 6(6):596–604. 10.1111/j.1462-2920.2004.00593.xView ArticleGoogle Scholar
- Schroder U: Anodic electron transfer mechanisms in microbial fuel cells and their energy efficiency. Phys Chem Chem Phys 2007, 9(21):2619–2629. 10.1039/b703627mView ArticleGoogle Scholar
- He Z, Angenent LT: Application of bacterial biocathodes in microbial fuel cells. Electroanalysis 2006, 18(19–20):2009–2015.View ArticleGoogle Scholar
- Kargi F, Eker S: Electricity generation with simultaneous wastewater treatment by a microbial fuel cell MFC with Cu and CuAu electrodes. Journal of Chemical Technology 38; Biotechnology 2007, 82(7):658–662. 10.1002/jctb.1723View ArticleGoogle Scholar
- Logan BE, Call D, Cheng S, Hamelers HVM, Sleutels THJA, Jeremiasse AW, Rozendal RA: Microbial Electrolysis Cells for High Yield Hydrogen Gas Production from Organic Matter. Environ Sci Technol 2008, 42(23):8630–8640. 10.1021/es801553zView ArticleGoogle Scholar
- Kim JR, Zuo Y, Regan JM, Logan BE: Analysis of ammonia loss mechanisms in microbial fuel cells treating animal wastewater. Biotechnol Bioeng 2008, 99(5):1120–1127. 10.1002/bit.21687View ArticleGoogle Scholar
- Rozendal RA, Hamelers HVM, Buisman CJN: Effects of membrane cation transport on pH and microbial fuel cell performance. Environ Sci Technol 2006, 40(17):5206–5211. 10.1021/es060387rView ArticleGoogle Scholar
- Harnisch F, Schröder U: Selectivity versus mobility: separation of anode and cathode in microbial bioelectrochemical systems. Chem Sus Chem 2009, 2(10):921–926. 10.1002/cssc.200900111View ArticleGoogle Scholar
- Wilsenach JA, Schuurbiers CAH, van Loosdrecht MCM: Phosphate and potassium recovery from source separated urine through struvite precipitation. Water Res 2007, 41(2):458–466. 10.1016/j.watres.2006.10.014View ArticleGoogle Scholar
- Rozendal RA, Hamelers HVM, Molenkamp RJ, Buisman CJN: Performance of single chamber biocatalyzed electrolysis with different types of ion exchange membranes. Water Res 2007, 41(9):1984–1994. 10.1016/j.watres.2007.01.019View ArticleGoogle Scholar
- Chae KJ, Choi M, Ajayi FF, Park W, Chang IS, Kim IS: Mass Transport through a proton exchange membrane (nafion) in microbial fuel cells. Energy Fuel 2007, 22(1):169–176.View ArticleGoogle Scholar
- Kim JR, Cheng S, Oh S-E, Logan BE: Power generation using different cation, anion, and ultrafiltration membranes in microbial fuel cells. Environ Sci Technol 2007, 41(3):1004–1009. 10.1021/es062202mView ArticleGoogle Scholar
- Thurston RV, Russo RC, Emerson K: Aquenous Ammonia Equilibrium-Tabulation of Percent Un-Ionizied Ammonia. United States: Environ Protection Agency; 1979.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.