stacked microbial desalination cells to enhance water desalination efficiency

Published: February 15, 2011

r 2011 American Chemical Society 2465 dx.doi.org/10.1021/es103406m | Environ. Sci. Technol. 2011, 45, 2465–2470

ARTICLE

pubs.acs.org/est

Stacked Microbial Desalination Cells to Enhance Water DesalinationEfficiencyXi Chen, Xue Xia, Peng Liang, Xiaoxin Cao, Haotian Sun, and Xia Huang*

State Key Joint Laboratory of Environment Simulation and Pollution Control, Department of Environmental Science and Engineering,Tsinghua University, Beijing, 100084, P.R. China

bS Supporting Information

ABSTRACT: Microbial desalination cell (MDC) is a new method to obtain clean water frombrackish water using electricity generated from organic matters by exoelectrogenic bacteria. Anionsand cations, derived from salt solution filled in the desalination chamber between the anode andcathode, move to the anode and cathode chambers under the force of electrical field, respectively. Onthe basis of the primitive single-desalination-chambered MDC, stacked microbial desalination cells(SMDCs) were developed in order to promote the desalination rate in the present study. The effectsof desalination chamber number and external resistance were investigated. Results showed that aremarkable increase in the total desalination rate (TDR) could be obtained by means of increasingthe desalination cell number and reducing the external resistance, which caused the charge transferefficiency increased since the SMDCs enabledmore pairs of ions separated while one electron passedthrough the external circuit. The maximum TDR of 0.0252 g/h was obtained using a two-desalination-chambered SMDC with an external resistance of 10 Ω, which was 1.4 times that of single-desalination-chamberedMDC. SMDCs proved to be an effective approach to increase the total water desalination rate if provided a proper desalinationchamber number and external resistance.

’ INTRODUCTION

The lack of clean water has become a bottleneck for thedevelopment of human society, accompanied by populationbooming and environmental pollution. Although total amountof water on earth is abundant, the useable freshwater is ratherlimited. Considering large capacities of seawater and brackishwater, desalination is considered as an important approach toproduce needed freshwater.1-3 The main desalination technol-ogies that have been commercialized include electrodialysis, re-verse osmosis (RO), and distillation. For the above approaches,the major concerns are the relatively intensive energy consump-tion in forms of heat or electricity.4,5 For example, the current ROsystems usually cost as much as 3-5 kWh/m3 of electricalenergy to produce drinking water.2,5 In order to reduce thetraditional energy consumption, renewable energies such assolar energy and wind power are under research to drivedesalination process.6 However, the costs of these technolo-gies, even higher than conventional ones, prevent them fromlarge-scale application.

A new method called microbial desalination cell (MDC),which can desalinate water using electricity generated by bacteriafrom wastewater, was first proposed by Cao et al. (2009). A threechambered reactor was constructed by inserting an anion ex-change membrane (AEM) next to the anode and a cationexchange membrane (CEM) next to the cathode of a microbialfuel cell (MFC), with the salt solution to be desalinated filled inthe middle chamber.7 The electricity generating mechanism ofMDC is similar to that of MFC. Current is generated by bacteria

on the anode from oxidizing organics, and electrons and protonsare released to the anode and anolyte respectively.8,9 As cationsare prevented from leaving the anode chamber by the AEM,anions (such as Cl-) move from the middle desalinationchamber to the anode. In the cathode, protons are consumedin the reduction reaction of oxygen, while cations (such as Naþ)in the middle chamber transfer across the CEM to the cathode.This process leads to water desalination in the middle chamber,without any external energy source required. In addition, elec-tricity is produced and organic matters in wastewater aredegraded by the anodic exoelectrogenic bacteria.

In the works of Cao et al., the concept of MDC wassuccessfully proved using a ferricyanide catholyte, and a desalina-tion ratio up to 90% was obtained in 24 h. The results demon-strated for the first time the great potential of MDCs as a low costdesalination process with environmental friendly benefits. Me-hanna et al. further developed an air-cathode MDC, which madeMDCs more useful for practical applications.10 Jacobson et al.constructed an upflow MDC for continuously operation anddiscussed the phenomena of bipolar electrodialysis and protontransport in the reactor.11 In practical desalination process, moreimportance is attached to desalination efficiency so as to reducethe economic cost. To promote the desalination rate of MDC, apossible solution is to increase electron transfer efficiency and

Received: October 8, 2010Accepted: January 27, 2011Revised: January 15, 2011

2466 dx.doi.org/10.1021/es103406m |Environ. Sci. Technol. 2011, 45, 2465–2470

Environmental Science & Technology ARTICLE

salt solution volume by developing a stacked desalination systemthat contains multiple desalination chambers, somewhat similarto electrodialysis. Another approach is to reduce the externalresistance, aimed at the energy produced by anodic bacteriaconsumed as much as possible on the internal desalination. In thepresent study, on the basis of the single-desalination-chamberedMDC reported by Cao et al., stacked microbial desalination cells(SMDCs) with an air-cathode were constructed to enhance thedesalination rate using multidesalination chambers combinedwith external resistance optimized. The effects of desalinationchamber number and external resistance on desalination ratewere examined.

’MATERIALS AND METHODS

SMDC Construction. SMDC were cubic-shaped bioreactors,which consisted of three blocks (anode chamber, cathodechamber and stacked desalination cells). The stacked desalina-tion cells were composed of desalination chambers and concen-trated chambers that were spaced by compartmental AEMs(DF120, Tianwei Membrane) and CEMs (Ultrex CMI7000,Membrane International). One desalination chamber was sepa-rated by one AEM and one CEM, respectively. For eachdesalination chamber, the AEM was on the side close to theanode while the CEM was on the other side close to the cathode.Between two adjacent desalination chambers laid one concen-trated chamber, which collects ions moving out from thedesalination chambers. One, two, and three desalination cham-bers, together with zero, one, and two concentrated chambers,respectively, were inserted between the anode and cathodechambers to form one-, two-, and three-desalination-chamberedSMDCs (referred to as 1-SMDC, 2-SMDC, and 3-SMDC,respectively). Figure 1 shows the photo and configuration of2-SMDC. Different chambers and ion exchange membranes(IEMs: AEM and CEM) were clamped together with gaskets

to provide a water seal between the chambers. The innerdiameter of the cross section of each chamber was 3.0 cm. Theeffective volumes of the anode, desalination, concentrated andcathode chambers were 21.2, 7.1, 7.1, and 7.1 mLwith width of 3,1, 1, and 1 cm, respectively. The anode chamber was filled withseveral pieces of carbon felt (each piece with the size of 1.0� 1.0� 1.0 cm),12 which served as electrode material. A graphite rod(5 mm diameter) was inserted into the felt to provide an externalelectrical contact. Prior to use, the carbon felt and graphite rodwere soaked in 1 M HCl for 48 h, followed by rinsing withdeionized water to remove trace metals. Cathode was based on30% wet-proofed carbon cloth (projected area: 7.1 cm2; E-Tek,Type B, BASF Fuel Cell, Inc., Somerset, NJ), which was coatedwith 0.5 mg/cm2 platinum on the water-facing side and fourlayers of polytetrafluoroethylene (PTFE) on the air-facingside.13,14 The carbon cloth was connected with external circuitusing a titanium wire (length: 1 cm, diameter: 1 mm).Microorganisms and Medium. The anodic carbon felt of

SMDCs, which had been covered with biofilm before use, wasobtained from an operating acetate-fed MFC. The anolyte was asolution of sodium acetate (1.64 g/L) in a nutrient buffersolution containing (per liter in deionized water): 4.4 g KH2PO4,3.4 g K2HPO4 3 3H2O, 1.5 g NH4Cl, 0.1 g MgCl2 3 6H2O, 0.1 gCaCl2 3 2H2O, 0.1 g KCl and 10 mL of trace mineral metalssolution,15 with conductivity of 10.8 mS/cm. The catholyte was abuffer solution containing (per liter in deionized water): 9.0 gKH2PO4 and 8.0 g K2HPO4 3 3H2O, with conductivity of 12.2mS/cm. The desalination and concentrated chambers were filledwith 20 g/L NaCl solution, which was a representative concen-tration of brackish water and seawater.SMDC Operation and Experimental Procedures. The re-

actors were operated in MFC mode before conducting desalina-tion experiments. The anodes were acclimated by running a MFCwith one single AEM between the anode and cathode chambersfor more than 10 cycles until the peak voltage was stable at around600 mV, with an external resistance of 1000Ω. All the three typesof SMDCs were transformed from MFC to MDC mode at thesame time and fedwith the same anolyte and catholyte, whichwerecontinuously circulated from a 500 mL and a 150 mL bottles,respectively, at a rate of 5 mL/min using a peristaltic pump(BT100-1 L, Lange, China). The desalination process startedwhen the 20 g/L NaCl solution was injected into desalination andconcentrated chambers. Since MDCs might be proposed as apretreatment process for RO,10 one desalination cycle in this studywas defined as the time needed for an SMDC to reach adesalination ratio of 70%, that is, the salt concentration in eachdesalination chamber was decreased to 6 g/L. The anolyte,catholyte, and salt solutions in both desalination and concentrationchambers were replaced when all the three types of SMDCsreached the 70% desalination ratio. The anolyte and catholytewere replaced twice in one desalination cycle in order to ensure asufficient supply of substrate and avoid pH change.Meanwhile, theconductivity of NaCl solution in each desalination and concen-trated chambers was measured using a conductivity meter (SG3-ELK,Mettler Toledo, Columbus, OH). In order to investigate theinfluence of external resistance on the performance of SMDCs, theexternal resistance was reduced from 1000 Ω to 1 Ω using aresistance box (0.1-99,999 Ω; ZX21, Tianshui, China). All theSMDCswere run at least three desalination cycles for each externalresistance and operated in duplicate simultaneously under ambienttemperatures (25 ( 1 �C). The results in this paper were theaverage of the two parallel samples.

Figure 1. The photo (A) and configuration (B) of 2-SMDC (AEM:anion exchange membrane; CEM: cation exchange membrane; C:concentrated chamber; D: desalination chamber).



Analyses and Calculation. Cell voltage was automaticallymonitored by a data acquisition system (DAQ2213, ADLINK,China) every 10 min during experiments. The current (I) in theelectrical circuit of an SMDC was determined from the cellvoltage (U) and external resistance (Re) according to I = U/Re.The concentration of NaCl was experimentally confirmed

having a linear correlation with the conductivity of NaCl solutionwithin the range of NaCl concentration tested in the study.Hence, the desalination ratio (η) of an SMDC with a number ofdesalination chambers of N at a certain desalination time, can beobtained by the following equation:

η ¼ 1-V1 � σ1 þ V2 � σ2 þ :::þ VN � σN

ðV1 þ V2 þ :::þ VNÞ � σ0ð1Þ

Here VK and σK (K = 1, 2, ..., N) stand for the volume andconductivity of salt solution in the Kth desalination chamber at acertain desalination time, respectively, and σ0 is the conductivityof the initial salt solution with a NaCl concentration of 20 g/L.When one desalination cycle terminated, both the specific

desalination rate (SDR) based on salt solution volume (i.e., onthe basis of desalination chamber volume (7.1 mL)) and totaldesalination rate (TDR) were calculated to evaluate desalinationperformance according to eqs 2 and 3. The former tells NaClremoval rate per unit desalinated water volume and the lattershows the total NaCl removal rate in a whole SMDC.

SDR ¼ 70%� C0

Tð2Þ

TDR ¼ 70%� C0 � ðV1 þ V2 þ :::þ VNÞT

ð3Þ

Here 70% and T are the desalination ratio and needed timerespectively when one desalination cycle terminated. C0 is theconcentration of the initial NaCl solution (20 g/L).

Charge transfer efficiency was calculated as the ratio Qth/Q,where Q is the coulombs harvested through the electrical circuitover one desalination cycle (Q =

RI dt), and Qth stands for the

theoretical amount of coulombs that is required for the move-ment of NaCl (Qth = 70%� 0.342mol/L� (V1þV2þ ...þVN)� F; 70%: the desalination ratio when one desalination cycleterminated; 0.342 mol/L: the initial molar concentration of saltsolution (20 g/L); F: the Faraday’s constant (96485 C/mol)).Current interrupt method was employed to determine the

ohmic resistance of SMDCs.16 A steep rise in cell voltage wasimmediately observed when the electrical circuit was cut offmeanwhile the real-time data of cell voltage was recorded by thedata acquisition system with a sampling frequency of 1000 Hz.The ohmic internal resistance was calculated from RΩ = ΔU/I0,where ΔU represents the steep rise in cell voltage and I0 is thecurrent before interruption.

’RESULTS AND DISCUSSION

Electricity Generation Performance of SMDCs. Threetypes of SMDCs were operated for more than two desalinationcycles at different external resistances. All SMDCs exhibited asimilar overall trend of current generation during one desalina-tion cycle as typically shown in Figure 2A-D. The current roseto a peak level (maximum current) immediately at the beginningof each desalination cycle, and then dropped slowly along thewhole cycle, although the peak magnitudes varied with differentexternal resistances and desalination chamber numbers. Slightfluctuation was caused by the replacement of anolyte andcatholyte during one cycle, indicating that the substrate levels(including substrate concentration and pH value) were not thelimiting factors of current generation. The decline of the currentwas mainly attributed to the increase of ohmic resistance thatresulted from changes in the conductivity of salt solution, whichhas been confirmed by Cao et al. using a single-desalination-chambered MDC (which indeed was referred to as 1-SMDC inthe current research).For an SMDC with a fixed desalination chamber number, the

maximum current varied with the change of external resistance.As shown in Figure 3, the maximum current values of 1-SMDC,2-SMDC, and 3-SMDC increased by 716%, 622% and 491%from 0.91 mA, 0.88 mA, and 0.79 mA to 7.43 mA, 6.35 mA, and4.67 mA, respectively, when the external resistance was reducedfrom 500 Ω to 10 Ω. However, when the external resistancecontinued to reduce from 10 Ω to 1 Ω, the maximum currentsdeclined sharply. In terms of the maximum current, the optimal

Figure 2. Current curves of SMDCs with an external resistance of (A)100Ω, (B) 50Ω, (C) 10Ω, and (D) 5Ω. Arrows indicate anolyte andcatholyte replacement during one desalination cycle.

Figure 3. Maximum currents of SMDCs under different desalinationchamber numbers and external resistances.



external resistance for all the three types of SMDCs was about10 Ω. These results elucidated that the current generated bySMDCs could be promoted by reducing the externalresistance.17,18 Nevertheless, exoelectrogenic bacteria could notproduce enough electrons and even behaved unstable when theexternal resistance became too low. Consequently, the maximumcurrent went downward.For an SMDC with a fixed external resistance, the maximum

current decreased with the increase of desalination chambernumber. Taking the case of 10 Ω external resistance for anexample (Figure 2C), the maximum currents were 7.43 mA, 6.35mA and 4.67 mA for 1-SMDC, 2-SMDC, and 3-SMDC, respec-tively. The desalination chamber number also influenced thecurrent curve evolution in one desalination cycle. As shown inFigure 2C, the currents decreased by 6.81 mA, 4.24 mA, and 2.40mA from their maximum values to 0.62 mA, 2.11 mA, and 2.27mA for 1-SMDC, 2-SMDC, and 3-SMDC after 12 h of desalina-tion, respectively, indicating that more desalination chambersinduced a slower decreasing rate of the current. In addition, theeffect of the desalination chamber number on electricity genera-tion was also related to the external resistance. The desalinationchamber number had little effect on electricity generationperformance when the external resistance was large enough.For example, when the external resistance was higher than 100Ω, only small differences in the maximum current production(Figure 3) and current curves between the three SMDCs(Figure 2A) were observed. In contrast, the desalination chambernumber showed more remarkable influence under the conditionof relatively low external resistance such as 10Ω (Figure 2C andFigure 3).Since a pair of AEM and CEM, together with salt solution of

two chambers (one desalination chamber and one concentratedchamber), were increased into the system when another desali-nation cell was added, it is reasonable for the decrease of themaximum current due to the increased internal resistance causedby the extra added desalination cell. Meanwhile, as will bediscussed in later sections, the SDR decreased with the increaseof desalination chamber number, leading to a slower rising inohmic resistance and correspondingly a slower decrease incurrent. The total resistance in an SMDC consisted of internalresistance and external resistance. The former was affected by thedesalination chamber number. Therefore, the variation of inter-nal resistance, resulting from the change of the desalinationchamber number, would have a significant impact on currentgeneration when the external resistance was relatively low.Desalination Performance of SMDCs. The conductivity of

salt solution in each desalination chamber of SMDCs wasmeasured during one desalination cycle. The average desalina-tion ratio was calculated according to eq 1. As shown inSupporting Information (SI) Figure S1, the desalination ratiosof all the three types of SMDCs went up along with onedesalination cycle and increased slower when more desalinationcells were set in an SMDC. In the case of 50Ω external resistance,the desalination ratios for 1-SMDC, 2-SMDC, and 3-SMDCwere 99.4%, 85.6%, and 72.1% after 18 hours of operation,respectively (SI Figure S1B). In addition, the time needed toreach a certain desalination ratio was shortened with the externalresistance reduced.The SDR values of the three types of SMDCs during one

desalination cycle were calculated according to eq 2 (Figure 4A).The SDR increased with the decrease of external resistance whenthe desalination chamber number was fixed. The SDR for

2-SMDC increased from 0.25 g/(L 3 h) to 1.96 g/(L 3 h) whenthe external resistance decreased from 500Ω to 10Ω. However,the SDR decreased by 6.2% when the external resistance wasfurther reduced from 10Ω to 5Ω. On the other hand, the SDRdecreased with the increase of desalination chambers. For anexternal resistance of 10 Ω, the SDR values were 2.54 g/(L 3 h),1.96 g/(L 3 h), and 1.24 g/(L 3 h) for 1-SMDC, 2-SMDC, and3-SMDC, respectively. The increase in internal resistance causedby the increase of desalination chambers is one of main reasonsfor the decrease of SDR. Meanwhile, the electric potentialcreated by the salt gradient between desalination chamber andconcentrated chamber may contribute the decrease of SDR sincethe electric potential is impeditive to ion transfer from desalina-tion chambers to concentration chambers and will increase alongwith the desalination process. The effects of the external resis-tance and desalination chamber number on the SDR wereconsistent with those on the current. Since the current deter-mined the number of ions that transferred across eachmembraneper second and the volume of each single desalination chamberclipped between a pair of IEMs was fixed at 7.1 mL, the currentalso mainly determined the SDR.Figure 4B shows the TDR of SMDCs under different condi-

tions, which were obtained from eq 3. The influence of theexternal resistance on the TDR was the same as that on the SDR,since the TDR depended only on the current when the desalina-tion chamber number was fixed. Taking the case of 2-SMDC foran example, the maximum TDR of 0.0252 g/h was obtainedunder an external resistance of 10 Ω, meanwhile the maximumcurrent of 7.43 mA was also generated (see Figure 3). Theinfluence of the desalination chamber number on the TDR wasassociated with the external resistance. For an external resistancelarger than 10Ω, the TDR rose with the addition of desalinationchambers. However, when the external resistance was not largerthan 10 Ω, the TDR first increased then decreased with theincreasing desalination chamber number. Taking the externalresistance of 20Ω as an example, the TDR rose from 0.014 g/h to0.0187 g/h while the desalination chamber number was

Figure 4. (A) Specific desalination rate (SDR) based on the saltsolution volume and (B) total desalination rate (TDR) of SMDCsunder different desalination chamber numbers and external resistances.



increased from one to three. For an external resistance of 10 Ω,the maximum TDR of 0.0252 g/h, however, occurred in2-SMDC, which was 1.4 times that of 1-SMDC (0.0178 g/h)and 1.3 times that of 3-SMDC (0.0196 g/h).In an SMDC, when one electron passed through the external

circuit, a pair of cation and anion would be separated from eachdesalination chamber simultaneously to accomplish a closedloop. Hence, the TDR depended not only upon the current,which determined the desalination rate of one single desalinationchamber, but also upon the desalination chamber number, whichdetermined the number of ion-pairs separated at the same timealong with each electron that flows through the external circuit.This exactly was why the current decreased with the increase ofdesalination chambers but the TDR still went up when theexternal resistance was above 10 Ω. Meanwhile, when theexternal resistance was no larger than 10Ω, the current reductioncaused by the increase of desalination chambers would be so largethat the simultaneous separation of ions in multiple desalinationchambers could not compensate the reduction of the TDRcaused by current decrease.All the discussion above proved the existence of an optimal

desalination chamber number along with an external resistanceoptimized for salt removal. In the present study, the optimaloperation condition was two desalination chambers with anexternal resistance of 10 Ω, under which a maximum TDR of0.0252 g/h was obtained.Charge Transfer Efficiency. As expected, charge transfer

efficiency was increased significantly by increasing desalinationchambers between anode and cathode chambers (Figure 5),since multiple ion-pairs were separated along with one electronpassing through the external circuit. For the external resistance of10Ω, the charge transfer efficiencies were 120%, 223% and 283%for 1-SMDC, 2-SMDC, and 3-SMDC, respectively. Moreover,the charge transfer efficiency increased when the external resis-tance declined, despite a few peculiar values. For example, as for3-SMDC, the charge transfer efficiency rose from 170% to 323%when the external resistance decreased from 500Ω to 5Ω. Thisindicated that the reduction of the external resistance wouldbenefit the charge transfer efficiency. It was proposed that theenlarged current, resulting from the decline in external resistance,offset the reverse diffusion of ions from concentrated chambersto desalination chambers. Moreover, the above electron transferefficiencies of SMDCs, larger than their theoretical maximumvalue (100% for 1-SMDC, 200% for 2-SMDC, and 300% for3-SMDC), may be attributed to the diffusion of ions fromdesalination chambers to anode and cathode caused by the salt

concentration gradient, which was proved by Mehanna et al.10

and Jacobson et al.11 The quantitative analysis on the diffusion ofions needs further study.Change in Ohmic Resistance in a Desalination cycle. The

ohmic resistance of an SMDC was made up of anodic ohmicresistance, cathodic ohmic resistance and ohmic resistance of thedesalination section. The first two components were relativelyconstant by frequent replacement of anolyte and catholyte. Theohmic resistances of SMDCs with an external resistance of 10Ωwere measured along one desalination cycle. The comparison ofthe variation of ohmic resistances (Figure 6) and currents(Figure 2C) reflected a reverse coordination, indicating theinternal resistance of an SMDC was significantly affected bythe ohmic resistance. At the beginning of one desalination cycle,the ohmic resistances were 21Ω, 39Ω, and 56Ω for 1-, 2-, and3-SMDC, respectively. The addition of one desalination cellresulted in an increase of about 18 Ω of the ohmic resistance,which was equivalent to the ohmic resistance of a pair of AEMand CEM and two chambers of 20 g/L NaCl solution. Ohmicresistance increased along the cycle, and got a sharp rise in the last4 h. The ohmic resistance increased the fastest for 1-SMDC(from 21Ω to 312Ω), medium for 2-SMDC (from 39Ω to 256Ω) and the slowest for 3-SMDC (from 56Ω to 117Ω). As thedesalinating process continued, the concentration of salt solutionin desalination chambers decreased, leading to the increase ofohmic resistance. The SDR determined the change in ohmicresistances. For 1-SMDC, the SDR was the fastest, the change inohmic resistance was correspondingly the most remarkable.Similarly, the change in the ohmic resistance of 3-SMDC wasthe slowest due to its least SDR.

’OUTLOOK

In this study, the TDR of MDCs was remarkably enhanced byconstructing a stacked MDC. Although the current and corre-sponding SDR declined with the increase of desalination cham-ber number, both electron transfer efficiency and salt solutionvolume increased, which proved beneficial to promote the TDR.Nevertheless, the desalination chamber number should not beincreased infinitely, otherwise the current would be very low dueto the increase of internal resistance caused by addition ofdesalination chambers, leading to the decrease of TDR. As theeffects of the desalination chamber number and external resis-tance on TDR were mutually influenced with each other, thereshould be an optimal condition for TDR. Expanding desalinationchamber number together with reducing external resistance

Figure 5. The charge transfer efficiencies of SMDCs under differentdesalination chamber numbers and external resistances. Figure 6. Ohmic resistance in one desalination cycle of SMDCs with an

external resistance of 10 Ω.



proved the feasible approaches to elevate salt removal rate ofMDC. Therefore, exploring IEMs of smaller resistance andoptimizing the structure of SMDCs, such as the dimension andconfiguration of desalination and concentrated chambers, are theissues needed to be investigated in future works.

As the results shown in this study, desalination ratio ofSMDCs increased higher (SI Figure S1C) and ohmic resistanceincreased faster (Figure 6) in the later period of desalination,which would cause a low desalination rate and a poor electricitygeneration performance in this period. To determine the degreeof desalination, a study on the relationship between desalinationperformance and ohmic resistance would be of great help. Asrecommended by Mehanna et al.,10 MDCs may be used as apretreatment technology for electrodialysis or RO, or be used toproduce water that needs not to be completely desalinated.

’ASSOCIATED CONTENT

bS Supporting Information. One additional figure. Thismaterial is available free of charge via the Internet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*Phone: þ86 10 62772324; e-mail: [email protected].

’ACKNOWLEDGMENT

This work was supported by the National High TechnologyResearch and Development Program (863 Program) (No.2009AA06Z306) and the Program of Introducing Talents ofDiscipline to Universities (the 111 Project).

’REFERENCES

(1) Mohtada, S; Toraj, M Treatment of sea water using electro-dialysis: Current efficiency evaluation.Desalination. 2009, 249, 279–285.(2) Shannon, M. A.; Bohn, P. W.; Elimelech, M.; Georgiadis, J. G.;

Marinas, B. J.; Mayes, A. M. Science and technology for water purifica-tion in the coming decades. Nature 2008, 452, 301–310.(3) Khawaji, A. D.; Kutubkhanah, I. K.; Wie, J. M. Advances in

seawater desalination technologies. Desalination 2008, 221, 47–69.(4) Zhou, Y.; Tol, R. S. J. Evaluating the costs of desalination and

water transport. Wat. Resour. Res. 2005, 41, 1–16.(5) Veerapaneni, S.; Logan, B. E.; Bond, R. Reducing energy

consumption for seawater desalination. J. Am. Water Works Assoc.2007, 99, 95–106.(6) Mathioulakis, E.; Belessiotis, V.; Delyannis, E Desalination by

using alternative energy: Review and state-of-the-art.Desalination. 2007,203, 346–365.(7) Cao, X. X.; Huang, X.; Liang, P.; Xiao, K.; Zhou, Y. J.; Zhang,

X. Y.; Logan, B. E. A new method for water desalination using microbialdesalination cells. Environ. Sci. Technol. 2009, 43, 7148–7152.(8) Logan, B. E.; Hamelers, B.; Rozendal, R.; Schroder, U.; Keller, J.;

Freguia, S.; Alterman, P.; Verstraete, W.; Raraey, K. Microbial fuel cells:methodology and technology. Environ. Sci. Technol. 2006, 40, 5181–5192.(9) Lovley, D. R. The microbe electric: conversion of organic matter

to electricity. Curr. Opin. Biotechnol. 2008, 19, 564–571.(10) Mehanna, M.; Saito, T.; Yan, J.; Hickner, M.; Cao, X.; Huang,

X.; Logan, B. E. Using microbial desalination cells to reduce watersalinity prior to reverse osmosis. Energy Environ. Sci. 2010, 3, 1114–1120.

(11) Jacobson, S. K.; David, M. D.; Zhen, H. Efficient salt removal ina continuously operated upflow microbial desalination cell with an aircathode. Bioresour. Technol. 2011, 102, 376–380.

(12) Cao, X. X.; Huang, X.; Boon, N.; Liang, P.; Fan,M. Z. Electricitygeneration by an enriched phototrophic consortium in a microbial fuelcell. Electrochem. Commun. 2008, 10, 1392–1395.

(13) Liu, H.; Logan, B. E. Electricity generation using an air-cathodesingle chamber microbial fuel cell in the presence and absence of aproton exchange membrane. Environ. Sci. Technol. 2004, 38, 4040–4046.

(14) Chen, S. A.; Liu, H.; Logan, B. E. Increased power generation ina continuous flow MFC with advective flow through the porous anodeand reduced electrode spacing. Environ. Sci. Technol. 2006, 40, 2426–2432.

(15) Lovley, D. R.; Phillips, E. J. P. Novel mode of microbial energymetabolism: Organic carbon oxidation coupled to dissimilatory reduc-tion of iron or manganese. Appl. Environ. Microbiol. 1988, 54, 1472–1480.

(16) Liang, P.; Huang, X.; Fan, M. Z.; Cao, X. X.; Wang, C.Composition and distribution of internal resistance in three types ofmicrobial fuel cells. Appl. Microbiol. Biotechnol. 2007, 77, 551–558.

(17) Aelterman, P.; Versichele, M.; Marzorati, M.; Boon, N.; Ver-straete, W. Loading rate and external resistance control the electricitygeneration of microbial fuel cells with different three-dimensionalanodes. Bioresour. Technol. 2008, 99, 8895–8902.

(18) Jang, J. K.; Phama, T. H.; Changa, I. S.; Kang, K. H.; Moon, H;Cho, K. S.; Kim, B. H. Construction and operation of a novel mediator-and membrane-less microbial fuel cell. Process. Biochem. 2004, 39, 1007–1012.

stacked microbial desalination cells to enhance water desalination efficiency

Documents