a new method for water

8/13/2019 A New Method for Water

1/5


2/5

salt removal during a single batch desalination cycle, whileat the same time producing useful amounts of electricity.

Materials and MethodsMDC Construction. The design of the MDC was based ona cubic-shaped MFC, where the electrode chamber isproduced by drilling a hole in a solid block of polycarbonate(13). The MDC consisted of three chambers (anode, middledesalination, and cathode), separated using AEM and CEMmembranes, clamped together with gaskets that provide a

waterseal betweenthe chambers (Figure 1). An AEM (DF120,Tianwei Membrane) was used to separate the anode andmiddle chambers, and a CEM (Ultrex CMI7000, MembranesInternational) was used to separate the middle and cathodechambers.The cross section of theworkingareaof these twochambers was 9 cm2. Theinside volumes of theanode, middledesalination, and cathode chamber were 27, 3, and 27 mL,respectively. When the anode and cathode chambers werefilled with electrode material, which consisted of carbon felt(19), theliquid chamber volume decreased to 11 mL.Externalelectrical contact to the electrodes was producedby pushinga graphiterod (5 mmdiameter) intothe felt. Prior touse, thecarbon felt and rod were washed for 48 h in 1 M HCl andrinsed with water to remove trace metals.

Medium. The anode chamber of the MFC was fed asolution of sodium acetate (1.6 g/L) in a nutrient buffersolution containing (per liter in deionized water): 4.4 g

KH2PO4, 3 . 4 g K 2HPO43H2O,1.5gNH4Cl, 0.1 g MgCl26H2O,

0.1gCaCl22H2O,0.1gKCl,and10mLoftracemineralmetalssolution (20). The cathode chamber was fed a ferricyanidecatholeyte, containing (per liter in deionized water): 16.5 gK3Fe(CN)6, 9.0 g KH2PO4, 8.0 g K2HPO43H2O. The middlechamber was filled with the water to be desalinated, at NaClconcentrations of 5, 20, or 35 g/L. These concentrationsrepresent a reasonable range of salinities for brackish waterand seawater (4).

MDC Operation and Experimental Procedures.Three-chamber MDCs were inoculated with a mixed bacterialculture from the anode of an active acetate-fed laboratory

MFC. Before conducting desalination experiments, theanodeswereacclimatedby running thereactors in MFCmodeusing only a single CEM, until peak voltage was stable atabout 600 mV and reproducible over 10 cycles. Whenoperated in MDC mode (Supporting Information), solutionsfrom individual feed reservoirs (100 mL each) were continu-ouslyrecirculated through the anode and cathode chambersat a rate of 5 mL/min using a peristaltic pump (BT00-300T,Lange, China). The anolyte in the feed bottle was replacedevery 12 h to ensure a sufficient supply of substrate for thebacteria, and avoid a drop in pH (Supporting Information).Due to the large difference in the anode and feed reservoirvolume (100 mL) compared to that of the desalinationchamber (3 mL), the total change in conductivity of the feedsolutions, if not changed, would have been


3/5

iscomparableto that obtainedin previoustestsusing a similarMFC design (13). The maximum voltage produced duringMDC operation with an initial salt concentration of 20 g/L(200 external resistor) was 600 mV, with a maximumcurrent of 3 mA (Figure 2). Under this fixed resistancecondition, this translates to a maximum power output of 2

W/m2 (based on the cross section area of 9 cm2), and 31W/m3 based on thetotal reactor volume (all threechambers,57 mL). The anolyte solution was frequently replaced (every12 h) to avoid substrate limitations forbacteria on theanodeor changes in pH. When the anolyte was replaced there waslittlechange in voltage, demonstrating the current generation

was not affectedby substratelevels. In a typical MFC, voltageproduction is usually constant over a cycle until the acetateconcentration reaches a low level. However, herethe voltageimmediatelyand continuously decreased overthe cycle. Thisshows that the main impact on the voltage produced in theMDC was the decrease in the conductivity of the solution inthedesalination chamber over the full cycle, which changedthe ohmic resistance (see below).

The water in the middle chamber was efficiently desali-nated at all three initial salt concentrations. Based on thechange in solution conductivity, the salt removals at eachinitial salt concentration were about 88 ( 2% (5 g/L), 94 (3% (20 g/L), and 93 ( 3% (35 g/L) (Figure 3). There was littlechange in solution conductivity in the open circuit control.

The total NaCl removal at different initial salt concentra-

tions and two different external resistances (200

and 800

) was compared to the total amount of charge transferred(Figure 4). Overall, there was good agreement betweenelectrons harvested and NaCl removal, and therefore the

charge transfer efficiency was nearly 100%. This transfer ofionic species relative to charge is high in the MDC whencompared to traditional electrodialysis. Lower efficienciesare observed in electrodialysis as a result of water splittingin thediluate,shuntcurrents between theelectrodes,or back-diffusion of ions from the concentrate to the diluate due tohigh current densities. In the MDC, however, the currentdensities are relatively small and therefore there wasnegligible water splitting to compensate for the limitingcurrent. Back-diffusion of ions from electrode chambers tothe middle chamber was negligible. At salt concentration of35 g/L, thetheoretical chargetransfer(Qth) wasslightly higherthan Q. This likelyindicatesthattherewassomesalt removaldue to the large concentration gradient between the middlechamber and the anode (cathode) chamber.

The transfer of the ions out of the middle chamber wasalso examined on the basis of Na+ accumulation in thecathode chamber relative to that of K+. The solution in thecathode chamber contained only potassium and no sodiumions, making it possible to monitor chemical species re-sponsible forchargetransfer.When thesystemwas operated,sodiumionsaccumulatedin thecathodechamberup tolevelsof 0.9 g/L (10 batches) and 2.4 g/L (20 batches) (Figure 5).These values were within 96% of that calculated based ontotal Coulombs transferred over the multiple cycles ofoperation. Variations in the concentration of potassium inthe cathode chamber during the experimental runs werenegligible. This demonstrates that there was negligiblepotassiumcross over from thecathodechamberto themiddle

chamber by back diffusion.

FIGURE 2. Voltages generated in tests using the three-chamberMDC with an initial salt concentration in the middle chamberof 20 g/L (Large arrows indicate salt solution replacement;small arrows indicate anolyte replacement.).

FIGURE 3. Change of solution conductivity in the middledesalination chamber over complete batch cycles with different

initial salt concentrations (The control is the same system butoperated with an open circuit).

FIGURE 4. NaCl removal compared to total Coulombstransferred (electrons harvested) for different initial saltconcentrations and external resistances (open symbols, 800 ;filled symbols, 200 ).

FIGURE 5. Initial and final concentrations of sodium in thecathode chamber after multiple batch cycles compared to thatcalculated based on total Coulombs of charge transferred.

7150 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 18, 2009
http://pubs.acs.org/action/showImage?doi=10.1021/es901950j&iName=master.img-004.png&w=197&h=142http://pubs.acs.org/action/showImage?doi=10.1021/es901950j&iName=master.img-003.png&w=184&h=138http://pubs.acs.org/action/showImage?doi=10.1021/es901950j&iName=master.img-002.png&w=206&h=164http://pubs.acs.org/action/showImage?doi=10.1021/es901950j&iName=master.img-001.png&w=193&h=156


4/5

Changesin Ohmic Resistance Over a Desalination Cycle.As the water is desalinated over a cycle, the ohmic resistanceshould increase as the conductivity of the middle chamberdecreases. To follow this process, the ohmic resistance oftheMDC was evaluatedover a cycle using EIS(Figure 6).Theohmic resistance increased from 25 at the beginning ofthedesalination cycle to 970 atthe end ofthe cycle intests

with an initial salt concentration of 5 g/L. This shows anincrease in resistance by a factor of 40. In contrast, theresistance of the electrolyte solutions increased only by afactor of 8 based on measured conductivity changes (Figure3). This result means that under low electrolyte conditions,theresistanceof themembranesolution interface or theionexchange membrane resistance must have increased. It wasreported previously that a higher electrolyte concentrationdecreased the membrane solution interface resistance dueto the compression of electrical double layer (22). Thisincrease in membrane resistance over a cycle is alsoconsistent with previous results obtained in MFC tests withNafion 117at lowelectrolyte concentrations(17). Thechangein ohmic resistance explains thevariationin voltageobservedover a single desalination cycle. The voltage rapidly reachedan initial peak, and then continuously decreased over thecourse of the cycle (Figure 2), consistent with the observedincrease in ohmic resistance (Figure 6).

Discussion

The results obtained in this study show that in principle, anMDC can be used to desalinate water. About 90% of the salt

was removed from the water over a single desalinationcycle,and there was no need to pressurize the water or use anexternal source of electricity. The process was effective fordesalinating water even at a salt concentration as high as 35g/L. Thiscanbe compared to traditional electrodialysis whichis recommended for use at salt concentrations up to 6 g/Lof dissolved solids (23) due to the high energy demands ofthe process. It was found that the internal resistance of theMDC increased more rapidly than we expected based onchanges in solution conductivity, likely as a result of theincreased membrane/solution interface resistance due tothe electrical doublelayer. This increase in internal resistance

will limit the efficiency and performance of the system, andthus it needsto be further studied and minimized toincreaseMDC performance.

The experiments conducted here were designed todemonstrate the proof-of-concept of an MDC, but further

work will be needed to make the MDC useful for practicalapplications. For example, the volume of water used inthe electrode chambers to achieve desalination was largerelative to the desalination chamber, and it was not

optimized here. The anode and feed solution (100 mL)

was changed many times over a si ngle desalination cyclein order to study the effects of the water desalination oncurrent (andinternal resistance), and to eliminate possibleeffects of substrate limitation on bacteria in the anodechamber. This resulted in the use of much more waterthan wouldbe needed tominimizethe volume ofthe anodesolution. In addition, we used a ferricyanide catholeytebecause as it allowed us to control the cathode potentialandto easily measure sodiumion transfer into thecathodechamber. While using ferricyanide allowed us to demon-strate the feasibility of the MDC process, its use would not

be acceptablein practice. Furtherdevelopment of theMDCprocess using chemical air cathodes or biocathodes isneeded (24-26).

While we used an acetate substratehere, the use of actualwastewaters should also be examined as low cost fuels infuture studies. The low salinity of these wastewaters maybenefit process efficiency due to passive ion transfer fromthe saline water into the wastewater, which could help toreduce ohmic losses and charge accumulation in themembranes. In addition it mightsolve a problem of extractingenergy from wastewater using MFCs as the conductivity ofmost domestic wastewaters is low(1 mS/cm), which limitsachieving higher power densities (27).

Many wastewaters have a low alkalinity, and thereforea low buffering capacity. The extraction of ions into thecathode and cathode chambers increases the salinity ofthese solutions, and can also cause imbalances in pH.Changes in solution pH in electrode chambers in MFCs iscommon when membranes are placed between theelectrodes (14, 15). These pH changes were avoided hereby using a buffer and frequently changing the anodesolution. Through adjustment of the flow rates intothe electrode (anode and cathode) chambers, and thedesalination chamber, it should be possible to limit pHchanges to desired levels. The addition of inexpensivebicarbonate buffer (28) or the use of anode to cathoderecirculation (29) are two possible approaches for ad-dressing this problem.

Anotherchallenge will be to develop stackeddesalination

systems.While onlya single stack (three chambers) wasusedhere, increasing the number of cells will increase the diluateproduction, although at a cost of producing higher internalresistances. We have examined only batch processing of the

water for desalination, but future studies are needed withcontinuous flow systems. The MDC process can be lookedat as a new type of BES, and its further development willbenefit from advances being made in other electrochemicalsystems such as MFCs.

AcknowledgmentsThis research was supported by International Program ofMOST (2006DFA91120) and 863 Project (2006AA06Z329) inChina andAward KUS-I1-003-13 (to B.E.L.) by KingAbdullahUniversity of Science and Technology (KAUST).

Supporting Information AvailableAdditional information includingone figureand threetables.This material is available free of charge via the Internet athttp://pubs.acs.org.

Literature Cited

(1) UnitedNations. Water in a ChangingWorldsThe UnitedNationsWorld Water Development Report 3, http://www.unesco.org/water/wwap/wwdr/wwdr3/, 2009.

(2) Shannon, M. A.; Bohn, P. W.; Elimelech, M.; Georgiadis, J. G.;Marinas, B. J.; Mayes, A. M. Science and technology for waterpurificationin the coming decades. Nature2008, 452, 301310.

(3) Matsuura, T. Progress in membrane scienceand technologyforseawater desalinationsA review. Desalination2001, 134, 47

54.

FIGURE 6. Nyquist plots used to determine ohmic resistance atthe start and end of a batch cycle.

VOL. 43, NO. 18, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY9 7151
http://pubs.acs.org/action/showImage?doi=10.1021/es901950j&iName=master.img-005.png&w=190&h=149


5/5

(4) Mathioulakis, E.; Belessiotis, V.; Delyannis, E. Desalination byusing alternative energy: Review and state-of-the-art. Desali-nation2007, 203, 346365.

(5) Cath, T. Y.; Childress, A. E.; Elimelech, M. Forward osmosis:Principles, applications, and recent developments. J. Membr.Sci. 2006, 281, 7087.

(6) Logan, B. E.; Regan, J. M. Microbial fuel cell- challenges andapplications.Environ. Sci. Technol. 2006, 40, 51725180.

(7) Logan, B. E.; Call, D.; Cheng, S.; Hamelers, H. V. M.; Sleutels,T. H. J. A.;Jeremiasse, A. W.;Rozendal, R. Microbialelectrolysiscells forhigh yieldhydrogen gasproductionfromorganic matter.Environ. Sci. Technol.2008, 42, 86308640.

(8) Clauwaert, P.; Aelterman, P.; Pham, T. H.; De Schamphelaire,

L.; Carballa, M.; Rabaey, K.; Verstraete, W. Minimizing lossesin bio-electrochemical systems: the road to applications.Appl.Microbiol. Biotechnol.2008, 79, 901913.

(9) Cheng, S.; Xing, D.; Call, D. F.; Logan, B. E. , Direct biologicalconversion of electrical current into methane by electrometha-nogenesis.Environ. Sci. Technol.2009, 43, 3953-3958.

(10) Rabaey, K.; Rodriguez, J.; Blackall, L. L.; Keller, J.; Gross, P.;Batstone, D.; Verstraete, W.; Nealson, K. H. Microbial ecologymeets electrochemistry: electricity-driven and driving com-munities.ISME J. 2007, 1, 918.

(11) Lovley, D. R. The microbe electric:conversion of organic matterto electricity. Curr. Opin.Biotechnol.2008, 19, 564571.

(12) Rabaey, K.; Boon, N.; Siciliano, S. D.; Verhaege, M.; Verstraete,W. Biofuel cells select for microbial consortia that self-mediateelectron transfer.Appl. Environ. Microbiol. 2004, 70, 53735382.

(13) Liu, H.; Logan,B. E. Electricity generation using an air-cathodesinglechamber microbial fuel cell in thepresenceand absence

of a proton exchange membrane. Environ. Sci. Technol.2004,38, 40404046.

(14) Ter Heijne, A.;Hamelers, H. V. M.;De Wilde,V.; Rozendal,R. A.;Buisman, C. J. N. A bipolar membrane combined with ferriciron reduction as an efficient cathode system in microbial fuelcells.Environ. Sci. Technol.2006, 40, 52005205.

(15) Rozendal, R. A.; Hamelers, H. V. M.; Buisman, C. J. N. Effectsof membrane cation transport on pH and microbial fuel cellperformance.Environ. Sci. Technol.2006, 40, 52065211.

(16) Kim, J. R.; Cheng, S.; Oh, S. E.; Logan, B. E. Power generationusing different cation, anion, and ultrafiltration membranes inmicrobial fuel cells. Environ. Sci. Technol. 2007, 41, 10041009.

(17) Harnisch,F.; Schroder,U.; Scholz, F.The suitability ofmonopolarandbipolar ion exchange membranes as separators for biologicalfuel cells. Environ. Sci. Technol. 2008, 42, 17401746.

(18) Sleutels,T. H.J. A.;Hamelers,H. V.M.; Rozendal, R.A.; Buisman,C. J. N. , Ion transport resistance in Microbial Electrolysis Cellswith anion and cation exchange membranes.Int. J. HydrogenEnergy2009, 34, 36123620.

(19) Cao, X. X.; Huang, X.; Boon, N.; Liang, P.; Fan, M. Z. Electricitygeneration by an enriched phototrophic consortium in amicrobial fuelcell. Electrochem. Commun.2008, 10, 13921395.

(20) Lovley, D. R.; Phillips, E. J. P. Novel mode of microbial energymetabolism:organic carbon oxidation coupled to dissimilatoryreductionof ironor manganese.Appl. Environ. Microbiol. 1988,54, 14721480.

(21) He, Z.; Mansfeld, F. Exploring the use of electrochemicalimpedance spectroscopy (EIS) in microbial fuel cell studies.

Energy Environ. Sci. 2009, 2, 215219.(22) Sang, S.; Huang, K.; Li, X. The influence of H2SO4 electrolyte

concentration on proton transfer resistance of membrane/solution interface. Eur. Polym. J. 2006, 42, 28942898.

(23) Banasiak, L. J.; Kruttschnitt, T. W.; Schafer, A. I. Desalinationusing electrodialysis as a function of voltage and salt concen-tration. Desalination2007, 205, 3846.

(24) Cheng, S.; Liu, H.; Logan, B. E. Power densities using differentcathodecatalysts(Pt andCoTMPP)and polymerbinders (Nafionand PTFE) in single chamber microbial fuel cells. Environ. Sci.Technol.2006, 40, 364369.

(25) Liang,P.; Fan, M. Z.;Cao, X. X.;Huang, X. Evaluation of appliedcathodepotential to enhancebiocathode in microbialfuel cells.J. Chem. Technol. Biotechnol.2009, 84, 794799.

(26) Cao, X. X.; Huang, X.; Liang, P.; Boon, N.; Fan, M. Z.; Zhang, L.;Zhang, X. Y. A completely anoxic microbial fuel cell using aphoto-biocathode for cathodic carbondioxide reduction. Energy

Environ. Sci.2009, 2, 498501.(27) Liu, H.;Cheng, S. A.;Logan, B. E. Power generation in fed-batch

microbialfuel cellsas a function of ionicstrength, temperature,and reactorconfiguration.Environ.Sci. Technol.2005, 39, 54885493.

(28) Torres, C.; Lee, H.-S.; Rittmann, B. E. Carbonate species as OHcarriers for decreasing the pH gradient between cathode andanode in biological fuel cells. Environ. Sci. Technol.2008, 42,87738777.

(29) Freguia, S.; Rabaey, K.; Yuan, Z. G.; Keller, J. Sequential anode-cathode configurationimprovescathodic oxygen reduction andeffluent quality of microbial fuel cells. Water Res. 2008, 42,13871396.

ES901950J

7152 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 18, 2009

a new method for water

Documents