a desalination battery

5
A Desalination Battery Mauro Pasta, Colin D. Wessells, Yi Cui, ,§ and Fabio La Mantia , * Analytische Chemie Zentrum fü r Elektrochemie, Ruhr-Universitä t Bochum, Universitä tsstrasse 150, D-44780 Bochum, Germany Department of Materials Science and Engineering, Stanford University, Stanford, California 94305, United States § Stanford Institute for Materials and Energy Sciences, SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, California 94025, United States * S Supporting Information ABSTRACT: Water desalination is an important approach to provide fresh water around the world, although its high energy consumption, and thus high cost, call for new, efficient technology. Here, we demonstrate the novel concept of a desalination battery, which operates by performing cycles in reverse on our previously reported mixing entropy battery. Rather than generating electricity from salinity differences, as in mixing entropy batteries, desalination batteries use an electrical energy input to extract sodium and chloride ions from seawater and to generate fresh water. The desalination battery is comprised by a Na 2x Mn 5 O 10 nanorod positive electrode and Ag/AgCl negative electrode. Here, we demonstrate an energy consumption of 0.29 Wh l 1 for the removal of 25% salt using this novel desalination battery, which is promising when compared to reverse osmosis (0.2 Wh l 1 ), the most efficient technique presently available. KEYWORDS: Seawater desalination, mixing entropy battery, reverse osmosis, ion selectivity I ncreasing amounts of fresh water will be required in the future as a result of the increase in population and enhanced living standards, as well as the expansion of industrial and agricultural activities. 1,2 At the current growth rate, humans will consume 90% of available fresh water by 2025, by which time the population living in water-stressed areas is expected to increase to 3.9 billion. 3 Seawater desalination is becoming a feasible source for fresh water, free from local and global variations in rainfall. For this reason, the use of seawater desalination has steadily increased in recent years. Seawater desalination processes require electric or thermal energy to separate saline seawater into two streams, a fresh water stream containing a low concentration of dissolved salts and a concentrated brine stream. A variety of desalination technologies have been developed over the years. 2,410 Reverse osmosis requires a large electrical energy input, which accounts for 44% of its cost, 1 and it is based on selective membranes, which are prone to fouling and require frequent replacement. 11 Recent research on high-flux membranes based on carbon nanotubes 12 could potentially reduce the energy consumption, while alternative electrical-based desalination technology such as ion concentration polarization in nanodevices 13 can potentially avoid the fouling. Thermal energy based multistage flash distillation requires energy intensive heating to temperatures above 90 °C, accounts for 50% of its cost. 1 Multiple effect distillation is gaining popularity due to its higher efficiency and lower top brine temperatures (about 70 °C) than multistage flash technology. 14 Forward osmosis is a promising new process that utilizes lower temperatures (60 ° C) but still requires the use of membranes. 15,16 Solvent extraction using directional solvents like decanoic acid at mild temperatures (3050 °C) has been demonstrated recently by Bajpayee and co-workers. 3 Previous electrochemical approaches relying on the electrical double- layer built at the surface of high surface area carbonaceous electrodes are capacitive deionization (CDI) 17 and the newly developed capacitive double layer expansion (CDLE). 18,19 Despite these recent advances, the 5080% target for reduction in desalination costs by 2020 set by the National Research Council roadmap will not be achieved by incremental improvements to existing technologies, and therefore, a new approach is needed. 1 Here we introduce a new concept of a desalination batterywhich is based on a previously described device known as the mixing entropy battery. 20 This desalination battery operates in a similar way to the capacitive desalination techniques 21 but instead of storing charge in the electrical double layer (built at the surface of the electrode) it is held in the chemical bonds (bulk of the electrode material). Battery electrodes offer higher specific capacity and lower self-discharge than capacitive electrodes. 22 In this work, we reversed the previously proposed four step cycle of the mixing entropy battery, in a fashion similar to the four cycles of the CDI and CDLE. This desalination battery Received: November 4, 2011 Revised: December 27, 2011 Published: January 23, 2012 Letter pubs.acs.org/NanoLett © 2012 American Chemical Society 839 dx.doi.org/10.1021/nl203889e | Nano Lett. 2012, 12, 839843

Upload: marco-villegas-bermudez

Post on 18-Aug-2015

220 views

Category:

Documents


2 download

DESCRIPTION

A Desalination Battery

TRANSCRIPT

A Desalination BatteryMauro Pasta,Colin D.Wessells,Yi Cui,,and Fabio La Mantia,*Analytische Chemie Zentrum fur Elektrochemie, Ruhr-Universita t Bochum, Universita tsstrasse 150, D-44780 Bochum, GermanyDepartment of Materials Science and Engineering, Stanford University,Stanford,California 94305,United StatesStanford Institute for Materials and Energy Sciences,SLAC National Accelerator Laboratory,2575 Sand Hill Road,Menlo Park,California 94025,United States*S Supporting InformationABSTRACT: Water desalination is an important approach toprovide fresh water around the world, although its high energyconsumption, and thus high cost, call for new, efficienttechnology. Here, we demonstrate the novel concept of adesalinationbattery, whichoperatesbyperformingcyclesinreverse onour previously reportedmixing entropy battery.Ratherthangeneratingelectricityfromsalinitydifferences, asin mixing entropy batteries, desalination batteries use anelectrical energy input toextract sodiumandchloride ionsfrom seawater and to generate fresh water. The desalination battery is comprised by a Na2xMn5O10 nanorod positive electrodeand Ag/AgCl negative electrode. Here, we demonstrate an energy consumption of 0.29 Wh l1for the removal of 25% salt usingthis novel desalinationbattery, whichis promisingwhencomparedtoreverseosmosis (0.2Whl1), themost efficienttechnique presently available.KEYWORDS: Seawater desalination,mixing entropy battery,reverse osmosis,ion selectivityIncreasingamounts of freshwater will be requiredinthefuture as a result of the increase in population and enhancedliving standards, as well as the expansionof industrial andagricultural activities.1,2At the current growth rate, humans willconsume 90% of available fresh water by 2025,by which timethe population living inwater-stressed areas is expected toincreaseto3.9billion.3Seawater desalinationis becomingafeasible source for fresh water, free fromlocal and globalvariations in rainfall. For this reason, the use of seawaterdesalination has steadily increased in recent years.Seawater desalinationprocesses requireelectricor thermalenergytoseparatesalineseawater intotwostreams, a freshwater stream containing a low concentration of dissolved saltsand a concentrated brine stream. Avariety of desalinationtechnologies have been developed over the years.2,410Reverseosmosis requires a large electrical energy input, which accountsfor44%of itscost,1anditisbasedonselectivemembranes,which are prone to fouling and require frequent replacement.11Recent research on high-flux membranes based on carbonnanotubes12could potentially reduce the energy consumption,whilealternativeelectrical-baseddesalinationtechnologysuchas ion concentration polarization in nanodevices13canpotentially avoid the fouling.Thermal energybasedmultistageflashdistillationrequiresenergy intensive heating to temperatures above 90 C, accountsfor 50%of its cost.1Multiple effect distillation is gainingpopularity due toits higher efficiency andlower topbrinetemperatures (about 70 C) than multistage flash technology.14Forward osmosis is a promising new process that utilizes lowertemperatures (60 C) but still requires the use ofmembranes.15,16Solvent extractionusingdirectional solventslikedecanoicacidat mild temperatures(3050 C) has beendemonstratedrecentlybyBajpayeeandco-workers.3Previouselectrochemical approaches relying onthe electrical double-layer built at the surface of highsurface area carbonaceouselectrodesarecapacitivedeionization(CDI)17andthenewlydeveloped capacitive double layer expansion (CDLE).18,19Despite these recent advances, the 5080% target for reductionindesalinationcosts by 2020set by the National ResearchCouncil roadmap will not be achieved by incrementalimprovements toexistingtechnologies, andtherefore, anewapproach is needed.1Here we introduce a new concept of a desalination batterywhich is based on a previously described device known as themixingentropybattery.20Thisdesalinationbatteryoperatesin a similar way to the capacitive desalination techniques21butinstead of storing charge in the electrical double layer (built atthesurfaceof theelectrode)itisheldinthechemical bonds(bulk of the electrode material). Battery electrodes offer higherspecific capacity and lower self-discharge than capacitiveelectrodes.22In this work, we reversed the previously proposed four stepcycle of the mixing entropy battery, in a fashion similar to thefour cyclesof theCDI andCDLE. ThisdesalinationbatteryReceived: November 4,2011Revised: December 27,2011Published: January 23,2012Letterpubs.acs.org/NanoLett 2012 American Chemical Society 839 dx.doi.org/10.1021/nl203889e | Nano Lett. 2012,12, 839843consists of a cationic sodium insertion electrode and a chloride-capturing anionic electrode. A four-step charge/dischargeprocess allows these electrodes to separate seawater into freshwater andbrinestreams. Inthefirst step, thefullychargedelectrodes, whichdonot containmobilesodiumor chlorideions when charged, are immersed in seawater. Aconstantcurrentisthenappliedinordertoremovetheionsfromthesolution (Figure 1, Step 1). In the second step (Figure 1, Step2), thefreshwater solutioninthecell isextractedandthenreplaced with additional seawater. The electrodes are thenrechargedinthis solution, releasing ions andcreating brine(Figure 1, Step 3). In the final, fourth step (Figure 1, Step 4),the brine solution is replaced with newseawater, and thedesalinationbatteryisreadyforthenextcycle. Freshwaterisproducedduringtheinitial dischargeof theelectrodes(Steps12), while recharging the electrodes results in the productionof a brine stream (Steps 34). Figure 1b schematically shows atypical potential profile of a desalination battery as a function ofchargestate. The circularintegralof thiscurve isequal to thenet amount of electrical energy needed to drive the desalinationof the seawater.23To demonstrate the validity of our desalination batterydesign, electrochemical cell employingthefollowingreactionwere constructed+ + + 5MnO 2Ag 2NaCl Na Mn O 2AgCl2 2 5 10Anelectrodemadeof Na2xMn5O10(NMO)nanorods20wasusedtocapture Na+ions, anda Ag electrode was usedtocapture Clions (see Supporting Information for detailedelectrode preparation). Among the few good water compatibleNa+insertion materials now available, we selected Na2Mn5O10Figure 1. (a) Schematic representation of the working principle behind a complete cycle of the desalination battery, showing how energy extractioncanbeaccomplished: step1, desalination; step2, removal ofthedesalinatedwaterandinletofsewwater;step3, dischargeofNa+andClinseawater; step4, exchangetonewseawater. (b)Typical formof acycleof batterycell voltage(E)vscharge(q)inthedesalinationbattery,demonstrating the energy consumed.Nano Letters Letterdx.doi.org/10.1021/nl203889e | Nano Lett. 2012,12, 839843 840because of its higher specific charge storage capacity (35 mAhg1), low cost, and benign environmental impact.24,25Identifyinga suitableelectrode for chloridecapture is moredifficult. Silver has been used because it forms AgCl, bycapturing chloride ions, which is stable inpotential and isinsoluble.The NMOhas been prepared by a polymer synthesismethod,26with which nanorods with a mean diameter of 300nm and length of about 2 m were obtained (Figure 2a-b). Thenanoscale morphology of the NMO results in the exposure of ahigh surface area to the solution, and therefore, fast exchange ofionsbetweenthesolidandtheliquidphaseoccurs. Also, theshort diffusionlengthwithinthe individual NMOnanorodspermitsafasttransferofsodiumionsfromtheirsurfacesintotheir bulk interiors. The NMO electrode was then prepared bydrop casting a NMO-based ink on a carbon cloth currentcollector; these electrodes showed typical electrochemicalbehavior when in contact with sodiumcontaining electro-lytes.24,27A similar approach was used for the Ag electrode, andin this case the ink contained silver microparticles (0.51.0 mdiameter). Thechoiceof acarbonclothcurrent collector isdictated by its corrosion resistance in highly saline seawater. Inaddition, the superior adherence and soaking of the ink into thecarbonfibersresultsinhighermassloading, improvingdeviceperformance. The as-prepared Na2Mn5O10/AgCl cell wassubjected to galvanostatic cycling in an aqueous solutioncontaining 0.6 MNaCl in order to evaluate the optimaloperational potential range (Figure 3a). Three reaction plateausare observedat 0.3, 0.6, and1.0V, andeachone of themcorresponds to the intercalation of Na+into a crystallo-graphicallydistinct siteintheNMOstructure. Thereactionplateauat 0.3Vis most desirable for use ina desalinationbatterybecauseitishighlyreversiblewithlittleself-dischargeand has a low enough potential to avoid oxygen gas evolution inpH-neutral electrolytessuch as seawater.Electrochemical impedancespectroscopy(EIS)wasusedtoidentify the limiting step in the electrochemical reactionsoccurring in the cell.As shown in Figure 3b,it is evident thatthe formation of AgCl on the surface of the silver microparticlessubstantially increases the impedance of the system. In contrast,the NMO electrode, which has a high surface area, has a smallcharge transfer resistance, andis diffusion-limited. It is notpossibletoaddress throughthesemeasurements whether ornot the diffusion control is due to the solution or to the solidstate. Theresistanceofthesolutioninthedesalinationcell ismore than50%lower thaninthe floodedcell due tothepositioning of the electrodes.Desalination battery cycling was performed in a custom-made plexiglass cell (see Supporting Information). The cellpermits onlya two-electrodegeometry, withthreecompart-ments, two for the two electrodes, and one for the electrolyte.The total volume of the cell is around 300 L, and it exposes tothe solution 2 cm2of the electrodes. This cell geometry has theFigure2. SEMimages of theas-preparedNa2Mn5O10showing(a)good uniformity of nanorod morphology throughout the sample, and(b) nanorods with an average size about 300 nm in width and 13 min length.Figure 3. (a) Galvanostatic (1 mA) cycling of the Na2Mn5O10 (WE)Ag (CE) system versus Ag/AgCl/KCl 3M. (b) Nyquist plot of the EISspectrainthefrequencyrange100kHzto1Hzof: NMOworkingelectrode (red curve), silver counter electrode before (blue curve) andafter (green curve) chloride formation in a in seawater electrolyte witha three electrode setup.Black curve is the spectra of the desalinationcell.Nano Letters Letterdx.doi.org/10.1021/nl203889e | Nano Lett. 2012,12, 839843 841advantage of a reducedohmic dropandminimal processedvolume.The desalination cycle is started with the electrode materialsin their charged state in seawater. During the first step a 500Acm2current is applied, andions areremovedfromthesolution. Through control of the total discharge, the quantity ofions removedcan be controlled. The desalinatedsolution wasthenremovedfromthecell andreplacedwithnewseawater(Step 2).Then the desalinatedsolution was analyzed by ICP-MS in order to evaluate the Coulombic efficiency of thedesalinationprocess(Table1). Inthethirdstep, a+500Acm2current is applied, charging the electrodes by releasing theionsincorporatedinthefirststep. Theresultingconcentratedseawater solution is then replaced with new seawater (Step 4),completingthecycle;thesystemisthenreadyforthefurthercycling.The cycle for a 25%removal of chlorides is reportedinFigure4. Thecircularintegral oftheEvsQcurverepresentstheenergyrequiredfor thedesalinationprocess (removal of25% of salt) to take place, which is on the order of 0.29 Wh l1.This valueis verypromisingwhencomparedtothereverseosmosis under similar conditions (0.2 Wh l1).13In Table 1, theresults of the ICP-MS analysis of the seawater and thedesalinatedseawaterarereportedforatheoretical removal of25 and 50% of chlorides, respectively. To address uncertainty intheextractedvolumefromthedesalinationcell, sulfateionswere used as a control in the experiment. Sulfate ions cannot beremoved by using silver because Ag2SO4 is formed only at highvoltages (0.65 V/NHE) that are never achieved in thedesalination cell. All of the other concentrations were obtainedbybringingtheconcentrationof sulfateionsequal tothatinseawater. TheCoulombicefficiencies, C, reportedinTable1are defined as = z Fn nQ( )C,ii I Fwhere ziis the valence state of the ion, F is the Faradayconstant, nI and nF are the initial and final number of moles inthe solution, and Qis the total charge flown. Coulombicefficiencies, whichshould not be confused withthe energyefficiency of the device, provide information regarding theselectivity of the electrode material toward the ions in solution.Infact, itisnotablethattheextractionof cationsisnotNa+selectivebutalsoCa2+, Mg2+, andK+areabletoinsertintheNMO structure. Comparison of the crystal ionic radii of thesecations shows that Ca2+andNa+haveaverysimilar radius(around115pm), Mg2+is smaller (around86), whileK+isclearly bigger (ca. 150 pm). As consequence of this, it isobserved(Table1)thatcalcium, magnesium, andsodiumarealsoremovedfromseawater, whilepotassiumisremovedtoalesser extent. The total Coulombic efficiency of the desalinationcan be obtained from the value of chlorides and is ca. 80%. Theremainingcharge(about20%)islostduringthereductionofoxygen and formation of OH.The desalination battery has simple construction, uses readilyavailablematerials, has apromisingenergyefficiency, operatesat roomtemperature with fewer corrosion problems thanexisting desalination technology, and it could potentially be Na+and Clselective, which would end the need for resalination. Itsprimarylimitationof lowtotal ionextractionarisesfromthelow specific charge capacity of the NMO sodium ion electrode(35 mAhg1vs anaverage value of 160 mAhg1for Li-intercalating cathodes). This lowcharge capacity limits thevolume of water that can be desalinated within one cycle, andthereforetheoverall efficiencyof desalination. However, thebattery research community has recently shown increasinginterest inaqueoussodiumionbatteriesforgridscalepowerstorage applications. In the near future, higher capacity sodiumionelectrodes, as well as improvedchloride electrodes willmakethedesalinationbatteryafeasiblemethodforseawaterdesalination. With regard to the chloride intercalation/capturingelectrodematerial, silver shows several advantages:stability of potential, corrosion resistance, and bactericidalproperties. Nevertheless, the high price of silver, and poorelectronicconductivityof AgCl, whichistheprimarykineticlimitation of the desalination battery (see SupportingInformation)will limit the utility of theAg/AgCl electrodeinpractical devices. The workreportedhere demonstrates theconcept of adesalinationbattery. Further developments willresult in a versatile technology for the desalination of seawater,either independently or through integration with otherdesalination methods.ASSOCIATED CONTENT*S Supporting InformationAdditional information, figures, and table. This material isavailable free of charge via the Internet at http://pubs.acs.org.AUTHOR INFORMATIONCorresponding Author*E-mail: [email protected] 1. ICP-OES/ICP-MS:Coulombic Efficiencies andSelectivity for Anions and Cationsion Na+K+Mg2+Ca2+ClSO42sea water (mg/L) 11250 450 1400 450 18500 275025% removal(mg/L)9840 430 1130 280 14470 2750C,2547%