36 TheElectrochemicalSocietyInterface•Winter2008

Estimated production costs of PEFC’s

today: $1833/kWmass production:

$40/kW

Fig. 1. Estimated costs per kW for fuel cell stack. Cost estimates from Tsuchiya etal. (Reference 5).

Getting Back into Gear: Fuel Cell Development after the Hypeby Jeremy P. Meyers

Just over a decade ago in thismagazine, the question was posed,“IsThereaFuelCellinYourFuture?”

–.Whiletheauthorofthatarticlewiselyrefrainedfromexplicitpredictionsandcarefully spelled out a lot of the keychallenges that lay ahead for fuel celldevelopment at the time, readers whoread the article and who read aboutfuel cells in the mainstream presscouldbeforgivenforassumingthattheanswermightwellbe“Yes.Andsoon.”The late 1990s saw many companiesinvesting heavily in fuel cells withtheclearimplicationthatcleanenergytechnologywasabouttotakeoffinthesamewaythosecontemporaryinternetstartups had. Fuel cell developers sawtheirstockstakeoffasinvestorssoughtthe next big thing, and clean energyseemedlikeitwaspoisedtobethatveryopportunity. The stock market bubbleburst,however,and,notlongafter,thesuccess of hybrid vehicles suggestedthat they might achieve prominencein the marketplace before we mightexpect fuel cell vehicles to becomewidespread.

Fuel cells promise clean andefficient energy conversion, low-to-zero emissions at the point of use (amajor boon for urban environmentsin particular), and flexibility in termsof the primary source of power thatis used to generate the hydrogen orhydrogen-rich fuel. Industry andgovernment have partnered in NorthAmerica, Europe, and Asia to developproton-exchange membrane fuel cells(PEFCs) for stationary, portable, andtransportationapplications.1,2Thisclassof fuel cells has benefited from morethan a decade of industrial researchand development for these uses, withthelion’sshareofdevelopmentdevotedtotransportationapplications.Becauseof this prolonged development effort,we have seen marked improvementsin power density, energy conversionefficiency, and lifetime, and yet, veryfewofusareusingfuelcellvehiclesforourdailycommutes.

The fuel cell industry is regroupingand re-directing its energies after aperiod of exuberance and subsequentdisappointment. Fuel cell vehicledemonstration programs continue,albeit not at the pace that they wereonce projected to achieve. We findfuel cells moving closer to enteringinto specific niche markets such ascogeneration, forklift traction power,backup power/industrial batteryreplacementandconsumerelectronics,allofwhichhavegreatpromise,albeitsmaller markets than automotivemarketsmightsomedaybecome.Whileinvestors have become more cautious

about fuel cells, given the longperiodin which investments did not yieldlargegrowthashoped,thecurrentstateofthetechnologycontinuestoadvance,and is certainly more mature thanit was when this wave of investmentbegan.Nowisagoodtimeforareviewof fuel cell technology, its promise,and its remaining challenges. ECS isan excellent forum to discuss thesechallenges,asmanyoftheopportunitieslieattheinterfaceofelectrochemistry,electrochemical engineering, masstransport,andmaterialsscience.

Where Are the Fuel Cell Vehicles?

So, in the year 2008, at a time ofrecord gasoline prices and increasedenvironmental awareness, why aren’twe driving around in fuel cell cars?Firstofall,fuelcellsandthehydrogendelivery and storage infrastructureneededtosupportthemstillcostfartoomuch to be competitive with internalcombustion engines. Secondly, theysimply don’t last long enough. Whentrying to compete in the automotivemarketwherecarsroutinelyrunwellinexcessof100,000miles,durabilityisadifficultchallenge.

What can be done? We must firstidentify where the sticking points are,and figure out what we can do to getthem unstuck. We will consider theproblemofmarketacceptance,andtheimprovements that are still needed incostanddurability.

Match-making: Technology and Marketplace

PEFCs have been the technology ofchoice for transportation because oftheir low-temperature operation, rapidstartup capability, and flexibility interms of the source of the hydrogenfuel. After researchers at Los Alamosdemonstrated that a low-loadingcatalyst layer could be fabricated,significantinvestmentwaspouredintothe technology.3 While PEFCs haveyettomeetallof therequirements forautomotiveapplications,thepastdecadehasseenconsiderableprogressinpowerdensity anddurability. Itmakes sense,however, to consider the opportunityforfuelcellsinothermarkets.

Fuelcellsarenotre-charged,aswitha rechargeable battery; instead, theyare simply fed fuel and air, just like acombustion engine. The fuel must bereplenished after extended operationbecausethedurationofpowerprovidedis limitedbythefuelstorageavailable.However, the fuel can potentially bereplenished during operation, whichmakes fuel cells and generators thebestoptionsduringprolongedelectric-grid outages. Additionally, fuel cellsare inherently simpler than internalcombustion engines, although thebalance-of-plant equipment requiredis not produced in the volumes thatthe accessories for gasoline enginesare, and are not as integrated in theirfunctionalityasaresult.

TheElectrochemicalSocietyInterface•Winter2008 37

In general, the electrodes forelectrochemical energy conversiondevices must be placed very close tooneanothertominimizeOhmiclossesassociatedwiththepassingofcurrent.Giventheslowdiffusionandmigrationrates in electrolytes, all of the activematerial in a closed system such asa battery must be placed very closeto the electrodes. The distances overwhichreactantscanbecarriedareverysmall relative to convective flow, sofuel cells—with external fuel storagethat is conveyed to the electrodesin the gaseous of liquid state—cangenerallyprovidemuchlongerdurationof power for a given size and costthanconventionalbatteries.Therefore,applications with extended timesbetween refueling are places wherebatteriesstruggletomeetrequirements,andarefrequentlysupplementedwithagenerator,runningonfuel.

Whilefuelcellsareefficientrelativetocombustionengines,theyarenotasefficient as batteries, due primarily totheinefficiencyoftheoxygenreductionreaction (and, it must be pointed out,the oxygen evolution reaction, shouldthehydrogenbeformedbyelectrolysisofwater).Atthisstageofdevelopment,theymakethemostsenseforoperationdisconnected from the grid, or whenfuelcanbeprovidedcontinuously.Thedevice must be reliable, and it mustbe readilyavailablewhencalledupon.Batteriestendtodominatethesemarketsbecauseoftheirfavorableattributes,intermsofresponsetimeandcomplexityof operation. They start up essentiallyinstantaneously, most require low tozero maintenance, and have no localemissionsornoise.Untilquiterecently,fuel cells were too expensive even forthese markets, and had durability and

reliabilityproblemsthatprecludedthemfrom consideration. However, as PEFCtechnologyhasmaturedinanticipationof the automotive markets, they havebecome much more favorable. Forapplications that require frequent andrelatively rapid start-ups (on the orderofseconds)wherezeroemissionsarearequirement,asinenclosedspacessuchas warehouses, and where hydrogenis considered an acceptable reactant,a PEFC is becoming an increasinglyattractivechoice.

By targetingnichemarkets thatpayhigh costs for their power or haveissues with their current energy-storage systems, PEFCs can be sold atcommerciallyviablepricesinrelativelylow volumes. An example of such atargetapplicationisforindustrialutilityvehicles that can be rapidly refuelled.Theseuserscurrentlyrelyonthelabor-intensive process of mechanicallyexchanging battery sets of batterysets. Such niche markets can providethe production volumes necessaryto achieve further cost reductions,which will open up additional marketsegments. This is shown conceptuallyinFig.2.Theautomotivetargetof$40/kWisanaggressivetarget,andasamplecommercializationrolloutas shown inFig. 2a reveals the rate of adoption ofthetechnologyandtheaccompanyingcostreductions.Thedashedlineshowsthe competitive cost per kW that themarket is likely to bear. As long asproduction costs are higher than theacceptable price, then someone—either the producer or a governmentagency—must make up the differencebetween the cost of production andthepriceatwhich it is sold.Figure2bsuggests that for such an approach tothemarket,itwilltakenearly20years

before producers start to see profits,and will have 20 years of sunk coststoovercome,ormust relyuponaverypatient government subsidy programtomake it toprofitability. If,however,other niche markets can be entered athigher price points (and the fuel cellcancompetefavorablywithincumbenttechnologies at those points), thenprofits can be realized much sooner.Recentmarketstudiessuggestthatfuelcells can compete favorably with lead-acidbatteriesforforkliftapplicationsattoday’scosts,atleastinsomemarkets.4If they can enter this market, thenmanufacturerscanachieveprofitabilityand prove out longer lifetimes andlower costs for other markets whileturningaprofit.

It’s the Economics

Ifcostisaproblem,thenitbehoovesus to ask why fuel cells cost so much.Asksomeonewhohasn’tworkedintheindustryandheor she is likely to say,“It’s the platinum.” It is true that, todate,allfuelcellsuseplatinumcatalysts,and platinum is a very expensivematerial, one that has only becomemore expensive in recent years. It isperhaps surprising to learn, however,that, even if the platinum were free,thefuelcellswouldstillcosttoomuchto enter into markets. As shown inFig. 1, from a recent survey of fuelcell manufacturing,5 the low-volumecosts for fuelcellsarestill inexcessof$1800/kWandthelion’sshareofcostisestimatedtolieineitherthemembrane-electrode assembly or bipolar platemanufacturing, even neglecting thematerial costs. To date, manufacturersstillhaven’tmadeverymanyfuelcells,and they have neither the experience

Fig. 2. (a) Projections for cost reduction and product introduction from Tsuchiya etal. (Reference 5). (b) Projected losses/ profits assuming only a single point of entry into the market. Assumes $40/kW is price market will bear and cost projections from Figure 2(a).

(a) (b)

Meyers(continued from previous page)

with, nor the capital equipment for,high-volume manufacturing to bringthe costs down. While material costsshouldremainfixedwithvolume, it isanticipated that production costs willcomedownwithvolume.

AsPEFCtechnologytriestoenterthemarketplace,itisimportanttoidentifyprocesses that can be scaled up formass production of MEAs. There iscurrently little information availableabout the actual MEA fabricationprocesses that are used by industry.Theprocesses currentlyusedareoftenmoreamenabletobatchmanufacturingrather than continuous web processes.CommercialMEAfabricationprocessesaim to scale and improve on theabove mentioned thin-film electrodefabricationprocesses,theforemostgoalbeing to reduce the costs associatedwith manufacturing a MEA, whilemaintaining the performance levels.It behooves developers to understandhow to manufacture an MEA for highperformance, especially since higherpower density increases the numberof applicationswhere fuel cells canbeconsidered, but also because higherpower density means that less totalmaterial is ultimately required. Wetherefore want to design cells thatdeliverhighpowerperunitarea.

The proper construction of a stablecatalystlayerisoneofthemostcriticaldeterminants of performance for aproton-exchangemembrane(PEM)fuelcell. For any electrochemical devicethat needs to sustain high rates ofreaction, electrocatalysis plays anessentialrole,butmaintainingaccesstothatcatalyst isalsocrucial.Atpresent,all practical fuel cells operating withacid electrolytes employ expensiveplatinum or platinum-alloy catalyststo ensure a high reaction rate at thecathode.6Inordertoenhancereactionrates, the mass activity of the catalystmust be enhanced, either throughmodifications to the catalyst surfaceor by enhancement of the surface-to-volume ratio; in order for a fuel celltomaintainhighperformanceover itslifetime,thecatalystsmustretaintheirspecific activity and microstructure.The catalytic reaction in a fuel celldepends critically on the nature oftheelectrocatalyst,thecatalystsupport,and on the electrode and membrane-electrodeassembly(MEA)structures.

Optimized MEAs requirearchitectures that establish reactanttransport pathways to the dispersedelectrocatalysts:intheoxygenreductionreaction, for instance, one mustmaintain pathways for both protonicand electronic conduction, while stillpermittingefficientmoleculartransportof gas- or liquid-phase reactants andproducts to and from the carbonsupported nanoscale electrocatalyst.7

Catalystlayersarecarefullyconstructedto maintain these interfaces andtransport pathways; degradation thatresultsinchangestotheseinterfaceorininterruptionofthetransportpathwayswillresultinloweredperformanceandcanpotentiallyobscure anyadvantageof catalysts with inherently superioractivity. Understanding how to createthesecatalystlayersandhowtospecifythemanufacturingprocessesthatdeliverthese is a critical challenge.We see inFig. 3 an imageof a catalyst layer andcan see the multiple phases that mustbe brought into intimate contact: theionomer,thecatalystandsupports,andvoidsorphaseboundariesthatallowforgasreactanttransport.8Ultimately,itisimportanttounderstandhoweffectivetransport properties depend uponcatalyst layer morphology and how tospecify the manufacturing processesin such a way that they will yield anoptimized,stablestructure.

While perfluorinated membranesand precious-metal platinum groupmetalcatalystsare likely tobeused tomeet near-term cost reduction goals,theultimaterequirementssuggest thateither drastically lowered loadingsor non-precious metal catalysts arerequired. The search for alternativematerials must be guided by a greaterunderstandingoftheperformance,anddevice-level targets must be translatedto challenges that can be met bymaterial science by carefully isolatingand understanding the mechanismsthatgiverisetohighperformance.

Built to Last

While performance is importantboth for product requirements and asan enabler of cost reduction, lifetimeis also critical. The PEFC operatingenvironment exposes all of the cellelements to a battery of potentiallyharmful conditions: extremes inpotential, liquid water in a stronglyacidic environment, and possiblyreactivereactionintermediates.6

Membranes are critical componentsof the fuel cell stack and must beable to perform over the full rangeof systemoperating temperatureswithless than 5% loss of performance bythe end of life and without externalhumidification. Membranes can failsuddenly,withoutgradualperformancedegradation.9Thedurabilityofcatalystsis a critical problem, particularly asPEFCsstarttoincorporatelowercatalystloadings;smalllossesinelectrocatalyticactivity become proportionatelygreater as the initial area for theelectrochemical reactions decreases.The performance of the catalyst layerdegrades by platinum sintering anddissolution,especiallyunderconditionsof load-cycling and high electrodepotentials. Furthermore, the carbonsupportsthatprovideelectricalcontactto the platinum nanoparticles can beelectrochemicallyoxidizedundersomeoperatingconditions.10

Fuel cells for transportationapplications will need to show>150,000miles,orroughly5,000hoursof operation under aggressive cyclicconditions.Performanceoffuelcellsfor

Fig. 3. Image of a catalyst layer indicating the multiple phases that must be brought into intimate contact: the ionomer, the catalyst and supports, and voids or phase boundaries that allow for gas reactant transport. Image from Reference (8).

38 TheElectrochemicalSocietyInterface•Winter2008

stationaryapplicationsofupto20,000hours has been demonstrated, butfor fuel cells to displace conventionalgenerators, developers will have tomore than 40,000 hours of reliableoperation over the full range ofexternal environmental conditions(-35°C to 40°C). Demonstrationof these lifetimes suggests either anintimate understanding of all of thepossible modes of failure, or a verylongdevelopmentandvalidationcycle.Greater understanding of all of theprocesses leading to failure—chemicaland mechanical attack of the mem-brane, dissolution, corrosion, etc.—isnecessarytoensurethatdeveloperscanrationallytradecostversusperformanceand lifetime.Manymore fundamentalstudiesareneeded,andaccesstolong-termperformancedataatthescaleandunderthetypicaloperatingconditionsof end-use applications are required.While we anticipate that the nicheapplications outlined above will beaccessible with the current state ofthe technology, it still must be shownthat the peculiarities of the operatingenvelopeinthesenewapplicationswillnotopenupnewmodesoffailurethatarelessimportantinthetransportationand stationary applications thathave been tested already. Soon, timewilltell.

The Way Forward

The initial stages of PEFCdevelopmenthavegivenwaytoamoresober, practical approach wherein itis acknowledged that less attention-grabbing applications are likely tolead the way to commercialization,and where fundamentals are beingstudied tounlockmechanisms,designguidelines, and templates for newmaterials.Battery-basedhybridvehicletechnology will likely exist alongsidefuel cells, and may well win out insome transportation applications, butfuel cells have a list of characteristicsthat suggest that they will find theirownnichesaswell.Nowisatimeripefor collaboration between researchers,to look at the wealth of data that hasbeencreatedintheearlystages,totakeahardlookatwhich(andtoseize)theearly market opportunities that fuelcellsarepoisedtoenter.

References

1. U.S. DOE, Multi-Year Research,Development and DemonstrationPlan:PlannedProgramActivities for2005-2015,http://www1.eere.energy.gov/hydrogenandfuelcells/mypp/.

2. European Hydrogen and FuelCell Technology Platform Imple-mentation Plan - Status 2006, inEuropean Hydrogen and Fuel CellTechnologyPlatform(HFP)(2007).

3. M. S. Wilson, F. H. Garzon, K. E.Sickafus,S.Gottesfeld,J. Electrochem. Soc.,140,2872(1993).

4. B.Cook,FuelCell Industry:Longer-Term Potential, But High RelativeCostsLimitNear-TermOpportunity,in J. P. Morgan North AmericanEquityResearch(2007).

5. H.TsuchiyaandO.Kobayashi,Int. J. Hydrogen Energy,29,985(2004).

6. R. Borup, J. Meyers, B. Pivovar, Y.S. Kim, R. Mukundan, N. Garland,D. Myers, M. Wilson, F. Garzon, D.Wood,P.Zelenay,K.More,K.Stroh,T. Zawodzinski, J. Boncella, J. E.McGrath,M.Inaba,K.Miyatake,M.Hori,K.Ota,Z.Ogumi,S.Miyata,A.Nishikata, Z. Siroma, Y. Uchimoto,K. Yasuda, K. I. Kimijima, and N.Iwashita,Chemical Reviews,107,3904(2007).

7. T.E.Springer,T.A.Zawodzinski,andS.Gottesfeld,J. Electrochem. Soc.,138,2334(1991).

8. J. Xie, D. L. Wood, K. L. More,P. Atanassov, and R. L. Borup, J. Electrochem. Soc.,152,A1011(2005).

9. K.D.Kreuer,J. Membrane Science,185,29(2001).

10. C.A.Reiser,L.Bregoli,T.W.Patterson,J.S.Yi,J.D.Yang,M.L.Perry,andT.D. Jarvi,Electrochem. Solid-State Lett.,8,A273(2005).

11. J.F.ChalkS.G.Miller,J. Power Sources,159,73(2006).

12. M. Fowler, R. F. Mann, J. C.Amphlett, B. A. Peppley, and P. R.Roberge, Handbook of Fuel Cells:Fundamentals, Technology andApplications, Vol. 3, Fuel CellTechnology and Applications, W.Vielstich, H. A. Gasteiger, and A.Lamm, Eds., p. 663, John A. WileyandSons,(2003).

About the Author

Jeremy P. meyers is an assistant pro-fessor in Mechanical Engineering andMaterials Science & Engineering at theUniversityofTexas atAustin. Prior tojoining the faculty at the Universityof Texas, he managed the advancedtransportation technology group atUTC Power, where he was responsiblefor performance and durabilityimprovements for PEM fuel cells. In2006, he and his co-workers receivedtheGeorgeMeadMedalforEngineeringAchievement at UTC, and in 2008, hereceived the DuPont Young FacultyAward.JeremyiscurrentlythetreasureroftheECSEnergyTechnologyDivisionandcanbereachedatjeremypmeyers@mail.utexas.edu.

General Topics

F Batteries,FuelCells,andEnergyConversion,includingSOFC11

F BiomedicalApplicationsandOrganicElectrochemistry

F Corrosion,Passivation,andAnodicFilmsF DielectricandSemiconductorMaterials,

Devices,andProcessingF Electrochemical/ChemicalDepositionand

Etching,includingEUROCVD16F ElectrochemicalSynthesisandEngineeringF Fullerenes,Nanotubes,andCarbon

NanostructuresF PhysicalandAnalyticalElectrochemistryF SensorsandDisplays:Principles,Materials,

andProcessing

DeadlinesAbstract SubmissionAbstractsaredueNOLATERthanApril24,2009.

EUROCVDabstractsaredueNOLATERthanApril3,2009.

Pleasecarefullycheckeachsymposiumlistingforanyalternativeabstractsubmissiondeadlines.

RegistrationDeadlineforadvancemeetingregistrationisSeptember4,2009.

HotelsSeveralhoteloptionswillbeavailablefromtheECSwebsite.ReservationsaredueAugust24,2009.

Short CoursesAnumberECSShortCourseswillbeheldattheViennameeting;todate,thecourselistincludesBasicImpedanceSpectroscopyandBasicsofCleaningProcessingforIntegratedCircuitManufacturing.

ECS TransactionsFullpaperspresentedatECSmeetingswillbepublishedinECS Transactions(ECST).VisittheECSwebsiteformoredetails.

Travel GrantsTravelgrantsareavailableforstudentattendeesandforyoungfacultyandearlycareerattendees.

Abstract Deadline: April 24, 2009EUROCVD ONLY: April 3, 2009

ViennaVienna

216th ECS Meeting

Austria FOctober 4-9, 2009

w w w . e l e c t r o c h e m . o r gVisit the ECS website for meeting details: