- Email: [email protected]
Pem fuel cells for transportation and stationary power generation applications
Int
Pergamon PII: s0360-3199(97poo16-5
./. Hrdriroqm Enrr,q~, ( 1997 lnternatimal
Vol.
22, No. Association
11. pp. I I:7 11.44, 19Yhr Hydropen Energ) Elsevter Sctence Ltd
411rights reserved.Printed in Great Britain 0360~3199!97
PEM FUEL CELLS FOR TRANSPORTATION AND STATIONARY GENERATION APPLICATIONS S. J. C. CLEGHORN. X. REN, T. E SPRINGER. M. S. WILSON, C. ZAWODZINSKI, and S. GOTTESFELD
61 -.OOi
f).O!:
POWER
T. A. ZAWOD%INSKI
Electronic and Electrochemical Materials and Devices Group, MST-l I, MS D429, Los Alamos National Lahoralory Los Alamos, NM 87545, U.S.A.
describe recent activities at Los Alamos National Laboratory devoted to polymer electrolyte fuel ceils in the contexts of stationary power generation and transportation applications. A low cost/high performance hydrogen or reformateiair stack technology is being developed based on ultra-low platinum loadings and non-machined. inexpensive elements for flow-fields and bipolar plates. On-board methanol reforming is compared to the option of direct methanol fuel cells in light of recent significant IDower density increases demonstrated in the latter. (‘. 1997 International Association for fiydrogen Energ; Abstract---We
INTRODUCTION
technology for utility and stand-alonepower generation applications,as well asfor transportation, where It has The fuel cell is the most efficient meansof converting previously held the strongestattraction. hydrogen to electric power. As such, the fuel cell repEventhough theseandother recentdevelopmentshave resentsa key element in efforts to demonstrateand beenvery effective in raisingthe profile of PEM fuel cells implement hydrogen as a fuel for electric power asa viable technologyfor transportation, the commercial generation. implementationof hydrogen fueled PEM fuel cell stacks For the last 30 years,the benefitsprovided by fuel cell is likely to occur first for stationary power generation technologies,including high efficiency, low environ- applications, becauseof the somewhatless stringent mental emissionsand modular construction, have fos- requirementson overall cost. space,and the nature tered continued interest in their development for sourceof the hydrogen fuel. transportation and customer-side-of-the-meterutility Somerecent achievementsat Los Alamos that have applications.Among the family of fuel cell technologies, advancedour PEM fuel cell technology to the point of polymer electrolyte membrane(PEM) fuel cells have readinessfor the demonstrationof a low cost/high persomeuniquely attractive features. They provide high formancehydrogen fueledstack are summarizedbelow. power density, high chemical-to-electricalenergy con(MEAs) with versionefficiency, and fast and easystart-up. Addition- l Low costmembraneelectrodeassemblies ultra-low platinum loadingsof 0.2 g Pt/kW have been ally, they are very reliable and are constructed from demonstratedin 5000h life tests. generating area1 durable and benign materials. Until recently, however, power densitiesof 6 kW/m’ on pressurizedhydrothe large-scale,terrestrial application of PEM fuel cell gen/air [I -41. These MEAs have recently been technology for transportation and utility or stand-alone implementedsuccessfully in stacksby somedeveloper?. powergenerationwasconsideredto betoo expensiveand l Effective water managementprinciples have been too sensitive to the low levels ofcarbon monoxide present identified, basedto a large extent on advancedmemin reformed carbonaceousfuel feed streams.Thesebarbrane electrolytes evaluated jointly with membrane riers to the practical implementationof PEM fuel cells manufacturers[5, 61. have been significantly lowered in the last few years, in l Detailedevaluationsof losses in the PEM fuel cellhaye part as a result of work done at Los Alamos National beenbasedon long-terminvestigationof fundamental Laboratory. This paper describessomeof our efforts to lower thecost of PEM fuel cell stacksandto demonstrate processes in the cell, including all electrocataiytic and someeffective approachesfor the mitigation of anode transport phenomena[6J. catalyst deactivation
effects resulting
from low levels of
l
CO or other organic impurities. Theserecent technical developmentsincreasethe attractivenessof PEM fuel cell 1137
Based on the above evaluations,
an effective computer
codeand a setof experimentaldiagnosticsfor the PEM fuel cell have been established [7).
1138 l
S. J. C. CLEGHORN
The toleranceto carbonmonoxidelevelsexpectedfrom steamreforming of carbonaceousfuels has beensignificantly improved. In previouswork we have shown that air bleedinginto the anodecorrectsfor CO levels in the anode feed streamas high as 50-100 ppm [8, 91. However, we have recently identified alternative methodsto increasethe CO level tolerance to well above 100ppm.
Theseadvancementsin PEM fuel cell technology at Los Alamos have contributed significantly to the understandingand developmentof high performance,low cost PEM fuel cell stacks. Activities basedon thesetechnologiesare taking place at severalstack development sitesin North America. Most recently,the Fuel Cell CoreTechnologyProgram at Los Alamos addressedPEM fuel cell issuesrelated to both transportation and stationary power generation applications. In the transportation arena, providing hydrogen/air and reformate/air PEM fuel cell stacksat low cost and coupling the PEM fuel cell stack to reformersof carbonaceousfuels remainimportant challengesto overcome.Fuels being consideredrange from methanol to petroleum mid-distillates. In a transportation systembasedon the reforming of liquid hydrocarbons,the challengeof operatingPEM fuel cell stacks on hydrogen that is diluted by CO* (and nitrogen), in addition to containing significantlevelsof CO, is similar to that for stationary powerapplicationswherethe stack is fed by reformednatural gas.The situation in the latter caseis significantlylesssevere,however, becausestationary power requireslessstringentcost and power density targets. This paper describessome of our researchefforts devoted to low cost/high performance PEM fuel cells for stationary power generation.It alsobriefly mentions somerecentachievements with direct methanolPEM fuel cellsat Los Alamos. STATIONARY Background
POWER APPLICATIONS [l&12]
et al.
over PAFC technology in terms of energy conversion efficiency (higher voltage at samecurrent density) and power density. In the long term and in light of recent advancements,the life-cycle cost of PEFC technology could be more favorable. A stable,inert polymer electrolyte and an easy start-up/shut-down characteristic may make the reliability and longevity of the PEFC superior. The PEFC could compete for the near-term marketsin which the PAFC is expectedto be successful. As such, the following background discussionsare extracted from studiesoriented towards the PAFC [1316]. Many applicationsin the U.S. utility sector require power capacitiesin the range of 1 to IOMW. This is particularly true for peak demandin someof the small public utility systemsthat are requiredto pay high charges for peak demand, usually to a private utility grid. Thesemodestutility applicationscould be more suitable for the PAFC and PEFC. Indications are that at US$l lOO/kW, fuel cell technologiescould be attractive for peak power requirements.A conservativeestimateof this market in the U.S. is 14000to 17000MW for fuel cells. The efficiency, compact design,low maintenance and potentially lower cost of the PEFC could allow it to capture a substantialportion of this market. Ideally, the PEFC systemwould be in a cogenerationunit, with the wasteheat utilized for its thermal value. Increasingthe PEFC operatingtemperaturereducesthe effect of CO on the anodecatalyst and would allow the wasteheat to be usedmore effectively. Someincreasein operating temperatureis possibleevenin presentday PEFCs. It is becoming more difficult to obtain regulatory approval for large-scalepower stations or to secure rights-of-way for high tensionpower lines.Thus, in the relatively near future, it is reasonableto expect that a substantialamount of additional power capacity may be from neighborhoodor apartmentbuildingpowerstations that tap into the massivenatural gasdistribution grid in the U.S. Efficient, low emission,and quiet fuel cellsare ideal for this type of application. Though gas turbine systemsby comparisonarelessexpensive,it may bemore cost effective to pay more for an environmentallybenign fuel cell systemand bypassthe regulatory processes and the associatedcosts.Industrial and commercialfacilities will probably take the lead in implementingfuel cell systems.According to the Gas ResearchInstitute (GRI), approximately 18000MW of the latter power market in the U.S. shouldbe economicalfor fuel cell cogeneration systems,provided the installedcost is USS1OOOjkW[ 151. The industrial sectoris similarto the commercialsector in sizeand in requiredfuel cell characteristics.Approximately 10% of this potential market is for small (less than 1MW) cogenerationsystems.Waste heat is at a premiumin theseapplications,thus the PEFC may need to operateasa cogenerationsystemto be competitive.
While large utility applications are envisioned as belongingto the domain of molten carbonatefuel cells (MCFCs) and solid oxide fuel cells(SOFCs), there are many applicationssuchasutility peakpower generation, demand-side-management, and dispersedpower generation for which the phosphoricacid fuel cell (PAFC) and the polymer electrolyte fuel cell (PEFC) could be moreattractive. Natural gasfueledPAFC technologyfor such applications has becomecommercialized,with a 200kW PAFC unit (from ONSI) costing approximately US$2500/kW. The unit operates at 40% electrical efficiency and 85% overall thermal efficiency (including heated water). Ballard, originally in conjunction with Dow, is developinga 250kW natural gas fueled PEFC Low cost stack components power plant after the successfuldemonstration of a With a target of only US%lOOO/kW, a PEFC stationary 10kW plant [ 121. The PEFC appearsto havesomeimportant advantages power plant, with significant power conditioning and
PEM
FUEL
CELLS
FOR
TRANSPORTATION
AND
STATIONARY
natural gas reforming costs, will require relatively low cost fuel cell stacks. At Los Alamos, we are experimenting with the scale-up and automated fabrication of the low platinum loading catalyzed membranes based on “thinfilm” catalyst layers [l+]. A recently developed scheme for the automated preparation of lab-scale (areas up to 200cm’) membrane electrode assemblies is depicted in Fig. 1. The fabrication process of applying thin-film catalysts directly to the membrane is computer-controlled and provides high repeatability [ 171. At present, some of our work concentrates on the development of low cost gas diffusion backing, flow-field and bipolar plate materials. We have investigated metalbased hardware components as part of our PEFC stack development for stationary power applications, where the weight density of the stack is not as critical as for transportation applications. One promising stack contiguration is based on the use of stainless steel screens as flow-fields. Researchers elsewhere [18] have reported the use of stainless steel flow-fields with no apparent effects of corrosion after 1000 h of testing. Though the stainless steel element is not in direct contact with the membrane electrolyte, the surface of the metal is exposed to the corresponding humidified gas streams. Therefore, some form of surface treatment may eventually be required to achieve extended long-term stability. lnitial short-term tests of PEFCs based on ultra-low Pt catalyst loadings [l-4] and stainless steel screen flowfields show performance equal to or exceeding that obtained under similar conditions with machined metal flow-fields [17]. The ability to eliminate machining costs of flow-fields by using off-the-shelf metal screens would be advantageous to a stack manufacturer attempting to lower the cost of PEFC stacks. Fig. 2 depicts a polarization curve for a 100 cm’ active area PEFC, employing it Los Alamos MEA with 0.14mg Pt/cm’/electrode and stainless steel screen flow-fields. The screens seem to be effective flow-fields under ordinary PEFC conditions of
mixed gas/liquid gas distribution surface areas.
fabrication
using a computer
GENERATION
APP1.1(‘ATIOK\I
1 I39
(water) feed streams, enabling cffectlve across relatively large electrode
Stable PEFC anode operation on a feed stream oi natural gas reformate is a challenging requirement for utilization of PEFCs for stationary power generation. Some of the problems associated with the operation ofPEFCs on realistic fuel feed streams have been addressed recently by operation of PEFCs on methanol reformate. The roughly 1% CO in a methanol reformate can be decreased to about 10 ppm using shift reactors I’ollowed by a preferential oxidation reactor (PROX) utilizing a Pt/alumina catalyst [19]. The sensitivity of the PEFC to such residual levels of CO (10 ppm) is still very significant at the normal operation temperature of 80 ‘C. We have demonstrated the ability to remediate effects of residual CO in the fuel feed stream by bleeding small levels of 0, or air into the anode compartment of the fuel cell 18. 91. There have been recenl indications that others have successfully implemented this approach in PEFC‘ stacks. We have also shown [20] that a Pt--Ru anode catalyst does not catalyze the reverse shift process that generates CO from CO? within the fuel cell anode, but the high level of CO2 in the reformate can still create dilution effects. A combination of (i) shift reactors to lower the roughly 10% CO level of the methane reformate down to about l%, (ii) a PROX reactor upstream of the fuel cell to lower the CO to the I-~10 ppm level. and (iii) limited air bleeding into the anode to eliminate the residual CO effect, could conceivably provide fuel cell performance very nearly equal to that of a neat hydrogen cell. ‘The general system is depicted in Fig. .?a. An overall loss oi fuel caused by the oxidative treatments rn quch 3 cvstcm should not exceed 4%. Several recent efforts with Pt- Ru:(‘ c:~talvq:l\ [?(I I!?]
A: x-position B: y-position
Fig. 1. MEA
POWER
C: pen drop D: pump on
controlled process for applying a layer of catalyst blank positioned on a heated vacuum table.
ink to either
a membrane
tx a dcca I
1140
S. J. C. CLEGHORN
0.2
I-
0.0
l-Y 0.0
f
I
I
I
I
I
0.2
0.4
0.6
0.8
1.0
1.2
CURRENT Fig. 2. Internally
manifolded
et al.
DENSITY
(A/cm*)
100 cm2 cell with 0.14 mg Pt/cm2 thin-film electrodes on Nafion TM 112 membrane screen flow-fields. Hz/air = 3/3 atm, 50% air utilization at 1 A/cm’, T,,, = 80°C.
have indicated that some of the problems caused by residualCO within the PEFC anodecan be solvedby the use of appropriate Pt-Ru anode catalysts that appear to exhibit sufficient CO electro-oxidative activity. The identification of a CO tolerant, or rather CO consuming, anodecatalyst would naturally have a significantimpact on the designand efficiency of the system.
a) FWl +
Purifving naturalgas
--)
SHUT REACTOR(S)
steel
reformate with a hydrogen separator
As discussed above, a major issueimpactingthe coupling of the PEFC to a natural gas(methane)reformer is the effect of contaminants,particularly CO, on the fuel cell anode.A possibleapproachto circumvent this difficulty isto utilize someform of gasseparator,which would
Air Inject I
REFORMER
and stainless
Air Bleed I PROX REACTOR
PEFC
BURNER I Anode
Off-gas
\Comb;stio” Air
Combustion Air
Fig. 3. Reformer/fuel
processors for PEFCs: (a) a more conventional system with shift and PROX points, and (b) an alternative concept incorporating a hydrogen permselective
reactors and air injection separator.
at two
PEM
FUEL
CELLS
FOR
TRANSPORTATION
AND
STATIONARY
supply pure hydrogen to the fuel cell. In this section, we describe a recent effort at Los Alamos to develop an effective hydrogen separator to provide a pure hydrogen feed stream from simulated reformates to a polymer electrolyte fuel cell. This approach, depicted in Fig. 3b, not only eliminates the anode contamination difficulties, but decouples to some extent the operation of the fuel processor and the fuel cell. The hydrogen feed to the fuel cell could be effectively dead-ended, as is done in a number of fuel cell systems that use a neat hydrogen feed. This simplifies control of the complete system and allows stockpiling of hydrogen during low power demand. Thus, the fuel cell stack could then be brought up instantaneously with a “cold” reformer. Commercial hydrogen purifiers traditionally use Pd!‘Ag films, either free-standing or supported, which have significant cost and integrity drawbacks. Several groups are studying ways to increase the stability and lower the Pd content of hydrogen separator membranes. A fellow group at Los Alamos is developing Pd clad Ta films [23]. Tantalum supplies phase and structural stability. while providing high hydrogen permeability. The Pd is applied to either face of the Ta film to facilitate the interfacial process of hydrogen dissociation. At this point, the total Pd loading is only about 1 mg Pd/cm’ of separator membrane. We have operated a Scm’ single cell PEFC on the permeate of a 4cm2 Pd/Ta membrane separator operating at 3 15°C with a simulated methanol reformate feed (I % CO, 24% CO,, 74% H,). Fig. 4 indicates that about 600 mA/cm’ is obtained at 0.6 V. In contrast, the presence of only 100 ppm of CO (in H2) fed directly to the anode suppressed the current density at this voltage to about 60 mA/cm*, one-tenth the cell performance on the permeate. It is anticipated that the separator will be equally efTective with the much higher CO
POWER
GENERATION
.4PPLK’4TIOT\:S
! IJI
levels expected from a natural gas reformer. On-going development objectives for this membrane separator include improving its throughput and further lowering the Pd loadings. Currently, the amount of Pd required is comparable to adding 1 mg Pd/cm2 fuel cell membrane area. Preliminary process simulations for the reformer: separator/fuel cell scheme depicted in Fig. 3b are quite promising, particularly for a methanol fueled system. If a permselective separator is effectively paired with a methanol reformer, the need for high and low temperature shift and partial oxidation reactors would be eliminated because the roughly 1% CO in the reformate product is useful as burner fuel for the steam reformer, The natural gas fueled system is more problematic when using a separator because of the lower hydrogen concentration in the reformate, however, a number 01 approaches to increase the separator throuphput are possible.
DIRECT
METHANOL
PEM
FUEL
(‘EL.LS
Until very recently, the PEFC systems considered for transportation applications have been based on either hydrogen carried on board the vehicle. or on the steam reforming of methanol to generate a mixture of hydrogen and CO?, as the fuel feed stream for the fuel cell stack. A primary reason for resorting to methanol reforming onboard the vehicle has been the much more limited per.. formance of the direct methanol fuel cell (DMFC). The demonstrated power density for the latter has been ter) low indeed, and until recently its application for transportation was not a serious option at all. However. recent advancements in DMFC K&T) have
Pseudo-Reformate
(1% Membrane
CO) Separator
WI
1
i
1
0.0
’ 0.0
sp”I’*Flp
,
,
I
I
I
0.20
0.40
0.60
0.80
1.0
1.2
CURRENT DENSITY (A/cm2) Fig. 4 Polarization curves depicting a single cell PEFC operating at 80°C on(i) pseudo-reformate (I % CO) fed via a high tht oughput hydrogen permselective separator, (ii) 20 ppm CO in hydrogen, and (iii) 100 ppm CO in hydrogex
S. J. C. CLEGHORN et al.
1142
been quite dramatic, achieving power densities which are significant fractions of that provided by the reformate/air fuel cell (RAFC) [24-281. The more applied DMFC R&D work, aimed primarily at the demonstration of enhanced stable performance of DMFC single cells or stacks, has been supported in the U.S. by the Department of Energy (Office of Transportation Technology), the Advanced Research Projects Agency (ARPA) and the U.S. Army. Similar efforts in Europe resulted in a demonstration at Siemens of a high performance polymer electrolyte DMFC operating at temperatures approaching 130°C ~241.
prisinga methanolreformer and PEM reformate/air fuel cell stack. Basedon that evaluation, it seemsthat the DMFC could provide comparablepower densitiesand energy conversionefficiencies,thusbecominga seriouscandidate for transportation applications,provided the following requirementsare alsomet:
As in the previouscaseof the hydrogen/air PEFC or the RAFC, the strong recent advancesin DMFC performance have been achieved without any dramatic breakthroughsin electrocatalysis.The useof established Pt-Ru anodeand Pt cathodeelectrocatalystsin polymer electrolyte DMFCs operated at temperaturesof ll(r 130°Cprovide very significantenhancements in DMFC performancewhen catalyst layer compositionand structure areoptimized. Operation at elevatedtemperaturesis facilitated in such polymer electrolyte fuel cells by the anodebeing continuously in contact with liquid methanol-water mixtures. Recent experimentalwork at Los Alamos National Laboratory resultedin record highperformancesof polymer electrolyteDMFCs [28], demonstratedby Figs 5 and 6, which showpolarization curvesfor a methanol/air cell and power density curvesfor DMFCs with oxygen and with air cathodes,achievedin cellswith NafionTMmembranesoperatedat elevatedtemperatures.Theserecent resultshave been subsequentlyused in a comparative evaluation [29] for transportation applications of (i) a polymer electrolyte DMFC stack, and (ii) a systemcom-
0.8
1. catalystloadingsfurther reduced(or alternativeanode catalystsdeveloped); 2. long-termstableperformance(1000h time scale)demonstrated; 3. strategiesto minimizemethanolcross-overto the cathode[30] increasethe DMFC fuel efficiencyto the 90% level. CONCLUSION We have investigatedthe use of ultra-low Pt loading MEAs (0.14mg Pt/cm2)and stainlesssteelscreenflowfields in lOOcm* single cells, as part of our efforts to develop low cost/high performance hydrogen or reformate/air PEFC stacksfor stationary power applicationssuchas utility peak power generation,demandside-management and dispersedpower generation. In general, theselow cost component cells provide performances(approximately 0.6A/cm2 at 0.7V) approaching thoseobtainedwith commercialcells,which typically utilize much higher Pt loadings(4mg/cm2) and more expensiveflow-field materials (i.e. machined graphite plates).On-going researchincludesfurther development of inexpensiveflow-field, bipolar plate and gasdiffusion backingmaterials.Additional researchat Los Alamos is focusedon reducing the Pd content of hydrogen mem-
u N117 Pt-RuOx, 22 1 M methanol,
mg cm* 2.0 mL min,
12 psig
0.8 Air, 30 psi,
0.6 L m in
0.4
CURRENT
DENSITY
(A cmm2)
Fig. 5. Polarization curves for 1 lO”C, air cathode DMFC based on thin-film catalyzed NafionTM 112, 115 and 117 membranes. 2.3 mg/cm* Pt-black, 3 atm air at 0.6 l/min. Anodes: 2.2 mg/cm* Pt-RuOx, 1 M methanol at 2 ml/min and 1.8 atm. Cathodes:
PEM
FUEL
CELLS
FOR
Fig. 6. Power
density
curves
TRANSPORTATION
AND
STATIONARY
CURRENT for a thin-film
catalyzed
Nafion””
brane separators for RAFC applications and improving overall DMFC performance, for which greatly increased power densities have recently been obtained. Ac.knoIvlealqrmenrs-Nathaniel Peachey and Robert Dye of MST-7 at Los Alamos provided the membrane separator and assisted in coupling it with the fuel cell. Funding was provided by two program offices within the U.S. Department of Energy, the Office of Transportation Technology and the Hydrogen Program in the Office of Utility Technologies.
REFERENCES 1. Wilson, M. S. and Gottesfeld. S.. Journal of’ Applied Electrochf~mistry, 1992,22. 17. 2. Wilson, M. S. and Gottesfeld, S., Journal of’ the Electrochemical Sock/y, 1992, 139, L28. 30. 3 Wilson, M. S., U.S. Patent Nos. 5.211,984 and 5,234,777. 1993. 4 Wilson, M. S., Valerio. J. and Gottesfeld. S., Electrochimica i4<.tu, 1995. 40, 355.-363. T. A. and Gottesfeld, S., Jour5 Springer, T. E., ZawodLinski. nul qf‘the Electrochemicul Society, 1991, 138, 2334-2442. 6. Zawodzinski? T. A., Springer, T. E., Uribe, F. A. and Gottesfeld. S.. Solid State lonics, 1993, 60, 199-21 I. 7. Springer. T. E., Wilson. M. S. and Gottesfeld, S., Journal of the Electrochemical Society, 1993. 140, 3513. 8. Gottesfeld. S. and Pafford, J., Journal of the Electrochemical Society, 1988. 135, 2651L2652. 9. Gottesfeld, S., U.S. Patent No. 4.910,099, March 1990. 10. Prater, K. B., Journal of’ Power Sources, 1994, 51, 129. 11. Landgrebe, A. and Gottesfeld, S. In The Electrochemical Society Extended Abstracts. Vol. 94-2. Abstract No. 604, p. 944. Miami Beach, FL. October 1994. 12. Watkins, D. S. and Peters, B. C. In Program and Absfracts, Fuel Cell Seminar, San Diego, CA, 28 November -I December 1994. p. 250. 13. Impact of fundamental technical improvement. GRI Report No. GRI-8910137, International Fuel Cells Corporation, 1989
POWER
GENERATION
APPLI(‘ATiOhi
! 13:
DENSITY (Ahn*) I 12 assembly
operating
at 130 (’ on oxygen
and 1 IO C’ ,,n ‘lit
14. The financial and strategic benefits of fuel cell power piants. Vols I and II. EPRI Report No. EPRI EM-4511. ‘Temple. Parker and Solane. Inc., 1986. 15. Trimble, K. and Woods, R.. .lourrta/of’Poil’l~r .%ur~ es. 19’90, 29, 37-45. 16. Serfass, J. A., Journal of Power Sources. 1990.29. I 19 130. 17. Zawodzinski. C., Wilson, M. S. and Gottesfeld, S. In Proton Conducting Membrane Fuel Cc& I. The Electrochemical Society Proceedings Series, PV 95-23, eds S. Gottcsfeld. G. Halpert and A. Landgrehe. The Electrochemical Sncietv. Pennington. NJ, 1995, p. 57. 18. Mallant, R. K. A. M.. Koene, F. G. H.. Verhoeve, C. W G. and Ruiter, A. In Proywm and Ah.vracr.\. Fuel Cell Seminar, San Diego. CA. 28 Novembers 1 December 1994. p. 503. 19. Vanderborgh, h. E., Guante, J., Dean. 12. E. and Sutton. R. D. In Program and .46stracts. Fuel Cr# Semirw. Long Beach, CA, October 1988. p. 52. C. R., Valerio. J. A. and Gottesfeld. 20. Wilson. M. S., Derouin, S. In Proceedings of’ the 28th Intersociety FZnrr,g~~ C’rvwrsion Engineering C’onfcrenc,c (IEC‘EC). Vol. 1. Atlanta. (;.A. August 1993, p. 1203 21. Iwase. M. and Kawatsu, S. In Proton C‘onductq Menhrrrrrt Fuel C&s /. The Electrochemical Society Proceedmgs Series. PV 95-23. eds S. Gottesfeld, G. Halpert and A. La!xdgrehc. The Electrochemical Society. Pennington, NJ, 1995, p. I?. 22. Schmidt. V. M., lanniello. R.. Oetjen. H.-F., Reger. f-1.. Stimming, U. and Trila. F. In Proton Conducting Mrmhrunv Fuel c’rl1.s I. The Electrochemical Society Proceedings Series. PV 95-23, eds S. Gottesfeld, G. Halpert and A. Landprehe. The Electrochemical Society: Pennington. NJ, 199.5. p. 1. 23. Peachey, N. M.. Snow, R. C. and Dye. R. (‘., .Ic~urr~cri 01 Membrane Science, 1996. Ill, 123~. 133. 24. Griine, H.. Luft, G.. Mund, K. and Waidhas. M. In /+ryrm~ und Abstrwts, Fuel Cell Semmar. San I>iegn. CA. 2X Novembe---l December 1994, p. 474. 25. Surampudi. S.. Narayanan. S. R., Vamos. E., Fr.rnk. If.. Halpert, G.. LaConti. A., Kosek, J.. Surya Prakash, G. K and Olah, G. A., Journal of Power Sourcc~s. 1994. 47. 37” 26. Kosek, J. A.. Cropley, C. C.. Wilson. G . LaConti. 4 H.. Narayan. S Vamo\. F. ‘Surampudi. S :~ntl t-rai:h tl II:
1144
S. J. C. CLEGHORN
Proceedings of the 28th Intersociety Energy Conversion Engineering Conference (ZECEC), Vol. 1, Atlanta, GA, 1993, p. 1209. 27. Savinell, R. F., Yeager, E., Tryk, D., Landau, U., Wainright, J. S., Weng, D., Lux, K., Litt, M. and Rogers, C., Journal of the Electrochemical Society, 1994, 141, L46. 28. Ren, X., Wilson, M. S. and Gottesfeld, S., Journal of the Electrochemical Society, 1996, 143, L12. 29. Ren, X., Wilson, M. S. and Gottesfeld, S. In Proton Con-
et al.
ducting Membrane Fuel Cells I, The Electrochemical Society Proceedings Series, PV 95-23, eds S. Gottesfeld, G. Halpert and A. Landgrebe. The Electrochemical Society, Pennington, NJ, 1995, p. 252. 30. Ren, X., Zawodzinski, T. A., Uribe, F., Dai, H. and Gottesfeld, S. In Proton Conducting Membrane Fuel Cells I, The Electrochemical Society Proceedings Series, PV 95-23, eds S. Gottesfeld, G. Halpert and A. Landgrebe. The Electrochemical Society, Pennington, NJ, 1995, p. 284.
Pergamon PII: s0360-3199(97poo16-5
./. Hrdriroqm Enrr,q~, ( 1997 lnternatimal
Vol.
22, No. Association
11. pp. I I:7 11.44, 19Yhr Hydropen Energ) Elsevter Sctence Ltd
411rights reserved.Printed in Great Britain 0360~3199!97
PEM FUEL CELLS FOR TRANSPORTATION AND STATIONARY GENERATION APPLICATIONS S. J. C. CLEGHORN. X. REN, T. E SPRINGER. M. S. WILSON, C. ZAWODZINSKI, and S. GOTTESFELD
61 -.OOi
f).O!:
POWER
T. A. ZAWOD%INSKI
Electronic and Electrochemical Materials and Devices Group, MST-l I, MS D429, Los Alamos National Lahoralory Los Alamos, NM 87545, U.S.A.
describe recent activities at Los Alamos National Laboratory devoted to polymer electrolyte fuel ceils in the contexts of stationary power generation and transportation applications. A low cost/high performance hydrogen or reformateiair stack technology is being developed based on ultra-low platinum loadings and non-machined. inexpensive elements for flow-fields and bipolar plates. On-board methanol reforming is compared to the option of direct methanol fuel cells in light of recent significant IDower density increases demonstrated in the latter. (‘. 1997 International Association for fiydrogen Energ; Abstract---We
INTRODUCTION
technology for utility and stand-alonepower generation applications,as well asfor transportation, where It has The fuel cell is the most efficient meansof converting previously held the strongestattraction. hydrogen to electric power. As such, the fuel cell repEventhough theseandother recentdevelopmentshave resentsa key element in efforts to demonstrateand beenvery effective in raisingthe profile of PEM fuel cells implement hydrogen as a fuel for electric power asa viable technologyfor transportation, the commercial generation. implementationof hydrogen fueled PEM fuel cell stacks For the last 30 years,the benefitsprovided by fuel cell is likely to occur first for stationary power generation technologies,including high efficiency, low environ- applications, becauseof the somewhatless stringent mental emissionsand modular construction, have fos- requirementson overall cost. space,and the nature tered continued interest in their development for sourceof the hydrogen fuel. transportation and customer-side-of-the-meterutility Somerecent achievementsat Los Alamos that have applications.Among the family of fuel cell technologies, advancedour PEM fuel cell technology to the point of polymer electrolyte membrane(PEM) fuel cells have readinessfor the demonstrationof a low cost/high persomeuniquely attractive features. They provide high formancehydrogen fueledstack are summarizedbelow. power density, high chemical-to-electricalenergy con(MEAs) with versionefficiency, and fast and easystart-up. Addition- l Low costmembraneelectrodeassemblies ultra-low platinum loadingsof 0.2 g Pt/kW have been ally, they are very reliable and are constructed from demonstratedin 5000h life tests. generating area1 durable and benign materials. Until recently, however, power densitiesof 6 kW/m’ on pressurizedhydrothe large-scale,terrestrial application of PEM fuel cell gen/air [I -41. These MEAs have recently been technology for transportation and utility or stand-alone implementedsuccessfully in stacksby somedeveloper?. powergenerationwasconsideredto betoo expensiveand l Effective water managementprinciples have been too sensitive to the low levels ofcarbon monoxide present identified, basedto a large extent on advancedmemin reformed carbonaceousfuel feed streams.Thesebarbrane electrolytes evaluated jointly with membrane riers to the practical implementationof PEM fuel cells manufacturers[5, 61. have been significantly lowered in the last few years, in l Detailedevaluationsof losses in the PEM fuel cellhaye part as a result of work done at Los Alamos National beenbasedon long-terminvestigationof fundamental Laboratory. This paper describessomeof our efforts to lower thecost of PEM fuel cell stacksandto demonstrate processes in the cell, including all electrocataiytic and someeffective approachesfor the mitigation of anode transport phenomena[6J. catalyst deactivation
effects resulting
from low levels of
l
CO or other organic impurities. Theserecent technical developmentsincreasethe attractivenessof PEM fuel cell 1137
Based on the above evaluations,
an effective computer
codeand a setof experimentaldiagnosticsfor the PEM fuel cell have been established [7).
1138 l
S. J. C. CLEGHORN
The toleranceto carbonmonoxidelevelsexpectedfrom steamreforming of carbonaceousfuels has beensignificantly improved. In previouswork we have shown that air bleedinginto the anodecorrectsfor CO levels in the anode feed streamas high as 50-100 ppm [8, 91. However, we have recently identified alternative methodsto increasethe CO level tolerance to well above 100ppm.
Theseadvancementsin PEM fuel cell technology at Los Alamos have contributed significantly to the understandingand developmentof high performance,low cost PEM fuel cell stacks. Activities basedon thesetechnologiesare taking place at severalstack development sitesin North America. Most recently,the Fuel Cell CoreTechnologyProgram at Los Alamos addressedPEM fuel cell issuesrelated to both transportation and stationary power generation applications. In the transportation arena, providing hydrogen/air and reformate/air PEM fuel cell stacksat low cost and coupling the PEM fuel cell stack to reformersof carbonaceousfuels remainimportant challengesto overcome.Fuels being consideredrange from methanol to petroleum mid-distillates. In a transportation systembasedon the reforming of liquid hydrocarbons,the challengeof operatingPEM fuel cell stacks on hydrogen that is diluted by CO* (and nitrogen), in addition to containing significantlevelsof CO, is similar to that for stationary powerapplicationswherethe stack is fed by reformednatural gas.The situation in the latter caseis significantlylesssevere,however, becausestationary power requireslessstringentcost and power density targets. This paper describessome of our researchefforts devoted to low cost/high performance PEM fuel cells for stationary power generation.It alsobriefly mentions somerecentachievements with direct methanolPEM fuel cellsat Los Alamos. STATIONARY Background
POWER APPLICATIONS [l&12]
et al.
over PAFC technology in terms of energy conversion efficiency (higher voltage at samecurrent density) and power density. In the long term and in light of recent advancements,the life-cycle cost of PEFC technology could be more favorable. A stable,inert polymer electrolyte and an easy start-up/shut-down characteristic may make the reliability and longevity of the PEFC superior. The PEFC could compete for the near-term marketsin which the PAFC is expectedto be successful. As such, the following background discussionsare extracted from studiesoriented towards the PAFC [1316]. Many applicationsin the U.S. utility sector require power capacitiesin the range of 1 to IOMW. This is particularly true for peak demandin someof the small public utility systemsthat are requiredto pay high charges for peak demand, usually to a private utility grid. Thesemodestutility applicationscould be more suitable for the PAFC and PEFC. Indications are that at US$l lOO/kW, fuel cell technologiescould be attractive for peak power requirements.A conservativeestimateof this market in the U.S. is 14000to 17000MW for fuel cells. The efficiency, compact design,low maintenance and potentially lower cost of the PEFC could allow it to capture a substantialportion of this market. Ideally, the PEFC systemwould be in a cogenerationunit, with the wasteheat utilized for its thermal value. Increasingthe PEFC operatingtemperaturereducesthe effect of CO on the anodecatalyst and would allow the wasteheat to be usedmore effectively. Someincreasein operating temperatureis possibleevenin presentday PEFCs. It is becoming more difficult to obtain regulatory approval for large-scalepower stations or to secure rights-of-way for high tensionpower lines.Thus, in the relatively near future, it is reasonableto expect that a substantialamount of additional power capacity may be from neighborhoodor apartmentbuildingpowerstations that tap into the massivenatural gasdistribution grid in the U.S. Efficient, low emission,and quiet fuel cellsare ideal for this type of application. Though gas turbine systemsby comparisonarelessexpensive,it may bemore cost effective to pay more for an environmentallybenign fuel cell systemand bypassthe regulatory processes and the associatedcosts.Industrial and commercialfacilities will probably take the lead in implementingfuel cell systems.According to the Gas ResearchInstitute (GRI), approximately 18000MW of the latter power market in the U.S. shouldbe economicalfor fuel cell cogeneration systems,provided the installedcost is USS1OOOjkW[ 151. The industrial sectoris similarto the commercialsector in sizeand in requiredfuel cell characteristics.Approximately 10% of this potential market is for small (less than 1MW) cogenerationsystems.Waste heat is at a premiumin theseapplications,thus the PEFC may need to operateasa cogenerationsystemto be competitive.
While large utility applications are envisioned as belongingto the domain of molten carbonatefuel cells (MCFCs) and solid oxide fuel cells(SOFCs), there are many applicationssuchasutility peakpower generation, demand-side-management, and dispersedpower generation for which the phosphoricacid fuel cell (PAFC) and the polymer electrolyte fuel cell (PEFC) could be moreattractive. Natural gasfueledPAFC technologyfor such applications has becomecommercialized,with a 200kW PAFC unit (from ONSI) costing approximately US$2500/kW. The unit operates at 40% electrical efficiency and 85% overall thermal efficiency (including heated water). Ballard, originally in conjunction with Dow, is developinga 250kW natural gas fueled PEFC Low cost stack components power plant after the successfuldemonstration of a With a target of only US%lOOO/kW, a PEFC stationary 10kW plant [ 121. The PEFC appearsto havesomeimportant advantages power plant, with significant power conditioning and
PEM
FUEL
CELLS
FOR
TRANSPORTATION
AND
STATIONARY
natural gas reforming costs, will require relatively low cost fuel cell stacks. At Los Alamos, we are experimenting with the scale-up and automated fabrication of the low platinum loading catalyzed membranes based on “thinfilm” catalyst layers [l+]. A recently developed scheme for the automated preparation of lab-scale (areas up to 200cm’) membrane electrode assemblies is depicted in Fig. 1. The fabrication process of applying thin-film catalysts directly to the membrane is computer-controlled and provides high repeatability [ 171. At present, some of our work concentrates on the development of low cost gas diffusion backing, flow-field and bipolar plate materials. We have investigated metalbased hardware components as part of our PEFC stack development for stationary power applications, where the weight density of the stack is not as critical as for transportation applications. One promising stack contiguration is based on the use of stainless steel screens as flow-fields. Researchers elsewhere [18] have reported the use of stainless steel flow-fields with no apparent effects of corrosion after 1000 h of testing. Though the stainless steel element is not in direct contact with the membrane electrolyte, the surface of the metal is exposed to the corresponding humidified gas streams. Therefore, some form of surface treatment may eventually be required to achieve extended long-term stability. lnitial short-term tests of PEFCs based on ultra-low Pt catalyst loadings [l-4] and stainless steel screen flowfields show performance equal to or exceeding that obtained under similar conditions with machined metal flow-fields [17]. The ability to eliminate machining costs of flow-fields by using off-the-shelf metal screens would be advantageous to a stack manufacturer attempting to lower the cost of PEFC stacks. Fig. 2 depicts a polarization curve for a 100 cm’ active area PEFC, employing it Los Alamos MEA with 0.14mg Pt/cm’/electrode and stainless steel screen flow-fields. The screens seem to be effective flow-fields under ordinary PEFC conditions of
mixed gas/liquid gas distribution surface areas.
fabrication
using a computer
GENERATION
APP1.1(‘ATIOK\I
1 I39
(water) feed streams, enabling cffectlve across relatively large electrode
Stable PEFC anode operation on a feed stream oi natural gas reformate is a challenging requirement for utilization of PEFCs for stationary power generation. Some of the problems associated with the operation ofPEFCs on realistic fuel feed streams have been addressed recently by operation of PEFCs on methanol reformate. The roughly 1% CO in a methanol reformate can be decreased to about 10 ppm using shift reactors I’ollowed by a preferential oxidation reactor (PROX) utilizing a Pt/alumina catalyst [19]. The sensitivity of the PEFC to such residual levels of CO (10 ppm) is still very significant at the normal operation temperature of 80 ‘C. We have demonstrated the ability to remediate effects of residual CO in the fuel feed stream by bleeding small levels of 0, or air into the anode compartment of the fuel cell 18. 91. There have been recenl indications that others have successfully implemented this approach in PEFC‘ stacks. We have also shown [20] that a Pt--Ru anode catalyst does not catalyze the reverse shift process that generates CO from CO? within the fuel cell anode, but the high level of CO2 in the reformate can still create dilution effects. A combination of (i) shift reactors to lower the roughly 10% CO level of the methane reformate down to about l%, (ii) a PROX reactor upstream of the fuel cell to lower the CO to the I-~10 ppm level. and (iii) limited air bleeding into the anode to eliminate the residual CO effect, could conceivably provide fuel cell performance very nearly equal to that of a neat hydrogen cell. ‘The general system is depicted in Fig. .?a. An overall loss oi fuel caused by the oxidative treatments rn quch 3 cvstcm should not exceed 4%. Several recent efforts with Pt- Ru:(‘ c:~talvq:l\ [?(I I!?]
A: x-position B: y-position
Fig. 1. MEA
POWER
C: pen drop D: pump on
controlled process for applying a layer of catalyst blank positioned on a heated vacuum table.
ink to either
a membrane
tx a dcca I
1140
S. J. C. CLEGHORN
0.2
I-
0.0
l-Y 0.0
f
I
I
I
I
I
0.2
0.4
0.6
0.8
1.0
1.2
CURRENT Fig. 2. Internally
manifolded
et al.
DENSITY
(A/cm*)
100 cm2 cell with 0.14 mg Pt/cm2 thin-film electrodes on Nafion TM 112 membrane screen flow-fields. Hz/air = 3/3 atm, 50% air utilization at 1 A/cm’, T,,, = 80°C.
have indicated that some of the problems caused by residualCO within the PEFC anodecan be solvedby the use of appropriate Pt-Ru anode catalysts that appear to exhibit sufficient CO electro-oxidative activity. The identification of a CO tolerant, or rather CO consuming, anodecatalyst would naturally have a significantimpact on the designand efficiency of the system.
a) FWl +
Purifving naturalgas
--)
SHUT REACTOR(S)
steel
reformate with a hydrogen separator
As discussed above, a major issueimpactingthe coupling of the PEFC to a natural gas(methane)reformer is the effect of contaminants,particularly CO, on the fuel cell anode.A possibleapproachto circumvent this difficulty isto utilize someform of gasseparator,which would
Air Inject I
REFORMER
and stainless
Air Bleed I PROX REACTOR
PEFC
BURNER I Anode
Off-gas
\Comb;stio” Air
Combustion Air
Fig. 3. Reformer/fuel
processors for PEFCs: (a) a more conventional system with shift and PROX points, and (b) an alternative concept incorporating a hydrogen permselective
reactors and air injection separator.
at two
PEM
FUEL
CELLS
FOR
TRANSPORTATION
AND
STATIONARY
supply pure hydrogen to the fuel cell. In this section, we describe a recent effort at Los Alamos to develop an effective hydrogen separator to provide a pure hydrogen feed stream from simulated reformates to a polymer electrolyte fuel cell. This approach, depicted in Fig. 3b, not only eliminates the anode contamination difficulties, but decouples to some extent the operation of the fuel processor and the fuel cell. The hydrogen feed to the fuel cell could be effectively dead-ended, as is done in a number of fuel cell systems that use a neat hydrogen feed. This simplifies control of the complete system and allows stockpiling of hydrogen during low power demand. Thus, the fuel cell stack could then be brought up instantaneously with a “cold” reformer. Commercial hydrogen purifiers traditionally use Pd!‘Ag films, either free-standing or supported, which have significant cost and integrity drawbacks. Several groups are studying ways to increase the stability and lower the Pd content of hydrogen separator membranes. A fellow group at Los Alamos is developing Pd clad Ta films [23]. Tantalum supplies phase and structural stability. while providing high hydrogen permeability. The Pd is applied to either face of the Ta film to facilitate the interfacial process of hydrogen dissociation. At this point, the total Pd loading is only about 1 mg Pd/cm’ of separator membrane. We have operated a Scm’ single cell PEFC on the permeate of a 4cm2 Pd/Ta membrane separator operating at 3 15°C with a simulated methanol reformate feed (I % CO, 24% CO,, 74% H,). Fig. 4 indicates that about 600 mA/cm’ is obtained at 0.6 V. In contrast, the presence of only 100 ppm of CO (in H2) fed directly to the anode suppressed the current density at this voltage to about 60 mA/cm*, one-tenth the cell performance on the permeate. It is anticipated that the separator will be equally efTective with the much higher CO
POWER
GENERATION
.4PPLK’4TIOT\:S
! IJI
levels expected from a natural gas reformer. On-going development objectives for this membrane separator include improving its throughput and further lowering the Pd loadings. Currently, the amount of Pd required is comparable to adding 1 mg Pd/cm2 fuel cell membrane area. Preliminary process simulations for the reformer: separator/fuel cell scheme depicted in Fig. 3b are quite promising, particularly for a methanol fueled system. If a permselective separator is effectively paired with a methanol reformer, the need for high and low temperature shift and partial oxidation reactors would be eliminated because the roughly 1% CO in the reformate product is useful as burner fuel for the steam reformer, The natural gas fueled system is more problematic when using a separator because of the lower hydrogen concentration in the reformate, however, a number 01 approaches to increase the separator throuphput are possible.
DIRECT
METHANOL
PEM
FUEL
(‘EL.LS
Until very recently, the PEFC systems considered for transportation applications have been based on either hydrogen carried on board the vehicle. or on the steam reforming of methanol to generate a mixture of hydrogen and CO?, as the fuel feed stream for the fuel cell stack. A primary reason for resorting to methanol reforming onboard the vehicle has been the much more limited per.. formance of the direct methanol fuel cell (DMFC). The demonstrated power density for the latter has been ter) low indeed, and until recently its application for transportation was not a serious option at all. However. recent advancements in DMFC K&T) have
Pseudo-Reformate
(1% Membrane
CO) Separator
WI
1
i
1
0.0
’ 0.0
sp”I’*Flp
,
,
I
I
I
0.20
0.40
0.60
0.80
1.0
1.2
CURRENT DENSITY (A/cm2) Fig. 4 Polarization curves depicting a single cell PEFC operating at 80°C on(i) pseudo-reformate (I % CO) fed via a high tht oughput hydrogen permselective separator, (ii) 20 ppm CO in hydrogen, and (iii) 100 ppm CO in hydrogex
S. J. C. CLEGHORN et al.
1142
been quite dramatic, achieving power densities which are significant fractions of that provided by the reformate/air fuel cell (RAFC) [24-281. The more applied DMFC R&D work, aimed primarily at the demonstration of enhanced stable performance of DMFC single cells or stacks, has been supported in the U.S. by the Department of Energy (Office of Transportation Technology), the Advanced Research Projects Agency (ARPA) and the U.S. Army. Similar efforts in Europe resulted in a demonstration at Siemens of a high performance polymer electrolyte DMFC operating at temperatures approaching 130°C ~241.
prisinga methanolreformer and PEM reformate/air fuel cell stack. Basedon that evaluation, it seemsthat the DMFC could provide comparablepower densitiesand energy conversionefficiencies,thusbecominga seriouscandidate for transportation applications,provided the following requirementsare alsomet:
As in the previouscaseof the hydrogen/air PEFC or the RAFC, the strong recent advancesin DMFC performance have been achieved without any dramatic breakthroughsin electrocatalysis.The useof established Pt-Ru anodeand Pt cathodeelectrocatalystsin polymer electrolyte DMFCs operated at temperaturesof ll(r 130°Cprovide very significantenhancements in DMFC performancewhen catalyst layer compositionand structure areoptimized. Operation at elevatedtemperaturesis facilitated in such polymer electrolyte fuel cells by the anodebeing continuously in contact with liquid methanol-water mixtures. Recent experimentalwork at Los Alamos National Laboratory resultedin record highperformancesof polymer electrolyteDMFCs [28], demonstratedby Figs 5 and 6, which showpolarization curvesfor a methanol/air cell and power density curvesfor DMFCs with oxygen and with air cathodes,achievedin cellswith NafionTMmembranesoperatedat elevatedtemperatures.Theserecent resultshave been subsequentlyused in a comparative evaluation [29] for transportation applications of (i) a polymer electrolyte DMFC stack, and (ii) a systemcom-
0.8
1. catalystloadingsfurther reduced(or alternativeanode catalystsdeveloped); 2. long-termstableperformance(1000h time scale)demonstrated; 3. strategiesto minimizemethanolcross-overto the cathode[30] increasethe DMFC fuel efficiencyto the 90% level. CONCLUSION We have investigatedthe use of ultra-low Pt loading MEAs (0.14mg Pt/cm2)and stainlesssteelscreenflowfields in lOOcm* single cells, as part of our efforts to develop low cost/high performance hydrogen or reformate/air PEFC stacksfor stationary power applicationssuchas utility peak power generation,demandside-management and dispersedpower generation. In general, theselow cost component cells provide performances(approximately 0.6A/cm2 at 0.7V) approaching thoseobtainedwith commercialcells,which typically utilize much higher Pt loadings(4mg/cm2) and more expensiveflow-field materials (i.e. machined graphite plates).On-going researchincludesfurther development of inexpensiveflow-field, bipolar plate and gasdiffusion backingmaterials.Additional researchat Los Alamos is focusedon reducing the Pd content of hydrogen mem-
u N117 Pt-RuOx, 22 1 M methanol,
mg cm* 2.0 mL min,
12 psig
0.8 Air, 30 psi,
0.6 L m in
0.4
CURRENT
DENSITY
(A cmm2)
Fig. 5. Polarization curves for 1 lO”C, air cathode DMFC based on thin-film catalyzed NafionTM 112, 115 and 117 membranes. 2.3 mg/cm* Pt-black, 3 atm air at 0.6 l/min. Anodes: 2.2 mg/cm* Pt-RuOx, 1 M methanol at 2 ml/min and 1.8 atm. Cathodes:
PEM
FUEL
CELLS
FOR
Fig. 6. Power
density
curves
TRANSPORTATION
AND
STATIONARY
CURRENT for a thin-film
catalyzed
Nafion””
brane separators for RAFC applications and improving overall DMFC performance, for which greatly increased power densities have recently been obtained. Ac.knoIvlealqrmenrs-Nathaniel Peachey and Robert Dye of MST-7 at Los Alamos provided the membrane separator and assisted in coupling it with the fuel cell. Funding was provided by two program offices within the U.S. Department of Energy, the Office of Transportation Technology and the Hydrogen Program in the Office of Utility Technologies.
REFERENCES 1. Wilson, M. S. and Gottesfeld. S.. Journal of’ Applied Electrochf~mistry, 1992,22. 17. 2. Wilson, M. S. and Gottesfeld, S., Journal of’ the Electrochemical Sock/y, 1992, 139, L28. 30. 3 Wilson, M. S., U.S. Patent Nos. 5.211,984 and 5,234,777. 1993. 4 Wilson, M. S., Valerio. J. and Gottesfeld. S., Electrochimica i4<.tu, 1995. 40, 355.-363. T. A. and Gottesfeld, S., Jour5 Springer, T. E., ZawodLinski. nul qf‘the Electrochemicul Society, 1991, 138, 2334-2442. 6. Zawodzinski? T. A., Springer, T. E., Uribe, F. A. and Gottesfeld. S.. Solid State lonics, 1993, 60, 199-21 I. 7. Springer. T. E., Wilson. M. S. and Gottesfeld, S., Journal of the Electrochemical Society, 1993. 140, 3513. 8. Gottesfeld. S. and Pafford, J., Journal of the Electrochemical Society, 1988. 135, 2651L2652. 9. Gottesfeld, S., U.S. Patent No. 4.910,099, March 1990. 10. Prater, K. B., Journal of’ Power Sources, 1994, 51, 129. 11. Landgrebe, A. and Gottesfeld, S. In The Electrochemical Society Extended Abstracts. Vol. 94-2. Abstract No. 604, p. 944. Miami Beach, FL. October 1994. 12. Watkins, D. S. and Peters, B. C. In Program and Absfracts, Fuel Cell Seminar, San Diego, CA, 28 November -I December 1994. p. 250. 13. Impact of fundamental technical improvement. GRI Report No. GRI-8910137, International Fuel Cells Corporation, 1989
POWER
GENERATION
APPLI(‘ATiOhi
! 13:
DENSITY (Ahn*) I 12 assembly
operating
at 130 (’ on oxygen
and 1 IO C’ ,,n ‘lit
14. The financial and strategic benefits of fuel cell power piants. Vols I and II. EPRI Report No. EPRI EM-4511. ‘Temple. Parker and Solane. Inc., 1986. 15. Trimble, K. and Woods, R.. .lourrta/of’Poil’l~r .%ur~ es. 19’90, 29, 37-45. 16. Serfass, J. A., Journal of Power Sources. 1990.29. I 19 130. 17. Zawodzinski. C., Wilson, M. S. and Gottesfeld, S. In Proton Conducting Membrane Fuel Cc& I. The Electrochemical Society Proceedings Series, PV 95-23, eds S. Gottcsfeld. G. Halpert and A. Landgrehe. The Electrochemical Sncietv. Pennington. NJ, 1995, p. 57. 18. Mallant, R. K. A. M.. Koene, F. G. H.. Verhoeve, C. W G. and Ruiter, A. In Proywm and Ah.vracr.\. Fuel Cell Seminar, San Diego. CA. 28 Novembers 1 December 1994. p. 503. 19. Vanderborgh, h. E., Guante, J., Dean. 12. E. and Sutton. R. D. In Program and .46stracts. Fuel Cr# Semirw. Long Beach, CA, October 1988. p. 52. C. R., Valerio. J. A. and Gottesfeld. 20. Wilson. M. S., Derouin, S. In Proceedings of’ the 28th Intersociety FZnrr,g~~ C’rvwrsion Engineering C’onfcrenc,c (IEC‘EC). Vol. 1. Atlanta. (;.A. August 1993, p. 1203 21. Iwase. M. and Kawatsu, S. In Proton C‘onductq Menhrrrrrt Fuel C&s /. The Electrochemical Society Proceedmgs Series. PV 95-23. eds S. Gottesfeld, G. Halpert and A. La!xdgrehc. The Electrochemical Society. Pennington, NJ, 1995, p. I?. 22. Schmidt. V. M., lanniello. R.. Oetjen. H.-F., Reger. f-1.. Stimming, U. and Trila. F. In Proton Conducting Mrmhrunv Fuel c’rl1.s I. The Electrochemical Society Proceedings Series. PV 95-23, eds S. Gottesfeld, G. Halpert and A. Landprehe. The Electrochemical Society: Pennington. NJ, 199.5. p. 1. 23. Peachey, N. M.. Snow, R. C. and Dye. R. (‘., .Ic~urr~cri 01 Membrane Science, 1996. Ill, 123~. 133. 24. Griine, H.. Luft, G.. Mund, K. and Waidhas. M. In /+ryrm~ und Abstrwts, Fuel Cell Semmar. San I>iegn. CA. 2X Novembe---l December 1994, p. 474. 25. Surampudi. S.. Narayanan. S. R., Vamos. E., Fr.rnk. If.. Halpert, G.. LaConti. A., Kosek, J.. Surya Prakash, G. K and Olah, G. A., Journal of Power Sourcc~s. 1994. 47. 37” 26. Kosek, J. A.. Cropley, C. C.. Wilson. G . LaConti. 4 H.. Narayan. S Vamo\. F. ‘Surampudi. S :~ntl t-rai:h tl II:
1144
S. J. C. CLEGHORN
Proceedings of the 28th Intersociety Energy Conversion Engineering Conference (ZECEC), Vol. 1, Atlanta, GA, 1993, p. 1209. 27. Savinell, R. F., Yeager, E., Tryk, D., Landau, U., Wainright, J. S., Weng, D., Lux, K., Litt, M. and Rogers, C., Journal of the Electrochemical Society, 1994, 141, L46. 28. Ren, X., Wilson, M. S. and Gottesfeld, S., Journal of the Electrochemical Society, 1996, 143, L12. 29. Ren, X., Wilson, M. S. and Gottesfeld, S. In Proton Con-
et al.
ducting Membrane Fuel Cells I, The Electrochemical Society Proceedings Series, PV 95-23, eds S. Gottesfeld, G. Halpert and A. Landgrebe. The Electrochemical Society, Pennington, NJ, 1995, p. 252. 30. Ren, X., Zawodzinski, T. A., Uribe, F., Dai, H. and Gottesfeld, S. In Proton Conducting Membrane Fuel Cells I, The Electrochemical Society Proceedings Series, PV 95-23, eds S. Gottesfeld, G. Halpert and A. Landgrebe. The Electrochemical Society, Pennington, NJ, 1995, p. 284.