Direct methanol fuel cells progress in cell performance and cathode

Direct methanol fuel cells:progress in cell performance and cathode

research

Sharon C.Thomas 1,*,Xiaoming Ren 2,Shimshon Gottesfeld 2,Piotr Zelenay

Los Alamos National Laboratory,Los Alamos,NM 87545,USA Recei v ed 1April 2002;recei v ed in re v ised form 10May 2002

Abstract

Research presented in this paper focuses on se v eral factors affecting the performance of direct-methanol fuel cells (DMFCs)under de v elopment for either transportation or portable power applications.In particular,we discuss long-term stability of the anode,and crosso v er of methanol.We report on recent accomplishments in cell performance,fuel utilization,o v erall con v ersion efficiency,and lowering Pt loading in DMFCs designed for the potential use in automoti v e transportation.We describe a method of e v aluation of cathode performance based on cathode polarization cur v es,generated from DMFC data.With this method,we ha v e studied the effect of v ariation of temperature on DMFC cathode potential.Our results show that Pt-rich DMFC cathodes,operating on ambient air at 608C,can deli v er high performance of more than 0.85V v ersus RHE at 0.100A cm (2.#2002Published by Elsevier Science Ltd.

Keywords:Methanol;Fuel cell;Cathode performance

1.Introduction

Although direct-methanol fuel cells (DMFC)con-tinue to trail hydrogen áair polymer electrolyte mem-brane (PEM)fuel cells in power density,they do not require fuel processing subsystem,thus presenting an attracti v e alternati v e to the reformate-fed de v ices,further emphasized by significant impro v ements in DMFC performance,demonstrated in recent years [1á10].O v erall performance of DMFCs depends on se v eral factors,of which the most important are:(i)electro-catalytic acti v ity of the anode,(ii)ionic conducti v ity and resistance to methanol crosso v er of the proton conduct-ing membrane and (iii)water management on the cathode side of the cell (a function of flow-field and backing designs).As described pre v iously [1,10],opti-mization of different components of DMFCs may lead to a considerable increase in power density and fuel utilization.While much attention has been de v oted to

e v aluation and optimization o

f anode catalysts and structure,characterization of existin

g membranes and searching for alternati v es [11á24],relati v ely less effort has been dedicated to the contribution of the cathode to performance and performance stability of the DMFCs [25á33].

Recent research efforts at Los Alamos National Laboratory (LANL)ha v e focused on the potential applications of DMFCs as an either primary or aux-iliary power source for fuel cell v ehicles and as a portable power source at the 1á100W le v el.Different system constraints, e.g.stack power,temperature of operation,and catalyst loading,require substantially different approaches in the de v elopment of both sys-tems.For example,DMFCs for automoti v e transporta-tion are likely to operate at a temperature around 1008C and at maximally reduced precious metal loading.On the other hand,the temperature of DMFC operation in portable de v ices is expected not to exceed 808C.Due to generally lower power require-ments,reduction in precious metal loading in such de v ices may not be as crucial to their commercial v iability as to fuel cells under de v elopment for trans-portation applications.

*Corresponding author 1

On lea v e from Ballard Power Systems,9000Glenlyon Parkway,Burnaby,Canada BC V5J 5J9.2

Present address:Mechanical Technology,Inc.,30S.Pearl,Albany,NY 12207,USA.

Electrochimica Acta 47(2002)3741á

3748

In the present paper,we demonstrate enhanced acti v ity and stability of the anode,reduction of cross-o v er rate and continuing impro v ement in the stability of the cell performance.We also describe some of the efforts undertaken at Los Alamos to e v aluate cathode performance in a DMFC operating at the presence of methanol crosso v er.The results show that,in spite of the frequently mentioned lack of‘methanol tolerance’, Pt cathodes in an optimized DMFC can exhibit high performance,which is not much lower than its metha-nol-free performance.

2.Experimental

Pretreatment of Nafion TM117and1210membranes in the H'form,preparation of membrane electrode assemblies(MEAs)for a DMFC using decal technique, and the single-cell fuel cell hardware ha v e been pre-v iously described[11,34],as has the cell testing system used[35].Se v eral unsupported Pt?Ru blacks of a nominal1:1Pt:Ru atomic ratio(Johnson Matthey) were used for anode preparation.Anode inks were made by dispersing appropriate amounts of the Pt?Ru catalyst in Millipore water and adding5%Nafion TM solution(1200equi v alent weight,Solution Technology Inc.).Cathode inks contained unsupported Pt black(?30m2g(1,Johnson Matthey),Millipore water and5% Nafion TM solution(1100equi v alent weight,Solution Technology Inc).The catalyst layer preparation in-v ol v ed application of the catalyst inks either to con-v entionally PTFE-treated wet-proofed carbon-cloth backing,or to the membrane,either directly or by a decal transfer process(with neither method offering clear performance ad v antage in an operating DMFC). The geometric acti v e area of all MEAs was5cm2. When operated in fuel cell mode,the anode was fed with either hydrogen or methanol solution and the cathode with air.The methanol solutions,between0.25 and2.0mol l(1in concentration,were pumped through the DMFC anode flow field at precisely controlled rates (0.5á3.0ml min(1)using a Shimadzu LC-10AS HPLC pump.A back-pressure of between1.0and2.0atm was imposed upon the anode outlet flow to ensure that the membrane would be in contact with a liquid solution of methanol whene v er the cell temperature was in excess of 908C.When operated in dri v en cell mode to measure methanol anode polarization,the cathode gas was humidified hydrogen.When operated in‘hydrogen pump’mode,humidified hydrogen gas was supplied to both anode and cathode.The back-pressures,flow rates and humidification temperatures of the cathode gases, air(fuel cell mode)or hydrogen(dri v en mode),v aried with the operating temperature of the cell.

The crosso v er of methanol as a function of cell current density was recorded at the same time as the corresponding cell polarization plot(VáJ cell).The determination was based upon the amount of carbon dioxide present in the cathode exhaust,as measured using a GMM12Carbon Dioxide IR Sensor(Vaisala Oy,Finland)3.The IR detector was pre-calibrated with a gaseous mixture of4%CO2and96%N2.The CO2 content in the cathode exhaust was then con v erted into the corresponding amount of methanol that had crossed through the membrane,expressed as current density of methanol oxidation on the cathode(i x).The crosso v er of methanol at open-circuit was also determined electro-chemically by measuring oxidation current of methanol at the cathode in the presence of humidified nitrogen [36].

Computer programs custom-written in Lab v iew(Na-tional Instruments)were used to control the experiments as well as generate cathode polarization and cathode loss cur v es.

3.Results and discussion

3.1.Fuel utilization,anode and o v erall cell performance 3.1.1.Anode research and fuel utilization

Since electrocatalytic acti v ity of the Pt?Ru anode catalyst and long-term stability of the anode reflect directly on the fuel cell performance,we in v ested substantial effort in optimization of these factors. Currently a v ailable Pt?Ru catalysts allow preparation of acti v e DMFC anodes that can be operated for prolonged periods of time without noticeable loss in performance.A life test of such an anode at808C is shown in Fig.1.The test indicates that state-of-the-art anodes are capable of stable operation for weeks.The spikes on the current density v ersus time plot in Fig.1 are due to either opening of the cell circuit for30á60 min or stopping the feed of methanol for15á30min.

3The authors are aware of an uncertainty associated with the use of the CO2method to determine methanol crosso v er in DMFCs with CO2-permeable polymer membranes,such as Nafion TM.Most common errors result from differences in carbon dioxide acti v ity on the anode and the cathode side of the cell,which in turn depends on both the cell current and crosso v er current.These errors are usually quite small in the range of low and intermediate cell current densities [1],and can be estimated using‘open-cell’electrochemical crosso v er data from a‘dri v en’cell experiment as a reference(cf.Section2).The CO2method appears less reliable at high current densities,when significant part of the detected CO2may originate from gas diffusion from the anode to the cathode,leading to an o v erestimation of the ‘cathode-generated’CO2and,consequently,to an o v erestimation of methanol crosso v er.Due to that,the tedious and time-consuming methanol balance experiment is a better method of crosso v er determination at high current(power)density than the CO2method [1].On the other hand,it is at high cell current densities when the impact of methanol crosso v er on fuel utilization and cathode performance is the least important.

S.C.Thomas et al./Electrochimica Acta47(2002)3741á3748 3742

Although not absolutely necessary for stability of anode performance,opening of the cell circuit and/or cutting off the flow of methanol both lead to a temporary increase in anode acti v ity.This may be useful in any DMFC applications that will in v ol v e intermittent op-eration of the fuel cell.

Fuel utilization refers to the amount of fuel consumed to generate electricity to the total amount of fuel consumed in a DMFC and can be expressed as a ratio of the cell current density to the sum of the cell current density plus the equi v alent crosso v er current density,J cell /(J cell 'J x-o v er ).Partial loss of fuel because of fuel/oxygen recombination at the cathode and the resultant,often significant [37á44],cathode potential loss are considered a ‘soft’point of the DMFC and a major stumbling block on the path to making DMFC power systems practical.Howe v er,recent data pro v e that this difficulty can largely be o v ercome when the DMFC is operated under optimized conditions [1,45,46].

One effecti v e method for the reduction of methanol crosso v er through the membrane is to operate the fuel cell using relati v ely ‘lean’anode feed in terms of methanol concentration in the feed stream and,to a lesser extent,flow rate of the methanol solution (Figs.2and 3),which lowers the methanol concentration gradient across the membrane and thus the crosso v er rate.The same effect can be reached by increasing the cell current,i.e.by increasing the rate of methanol consumption at the anode.As before,that results in the lowering of the crosso v er and in impro v ed fuel utiliza-tion.The amount of crosso v er at the point of maximum cell power can be half or e v en less than that obser v ed at the open circuit conditions (the least fa v orable from the point of v iew of fuel utilization in DMFCs).At high current densities,fuel utilization can reach v alues in excess of 90%(Fig.3).

A possible implication of the effect of methanol concentration in the feed stream on fuel utilization and,consequently,on o v erall energy con v ersion effi-

ciency (v oltage efficiency )fuel utilization)is to operate a DMFC at v aried methanol feed concentration.This concept has earlier been illustrated by modeling simula-tions [45].

3.1.2.DMFC operation at higher temperature

(conditions rele v ant to potential automoti v e applications)Operation at higher DMFC cell temperature is key for approaching the power densities of reformate/air systems.Based on DMFC performance already demon-strated,we ha v e projected that the system power density and o v erall con v ersion efficiency of both direct and indirect methanol fuel cell systems would be similar [46,47].Recent demonstrations using optimized anodes focused on a combination of high power density,high fuel utilization,performance stability and,most re-cently,limited o v erall catalyst loading.High power density and fuel utilization as well as stable fuel cell operation ha v e been achie v ed in single DMFCs operat-ing at 1008C,with 0.5á1.0mol l (1methanol feed streams and 2.0atm air at high flow rate.Fig.4shows results of a 2000-h life test of a single cell

DMFC,

Fig.1.Stability test of a DMFC anode with unsupported Pt ?Ru catalyst at 0.35V (DHE)and 808C;c MeOH 01.0M,f MeOH 01.0ml min (1

.

Fig.2.Crosso v er of methanol through a Na?on TM 117membrane in a DMFC at open circuit;f 02.0ml min (1.Anode exhaust back-pressurized to 1.0atm at temperatures abo v e 908

C.

Fig.3.Fuel utilization as a function of cell current density at 1008C;Na?on TM 117membrane;anode exhaust back-pressurized to 1.0atm;air cathode at 2.0atm back-pressure and high ?ow stoichiometry.

S.C.Thomas et al./Electrochimica Acta 47(2002)3741á37483743

demonstrating a combination of peak power density near0.2W cm(2and fuel utilization o v er90%,both maintained continuously.One important feature estab-lished here is excellent stability of the anode at1008C (in addition to the anode stability already demonstrated at808C,Fig.1).Anode polarization data taken at the beginning and at the end of life test in Fig.4show v ery small anode acti v ity loss at0.35V,the potential close to the estimated potential of the anode during the life test.

The obser v ed drop in the rate of methanol oxidation at this potential o v er the2000-h life test was about12%, from0.285to0.250A cm(2at0.35V(DHE).This change in the anode performance agrees well with a12% change in the o v erall performance of the cell during the life test,thereby indicating that most performance loss incurred by the cell is due to a slow drop in the anode acti v ity.

Cell v oltage and power density v ersus current density plots obtained with total Pt loading lowered to2.6mg cm(2are gi v en in Fig.5.They demonstrate that e v en with relati v ely low concentration of methanol in the feed stream(0.5mol l(1)it is possible to achie v e with this reduced o v erall loading,at or near1008C,peak DMFC power density of at least0.15W cm(2and o v erall energy con v ersion efficiency of36%.As recently demonstrated in single-cell and short-stack testing,v ery similar maximum power densities,but at a penalty of somewhat reduced o v erall con v ersion efficiency,can be obtained with total Pt loading as low as1mg cm(2[48]. The keys to these achie v ements ha v e been the optimiza-tion of anode catalyst formulation and preparation, impro v ements in membrane/electrode assembly(MEA) and cell structure as well as operation conditions that ensure the highest ratio of cell current to crosso v er current.Progress in fuel utilization has been achie v ed by lowering concentration of methanol in the feed stream and,to a lesser degree,by controlling flow rate of the solution.

Our recent short-stack testing,performed using MEAs with total Pt loading of0.53mg cm(2and1.0 mol l(1methanol solution,re v ealed that merely5mg of platinum are required to generate1W of electricity in a direct methanol cell.Considering a pitch per cell of2 mm,as successfully demonstrated in our80W DMFC stack[49],the v olume power density projected for a DMFC stack operating at1008C can be expected to be between1and2kW l(1(as per stack acti v e v olume).

3.2.DMFC cathode

3.2.1.Cathode potential v ersus DMFC current density cur v es

To e v aluate a DMFC air cathode performance,it would ob v iously be desirable to know the potential of the cathode under DMFC operation conditions.Since the use of a reference electrode to obtain the cathode potential in an operating direct methanol fuel cell often poses problems[50],we attempted here to e v aluate the DMFC cathode potential from ordinary cell polariza-tion https://www.wendangku.net/doc/1f3518169.html,ing a two-electrode configuration within the operating cell,current density v ersus poten-tial cur v es for a DMFC air cathode were deri v ed from three iR-corrected v oltage/current density V*(J)mea-surements for(i)a methanol/air cell,(ii)a dri v en cell with the same methanol anode and a hydrogen-e v ol v ing cathode(hydrogen e v olution reaction,HER),and(iii)a dri v en‘hydrogen pump’cell with a

hydrogen-oxidizing

anode (hydrogen oxidation reaction,HOR)and a hydrogen-e v ol v ing cathode.

The iR -corrected potential differences measured in these three cells at gi v en J ,can be written as:

V MeOH =Air 1(J )0E MeOH

Air

(J )(E MeOH (J )(1a)V MeOH =H 21(J )0E MeOH (J )(E HER (J )(1b)V H 2=H 21(J )0E HOR (J )(E HER (J )

(1c)

here,the asterisk designates ‘iR -corrected’and the potential differences are written to yield positi v e v oltage v alues in each case.The E ’s are electrode potentials (v s.some arbitrary fixed reference potential)at current density J for a methanol anode at the rele v ant DMFC anode conditions,E MeOH (J ),a DMFC air cathode at rele v ant DMFC cathode conditions,E Air

MeOH (J ),and well humidified hydrogen-oxidizing electrode,E HOR (J ),and hydrogen-e v ol v ing electrode,E HER (J ),in contact with the ionomeric membrane.The superscript (MeOH)in the symbol for the DMFC air cathode potential,E Air

MeOH (J ),indicates that this cathode suffers the effects of metha-nol penetration.Each v alue of E is,ob v iously,a function of the concentration of reactants,pressure of gases and acti v ity of protons and water,but all of those are maintained identical in the three cell measurements (Eqs.(1a),(1b)and (1c)).Summing Eqs.(1a)and (1b),and subtracting Eq.(1c),one obtains:

E MeOH Air (J )(E HOR (J )

0V MeOH =Air 1(J )'V MeOH =H 21(J )(V H 2=H 21(J )

(2)

Eq.(2)explains the deri v ation of DMFC cathode potential at gi v en J v ersus the potential of a hydrogen-oxidizing anode operating at the same J .Fig.6displays the results of the process for such a cathode operating at 608C.

Unlike an ideal reference electrode,the potential of the hydrogen-oxidizing electrode v aries somewhat with J .Howe v er,one can readily compare e v aluated

E Air

MeOH (J )v ersus E HOR (J )to measured air cathode potential v ersus E HOR (J )in a hydrogen/air fuel cell *the latter v alue is simply the iR -corrected v oltage of the hydrogen/air fuel cell:E Air (J )(E HOR (J )0V H 2=Air

1(3)

The cathode loss specifically caused by methanol penetration to the DMFC air cathode can thus be e v aluated directly from the difference between E Air (J )and E Air

MeOH (J ),as deri v ed from Eqs.(3)and (2),respecti v ely,and demonstrated in Fig.7.

As Fig.6re v eals,the potential of this,hea v ily Pt loaded air cathode,operating in a DMFC at 608C on ambient air,is rather high:0.85V v ersus the hydrogen-oxidizing-electrode potential at 0.100A cm (2.That this is v ery close to the full acti v ity of this air cathode is seen in Fig.7,which confirms that the presence of methanol has only marginal effect on the air cathode potential at 0.100A cm (2.Depending on temperature,the potential shift of the cathode due to crosso v er can be estimated at 10á25mV for a DMFC operating near 0.5V.3.2.2.Effect of temperature on cathode potential

An interesting obser v ation has been the dissimilar increases in anode and cell performance with tempera-ture.This is illustrated in Figs.8and 9,where changes in the V áI cell plots o v er the temperature range from 60to 1008C are compared with the corresponding changes in the anode current in the same temperature range.The figures clearly indicate that DMFC performance in-creases with temperature less than does the electrocata-lytic acti v ity of the anode.For the sake of comparison,we can examine the ratio of anode current at 0.35V (v s.hydrogen e v ol v ing electrode)to cell current at 0.5V at different temperatures.If iR-corrected v alues are con-sistently used,this ratio assumes the v alues of 0.6,0.8and 1.1at 60,80and 1008C,respecti v ely.The obser v ed gradual increase in the anode-to-cell current ratio with temperature appears too high to be fully accounted

for

by the concurrent rise in the methanol crosso v er through the Nafion TM117membrane.

Computer-generated cathode polarization cur v es, based on three iR-corrected cell polarization cur v es,as described by Eq.(1)and(2),were used to assess the effect of methanol crosso v er on cathode potential at different temperatures.As shown in Fig.8,the increase in DMFC performance is not nearly as significant when the temperature change is from80to1008C as it is when the cell temperature is increased from60to808C. Polarization of the anode against a hydrogen-e v ol v ing electrode,Eq.(1b),indicates that anode performance increases monotonously with increase in temperature,as shown in Fig.9.The cathode polarization cur v es(Fig.

10)show that the cathode potential increases(cathode acti v ity increases)with temperature between60and 808C but the potential of the cathode at1008C is,in fact,below that at808C.This de v iation from the beha v ior expected based on monotonous oxygen reduc-tion rate enhancement with temperature must be explained by more substantial effects of methanol crosso v er on cathode performance at the ele v ated temperature of1008C[36].Under the simplest circum-stances,methanol entering the cathode from the mem-brane would cause a shift of air cathode potential,gi v en by:

D E

Air

0b log[(J

cell

'J

x-over

)=J

cell

]

0b log[1'(J

x-over

=J

cell

)](4) where b is the Tafel slope for the oxygen reduction reaction(ORR),J cell is the cell current density at the rele v ant cathode potential E and J x-o v er is the methanol crosso v er current density.(Noteworthy,the sum of J cell and J x-o v er is equal to the o v erall rate of ORR in the cell, J ORR).Eq.(4)describes the shift of the DMFC air cathode potential under the conditions described in[51] as‘no apparent interaction’.As the temperature in-creases,J ORR,an interfacial process with acti v ation energy of approximately15kcal mol(1[52]should increase proportionately more than J x-o v er which is determined by methanol diffusi v ity through the iono-meric membrane,associated with an acti v ation energy of the order of only5á6kcal mol(1[36].Hence,the drop of cathode potential with temperature between80 and1008C must be the result of additional effects of methanol penetration on cathode potential,possibly excessi v e flooding.

4.Summary and conclusions

The outcome of the work presented in this paper can be summarized as follows:

i)rate of the crosso v er of methanol through com-

monly used polymer electrolyte membranes can be greatly reduced,by50%or more,by adjusting the concentration of methanol in the anode feed stream according to the power requirement from the

cell.

At higher cell current densities,fuel utilization of 80á90%,or e v en higher,can be achie v ed.

ii)As a result of further optimization of the anode and the cathode structures,substantial impro v ement in cell performance has been accomplished that has allowed the Pt loading to be significantly reduced without se v ere drop in the cell performance.Power density of0.2W cm(2can be reached in DMFC operating at ele v ated temperatures with total Pt loading at or below3.0mg cm(2.Relati v ely high DMFC air cathode performance loss under these conditions,as re v ealed in this work for DMFC cathodes of high Pt loading,remains a challenging target for further impro v ements.

iii)DMFC cathode polarization cur v es,generated from polarization data for three cells,enable e v aluation of air cathode performance in the presence of the methanol‘leaking’through the membrane.The results show clearly that,when the Pt air cathode in a DMFC is of sufficient loading and cell operation conditions are optimized,cathode loss in the DMFC caused by methanol is marginal,less than20mV at0.100A cm(2.It is thus important to realize that the net performance of a Pt cathode in a DMFC could be really high,abo v e0.85V v ersus RHE at0.100A cm(2,despite the presence of methanol permeating from the anode side of the cell.This is the result of the intrinsically superior acti v ity of Pt as an ORR catalyst and the possibility to minimize,under some conditions,the penalty of the methanol in the cathode to only mixed potential effects[51].

Acknowledgements

This work was supported by the Defense Ad v anced Research Projects Agency through the Defense Sciences Of?ce and by the US.Department of Energy through the Of?ce of Ad v anced Automoti v e Technology. References

[1]X.Ren,P.Zelenay,S.Thomas,J.Da v ey,S.Gottesfeld,J.Power

Sources86(2000)111.

[2]T.Schultz,S.Zhou,K.Sundmacher,Chem.Eng.Technol.24

(2001)12.

[3]L.Carrette,K.A.Friedrich,U.Stimming,Fuel Cells1(2001)5.

[4]H.Dohle,J.Mergel,H.Scharmann,H.Schmitz,in:S.R.

Narayanan,S.Gottesfeld,T.Zawodzinski(Eds.),Direct Metha-nol Fuel Cells,The Electrochemical Society,Inc,Pennington,NJ, Proceedings Vol.2001á4,2001,p.209.

[5]A.S.Arico`,P.Cret?`,V.Baglio,E.Modica,V.Antonucci,J.

Power Sources91(2000)202á209.

[6]H.Dohle,J.Di v isek,R.Jung,J.Power Sources86(2000)469.

[7]D.Buttin,M.Dupont,M.Straumann,R.Gille,J.-C.Dubois,R.

Ornelas,G.P.Fleba,E.Raumanni,V.Antonuscci,A.S.Arico`,P.

Cret?`,E.Modica,M.Pham-Thi,J.-P.Ganne,J.Appl.Electro-chem.31(2001)275.

[8]S.R.Narayanan,F.Clara,T.I.Valdez,in:S.R.Narayanan,S.

Gottesfeld,T.Zawodzinski(Eds.),Direct Methanol Fuel Cells, The Electrochemical Society,Inc.,Pennington,NJ,Proceedings Vol.2001á4,2001,p.254.

[9]M.Baldauf,W.Preidel,J.Appl.Electrochem.31(2001)781.

[10]S.Gottesfeld,Polymer electrolyte fuel cells,in:C.Tobias,H.

Gerischer,R Alkire,D.Kolb(Eds.),Ad v ances in Electrochemical Science and Engineering,vol.5,Wiley,New York,1997. [11]X.Ren,M.S.Wilson,S.Gottesfeld,J.Electrochem.Soc.143

(1996)L12.

[12]S.Wainwright,J.-T.Wang,D.Weng,R.F.Sa v inell,M.Litt,J.

Electrochem.Soc.142(1995)L121.

[13]H.Hoster,T.Iwasita,H.Baumga¨rtner,W.Vielstich,Phys.Chem.

Chem.Phys.3(2001)337.

[14]P.M.Urban,A.Funke,J.T.Mu¨ller,M.Himmen,A.Docter,

Appl.Catal.A221(2001)459.

[15]P.Waszczuk,J.Solla-Gullon,H.-S.Kim,Y.Y.Tong,V.Montiel,

A.Aldaz,A.Wieckowski,J.Catal.203(2001)1.

[16]H.A.Gasteiger,N.Marko v ie′,P.N.Ross,Jr.,E.J.Cairns,J.Phys.

Chem.97(1993)12020.

[17]K.Wang,H.A.Gasteiger,N.M.Marko v ie′,P.N.Ross,Jr.,

Electrochim.Acta41(1996)2587.

[18]P.Waszczuk, A.Wieckowski,P.Zelenay,S.Gottesfeld, C.

Countanceau,J.-M.Le′ger,https://www.wendangku.net/doc/1f3518169.html,my,J.Electroanal.Chem.511 (2001)55.

[19]X.Ren,T.E.Springer,S.Gottesfeld,J.Electrochem.Soc.147

(2000)92.

[20]X.Ren,S.Gottesfeld,J.Electrochem.Soc.148(2001)A87.

[21]E.Peled,T.Du v de v ani,A.Aharon,A.Melman,Electrochem.

Solid-State Lett.3(2000)525.

[22]L.J.Hobson,H.Ozu,M.Yamaguchi,S.Hayase,J.Electrochem.

Soc.148(2001)A1185.

[23]B.Libby,W.H.Smyrl,E.L.Cussler,Electrochem.Solid-State

Lett.4(2001)A197.

[24]N.Jia,M.C.Lefeb v re,J.Halfyard,Z.Qi,P.G.Pickup,Electro-

chem.Solid-State Lett.3(2000)529.

[25]S.C.Roy,P.A.Christensen, A.Hamnett,K.M.Thomas,V.

Trapp,J.Electrochem.Soc.143(1996)3073.

[26]S.Gupta, D.Tryk,S.K.Zece v ic,W.Aldred, D.Guo,R.F.

Sa v inell,J.Appl.Electrochem.28(1998)673.

[27]T.R.Ralph,M.P.Hogarth,Platinum Metals Re v.46(2002)3.

[28]A.S.Arico`,A.K.Shukla,H.Kim,S.Park,M.Min,V.Antonucci,

Appl.Surf.Sci.172(2001)33.

[29]M.Neergat,A.K.Shukla,K.S.Gandhi,J.Appl.Electrochem.31

(2001)373.

[30]A.K.Shukla,M.Neergat,P.Bera,V.Jayaram,M.S.Hegde,J.

Electroanal.Chem.504(2001)111.

[31]N.Alonso-Vante,P.Bogdanoff,H.Tributsch,J.Catal.190

(2000)240.

[32]J.E.Hutchison,T.A.Postlethwaithe,C.Chen,K.W.Hathcock,

R.S.Ingram,W.Ou,R.M.Linton,R.W.Murray,Langmuir13 (1997)2143.

[33]W.Vielstich,V.A.Paganin, F.H.B.Lima, E.A.Ticianelli,J.

Electrochem.Soc.148(2001)A502.

[34]X.Ren,W.Henderson,S.Gottesfeld,J.Electrochem.Soc.144

(1997)L267.

[35]T.E.Springer,M.S.Wilson,S.Gottesfeld,J.Electrochem.Soc.

140(1993)3513.

[36]X.Ren,T.A.Zawodzinski,F.Uribe,H.Dai,S.Gottesfeld,in:S.

Gottesfeld,G.Halpert,https://www.wendangku.net/doc/1f3518169.html,ndgrebe(Eds.),Proton Conduct-ing Membrane Fuel Cells,The Electrochemical Society Proceed-ings Series,Pennington,NJ,1995,p.284.

[37]S.R.Narayanan,H.Frank,B.Jeffries-Nakamura,M.Smart,W.

Chun,G.Halpert,in:S.Gottesfeld,G.Halpert,https://www.wendangku.net/doc/1f3518169.html,ndgrebe (Eds.),Proton Conducting Membrane Fuel Cells,The Electro-

S.C.Thomas et al./Electrochimica Acta47(2002)3741á37483747

chemical Society Proceedings Series,Pennington,NJ,1995,p.

278.

[38]M.K.Ra v ikumar,A.K.Shukla,J.Electrochem.Soc.143(1996)

2601.

[39]A.Ku¨v er,W.Vielstich,J.Power Sources74(1998)211.

[40]J.Cruickshank,K.Scott,J.Power Sources70(1998)40.

[41]A.Heinzel,V.M.Barraga′n,J.Power Sources84(1999)70.

[42]H.Uchida,Y.Mizuno,M.Watanabe,Chem.Lett.11(2000)

1268.

[43]S.Hikita,K.Yamane,Y.Nakajima,JSAE Re v.22(2001)151.

[44]V.M.Barraga′n,A.Heinzel,J.Power Sources104(2002)62.

[45]R.M.Moore,S.Gottesfeld,P.Zelenay,Optimizing the o v erall

con v ersion ef?ciency of a direct-methanol fuel cell,in:S.

Gottesfeld,T.F.Fuller,G.Halpert(Eds.),Proton Conducting Membrane Fuel Cells(Second International Symposium),Elec-trochemical Society,Pennington,NJ,Proceedings Vol.98-27, 1998.

[46]R.M.Moore,S.Gottesfeld,P.Zelenay,Comparison of the

con v ersion ef?ciency of a direct-methanol v s.a hydrogen fuel cell,in:S.Gottesfeld,T.F.Fuller,G.Halpert(Eds.),Proton

Conducting Membrane Fuel Cells(Second International Sympo-sium),Electrochemical Society,Pennington,NJ,Proceedings Vol.

98-27,1998.

[47]S.Gottesfeld,S.J.C.Cleghorn,X.Ren,T.E.Springer,M.S.

Wilson,T.A.Zawodzinski,Polymer Electrolyte Direct Methanol Fuel Cells;An Option for Transportation Applications,Fuel Cell Seminar Abstract Book,1996,p.521.

[48]P.Zelenay, F.Guyon,S.Gottesfeld,in:S.R.Narayanan,S.

Gottesfeld,T.Zawodzinski(Eds.),Direct Methanol Fuel Cells, The Electrochemical Society,Inc.,Pennington,NJ,Proceedings Vol.2001á4,2001,p.123.

[49](a)X.Ren,J.Da v ey, B.S.Pi v o v ar,H.Dinh, C.Rice,S.

Gottesfeld,P.Zelenay,in:Extended Abstracts,AIChE2002 Spring National Meeting,New Orleans,LA,March10á14,2002.;

(b)X.Ren,J.Da v ey,B.S.Pi v o v ar,H.Dinh,C.Rice,P.Zelenay,

S.Gottesfeld,J.Power Sources,submitted.

[50]K.Ku¨v er,I.Vogel,W.Vielstich,J.Power Sources52(1994)77.

[51]D.Chu,S.Gilman,J.Electrochem.Soc.141(1994)1770.

[52]B.R.Scharifker,P.Zelenay,J.O’M.Bockris,J.Electrochem.Soc.

134(1987)2714.

S.C.Thomas et al./Electrochimica Acta47(2002)3741á3748 3748