Epe

AE

a

ARRAA

KMDCFP

1

btoot[atHtpead

lot

0d

Electrochimica Acta 56 (2011) 6078– 6083

Contents lists available at ScienceDirect

Electrochimica Acta

j ourna l ho me pag e: www.elsev ier .com/ locate /e lec tac ta

lectrocatalytic oxidation of methanol to soluble products on polycrystallinelatinum: Application of convolution potential sweep voltammetry in thestimation of kinetic parameters

run Murthy, A. Manthiram ∗

lectrochemical Energy Laboratory & Materials Science and Engineering Program, University of Texas at Austin, Austin, TX 78712, United States

r t i c l e i n f o

rticle history:eceived 12 April 2011eceived in revised form 20 April 2011ccepted 21 April 2011vailable online 4 May 2011

eywords:

a b s t r a c t

Irreversible oxidation of methanol on polycrystalline platinum leading to soluble products has beencarried out by fast scan voltammetry, and the reaction has been studied under diffusion controlled pro-cess. The conventional analysis of current–potential data, viz. dependence of peak potential on scan rateand peak width measurements, resulted in the estimation of apparent diffusion coefficient of methanoland the anodic transfer coefficient of the electrode reaction. However, from the convolution potentialsweep voltammetry, a more accurate and reliable kinetic data were obtained. Under the above conditions,

ethanol oxidationissociative adsorptiononvolution analysisast scan voltammetrylatinum electrode

methanol oxidation follows Butler–Volmer rate law with a linear variation of logarithmic heterogeneousrate constant with electrode potential. A constant apparent anodic transfer coefficient independent ofelectrode potential was observed pointing to the fact that the standard potential of the reaction cannotbe determined from the voltammetric experiments. The experimental current–potential curve was com-pared with a theoretical voltammogram and further oxidation of products at the electrode surface hasalso been analyzed using limiting convolution current.

. Introduction

The electrocatalytic oxidation of methanol over platinum haseen studied well by various electrochemical and spectroscopicechniques [1–13]. Over the past decades, a majority of the studiesn electrocatalytic decomposition of methanol has been focusedn engineering new electrocatalysts on various support materialshat are poison-tolerant, yet cost-effective for practical applications14–24]. Much effort has also been devoted on the mechanisticspects as well as on the detection of various intermediates andheir dependence on potential and time coordinates [2,5,7,10].owever, only a few studies have focused on the estimation of

he kinetic parameters of the heterogeneous electron transferrocess of methanol oxidation on platinum [8,25], which knowl-dge is important for a complete understanding of the reactionnd improving the performance of the new electrocatalysts beingeveloped.

The underlying reason to this effect can be traced to the fol-

owing limitations: (i) the kinetics of electrocatalytic oxidationf methanol is complicated involving multi-step electrode reac-ions and parallel pathways and (ii) formation of partial oxidation

∗ Corresponding author.E-mail address: rmanth@mail.utexas.edu (A. Manthiram).

013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.oi:10.1016/j.electacta.2011.04.078

© 2011 Elsevier Ltd. All rights reserved.

intermediates like –CH3O, –CH2O, –CHO and unwanted parallelreactions yielding poisonous carbon monoxide (CO) [2,26]. Theabove limitations preclude the application of mathematical mod-els developed for simple heterogeneous electron transfer reactions[27]. However, it was shown by Xu et al. [8] that by apply-ing potential pulses (>0.52 vs. RHE) of short-time potentiostaticelectrolysis (<30 ms) and with larger waiting periods betweensuccessive pulses, the total electrooxidation of methanol on a poly-crystalline platinum can be simplified to the reaction,

CH3OH + H2O → HCOOH + 4H+ + 4e− (1)

That has a rate determining step of initial dissociative adsorptionof methanol from solution with the concomitant electron transfer,

CH3OHsol → CH2OHad + H+ + e− (2)

The characteristic feature of short-time methanol decomposi-tion on platinum surface is that no significant CO formation isobserved. In the absence of surface poisons, mathematical mod-els developed for simple electron transfer process at the electrodesurface can be applied to the above reaction (1) wherein the num-ber of electrons involved in the rate determining step is one while

four electrons are transferred overall.

The same simplification of methanol oxidation process (reac-tion (1)) can be realized by employing fast scan voltammetry[10]. Lu et al. [10] reported that at sufficiently fast potential scan

rochimica Acta 56 (2011) 6078– 6083 6079

(tawn(atosiTdsvii

I

moBaaattaacptkvv

2

mcpIoTr(gtetwtswpw

tstt

A. Murthy, A. Manthiram / Elect

35 V s−1), no surface CO is observed and the methanol oxida-ion occurs in the potential range between hydrogen desorptionnd platinum surface oxide potential regions [10]. In this paper,e apply convolution analysis of fast scan voltammetric tech-ique to simplified methanol oxidation on polycrystalline platinumreaction (1)) with the rate determining step of initial dissociativedsorption of methanol with concomitant charge transfer (reac-ion (2)). Convolution potential sweep voltammetry (CPSV) is onef the most powerful electrochemical methods for analyzing theubtle details of heterogeneous electron transfer process and isndependent of a particular electrochemical method employed.he convolution method of analyzing cyclic voltammograms waseveloped independently by Oldham [28–30], who termed it asemi-integral analysis, and Saveant and co-workers [31–35] as con-olution potential sweep voltammetry. The convolution current Is related to the actual current i and time t through the convolutionntegral [29]

= 1√�

t∫

0

i(u)

(t − u)1/2du (3)

In this study, the heterogeneous electron transfer kinetics ofethanol oxidation under fast scan condition is analyzed with-

ut defining a priori the electron transfer rate law (for exampleutler–Volmer), and the same is obtained as a result of convolutionnalysis. Furthermore, unlike conventional voltammetric analysis,ll the experimental current–potential data are employed in CPSVnd hence more accurate kinetic parameters are obtained. Fromhe combination of fast scan voltammetry and convolution analysis,rue diffusion controlled reaction was observed at high scan ratesnd the apparent heterogeneous rate constant (kH) was obtaineds a function of electrode potential. The apparent anodic transferoefficient was also estimated from the derivative of ln kH vs.otential (E) plot and is in agreement with the one obtained fromhe conventional analysis of the voltammetric wave. The resultinginetic parameters are substituted in the current equation for irre-ersible electrode reactions and compared with the experimentaloltammetric wave.

. Experimental

Voltammetric experiments were carried out with a CH Instru-ents electrochemical workstation in a three-electrode single

ompartment cell. The working electrode was a polycrystallinelatinum disc of 2 mm diameter (active area = 0.0343 cm2; CH

nstruments) and was polished successively with finer gradesf alumina and sonicated for 3 min with high purity water.he electrochemical pretreatment rendered the electrodes highlyeproducible. The electrode was scanned between 20 and 1200 mVvs. RHE) for 20 times at such a scan rate (20 mV s−1) where aood reversible but not exactly mirror images of hydrogen adsorp-ion/desorption peaks appeared [36]. A mercury/mercury sulphatelectrode standardized with hydrogen electrode (pH 0.3) served ashe reference electrode and potentials were referred subsequentlyith respect to RHE. A platinum wire was used as a counter elec-

rode. ACS certified methanol (Fischer Scientific), ultra high pureulphuric acid (Fischer Scientific), and Millipore water (18 M� cm)ere used in the experiments. Solutions were purged with highurity nitrogen for 10 min before the experiments. All experimentsere conducted at ambient temperature.

The properties of the interfacial region affect both the magni-

ude and shape of the current–potential curves especially at highcan rates. For example, uncompensated solution resistance tendso shift peak potentials of oxidation waves in a positive direc-ion while the charging currents have deleterious effects both on

Fig. 1. Cyclic voltammograms recorded on polycrystalline platinum disc electrodesin (a) methanol (1 M) + H2SO4 (0.5 M) and (b) H2SO4 (0.5 M). Scan rate = 35 V s−1 andtemperature = 298 K.

the quality and magnitude of the voltammograms [37,38]. Thepositive feed back function of the potentiostat was employed tomeasure and compensate the iR drop without getting the circuitinto oscillatory modes. Furthermore, without assuming that thecharging current to be a linear function of potential, experimentalcharging current was estimated from the voltammogram measuredunder identical conditions but methanol. The current data obtainedfrom the potentiostat were subsequently transferred to Matlabworkspace for convolution analysis. The algorithm (Eq. (4)) usedto evaluate the semi-integral from the experimental data was fromLawson and Maloy [39] and was applied to the current–potentialdata free of background current,

I = 1√�

j=k∑j=1

i(j�t)�t√k�t − j�t + 0.5�t

(4)

3. Results and discussion

There are several literature studies conforming to the fact thatat short-time potentiostatic electrolysis of methanol above 0.5 V onsmooth polycrystalline platinum in sulphuric acid only the solubleproducts other than CO and CO2 are formed [8–10]. A maximumproduction of formaldehyde was observed near 0.5 V in perchloricacid [40]. The faster potentiodynamic electrolysis is advantageoussince all the thermodynamic/kinetic information of the interestedreaction is obtained in a single scan of short time. Formate wasfound to be an active and main product in the non-CO reactionpathway of methanol oxidation in a study based on in situ surfaceenhanced IR absorption spectroscopy (SEIRAS), and the other reac-tion intermediates proposed for this reaction pathway were notdetected [41].

Fig. 1 shows fast scan voltammograms at smooth polycrystallineplatinum disc electrode in 0.5 M sulphuric acid with and with-out methanol (1 M). No significant modifications of voltammogramin the hydrogen desorption/adsorption and surface oxide regionswere observed due to the presence of methanol. While in the dou-ble layer region, an irreversible oxidation wave exists in the case ofthe voltammogram with methanol. Increase in current after 1 V inthe positive going scan in the case of background voltammogramcould be because of oxidation of organic impurity. Similar resultswere obtained by Lu et al. [10] with fast scan voltammetry whereinat higher scan rates methanol oxidation does not involve CO for-mation pathway and soluble products proceeds actively. Moreover,methanol oxidation peak current depends on the isotopic composi-

tion indicating that a C–H bond cleavage (reaction (2)) is involved inthe rate determining step [10,42]. Voltammograms were measuredat various scan rates as shown in Fig. 2. According to the currentequation for irreversible diffusion controlled electrode reaction

6080 A. Murthy, A. Manthiram / Electrochimica Acta 56 (2011) 6078– 6083

F05

[r

i

wdda

ttiltimtalTtritDotAmf

Fl

ig. 2. Voltammograms of methanol (1 M) oxidation on polycrystalline platinum in.5 M H2SO4. Scan rates from top to bottom are 40, 38, 36, 34, 32, 30, 25, 20, 15, 10,, 3 and 2 V s−1 and temperature = 298 K.

43] (Eq. (5)), the peak current (ip) varies linearly with the squareoot of the scan rate �,

P = 2.985 × 105nAC√

DMeOHˇ� (5)

here denotes anodic transfer coefficient and DMeOH refers toiffusion coefficient of methanol. The other symbols n, A, and Cenote, respectively, number of electrons, area of the electrode,nd concentration of methanol.

Fig. 3 shows the variation of ip as a function of√

�, andhe plot is linear in the scan rate range of 30–40 V s−1 (correla-ion coefficient = 0.986), indicating the electrochemical oxidations under diffusion controlled in this range of scan rate [43]. Atower scan rate region, the products of the methanol dissocia-ive adsorption further oxidize to other intermediates and hencenitial methanol dissociative adsorption may not be a rate deter-

ining step. The intermediates occupy the active sites and blockhe methanol adsorption and hence lower peak current valuesre observed at lower scan rates. Methanol cannot occupy, ateast, till 0.2 V since the sites are occupied by hydrogen atoms.he voltammetric region thus falls between 0.2 and 0.9 V. Theime period (=potential range/scan rate) corresponding to a scanate of 30 V s−1 ((0.9 − 0.2)/30 = 23 ms) is 23 ms and for 40 V s−1

s ((0.9 − 0.2)/40 = 17.5 ms) 17.5 ms. From the slope of the plot inhe linear region, the apparent diffusion coefficient of methanol,MeOH, was obtained as 8.22 × 10−12 cm2 s−1. A similar lower valuef DMeOH was observed by Xu et al. [8], and they attributed this

o the different microscopic environments of methanol molecules.lternatively, the intervening adsorption of methanol after theass transport from the bulk solution and before the charge trans-

er may be invoked to interpret the lower DMeOH.

ig. 3. Variation of peak current (iP) with square root of scan rate (√

�). The solidine indicates a linear variation in the high scan rate region between 30 and 40 V s−1.

Fig. 4. Variation of anodic peak potential, EP, with logarithmic scan rate, log (�), inthe scan rate range 30–40 V s−1.

The transfer coefficient of the electrode reaction representsa derivative of activation (�G*) and standard (�G◦) free energiesof the electrode reaction. The voltammetric method of estimating

traditionally involves [44] (i) peak width measurement (Eq. (6))and (ii) scan rate dependence of peak potential (Ep) (Eq. (7)):

= 1.857RT

F(EP − EP/2)(6)

= 2.3RT

2F

∂log(�)∂EP

(7)

where R, T, and F denote, respectively, gas constant, temperature,and Faraday constant. Eqs. (6) and (7) were originally derived byNicholson and Shain for a totally irreversible electrode reaction[45]. The peak width (the potential difference between the poten-tial value at peak current and the potential value at half the peakcurrent) in the case of the voltammogram at 36 V s−1 is 150 mV,which is significantly higher than 60 mV, so a lower value (0.317) of

ensues. Furthermore, the plot Ep vs. log (�) is linear (Fig. 4) hav-ing a slope of 89 mV/decade, which is larger than 30 mV/decade,indicating a high degree of irreversibility of methanol oxidation atshort times (reaction (1)) and the corresponding is 0.332 (Eq. (7)).The value lower than 0.5 is indicative of a large activation barriersince the rate determining step involves a bond cleavage (reaction(2)).

There are two main advantages of convolution analysis over theconventional analysis of the voltammetric data. First, in contrastto conventional voltammetric analysis, the convolution approachdoes not require a prior assumption of the electron transfer kineticlaw (for example, Butler–Volmer or Marcus–Hush) but is deducedfrom the analysis [29]. The second merit is that all the data pointsof the voltammetric wave are employed, thus rendering the result-ing kinetic parameters more accurate. The convolution currents ofthe background subtracted voltammograms, expressed in A s1/2 orAmplomb [29], in the scan rate range 30–40 V s−1 is shown in Fig. 5.It may be noted that convolution analysis was successfully appliedto the voltammograms of scan rates in the range of kilovolts persecond by Saveant and Tessier [34]. The convolution voltammo-grams are sigmoid in shape and a plateau is reached when theelectrode process is diffusion controlled. Under this condition, theconvolution current reaches its limiting value IL (Eq. (8)), which isindependent of scan rate.

IL = nFA√

DMethanolCMethanol (8)

Eq. (8) offers a more precise way of determining the diffu-sion coefficient of methanol from IL and was calculated as1.38 × 10−11 cm2 s−1, which is close to that obtained from the ipvs. �1/2 plot (Fig. 3). From the convolution analysis, it is possible to

A. Murthy, A. Manthiram / Electrochimica Acta 56 (2011) 6078– 6083 6081

Fso

oasorti

l

bleo

ˇ

kSdvdltl

F(T

ig. 5. Dependence of convolution current (I) on electrode potential (E) at variouscan rates between 30 and 40 V s−1. The curves collected at various scan rates almostverlap with each other.

btain all the kinetic information contained in the voltammograms opposed to only the peak characteristics of conventional analy-is, and hence the data are highly reliable. Quantitative informationn the potential dependence of the heterogeneous charge transferate constant kH is obtained by logarithmic analysis of the convolu-ion current in conjunction with the voltammetric current for therreversible electrode reaction (Eq. (9)) [46–48].

nkH = ln√

DMeOH − lnIL − I(t)

i(t)(9)

Fig. 6 shows the variation of ln kH vs. E at various scan ratesetween 30 and 40 V s−1; the plots overlap one another and are

inear which is typical of Butler–Volmer kinetics. A more accuratestimation of transfer coefficient can be obtained from the slopef ln kH vs. E plot since is related to ln kH as [47,48]

= RT

F

∂ lnkH

∂E(10)

As the ln kH vs. E plot is linear, is a constant (Butler–Volmerinetics) and using Eq. (10), its value was calculated as 0.312.ince the methanol oxidation at short time is irreversible the stan-ard potential E◦ of the reaction cannot be determined from theoltammetric experiments [47]. The linear ln kH vs. E plot is best

escribed by Butler–Volmer kinetics despite the expected non-

inear activation-driving force relationship since during the chargeransfer a change in solvent reorganization energy was postu-ated for methanol oxidation [8] which can be described rather

ig. 6. Dependence of logarithmic heterogeneous electron transfer rate constantln kH) upon electrode potential (E) at various scan rates between 30 and 40 V s−1.he curves collected at various scan rates overlap with each other.

Fig. 7. Comparison between experimental and theoretical voltammograms. Circlesdenote experimental voltammogram at 36 V s−1 while the solid line was obtainedfrom Eq. (11).

by Marcus model [49,50] in conjunction with Saveant model [27].Marcus–Hush–Saveant model could have been invoked involvingMarcus quadratic equation if the ln kH vs. E plot (Fig. 6) was foundnon-linear (quadratic) [51]. In that case, transfer coefficient wouldhave been a linear function of potential and the standard potentialof the reaction can be calculated using the method described byAntonello and Maran [47]. Despite the above limitation Lu and co-workers [8] hypothesized E◦ as 0.3 V vs. RHE. The heterogeneouselectron transfer rate constant kH around this potential, i.e. thestandard rate constant k0

H, was calculated as 3.1 × 10−8 cm s−1 andthe value points to a typical irreversible reaction with very slowkinetics.

A comparison between experimental and theoretical voltam-mograms was carried out (Fig. 7). Voltammetric current forirreversible electrode reaction is given by [43]

i = nFAC√

�bDMeOH�(bt) (11)

where b = ˇF�/RT and√

��(bt) is the current function for irre-versible reaction. An analytical expression for

√��(bt) is given by

the equation [52]

√��(bt) = a1� + a2�2

b1 + b2� + b3�2(12)

wherein the coefficients a1, a2, b1, b2 and b3 are respectively,1.7807, 0.3361, 1.0000, 2.0492 and 1.2705 and � = exp(− Fx/RT). Thepotential scale x is referenced with respect to peak potential Ep, viz.

x = ˇ(E − EP) (13)

The circles in Fig. 7 show the current values obtained experi-mentally at 36 V s−1 and the solid line denotes Eq. (11) substitutedwith parameters obtained from the voltammetric analysis andother experimental variables. There is a very good agreementbetween the two-voltammetric waves, indicating a true irre-versible electrocatalytic oxidation of methanol at short times(reaction (1)) with diffusion control.

Methanol oxidation on a polycrystalline platinum leads to aserial pathway mechanism through the formation of adsorbed CO

CH3OH → CO(ad) + 4e− + 4H+ (14)

CO(ad) + H2O → CO2 + 2e− + 2H+ (15)

There also exists a parallel (direct) pathway without involvingCO(ad), viz. through the formation of CO2, or other stable oxidation

products such as formic acid and formaldehyde [2]. Parallel path-way is a unique feature of short-time potentiostatic electrolysisor high scan rate voltammetric oxidation of methanol [8,10,53].Voltammetric oxidation of methanol decreases with decreasing

6 rochim

si

fivndasclip

C

C

C

ef(am

cbtot

H

H

tuprsp

I

w

I

nbalcb

vfwaatbtfa

[[

[[[

082 A. Murthy, A. Manthiram / Elect

can rate (Fig. 2), on the other hand, CO(ad) oxidation chargencreases with decreasing scan rates [53].

The apparent number of electrons transferred obtained bytting the experimental curve with a theoretical equation for irre-ersible electron transfer process is four (Fig. 7). From the apparentumber of electron transferred the products of the methanol oxi-ation can be inferred. It can be noted that the product is not andsorbed species since the peak currents were proportional to thequare root of the scan rate [54] and also the convolution limitingurrent reaches a plateau [55]. A kinetic delay for the CO adsorptioneads to negligible CO adsorption at short times [53]. The follow-ng reactions show possible products of methanol oxidation in aarallel pathway.

H3OH → HCHO + 2e− + 2H+ (16)

H3OH + H2O → HCOOH + 4e− + 4H+ (17)

H3OH + H2O → CO2 + 6e− + 6H+ (18)

The apparent four-electron transfer corresponds to three differ-nt product combinations, viz. (a) only formic acid formation, (b)ormation of formaldehyde and carbon dioxide at equal rates andc) formation of formic acid, formaldehyde and carbon dioxide allt equal rates. In all the above pathways dissociative adsorption ofethanol is the rate determining step.Furthermore, products such as formaldehyde and/or formic acid

an accumulate in the diffusion layer that can either diffuse to theulk solution or get oxidized at the electrode surface. The true pic-ure may be in between these two extreme cases. The oxidationf products at the electrode surface is indeed potential dependenthat takes place through a direct pathway

COOH → CO2 + 2e− + 2H+ (19)

CHO + H2O → CO2 + 4e− + 4H+ (20)

The oxidation of products at the electrode would contribute tohe voltammetric current the proportion of which can be analyzedsing convolution voltammogram at the plateau region where theroduct formation is at maximum. The limiting convolution cur-ent IL is directly proportional to concentration of the electroactivepecies. Eq. (8) can be modified by including the contribution ofroduct oxidation at electrode surface.

L = nFA√

DMethanolCMethanol + IL(products) (21)

here

L(products) = n1FA√

DFormic acidCFormic acid

+ n2FA√

DFormaldehydeCFormaldehyde (22)

1 and n2 in the above equation are 2 and 4, respectively. Eq. (8) cane evaluated using diffusion coefficient value obtained from Fig. 3nd other known experimental values as 3.8 × 10−5 A s−1/2 which isower than the experimental IL (4.4 × 10−5 A s−1/2). The differencean be attributed to the oxidation of products at the electrode giveny Eqs. (21) and (22).

As discussed above, two extreme cases can be considered,iz. (i) all the accumulated oxidation products diffuse awayrom the electrode surface, in which case, the experimental ILould be lower than 4.4 × 10−5 A s−1/2 (viz., 3.8 × 10−5 A s−1/2)

nd (ii) all the accumulated oxidation products get oxidizedt the electrode surface without escaping in to the bulk solu-ion, in which case, the number of electrons transferred would

e 6 and IL would be 1.5 (since n = 6 in Eq. (8)) times higherhan 3.8 × 10−5 A s−1/2, which is 5.7 × 10−5 A s−1/2. There-ore, the second term IL(products) in Eq. (21) can be evaluateds 5.7 × 10−5 A s−1/2 − 3.8 × 10−5 A s−1/2 = 1.9 × 10−5 A s−1/2.

[

[[

ica Acta 56 (2011) 6078– 6083

This quantity amounts to 100% efficiency for CO2 for-mation wherein the number of electrons transferredbecomes 6. However, the actual IL(products) is 4.4 × 10−5

A s−1/2 − 3.8 × 10−5 A s−1/2 = 6.0 × 10−6 A s−1/2. Therefore, theactual efficiency for CO2 formation in terms of convolution currentis at least [(6.0 × 10−6 A s−1/2)/(1.9 × 10−5 A s−1/2)] × 100 = 31.6%. Aquantitative evaluation of methanol oxidation products employingdifferential electrochemical mass spectrometry (DEMS) on smoothpolycrystalline platinum [11,56] and carbon-supported platinumnanoparticles at low catalyst loading [57] shows only 10–20%current efficiency for CO2 formation. The higher CO2 efficiencyobtained by convolution analysis here can be attributed to the highscan rates employed. The partial oxidation products formed at theelectrode surface have a higher chance to get oxidized further atthe electrode at a scan rate of 30–40 V s−1 in comparison to a scanrate of 10 mV s−1 [11,56,57]. In other words, at a lower scan rate,partial oxidation products have more chance to diffuse in to thebulk solution than at a higher scan rate, so a lower CO2 currentefficiency ensues in the former case.

4. Conclusions

A more accurate analysis of methanol oxidation to soluble prod-ucts on polycrystalline platinum has been carried out with fast scanvoltammetry in conjunction with convolution analysis. In the rangeof scan rates studied only at the higher end (30–40 V s−1), diffusioncontrolled voltammograms were observed. From the conventionalanalysis of the voltammogram, viz. using only the peak charac-teristics, apparent diffusion coefficient of methanol and anodictransfer coefficient were calculated. In contrast, by utilizing allthe current–potential data, a more reliable and accurate kineticparameters were obtained through convolution analysis. Withoutassuming a priori the kinetic law the analysis indicated that thereaction follows Butler–Volmer kinetics. The plot of logarithmicheterogeneous rate constant vs. electrode potential was linear atthe explored scan rates resulting in an anodic transfer coefficientindependent of electrode potential. Since the reaction was gov-erned by Butler–Volmer rate law the standard potential of thereaction could not be determined. Further oxidation of productshas been considered and its contribution to the over all current hasbeen estimated.

Acknowledgement

Financial support by the Office of Naval Research MURI grant no.N00014-07-1-0758 is gratefully acknowledged.

References

[1] J. Greeley, M. Mavrikakis, J. Am. Chem. Soc. 124 (2002) 7193.[2] E. Herrero, W. Chrzanowski, A. Wieckowski, J. Phys. Chem. 99 (1995) 10423.[3] H. Okamoto, W. Kon, Y. Mukouyama, J. Phys. Chem. B 109 (2005) 15659.[4] T. Iwasita, Electrochim. Acta 47 (2002) 3663.[5] D. Cao, G.-Q. Lu, A. Wieckowski, S.A. Wasileski, M. Neurock, J. Phys. Chem. B

109 (2005) 11622.[6] T.H.M. Housmans, M.T.M. Koper, J. Phys. Chem. B 107 (2003) 8557.[7] K. Franaszczuk, E. Herrero, P. Zelenay, A. Wieckowski, J. Wang, R.I. Masel, J.

Phys. Chem. 96 (1992) 8509.[8] W. Xu, T. Lu, C. Liu, W. Xing, J. Phys. Chem. B 109 (2005) 7872.[9] S. Sriramulu, T.D. Jarvi, E.M. Stuve, Electrochim. Acta 44 (1998) 1127.10] G.-Q. Lu, W. Chrzanowski, A. Wieckowski, J. Phys. Chem. B 104 (2000) 5566.11] H. Wang, C. Wingender, H. Baltruschat, M. Lopez, M.T. Reetz, J. Electroanal.

Chem. 509 (2001) 163.12] T.D. Jarvi, S. Sriramulu, E.M. Stuve, J. Phys. Chem. B 101 (1997) 3649.13] A. Cuesta, M. Escudero, B. Lanova, H. Baltruschat, Langmuir 25 (2009) 6500.14] B.J. Hwang, L.S. Sarma, J.M. Chen, C.H. Chen, S.C. Shin, Q.R. Wang, D.G. Liu, J.F.

Lee, M.T. Tang, J. Am. Chem. Soc. 127 (2005) 11140.15] B.J. Hwang, L.S. Sarma, G.R. Wang, C.H. Chen, D.G. Liu, H.S. Sheu,.J.F. Lee, Chem.

Eur. J. 13 (2007) 6255.16] Y. Ando, K. Sasaki, R. Adzic, Electrochem. Commun. 11 (2009) 1135.17] S.M. Golabi, A. Nozad, Electroanalysis 15 (2003) 278.

rochim

[[[[

[[[[[[[[[[[[[[[[[[[

[[

[

[[[

[[

[[[[[[

T.N. Veziroglu (Eds.), New Materials for Fuel Cell Systems, Editions de l’Ecole

A. Murthy, A. Manthiram / Elect

18] Z.-H. Wang, L.-L. Zhang, K.-Y. Qiu, J. Power Sources 161 (2006) 133.19] S. Liao, K.-A. Holmes, H. Tsaprailis, V.I. Birss, J. Am. Chem. Soc. 128 (2006) 3504.20] J. Kua, W.A. Goddard III, J. Am. Chem. Soc. 121 (1999) 10928.21] I. Gonzaılez-Gonzalez, D.A. Tryk, C.R. Cabrera, Diamond Relat. Mater. 15. (2006)

275.22] S. Wang, S.P. Jiang, T.J. White, J. Guo, X. Wang, J. Phys. Chem. C 113 (2009) 18935.23] S.T. Kuk, A. Wieckowski, J. Power Sources 141 (2005) 1.24] C. Zhou, H. Wang, F. Peng, J. Liang, H. Yu, J. Yang, Langmuir 25 (2009) 7711.25] S.L. Gojkovic, T.R. Vidakovic, D.R. Durovic, Electrochim. Acta 48 (2003) 3607.26] R. Parsons, T. VanderNoot, J. Electroanal. Chem. 257 (1988) 9.27] J.-M. Saveant, Acc. Chem. Res. 26 (1993) 455.28] K.B. Oldham, Anal. Chem. 44 (1972) 196.29] M. Grenness, K.B. Oldham, Anal. Chem. 44 (1972) 1121.30] K.B. Oldham, Anal. Chem. 41 (1969) 1904.31] J.C. Imbeaux, J.-M. Savéant, J. Electroanal. Chem. 44 (1973) 169.32] L. Nadjo, J.-M. Savéant, D. Tessier, J. Electroanal. Chem. 52 (1974) 403.33] J.-M. Savéant, D. Tessier, J. Electroanal. Chem. 65 (1975) 57.34] J.-M. Savéant, D. Tessier, J. Electroanal. Chem. 77 (1977) 225.35] F. Ammar, J.-M. Savéant, J. Electroanal. Chem. 47 (1973) 215.

36] B.E. Conway, H. A-Kozlowska, Acc. Chem. Res. 14 (1981) 49.37] J.C. Imbeaux, J.-M. Savéant, J. Electroanal. Chem. 28 (1970) 325.38] J.C. Imbeaux, J.-M. Savéant, J. Electroanal. Chem. 31 (1971) 183.39] R.J. Lawson, J.T. Maloy, Anal. Chem. 46 (1974) 559.40] C. Korzeniewski, C.L. Childers, J. Phys. Chem. B 102 (1998) 489.

[[[

ica Acta 56 (2011) 6078– 6083 6083

41] Y.X. Chen, A. Miki, S. Ye, H. Sakai, M. Osawa, J. Am. Chem. Soc. 125 (2003) 3680.42] S.T. Oyama, R. Radhakrishnan, M. Seman, J.N. Kondo, K. Domen, K. Asakura, J.

Phys. Chem. B 107 (2003) 1845.43] A.J. Bard, L.R. Faulkner, Electrochemical Method: Fundamentals and Applica-

tions, John Wiley & Sons, Inc., New York, 2000.44] R.J. Klingler, J.K. Kochi, J. Phys. Chem. 85 (1981) 1731.45] R.S. Nicholson, I. Shain, Anal. Chem. 36 (1964) 706.46] S. Antonello, F. Formaggio, A. Moretto, C. Toniolo, F. Maran, J. Am. Chem. Soc.

123 (2001) 9577.47] S. Antonello, F. Maran, J. Am. Chem. Soc. 121 (1999) 9668.48] S. Antonello, M. Musumeci, D.D.M. Wayner, F. Maran, J. Am. Chem. Soc. 119

(1997) 9541.49] R.A. Marcus, J. Phys. Chem. 67 (1963) 853.50] R.A. Marcus, Electrochim. Acta 13 (1968) 995.51] C. Costentin, M. Robert, J.-M. Saveant, J. Am. Chem. Soc. 126 (2004) 16051.52] M.A. Prasad, M.V. Sangaranarayanan, Electrochim. Acta 49 (2004) 2569.53] F. Seland, D.A. Harrington, R. Tunold, Electrochim. Acta 52 (2006) 773.54] A. Wieckowski, W. Chrzanowksi, E. Herrero, in: O. Savadogo, P.R. Roberge,

Polytechnique de Montrıeal, Montreal, 1995, p. 326.55] R. Bowling, R.L. McCreery, Anal. Chem. 60 (1988) 605.56] H. Wang, T. Loffler, H. Baltruschat, J. Appl. Electrochem. 31 (2001) 759.57] Z. Jusys, J. Kaiser, R.J. Behm, Langmuir 19 (2003) 6759.