methanol and formic acid electrooxidation on

Physical Chemistry Chemical Physics

This paper is published as part of a PCCP Themed Issue on: Electrocatalysis: theory and experiment at the interface Guest Editor: Andrea Russell Editorial Electrocatalysis: theory and experiment at the interface Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b808799g

Papers Structure of the water/platinum interface––a first principles simulation under bias potential Minoru Otani, Ikutaro Hamada, Osamu Sugino, Yoshitada Morikawa, Yasuharu Okamoto and Tamio Ikeshoji, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b803541e

A first-principles study of molecular oxygen dissociation at an electrode surface: a comparison of potential variation and coadsorption effects Sally A. Wasileski and Michael J. Janik, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b803157f

Electrochemical and FTIRS characterisation of NO adlayers on cyanide-modified Pt(111) electrodes: the mechanism of nitric oxide electroreduction on Pt Angel Cuesta and María Escudero, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b717396b

Electrodeposited noble metal particles in polyelectrolyte multilayer matrix as electrocatalyst for oxygen reduction studied using SECM Yan Shen, Markus Träuble and Gunther Wittstock, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802688b

CoPt nanoparticles and their catalytic properties in electrooxidation of CO and CH3OH studied by in situ FTIRS Qing-Song Chen, Shi-Gang Sun, Zhi-You Zhou, Yan-Xin Chen and Shi-Bin Deng, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802047g

Kinetic studies of adsorbed CO electrochemical oxidation on Pt(335) at full and sub-saturation coverages Prachak Inkaew and Carol Korzeniewski, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b804507k

Formic acid electrooxidation on Pd in acidic solutions studied by surface-enhanced infrared absorption spectroscopy Hiroto Miyake, Tatsuhiro Okada, Gabor Samjeské and Masatoshi Osawa, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b805955a

Voltammetric surface dealloying of Pt bimetallic nanoparticles: an experimental and DFT computational analysis Peter Strasser, Shirlaine Koh and Jeff Greeley, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b803717e

Unique activity of Pd monomers: hydrogen evolution at AuPd(111) surface alloys Y. Pluntke, L. A. Kibler and D. M. Kolb, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802915f

Electrochemical study on the adsorption of carbon oxides and oxidation of their adsorption products on platinum group metals and alloys Hanna Siwek, Mariusz Lukaszewski and Andrzej Czerwinski, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b718286b

Shape-dependent electrocatalysis: methanol and formic acid electrooxidation on preferentially oriented Pt nanoparticles J. Solla-Gullón, F. J. Vidal-Iglesias, A. López-Cudero, E. Garnier, J. M. Feliu and A. Aldaz, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802703j

Ethanol electrooxidation onto stepped surfaces modified by Ru deposition: electrochemical and spectroscopic studies V. Del Colle, A. Berná, G. Tremiliosi-Filho, E. Herrero and J. M. Feliu, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802683a

The role of adsorbed hydroxyl species in the electrocatalytic carbon monoxide oxidation reaction on platinum Anthony R. Kucernak and Gregory J. Offer, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802816h

The effects of the specific adsorption of anion on the reactivity of the Ru(0001) surface towards CO adsorption and oxidation: in situ FTIRS studies J. M. Jin, W. F. Lin and P. A. Christensen, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802701c

An unexpected enhancement in methanol electrooxidation on an ensemble of Pt(111) nanofacets: a case of nanoscale single crystal ensemble electrocatalysis Ceren Susut, George B. Chapman, Gabor Samjeské, Masatoshi Osawa and YuYe Tong, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802708k

Experimental and numerical model study of the limiting current in a channel flow cell with a circular electrode J. Fuhrmann, H. Zhao, E. Holzbecher, H. Langmach, M. Chojak, R. Halseid, Z. Jusys and J. Behm, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802812p

Surface Pourbaix diagrams and oxygen reduction activity of Pt, Ag and Ni(111) surfaces studied by DFT Heine A. Hansen, Jan Rossmeisl and Jens K. Nørskov, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b803956a

Electrocatalytic mechanism and kinetics of SOMs oxidation on ordered PtPb and PtBi intermetallic compounds: DEMS and FTIRS study Hongsen Wang, Laif Alden, F. J. DiSalvo and Héctor D. Abruña, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b801473f

Combinatorial screening of PtTiMe ternary alloys for oxygen electroreduction Ting He and Eric Kreidler, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802818b

Using layer-by-layer assembly of polyaniline fibers in the fast preparation of high performance fuel cell nanostructured membrane electrodes Marc Michel, Frank Ettingshausen, Frieder Scheiba, André Wolz and Christina Roth, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802813n

Stripping voltammetry of carbon monoxide oxidation on stepped platinum single-crystal electrodes in alkaline solution Gonzalo García and Marc T. M. Koper, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b803503m

PtxRu1-x/Ru(0001) surface alloys—formation and atom distribution H. E. Hoster, A. Bergbreiter, P. M. Erne, T. Hager, H. Rauscher and R. J. Behm, Phys. Chem. Chem. Phys., 2008 DOI: 10.1039/b802169d

PAPER

www.rsc.org/pccp | Physical Chemistry Chemical Physics

Shape-dependent electrocatalysis: methanol and formic acid electrooxidation on preferentially oriented Pt nanoparticles J. Solla-Gulloń,a F. J. Vidal-Iglesias,a A. Lo´pez-Cudero,a E. Garnier,ab J. M. Feliu*a and A. Aldaza Received 18th February 2008, Accepted 9th April 2008 First published as an Advance Article on the web 14th May 2008 DOI: 10.1039/b802703j Reactivity towards methanol and formic acid electrooxidation on Pt nanoparticles with well characterised surfaces were studied and compared with the behaviour of single crystal electrodes with basal orientations. Polyoriented and preferential (100), (111) and (100)–(111) Pt nanoparticles were synthesised, cleaned preserving its surface structure, characterised and employed to evaluate the influence of the surface structure/shape of the Pt nanoparticles on these two relevant electrochemical reactions. The results pointed out that, in agreement with fundamental studies with Pt single crystal electrodes, the surface structure of the electrodes plays an important role on the reactivity of both oxidation processes, and thus the electrocatalytic properties strongly depend on the surface structure/shape of the nanoparticles, in particular on the presence of sites with (111) symmetry. These findings open the possibility of designing new and better electrocatalytic materials using decorated shape-controlled Pt nanoparticles as previously described with Pt single crystal electrodes.

1. Introduction The properties of nanoparticles have been a subject of wide research both in catalysis and electrocatalysis.1 In the latter case, these studies have been carried out with two main objectives: (1) the understanding of fundamental aspects of surface electrochemical reactivity and (2) the development of new materials for practical applications, in particular, fuel cells. One of the most interesting questions in nanoparticle electrochemistry is how the electrocatalytic activity is influenced by the surface structure/shape of the nanoparticles. In addition, if we keep in mind that a majority of electrocatalytic reactions are structure sensitive or site demanding,2–7 it seems of great importance to achieve an effective control of the crystalline surface structure of the nanoparticles for a wellfounded comparison of the electrocatalytic activity of different electrocatalysts. In relation to this aspect, we have previously described the influence of the surface structure/shape of the nanoparticles in reactions of relevance such as ammonia oxidation, oxygen reduction and CO electrooxidation.8–13 Moreover, we have also shown than bismuth and germanium adsorbed on platinum are excellent tools to ‘‘in situ’’ characterize the surface structure of platinum electrodes, both for well-defined surfaces and nanoparticles.14–18 This approach to the surface characterization of Pt nanoparticles complements the use of the ‘‘so-called’’ reversible hydrogen adsorption/ a

Departamento de Quı´mica-Fı´sica and Instituto de Electroquı´mica, Universidad de Alicante, Apartado 99, 03080 Alicante, Spain. E-mail: [email protected] b Laboratoire de Catalyse en Chimie Organique, Equipe Electrocatalyse, UMR CNRS No 6503, 40 Avenue du Recteur Pineau, 86022 Poitiers Cedex, France

This journal is

c

the Owner Societies 2008

desorption process, which requires the deconvolution of the voltammograms to obtain a direct estimation of the amount of the different sites present at the surface,18 once the fraction of (111) two-dimensional domains has been determined by using Bi. This pure surface electrochemical approach allows discarding the use of ex situ techniques such as TEM or HRTEM that, albeit being extremely precise as shape and structure characterisation tools, are restricted to a limited number of nanoparticles. Methanol and formic acid electrooxidation are both known to be structure-sensitive reactions on platinum surfaces and many papers have been published dealing with the different processes involved in these electrocatalytic reactions. In the case of formic acid, the mechanism of its electrooxidation is reasonably well-established since the 70’s, via the so-called ‘‘dual-path’’ mechanism suggested on polycrystalline platinum.19,20 In addition, it is also well-known that both paths, the active intermediate reaction path and the poisoning intermediate one, are structure-sensitive processes.21–25 In this reaction, Pt(100) is considered as the more active surface, its maximum intrinsic activity (0.25 M HCOOH solution) being around 30 mA cm 2 at 0.46 V (RHE), which means a turnover rate of more than 70 molecules-(100) site 1 s 1, whereas Pt(111) is the least poisoned surface but the activity towards formic acid oxidation is also low at similar potential values, about 6 molecules-(111) site 1 s 1.26,27 This reactivity is masked by the fact that the Pt(100) surface is easily poisoned while at Pt(111) the poisoning rate is significantly lower. As for formic acid, the mechanism of methanol electrooxidation on Pt follows a ‘‘dual-pathway’’, with the additional complication that CO may be considered as reactive intermediate.28,29 Methanol electrooxidation on the three basal Phys. Chem. Chem. Phys., 2008, 10, 3689–3698 | 3689

planes of platinum has been shown as a structure sensitive reaction,22,30–32 where Pt(111) was found to be the least reactive towards methanol decomposition, while Pt(110) was considered to be the most active. In this way, the surface poisoning is slower on Pt(111) and faster on Pt(110).32 Nevertheless, in a very recent paper, Housmans et al. have investigated the selectivity and structure sensitivity of the methanol oxidation pathways on basal planes and stepped platinum single crystal electrodes by monitoring the mass fractions of CO2 (m/z 44) and methylformate (m/z 60) using online electrochemical mass spectrometry (OLEMS).33 They found that methanol oxidation on Pt basal planes showed an increase in maximum activity in the order Pt(111) o Pt(110) o Pt(100). The order obtained is different to that previously reported by Herrero et al., (Pt(111) o Pt(100) o Pt(110))32 and it was attributed to differences in the electrode pre-treatment which influences the surface structure of the Pt(110) electrode. In addition, on the basis of the onset of the reaction (o0.6 V), the Pt(110) was reported to be the most active, followed by the Pt(111) and Pt(100) surfaces.33 Besides the studies with the three basal planes, formic acid and methanol electrooxidations were also performed on stepped Pt surfaces in order to clarify how exactly the step density influences these reactions. The most complete study on formic acid structure sensitivity was done by Motoo’s group24 using a complete series of stepped surfaces around the stereographic triangle and also kinked electrode surfaces.34 For methanol electrooxidation, Shin and Korzeniewski suggested that an increase of the step density catalyzes methanol decomposition,35 whereas Tripkovic et al. showed that the increase in the step density leads to a decrease in the surface activity towards methanol electrooxidation, both using Pt[n(111)(100)] stepped surfaces.36 In addition, Housmans et al. studied methanol oxidation on Pt[n(111)(110)] stepped surfaces, and reported an increase in the activity with the step density, suggesting that the presence of steps with a (110) orientation catalyzes methanol decomposition, CO oxidation and also the direct methanol oxidation.37 These kind of studies are relevant not only from a fundamental point of view but also from a practical one, because in practical applications the stepped surfaces may be considered as models for surface defects always present on dispersed electrodes. In Pt nanoparticles, a markedly different behaviour between methanol and formic acid oxidation was found when the particle size decreased in the range of 2–9 nm which was explained in terms of ‘‘catalytic ensemble effect’’. A decrease of the methanol oxidation rate was observed for nanoparticles smaller than 4 nm, suggesting that the decomposition of methanol requires an ensemble of terrace sites,38 which is in agreement with Tripkovic et al.36 Conversely, for formic acid electrooxidation, an important enhancement of the oxidation reaction rate was observed as the particle size decreases below 4 nm. This was attributed to the lack of a Pt site ensemble requirement for the CO poisoning process. Finally, for Pt nanoparticles having a particle size higher than 4 nm, their methanol and formic acid electrooxidation reaction rates are similar to those observed on polycrystalline Pt. To avoid such size complications, the nanoparticles used in this work have mean sizes higher than 4 nm. 3690 | Phys. Chem. Chem. Phys., 2008, 10, 3689–3698

Thus, in the present paper we report recent results about the influence of the surface structure/shape of well characterised Pt nanoparticles on methanol and formic acid electrooxidation reactions. The aim is to establish links between reactivity previously established for Pt electrodes with well defined surface structures (both basal planes and stepped surfaces) and the reactivity of Pt nanoparticles with some preferential twodimensional surface domains.

2. Experimental To prepare the three working electrodes with basal orientations, single crystal platinum beads were oriented, ground until reaching a hemispherical shape and polished according to the procedure described in ref. 39. A polyoriented electrode surface was obtained as above by omitting the polishing step. The electrodes were cleaned by flame annealing cooled down in H2/Ar and protected with water in equilibrium with this gas mixture to prevent contamination before immersion in the electrochemical cell. It has been shown that this procedure leads to surface topographies reasonably close to the nominal ones.40,41 In this work four types of Pt nanoparticles were prepared. Polyoriented Pt nanoparticles were synthesised by the waterin-oil method (water/polyethylene glycol dodecyl ether (Brijs30)/n-heptane), using sodium borohydride as reducing agent and using a similar methodology to that previously reported.42–44 After the synthesis, the nanoparticles were cleaned employing the protocol described in ref. 44. This procedure allows cleaning the nanoparticles avoiding electrochemical adsorption of oxygen and thus preserving the initial surface structure of the nanoparticles. On the other hand, preferentially oriented nanoparticles were prepared by using a so-called colloidal method45,46 in which 1 ml of 0.1 M sodium polyacrylate solution was added to 100 ml of an aged 1 10 4 M solution containing the desired Pt precursor. As Pt source, K2PtCl4 was employed for the synthesis of the Ptnano(100) nanoparticles whereas H2PtCl6 was used for the Ptnano(100)+(111) and Ptnano(111) nanoparticles. The concentration ratio of K2PtCl4 or H2PtCl6 to polyacrylate was 1 : 5. In addition, in the case of Ptnano(100)+(111) and Ptnano(111) nanoparticles, the pH of the solution was adjusted to 7 with 0.1 M HCl whereas in the case of Ptnano(100) the pH was not adjusted. Finally the solutions were purged with Ar for 20 min and the Pt ions were reduced by bubbling H2 for 5 min, except in the case of Ptnano(111) where only 5 min of Ar bubbling and 1 min of H2 bubbling were used. The reaction vessel was then sealed and the solution was left overnight. After complete reduction (12–14 h) two NaOH pellets were added to produce the precipitation of the nanoparticles. After complete precipitation, the nanoparticles were washed 3–4 times with ultrapure water. Transmission electron microscopy (TEM) experiments were performed with a JEOL, JEM-2010 microscope working at 200 kV. The sample for TEM analysis was obtained by placing a drop of the dispersed solution onto a Formvar-covered copper grid and evaporating it in air at room temperature. For each sample, usually over more than 300 particles from This journal is

c


different parts of the grid were used to estimate the mean diameter and size distribution of the nanoparticles. The procedure used for the electrochemical characterisation of the nanoparticles has previously been reported.42–44 The electrochemical measurements were performed in a 0.5 M H2SO4 solution at room temperature. Electrolyte solutions were prepared from Milli-Qs water and Merck ‘‘p.a.’’ sulfuric acid every day an experiment was carried out. A threeelectrode electrochemical cell was used. The electrode potential was controlled using a PGSTAT30 AUTOLAB system. The counter electrode was a gold wire. The potentials were measured against a reversible hydrogen electrode (RHE) connected to the cell through a Luggin capillary. Before each experiment, the gold collector was mechanically polished with alumina and rinsed with ultra-pure water to eliminate the nanoparticles from previous experiments. Methanol and formic acid oxidation voltammograms and chronoamperometric experiments were carried out in 0.5 M H2SO4 + 0.5 M MeOH or 0.5 M HCOOH solutions at room temperature. Solutions were prepared from methanol and formic acid (Merck ‘‘p.a.’’). All experiments were made at room temperature. The active surface area of the Pt nanoparticles was determined by the charge involved in the so-called hydrogen UPD region assuming 0.21 mC cm 2 for the total charge after the subtraction of the double layer charging contribution. For the chronoamperometric experiments the electrode potential was initially set at 0.8 V from which the potential was shifted to 0.4 V in case of formic acid or 0.6 V in case of methanol electrooxidation. Irreversible adsorption of bismuth was performed by spontaneous deposition from a saturated solution of bismuth(III) oxide in 0.5 M sulfuric acid.47 After deposition the electrode surface was rinsed with water and immersed in the electrochemical cell. The adsorption of germanium was performed as described previously from 10 2 M solutions of GeO2 in 1 M NaOH.48 The electrode with the droplet attached was immersed in the cell at 0.1 V. Residual contamination of the cell by germanium ions was checked after each experiment.

3. Results and discussion 3.1 Characterization of the Pt nanoparticles Fig. 1 shows some representative TEM images of the Pt nanoparticles used in this work. Pt nanoparticles prepared in microemulsion show a semi-spherical shape (Fig. 1A) with a particle size of 4.5 0.8 nm. These nanoparticles can be considered as representative of polyoriented, non-specificallystructured material. On the other hand, the synthesis of Pt nanoparticles in presence of sodium polyacrylate gives larger Pt nanoparticles with somewhat preferential shapes as a function of the experimental conditions. A typical TEM image of the Pt nanoparticles prepared using K2PtCl4 is shown in Fig. 1B. As can be observed, a preferential cubic shape is obtained which suggests the existence of a (100) preferential surface structure. The particle size is 8.2 1.6 nm. Fig. 1C corresponds to the Pt nanoparticles prepared from H2PtCl6 as Pt precursor. As can be appreciated, the shape of the Pt nanoparticles is predominantly hexagonal suggesting the coexistence of (100) and (111) Pt surface domains. The particle This journal is

c


Fig. 1 TEM pictures of (A) polyoriented, (B) (100), (C) (100)–(111) and (D) (111) Pt nanoparticles.

size of this sample is 11.5 1.7 nm. Finally, Fig. 1D shows a representative TEM image of the Pt nanoparticles prepared from H2PtCl6 with Pt as precursor, but where both Ar and H2 purging times were significantly lower than in the previous synthesis. Pt nanoparticles with a preferential octahedral shape can be easily discerned (Fig. 1D). In addition, an Phys. Chem. Chem. Phys., 2008, 10, 3689–3698 | 3691

important number of Pt nanoparticles with a tetrahedral shape can also be observed. In any case, both shapes suggest the presence of a preferential (111) surface structure. The Pt particle size of this sample is 8.6 1.4 nm. TEM data supply valuable information about mean size and distribution of the nanoparticles as well as about its shape. This latter information, however, may not be representative of the surface structure. At this point, it is important to mention that due to the fact that the use of a transmission electron microscope is a time-consuming ex situ technique and that the number of nanoparticles examined in a given sample is relatively small, the results obtained may not be representative of the ‘‘average’’ surface structure of the sample. In fact, the surface structure characterization data should be obtained, if possible, under experimental conditions as similar as possible to those that will be used in the reaction for which the electrocatalysts have been prepared. Moreover, the characterization should involve a number of nanoparticles large enough to supply statistical information of the whole catalytic surface and not only from a set of representative nanoparticles. In order to determine the presence of different symmetry sites on each sample and, therefore, to characterize their surface structures, we have employed pure in situ electrochemical probes.18 First of all, in the case of platinum, hydrogen and anion adsorption may be qualitatively used as probe reactions to define the properties of the surface: (i) the overall adsorption charge is directly proportional to the amount of surface atoms and thus it can be used to calculate the real surface area if the surfaces are clean and (ii) the distribution of the charge among the different voltammetric peaks gives a ‘‘first estimation’’ of the presence of the different surface sites on the whole surface. An additional advantage is that all these measurements have to be performed in solution, i.e. the environment in which the electrochemical reactions occur. 3.2 Electrochemical characterization 3.2.1 Voltammetry in the supporting electrolyte. Fig. 2 shows the characteristic voltammetric profiles of the different Pt nanoparticles in 0.5 M sulfuric acid. In all samples, the sharpness and the symmetry of the adsorption states are clear evidence of the surface cleanliness, an indispensable prerequisite for a correct surface characterization. Fig. 2A corresponds to the cyclic voltammogram obtained with the semispherical Pt nanoparticles prepared in microemulsion. The voltammogram reported (Fig. 2A) looks very similar to that reported for polycrystalline platinum electrodes. Thus, the voltammogram shows the presence of adsorption states associated to (110) and (100) sites at 0.12 and 0.27 V, respectively. Moreover, a shoulder around 0.35 V is detected, being characteristic of a small number of short (100) terraces. In addition, the unusual adsorption state around 0.5 V characteristic of small (111) ordered surface domains can be identified. All these states would be observed with Pt nanoparticles with a different extent that reflects the different surface composition. In this way, Fig. 2B, C and D show the voltammograms obtained in the case of the Pt nanoparticles prepared in colloidal solution and where, according to the TEM images, some preferential shapes should be expected. Fig. 2B corres3692 | Phys. Chem. Chem. Phys., 2008, 10, 3689–3698

Fig. 2 Voltammograms corresponding to (A) polyoriented, (B) (100), (C) (100)–(111) and (D) (111) Pt nanoparticles. Test solution: 0.5 M H2SO4, sweep rate 50 mV s 1.

ponds to the preferentially cubic Pt nanoparticles. The main feature of the voltammogram is the sharpness of the peak at 0.27 V associated to (100) edge and corner Pt surface sites. In addition, the well-marked state at 0.37 V, typical of (100) Pt terrace sites, is now much more evident than in the polyoriented surface, Fig. 2A. Thus, the voltammetric profile clearly points out that these Pt nanoparticles have a (100) preferential surface structure and that a large part of these Pt(100) sites are located at the surface as relatively wide (100) terraces, as suggested by the peak at 0.37 V, followed by that at 0.35 V. That also reflects the existence of shorter (100) terraces. The electrochemical observations as well as the TEM images obtained are in agreement with previous results obtained by Ahmadi et al., who pointed out that the Pt nanoparticles obtained under these experimental conditions have a preferentially cubic shape.45,46 Fig. 2C shows other distinctive features that must be noted. First of all, the adsorption state around 0.5 V characteristic of small (111) ordered surface domains is much more clearly marked than in the previous cases. This feature is directly related to the presence of bidimensionally ordered (111) domains. On the other hand, the sharp peaks at 0.12 and 0.27 V are also present with a similar intensity in the voltammetric profile. Similar sharp peaks are observed on stepped surfaces vicinal to Pt(111) with monoatomic steps with (110) or (100) symmetry, respectively.49 Finally, well-marked This journal is

c


Table 1 Fraction of the (111) and (100) ordered domains determined for the different nanoparticles Sample

(111) sites/%

(100) sites/%

Ptpoly Pt(100) Pt(100)–(111) Pt(111)

9 18 30 42

18 40 32 3

adsorption states at 0.37 and 0.35 V are also observed pointing out that these nanoparticles also contain two-dimensional Pt(100) domains. In this way, and in agreement with TEM observations where a preferential hexagonal shape was found (Fig. 1C), it is possible to conclude that the surface structure of this type of Pt nanoparticles is mainly formed by (100) and (111) Pt surface domains. Similar voltammetric profiles were previously reported by Attard et al.50 after a sintering process of Pt/graphite samples. Finally, Fig. 2D shows the voltammetric profile obtained with the Pt nanoparticles characterised by the presence of preferential tetrahedral and octahedral shapes. In comparison with the previous case, the voltammogram presents two main features, a very sharp peak at 0.12 V and the largest and symmetrical contribution at 0.5 V of all samples. The first contribution is known to be related to the presence of (110) Pt surface sites and the second one to the presence of relatively large two-dimensionally ordered (111) Pt domains. Moreover, it is also important to note the small contributions at 0.27 V due to (100) surface sites, likely both at edges and corners between (111) domains. The same can be said about twodimensional (100) surface domains, located at 0.35–0.37 V. In addition, the practically negligible presence of (100) Pt surface contributions makes it possible to observe a better reversibility in the characteristic adsorption states related to anion adsorption on the bidimensionally ordered (111) Pt domains, which was more difficult to state in the previous samples due to the overlapping of contributions coming from large (100) surface domains. Surface active sites quantification. To get a more detailed surface structure characterization picture, and using the same approach previously reported,14–18 the percentage of the different ordered (100) and (111) domains was evaluated by adsorbing Ge and Bi. Very briefly, the charge involved in the oxidation process related to the presence of these irreversibly adsorbed adatoms can be correlated with the hydrogen charge obtained from the corresponding voltammogram (Fig. 2) by using the relationships deduced for series of welldefined stepped surfaces.14–18 Table 1 summarizes the results for the different nanoparticles. The results obtained in the analysis agree with those expected from the inspection of the hydrogen-anion cyclic voltammetry (Fig. 2) and TEM analyses (Fig. 1). This latter observation points out that, unlike gold,11 the surface structure is not different to that expected from the geometrical shape of the platinum nanoparticles. 3.2.2 Electrocatalytic reactions. For the sake of comparison and to establish some standard values, formic acid and methanol electrooxidation on Pt single crystals with basal orientations was performed on flame-annealed electrodes This journal is

c


Fig. 3 Formic acid electrooxidation on (A) Pt (100), (B) Pt(111), (C) Pt(110) and (D) polyoriented Pt single crystals electrodes. Test solution: 0.5 M H2SO4 + 0.5 M HCOOH, sweep rate 20 mV s 1. Inset: blank voltammograms in 0.5 M H2SO4, sweep rate 50 mV s 1 after annealing and cooling down under reducing atmosphere. For the sake of comparison current densities have been multiplied by the factor indicated. The dotted line corresponds to the polyoriented Pt electrode after 15 cycles up to 1.5 V.

cooled down in reducing atmosphere, the more convenient experimental conditions leading to ordered surfaces. Fig. 3 and 4 show the results obtained in 0.5 M H2SO4 + 0.5 M HCOOH or 0.5 M MeOH, respectively. In addition, insets on Fig. 3 show the voltammograms corresponding to the blank single crystal surfaces in 0.5 M H2SO4. These voltammograms are taken as fingerprints of the crystalline surface structure for each electrode. In the case of the polyoriented surface (Fig. 3D) this stationary voltammogram, with a characteristic fine structure (solid line)51 is substantially different from that reached after several cycles of electrochemical activation up to 1.5 V (dotted line) that has lost the adsorption states linked to widely ordered domains.52 Although both voltammograms are characteristic of clean surfaces (the charge remains constant), the difference lies in the different distributions of surface crystalline sites that also reflects different reactivity. The case of formic acid electrooxidation on the Pt(100) electrode, Fig. 3A, is the most representative of the dual-path mechanism. In this way, negligible current (90 mA cm 2 at 20 mV s 1 in 0.5 M HCOOH) is measured in the positive scan since the surface is almost completely poisoned by CO dissociatively formed since the first potential excursion from the Phys. Chem. Chem. Phys., 2008, 10, 3689–3698 | 3693

immersion potential (0.05 V). This immersion potential was selected in all cases as a compromise between the thermodynamic and the rest potential. At potentials above 0.8 V, the CO adlayer is oxidatively stripped, giving rise to a clean Pt(100) surface. In the negative scan, the current rises as the potential is made more negative, until a maximum current is obtained at ca. 0.42 V (E19 mA cm 2). From this point, adsorbed CO is formed again. The build up of the CO blocks the surface for the direct oxidation of formic acid and the oxidation current diminishes until negligible values are obtained at ca. 0.3 V. It has been shown that the intrinsic activity of this electrode surface is extremely high, and can be estimated to be around 30 mA cm 2 under the present conditions which proved that the poison could be continuously eliminated from the surface.26 In contrast, for the Pt(111) electrode, Fig. 3B, the most relevant feature is the small difference in the currents recorded in the positive and negative scans around 0.5 V, thus suggesting a low poisoning rate. In this way, in spite of the fact that Pt(111) is initially the least active surface (only around 2 mA cm 2 are measured in the present conditions), its activity has a lower rate of decay than the other faces. In this case, the intrinsic activity of the electrode surface would represent only a 30% increase.27 For the Pt(110), Fig. 3C, the profile is similar to that found with Pt(100). In the positive scan, negligible current is measured (13 mA cm 2 at 20 mV s 1 in 0.5 M HCOOH) due to the formation of a CO adlayer which poisons the surface. This CO adlayer starts to be oxidatively stripped at potentials above 0.8 V with a maximum current peak above 0.9 V. On the other hand, in the negative scan, the current increases until reaching a maximum current at 0.8 V (E 6.5 mA cm 2) from which CO is again formed blocking the surface. Thus, in spite of the fact that the profile is similar to that found for the Pt(100), the current density obtained is much lower than that for the Pt(100) electrodes, and also the potential at which the maximum current for oxidation of formic acid is obtained, is significantly shifted to a higher value. Unlike the other two basal planes, the intrinsic activity has not been determined for this electrode. The problem is complicated by the fact that the electrode becomes deactivated upon cycling. As this deactivation does not seem to be related to accumulation of CO in each cycle53,54 it points to a problem in which surface stability is involved. Finally, Fig. 3D shows the voltammetric profiles obtained with the polyoriented Pt electrodes. The continuous line represents the behaviour of the flame annealed electrode and the discontinuous one that of the electrode cycled 15 times up to 1.5 V. It is known that the latter treatment disrupts the longrange ordered domains initially present on the electrode surface. Both polyoriented Pt electrodes present similar voltammograms in terms of profiles, although, in term of current, the results are quite different. Thus the currents obtained in the positive-going sweep with the freshly annealed Pt surface are much higher than those obtained after several cycles of electrochemical activation up to 1.5 V, especially in the oxidation region between 0.6 and 0.9 V which clearly reflects the surface deactivation described for the Pt(110) orientation. Nevertheless, in both Pt surfaces and in the positive scan, the currents measured are again very low since the surface is almost completely poisoned by CO which starts to be again oxida3694 | Phys. Chem. Chem. Phys., 2008, 10, 3689–3698

tively stripped at potentials above 0.8 V. In fact, the current at 0.5 V in the positive-going sweep can be almost exclusively attributed to HCOOH oxidation on (111) sites. The values are around 15 times lower than on the basal plane, reflecting the fraction of reactive sites on the polycrystalline sample. It has to be mentioned that the effect of cycling has a minor effect on the whole electrode activity: only a decrease from 0.12 to 0.10 mA cm 2 is recorded. Incidentally, it has to be recalled that Pt(111) is not the most reactive surface for direct HCOOH oxidation, and that stepped Pt surfaces containing (111) terraces are significantly more active.34 Thus, although the electrochemical cycling destroys the long-range ordered domains, a significant fraction of active sites still remain on the electrode. In the negative scan, two main oxidation contributions can be observed at potential above 0.8 (E 8.4 mA cm 2 for the freshly annealed Pt surface and E 3 mA cm 2 for the electrochemically activated Pt surface, strongly affected by the deactivation of the Pt(110) sites) and 0.45 V (E 2 mA cm 2 and E 1.8 mA cm 2, almost exclusively due to (100) site contributions). Thus, the oxidation at higher potential may be related to the oxidation of formic acid in (110) domains whereas the oxidation at lower potential may be associated with the formic acid oxidation on (100) domains. However, the maximum current density obtained is still far away from that obtained for the Pt(100) surface electrode (E19 mA cm 2). In the case of methanol electrooxidation the fundamental information is scarcer and no intrinsic activity data are available. In the same way, only partial studies on stepped surfaces have been reported.36,37 As for oxidation of formic acid, a dual-pathway was frequently proposed, again with COad as a poison intermediate from the dehydrogenation of methanol, and an active intermediate, COOHads55 is responsible to the direct oxidation to CO2. Although the most interesting features dealing with reactivity are related to the appearance of the main peaks at high potentials or in the negative-going sweep, the real electrocatalytic activity is measured by the current density values at 0.5 V in the positive-going sweep (52, 5.8 and 2.5 mA cm 2 corresponding to Pt(111), Pt(100) and Pt(110) electrodes, respectively), which are strongly affected by surface poisoning. The voltammograms recorded on the basal planes, Fig. 4A, B and C agree well with those previously reported. Methanol oxidation on Pt basal planes show an increase in maximum activity in the order Pt(111) o Pt(110) o Pt(100) under the actual experimental conditions. The order obtained is different to that early reported by Herrero et al., (Pt(111) o Pt(100) o Pt(110))32 and similar to that reported by Housmans et al.33 This is likely due to the differences in the flame annealing step (specifically the cooling step in air or in hydrogen) that deeply affects the surface structure of the electrode. In the case of polyoriented Pt surfaces, the currents are higher than those obtained on the Pt basal planes except for the Pt(111) which is about 5 times more active. Again, the currents obtained with the freshly annealed Pt surface are higher than those obtained after several cycles of electrochemical activation up to 1.5 V although the difference is lower than for formic acid electrooxidation. 3.2.3 Nanoparticles. In order to be able to compare the different Pt nanoparticles in relation to their electrocatalytic This journal is

c


Fig. 5 (A) Voltammograms corresponding to three successive deposits of (111) Pt nanoparticles. Test solution: 0.5 M H2SO4, sweep rate 50 mV s 1. (B) Formic acid electrooxidation corresponding to three successive deposits of (111) Pt nanoparticles. Test solution: 0.5 M H2SO4 + 0.5 M HCOOH, sweep rate 20 mV s 1. (C) Hydrogen charge vs formic acid oxidation peak current plot.

surface area and the electrode activity is demonstrated as the plot is linear and crosses the origin. Similar results were found with the other types of particles and also for methanol electrooxidation. This proportionality ensures the absence of artifacts while evaluating the surface reactivity.

Fig. 4 Methanol electrooxidation on (A) Pt (100), (B) Pt(111), (for the sake of comparison current density has been multiplied by the factor indicated), (C) Pt(110) and (D) polyoriented Pt single crystal electrodes, the dotted line corresponds to the polyoriented Pt electrode after 15 cycles up to 1.5 V. Test solution: 0.5 M H2SO4 + 0.5 M CH3OH, sweep rate 20 mV s 1.

reactivity towards direct formic acid or methanol electrooxidation, it is fundamental to verify that there is a linear relationship between the amount of deposited nanoparticles and their electrocatalytic properties, without artifacts coming from other aspects such as the morphology of the deposit, thickness, pores, local pH effects. . . that could affect the uniform mass transfer to the active surface. To check this important aspect, we have performed experiments where the amount of nanoparticles was successively increased on top of the gold substrate. At least three deposits of Pt nanoparticles were evaluated for each type of sample. After formic acid, or methanol, electrooxidation in each deposit, which was previously characterised in 0.5 M H2SO4, we have plotted the current density obtained in each electrode in relation to the charge involved in the so-called hydrogen adsorption–desorption region which also includes anion desorption–adsorption. In absence of transport complications, it is expected to obtain a signal proportional to the surface area that is, in turn, proportional to the charge QH in the so-called hydrogen adsorption region. As a representative example, Fig. 5 shows the results found with the (111) preferential Pt nanoparticles in formic acid electrooxidation. The proportionality between the This journal is

c


Formic acid electrooxidation. Fig. 6 shows the voltammetric responses of the different types of Pt nanoparticles in 0.5 M H2SO4 + 0.5 HCOOH. The upper potential limit has been limited to 0.9 V, in all samples, in order to avoid surface structure perturbations due to electrochemical oxygen adsorption in the high potential range. As can be appreciated, some quantitative differences on the formic acid electrooxidation voltammetric profiles are observed. First of all, in the case of the polyoriented Pt nanoparticles (Fig. 6A), the voltammetric profile recorded is similar to that typically associated with a polyoriented surface and obtained in the same potential range. In this way, the positive going sweep exhibits a first step at 0.5 V (0.18 mA cm 2), better marked than on the bulk polycrystalline sample, followed by an increase in current at high potentials, at which CO is oxidized. As a result of this surface cleaning the activity in the negative-going sweep is higher, as observed with the massive electrodes. For the preferentially oriented Pt nanoparticles, the voltammetric profiles are more complex and require a more detailed discussion although the general features are maintained. In relation to the maximum oxidation currents obtained in the direct sweep, around 0.5 V, values of 0.18, 0.28, 0.30 and 0.40 mA cm 2 were obtained for polyoriented, and preferential (100), (100)–(111) and (111) Pt nanoparticles, respectively. The highest activity for direct oxidation is recorded with the preferential (111) electrodes and the lowest one with the polyoriented sample, which only has a small amount of twodimensional (111) domains. In an attempt to obtain more realistic electrocatalytic activities, chronoamperometric measurements were recorded at 0.4 V. The results are shown in Fig. 7. According to the current densities obtained after 600 s, Phys. Chem. Chem. Phys., 2008, 10, 3689–3698 | 3695

Fig. 6 Formic acid electrooxidation on (A) polyoriented, (B) (100), (C) (100)–(111) and (D) (111) Pt nanoparticles. Test solution: 0.5 M H2SO4 + 0.5 M HCOOH, sweep rate 20 mV s 1.

the electrocatalytic activity follows the series, (111) 4 (100)–(111) 4 (100) 4 polyoriented, with current density values of 55, 48, 35 and 31 mA cm 2, respectively. It is important to point out that in spite of the fact that the (100)

Fig. 7 Chronoamperometric measurements at 0.4 V for (A) polyoriented, (B) (100), (C) (100)–(111) and (D) (111) Pt nanoparticles. Test solution: 0.5 M H2SO4 + 0.5 M HCOOH.

3696 | Phys. Chem. Chem. Phys., 2008, 10, 3689–3698

preferential nanoparticles show the highest current density in the negative scan of the voltammogram, due to the fastest poisoning of this orientation, they have, in fact, the lowest electrocatalytic activity of the preferentially oriented Pt nanoparticles for HCOOH electrooxidation, in contrast to those surfaces containing (111) domains that give higher current densities. However, all Pt nanoparticles will be finally poisoned pointing out that bimetallic electrodes are required for long term use. From potential-perturbation dependent reactivity details, the first point to note is the hysteresis between the currents measured in the positive and negative sweeps. As can be observed, the smallest hysteresis corresponds to the (111) preferential nanoparticles, Fig. 6D, whereas the largest one is observed for the (100) Pt nanoparticles, Fig. 6B. An intermediate situation takes place in the case of the (100)–(111) Pt nanoparticles, Fig. 6C. This tendency on the hysteresis is in complete agreement with that obtained with Pt single crystal electrodes24,56 where the Pt(111) has been shown to be the least poisoned surface. The most relevant differences between the voltammetric profiles and in the current values are observed in the negative scans. For the (100) Pt nanoparticles, Fig. 6B, two clear oxidation peaks are observed at 0.73 and 0.46 V. The lower potential oxidation peak takes place at a similar potential to that observed with a Pt(100) surface whereas that at 0.73 V can be associated with the (110) contributions.56 In addition, the decrease of current after the peak (0.47 V) is very pronounced which is again related to a fast poisoning of the surface. Finally, the oxidation peak at higher potentials could be related to the oxidation of formic acid on the non-preferential Pt (100) surface sites. For the (111) Pt nanoparticles, Fig. 6D, a broad single oxidation peak is observed. The maximum current is located at 0.54 V and the re-poisoning of the surface qualitatively measured by the current decrease in the voltammogram is clearly much less pronounced than in the previous case, which indicates a lower degree of poisoning. Besides, in the positive scan and in contrast to the (100) Pt nanoparticles where a single oxidation appears, two different oxidation contributions are shown. The situation in the (100)–(111) Pt nanoparticles, Fig. 5C, causes again a clear intermediate behaviour between the (100) and (111) preferential nanoparticles. In the negative scan, an oxidation peak at 0.52 V and a marked shoulder at 0.73 V are detected. In addition, in the positive scan two oxidation peaks are again observed, although the oxidation at higher potential is much less intense. Methanol electrooxidation. Following the same methodology described in the previous section for formic acid, methanol electrooxidation experiments were performed. Fig. 8 shows the voltammetric responses for the different types of Pt nanoparticles in 0.5 M H2SO4 + 0.5 M MeOH. As previously stated, the upper potential limit was restricted to 0.9 V to avoid surface structure modifications. The results obtained show some relevant differences in the voltammetric profiles recorded by using different nanoparticles. In the case of the polyoriented Pt nanoparticles, the profile closely resembles that obtained with the polyoriented Pt bead in the same potential range. For the preferential Pt nanoparticles, and independently of their surface structure, two main This journal is

c


Fig. 9 Chronoamperometric measurements at 0.6 V for (A) polyoriented, (B) (100), (C) (100)–(111) and (D) (111) Pt nanoparticles. Test solution: 0.5 M H2SO4 + 0.5 M CH3OH.

Fig. 8 Methanol electrooxidation on (A) polyoriented, (B) (100), (C) (100)–(111) and (D) (111) Pt nanoparticles. Test solution: 0.5 M H2SO4 + 0.5 M CH3OH, sweep rate 20 mV s 1.

oxidation peaks are observed at 0.82–0.83 V in the positive scan and at 0.72–0.73 V in the negative scan. The relative intensity of both peaks is clearly dependent on the surface structure/shape of the Pt nanoparticles. Thus, in the (100) Pt nanoparticles, the oxidation peak in the positive scan has a higher current density value than the oxidation peak in the negative one, whereas for (111) Pt nanoparticles this tendency is the opposite and the oxidation peak in the negative scan is clearly higher than that obtained in the positive scan. The relative intensity in the (100)–(111) Pt nanoparticles is intermediate between that of (111) and (100) Pt nanoparticles. Turning to the maximum current densities in the positive scans and at 0.5 V, (111) Pt nanoparticles showed the highest current density (E17 mA cm 2) whereas the (100) Pt nanoparticles showed the lowest one (E 8 mA cm 2). Nevertheless, and taking into account that the upper potential limit was limited to 0.9 V to avoid surface modifications, chronoamperometric measurements were performed at 0.6 V to evaluate better the electrocatalytic activity of the Pt nanoparticles. The experimental results are reported in Fig. 9. As can be appreciated the electrocatalytic activity follows the series, (111) 4 (100)–(111) 4 polyoriented 4 (100) Pt nanoparticles, with current density values of 78, 44, 35 and 25 mA cm 2, respectively, after 600 s at 0.6 V. The experimental results reported in the present paper clearly indicate that, among the samples studied, the so-called This journal is

c


(111) Pt nanoparticles have shown the highest electrocatalytic activity towards methanol and formic acid electrooxidation in the low potential range. Nevertheless the discussion of the results should not only be focused on the nature of the Pt surface sites and other factors should also be taken into account due to the complexity of the nanoparticle surface. At least, two aspects should be discussed. The first one is related to the surface structure of the nanoparticles. In spite of the fact that both H adsorption–desorption as well as Bi and Ge analyses indicate the existence of some preferential surface structures, we should not forget the presence of heterogeneities on the surface. Thus, for instance, the presence of low coordination surface atoms and/or the existence of a size distribution of the Pt domains may also play an important role on the final electrocatalytic response. Previous experiments with Pt stepped surfaces have highlighted this crucial point. The second point is related to the effect of the particle size. It has previously been reported that the particle size plays a different role in methanol and in formic acid oxidation.38 For methanol a decrease of the oxidation rate was observed with a particle size lower than 4 nm, whereas for formic acid electrooxidation, and contrary to the methanol oxidation, an important enhancement of the oxidation reaction rate was observed as particle size decreased for d o 4 nm. However, it is also true that the particle size effect could be embraced in the surface structure effect because the smaller the particle size the shorter the surface domains are. In this way, further studies are required to fully understand all factors affecting the electrocatalysis for methanol and formic acid electrooxidation in the case of Pt nanoparticles having some preferential surface structures.

Conclusions Methanol and formic acid electrooxidation was performed on polyoriented and preferential (100), (111) and (100)–(111) Pt nanoparticles. The results pointed out that, in a similar way to that found for fundamental studies using Pt single crystal electrodes, the surface structure of the nanoparticles plays an important role in the reactivity and that the electrocatalytic properties strongly depend on the surface structure/shape of the Phys. Chem. Chem. Phys., 2008, 10, 3689–3698 | 3697

nanoparticles. The current/potential curves can be qualitatively analyzed from the reactivity properties shown by the single crystal electrodes with basal orientations, which can be taken as model surfaces. The different nanoparticles exhibit a weighted electrocatalytic activity, depending on the relative amount of the surface sites. Among the Pt nanoparticles studied, those containing a preferential (111) orientation are clearly the most active towards methanol and formic acid electrooxidation, because of their low poisoning rate. Other features can also be interpreted from single crystal data. These findings open the possibility of designing new and better electrocatalytic materials using decorated shape-controlled Pt nanoparticles, as previously described using Pt single crystal electrodes.

Acknowledgements This work has been financially supported by the Ministerio de Educacioń y Ciencia of Spain through projects CTQ200604071/BQU and NAN2004-09333-C05-05. E. Garnier is grateful for support from ‘‘The Re´gion Poitou Charentes’’, France and the University of Alicante.

References 1. Catalysis & Electrocatalysis at Nanoparticle Surfaces, ed. A Wieckowski, E Savinova and C Vayenas, Marcel Dekker, 2003, and references therein. 2. R. Adzic, in Modern Aspects of Electrochemistry, ed. R. E. White, J O’M. Bockris and B. E. Conway, Plenum Press, New York, 1990, vol. 21, p. 163. 3. N. M. Markovic and P. N. Ross, in Interfacial Electrochemistry, Theory, Experiments and Applications, ed. A Wieckowski, Marcel Dekker, New York, 1998, p. 821. 4. N. M. Markovic and P. N. Ross, Surf. Sci. Rep., 2002, 45, 117. 5. S. Sriramulu, T. D. Javi and E. M. Stuve, in Interfacial Electrochemistry, Theory, Experiments and Applications, ed. A Wieckowski, Marcel Dekker, New York, 1998, p. 793. 6. E. Herrero, J. M. Feliu and A. Aldaz, in Encyclopedia of Electrochemistry, ed. A. J. Bard and M Stratmann, Wiley-VCH Verlag, Weinheim Germany, 2003, vol. 2, p. 443. 7. M. T. M. Koper, R. A. Van Santen and M. Neurock, in Catalysis and Electrocatalysis at Nanoparticle Surfaces, ed. A Wieckowski, E. R. Savinova and C. G. Vayenas, Marcel Dekker, Inc., New York, 2003, p. 1. 8. F. J. Vidal-Iglesias, J. Solla-Gulloń, P. Rodrı´ guez, E. Herrero, V. Montiel, J. M. Feliu and A. Aldaz, Electrochem. Commun., 2004, 6, 1080. 9. J. Hernandez, J. Solla-Gulloń and E. Herrero, J. Electroanal. Chem., 2004, 574, 185. 10. J. Hernandez, J. Solla-Gulloń, E. Herrero, A. Aldaz and J. M. Feliu, J. Phys. Chem. B, 2005, 109, 12651. 11. J. Hernandez, J. Solla-Gulloń, E. Herrero, A. Aldaz and J. M. Feliu, J. Phys. Chem. C, 2007, 111, 14078. 12. J. Solla-Gulloń, F. J. Vidal-Iglesias, E. Herrero, J. M. Feliu and A. Aldaz, Electrochem. Commun., 2006, 8, 189. 13. J. Solla-Gulloń, F. J. Vidal-Iglesias, P. Rodrı´ guez, E. Herrero, J. M. Feliu and A. Aldaz, Proceedings-Electrochemical Society, PV, 2005, vol. 11, p. 1. 14. J. Solla-Gulloń, F. J. Vidal-Iglesias, P. Rodrı´ guez, E. Herrero, J. M. Feliu, J. Clavilier and A. Aldaz, J. Phys. Chem. B, 2004, 108, 13573. 15. P. Rodrı´ guez, E. Herrero, J. Solla-Gulloń, F. Vidal-Iglesias, A. Aldaz and J. M. Feliu, Electrochim. Acta, 2005, 50, 3111. 16. P. Rodrı´ guez, E. Herrero, J. Solla-Gulloń, F. Vidal-Iglesias, A. Aldaz and J. M. Feliu, Electrochim. Acta, 2005, 50, 4308. 17. P. Rodrı´ guez, J. Solla-Gulloń, F. J. Vidal-Iglesias, E. Herrero, A. Aldaz and J. M. Feliu, Anal. Chem., 2005, 77, 5317.

3698 | Phys. Chem. Chem. Phys., 2008, 10, 3689–3698

18. J. Solla-Gulloń, P. Rodrı´ guez, E. Herrero, A. Aldaz and J. M. Feliu, Phys. Chem. Chem. Phys., 2008, DOI: 10.1039/ b709809j. 19. A. Capon and R. Parsons, J. Electroanal. Chem., 1973, 44, 1. 20. A. Capon and R. Parsons, J. Electroanal. Chem., 1973, 44205. 21. J. Clavilier, R. Parsons, R. Durand, C. Lamy and J. M. Leger, J. Electroanal. Chem., 1981, 124, 321. 22. C. Lamy, J. M. Leger, J. Clavilier and R. Parsons, J. Electroanal. Chem., 1983, 150, 321. 23. R. R. Adzic, A. V. Tripkovic and W. O‘Grady, Nature, 1982, 296, 137. 24. S. Motoo and N. Furuya, Ber. Bunsen-Ges. Phys. Chem., 1987, 91, 457. 25. S. G. Sun, J. Clavilier and A. Bewick, J. Electroanal. Chem., 1987, 240, 147. 26. J. Clavilier, J. Electroanal. Chem., 1987, 236, 87. 27. A. Fernańdez-Vega, J. M. Feliu, A. Aldaz and J. Clavilier, J. Electroanal. Chem., 1991, 305, 229. 28. R. Parsons and T. VanderNoot, J. Electroanal. Chem., 1988, 257, 9. 29. T. D. Jarvi and E. M. Stuve, in Electrocatalysis, ed. J Lipkowski and P. N. Ross, Wiley-VCH, New York, 1998, p. 75. 30. J. Clavilier, R. Durand, G. Guinet and R. Faure, J. Electroanal. Chem., 1981, 127, 281. 31. X. H. Xia, T. Iwasita, F. Ge and W. Vielstich, Electrochim. Acta, 1996, 41, 711. 32. E. Herrero, K. Franaszczuk and A. Wieckowski, J. Phys. Chem. B, 1994, 98, 5074. 33. T. H. Housmans, A. H. Wonders and M. T. M. Koper, J. Phys. Chem. B, 2006, 110, 10021. 34. N. Furuya, T. Kawahasa and S. Motoo, 175 ECS Meeting, Los Angeles (EEUU), 1989, Abstr. 468. 35. J. Shin and C. Korzeniewski, J. Phys. Chem. B, 1995, 993419. 36. A. V. Tripkovic and K. D. Popovic, Electrochim. Acta, 1996, 41, 2385. 37. T. H. Housmans and M. T. M. Koper, J. Phys. Chem. B, 2003, 107, 8557. 38. S. Park, Y. Xie and M. J. Weaver, Langmuir, 2002, 18, 5792. 39. J. Clavilier, D. Armand, S. G. Sun and M. Petit, J. Electroanal. Chem., 1986, 205, 267. 40. K. Domke, E. Herrero, A. Rodes and J. M. Feliu, J Electroanal. Chem., 2003, 552, 115. 41. N. Garcı´ a-Araéz, V. Climent, E. Herrero and J. M. Feliu, Surf. Sci., 2004, 560, 269. 42. J. Solla-Gulloń, V. Montiel, A. Aldaz and J. Clavilier, J. Electronal. Chem., 2000, 491, 69. 43. J. Solla-Gulloń, A. Rodes, V. Montiel, A. Aldaz and J. Clavilier, J. Electroanal. Chem., 2003, 554, 273. 44. J. Solla-Gulloń, V. Montiel, A. Aldaz and J. Clavilier, J. Electrochem. Soc., 2003, 150, E104. 45. T. S. Ahmadi, Z. L. Wang, T. C. Green, A. Henglein and M. A. El-Sayed, Science, 1996, 272, 1924. 46. T. S. Ahmadi, Z. L. Wang, A. Henglein and M. A. El-Sayed, Chem. Mater., 1996, 8, 1161. 47. J. Clavilier, J. M. Feliu and A. Aldaz, J. Electroanal. Chem., 1988, 243, 419. 48. R. Go´mez, M. J. Llorca, J. M. Feliu and A. Aldaz, J. Electroanal. Chem., 1992, 340, 349. 49. J. Clavilier, in Electrochemical Surface Science (Molecular Phenomena at Electrode Surfaces), ACS Symp. Ser. 378, ed. M Soriaga, 1988, ch. 14, p. 202. 50. G. A. Attard, J. E. Gillies, C. A. Harris, D. J. Jenkins, P. Johnston, M. A. price, D. J. Watson and P. B. Wells, Appl. Catal. A, 2001, 222, 392. 51. J. Clavilier and D. Armand, J. Electroanal. Chem., 1986, 199, 187. 52. B. E. Conway, H. Angerstein-Kozlowska, W. B. A. Sharp and E. E. Criddle, Anal. Chem., 1973, 45, 1331. 53. J. M. Orts, A. Fernandez-Vega, J. M. Feliu, A. Aldaz and J. Clavilier, J. Electroanal. Chem., 1990, 290, 119. 54. S. G. Sun, Thesis, University of Paris VI, 1986. 55. Y. X. Chen, A. Miki, S. Ye, H. Sakai and M. Osawa, J. Am. Chem. Soc., 2003, 125, 3680. 56. J. M. Feliu and E. Herrero, in Handbook of Fuel Cells. Fundamentals and Application, ed. W Vielstich, A Lamm and H. A. Gasteiger, John Wiley & Sons Ltd, Chichester UK, 2003, vol. 2, p. 625, and references therein.

This journal is

c
