- Email: [email protected]
Irreversible adsorption of Sn adatoms on basal planes of Pt single crystal and its impact on electrooxidation of ethanol
Available online at www.sciencedirect.com
Electrochimica Acta 53 (2008) 6081–6088
Irreversible adsorption of Sn adatoms on basal planes of Pt single crystal and its impact on electrooxidation of ethanol Qing-Wei Zheng, Chun-Jie Fan, Chun-Hua Zhen, Zhi-You Zhou, Shi-Gang Sun ∗,1 State Key Laboratory of Physical Chemistry of Solid Surfaces, Department of Chemistry, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China Received 12 November 2007; received in revised form 18 January 2008; accepted 25 January 2008 Available online 5 February 2008
Abstract The behaviours of irreversible adsorption (IRA) of Sn adatoms on Pt(1 0 0), Pt(1 1 1) and Pt(1 1 0) electrodes were characterized using cyclic voltammetry. It has revealed that Sn can adsorb irreversibly on Pt(1 0 0) and Pt(1 1 1), while not significantly on Pt(1 1 0) electrode. Quantitative analysis of the relationship between 1 − θ H and θ Sn suggests that Sn adatoms may adsorb preferably on hollow sites of Pt(1 1 1) (threefold) and Pt(1 0 0) (fourfold) planes, which is in accordance respectively with the values 0.31 and 0.21 of coverage of IRA Sn adatoms in saturation adsorption determined on these electrodes. The IRA Sn adatoms on different basal planes of Pt single crystal yield different impact on the electrocatalytic oxidation of ethanol. It has revealed that the IRA Sn adatoms on Pt(1 0 0) electrode have declined the activity for ethanol oxidation, while IRA Sn adatoms on Pt(1 1 1) have enhanced remarkably the electrocatalytic activity with Sn coverage θ Sn between 0.09 and 0.18. The oxidation peak potential Ep and the current density jp of ethanol oxidation on Pt(1 1 1)/Sn were varied with θ Sn , and the highest jp (1258 A cm−2 ) as well as the lowest Ep (0.20 V) were measured simultaneously at θ Sn around 0.14. In comparison with the data obtained on a bare Pt(1 1 1), the Ep was shifted negatively by 65 mV and the jp has been enhanced to about 1.7 times on the Pt(1 1 1)/Sn (θ Sn = 0.14), which is ascribed to hydroxyl species adsorption at relatively low potentials on Pt(1 1 1)/Sn surfaces. The current study is of importance in revealing the fundamental aspects of modification of the basal planes of Pt single crystal using Sn adatoms, and the impact of this modification on electrocatalytic activity towards ethanol oxidation. © 2008 Elsevier Ltd. All rights reserved. Keywords: Irreversible adsorption; Sn adatoms; Pt(1 1 1); Pt(1 0 0); Pt(1 1 0); Ethanol electrooxidation
1. Introduction The study of the surface structure of Pt electrodes modified by metal adatoms and the role of adatoms in electrooxidation of small organic molecules is of both fundamental and technological importance in electrocatalysis and fuel cell applications [1]. It is well known that the modification of the electrode surface with adatoms such as Ru, Sn, Pb, As, Sb, Bi, etc. can enhance the catalytic properties of Pt electrode [2–6]. Different techniques to modify electrode surface using adatoms were developed [7], among them the under potential deposition (UPD) [1] and the irreversible adsorption (IRA) [4,6] are most convenient and were frequently employed. The presence of adatoms formed from
∗ 1
Corresponding author. Tel.: +86 592 2180181; fax: +86 592 2183047. E-mail address: [email protected] (S.-G. Sun). ISE member.
0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2008.01.078
UPD or IRA procedure [2,4,5] on electrode surface can modify the electronic and chemical properties of substrates, and as a consequence improve the catalytic properties of electrodes [8–12]. The electrochemical properties of Sn adlayers on Pt have been studied so far mostly on polycrystalline surfaces [13,14]. For a better understanding of the metallic adlayer structures, single crystal electrodes with well-defined atomic arrangement have been introduced. Stamenkovic et al. [15] have characterized bimetallic single crystals of Pt3 Sn(1 1 0) and Pt3 Sn(1 1 1) in ultra high vacuum by using Auger electron spectroscopy (AES), low energy ion scattering spectroscopy (LEISS) and low energy electron diffraction (LEED). Following characterization in the UHV, the Pt3 Sn(1 1 0) and Pt3 Sn(1 1 1) were transferred into an electrochemical cell to study their electrocatalytic properties for CO adsorption and oxidation by using in situ infrared spectroscopy. They demonstrated that continuous oxidative removal of adsorbed CO starts as low as E < 0.1 V (vs. RHE), which is an
6082
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
important property for CO-tolerant catalysts. Electronic effects, surface structure and intermolecular repulsion between adsorbed species are ascribed for the high catalytic activity of Pt3 Sn(h k l) alloys. To the best of our knowledge, the studies so far concerning modification of Pt single crystal electrodes using Sn adatoms were mainly focused on Pt (1 1 1) plane and employing UPD method [16–18]. Massong et al. [16] reported the electrochemical properties of UPD Sn on Pt(1 1 1) at potentials below 0.8 V and with low Sn coverage. A specific reversible redox current peak around 0.6 V (vs. RHE) was observed, ascribing to a surface redox processes of Sn adatoms on Pt(1 1 1) similar to those previously observed for As, Bi and other adatoms. The eps number (electrons transferred per surface site as obtained from the corresponding suppression of hydrogen adsorption) was determined at 1, which was interpreted to a one-electron process of hydroxyl species adsorption. Recently, the PtSn alloy nanomaterials were attractive and used as catalysts in electrooxidation of ethanol [17–19]. In the UPD experiments, the metal ions are always present in solutions, which will certainly influence the precise determination of adatom coverage and introduce complexity in studying electrochemical properties of the adatom modified electrodes [1]. In contrast with UPD, the irreversible adsorption presents advantages consisting in that the adsorption of adatoms can be done outside the electrochemical cell and does not require applying a reduction potential [6]. After the adsorption, the electrode is transferred into an electrochemical cell that contains a study solution without metal ions, so that the adatom coverage can be determined precisely, and the surface processes as well as the electrocatalytic properties of the adatom modified electrode may be finely investigated. Besides the UPD process on Pt single crystal electrode, Sn also shows irreversible adsorption properties [1]. Haner et al. [5] studied the geometric and electronic effects of tin adatoms by depositing spontaneously tin on Pt single crystal surfaces, but they did not observe any enhancement of activity for methanol oxidation. Similar results of irreversibly adsorbed tin on Pt single crystal electrodes were also reported by Campbell et al. [20]. In order to investigate further the surface modification process of Sn adatoms on Pt single crystal electrodes, three basal planes of Pt single crystal, Pt(1 1 1), Pt(1 0 0) and Pt(1 1 0) were prepared and modified by IRA Sn adatoms in the present study. The oxidation of ethanol was employed as a probe reaction to test the electrocatalytic properties of the Pt(h k l)/Sn electrodes. The study has put emphasis upon the surface processes of Pt single crystal basal planes towards Sn adatom modification, and the surface structure effects in electrocatalysis as well. 2. Experimental Pt(1 0 0), Pt(1 1 0) and Pt(1 1 1) were prepared in our laboratory through the method described previously [21]. Before each measurement the Pt single crystal electrodes were treated using Clavilier’s method [22], i.e. they were annealed in a hydrogen–oxygen flame, quenched with super pure water and transferred into electrochemical cell under the protection of a droplet of pure water. After a well-defined cyclic voltammo-
gram that characterizes the surface structure of each electrode has been recorded in cell 1 that contains 0.5 M H2 SO4 solutions, the Pt single crystal electrode was immersed in 1 M H2 SO4 solution containing 10−3 M Sn2+ ions for 1 min, then rinsed with super pure water and transferred back to cell 1 or cell 2 (perchloric acid solutions) at 0.0 V (vs. SCE). Coverage of IRA Sn adatoms (Snad ) in saturation adsorption was thus formed and determined. Starting from the saturation coverage of Snad , different sub-coverage of Snad was obtained by partially stripping Snad in cell 1 by applying a votammetry in controlling the upper limit of potential scan (Eu ) and the number of potential cycling. The calibration of the designated coverage of Snad was then carried out in cell 2, and the electrode was transferred finally into cell 3 that contains ethanol solution for further electrocatalytic study. The solutions were prepared using Millipore water (18.0 M cm) provided from a Milli-Q Lab apparatus (Nihon, Millipore Ltd.), super pure H2 SO4 or HClO4 , ethanol and SnSO4 of analytical grade. A saturated calomel electrode (SCE) was served as reference electrode. Potentials in this paper were reported versus the SCE scale. The solution was deaerated by bubbling pure N2 gas before experiment, and kept a flux of N2 over it during measurements to prevent possible interference of oxygen and impurities from the atmosphere. All experiments were carried out at room temperature around 20 ◦ C. 3. Results and discussion 3.1. Electrochemical characterization of surface modification of Pt(1 1 1), Pt(1 1 0) and Pt(1 0 0) with IRA Sn adatoms The adsorption and desorption of hydrogen is highly sensitive to the atomic arrangement of Pt single crystal electrodes, and has been conveniently employed to characterize the surface structure of the electrode. It is known that the surface structure of Pt single crystal electrodes remains stable unless a significant amount of oxygen is adsorbed on the surface, and a well-defined structure of Pt single crystal surface can be maintained in 0.5 M H2 SO4 solution at electrode potentials below 0.75 V. Fig. 1a shows cyclic voltammograms of Pt(1 1 1) (dot lines, Eu = 0.75 V) and the Snad modified electrode Pt(1 1 1)/Sn (solid lines, varying Eu ) recorded in 0.5 M H2 SO4 solution. It can observe that the electrochemical adsorption of hydrogen yields a current plateau between −0.20 and 0.0 V, and the sulfate adsorption gives a ‘butterfly’ peak in the potential region from 0.07 to 0.30 V on bare Pt(1 1 1), signifying a well-defined order structure of Pt(1 1 1). On the Pt(1 1 1)/Sn of saturation adsorption IRA Snad , the sulfate adsorption current is completely suppressed, and the hydrogen adsorption is inhibited to a large extent. Along with increasing progressively Eu , two oxidation current peaks appear respectively around 0.33 and 0.60 V in the positive going potential scan (PGPS), attributing respectively to adsorption of hydroxyl species [16,18] and Snad oxidation. The desorption of hydroxyl species in the negative-going potential scan (NGPS) gives rise to a sharp current peak near 0.30 V with Eu at 0.45 V, while the reduction of oxidized Snad adatoms occurs in sev-
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
Fig. 1. Cyclic voltammograms of Pt(1 1 1), Pt(1 0 0), Pt(1 1 0) (dashed lines) and Pt(1 1 1)/Sn, Pt(1 0 0)/Sn, Pt(1 1 0)/Sn (solid lines) electrodes in 0.5 M H2 SO4 solution. Initial voltammogram (- - -); recovery voltammogram after stripping Snad (· · ·), sweep rate 50 mV s−1 .
eral broad current peaks lying on potential region between 0.6 and 0.1 V when Eu is above 0.5 V. It can be seen that the Sn adatoms are stable on Pt(1 1 1) in 0.5 M H2 SO4 at potentials below 0.45 V, for which the hydrogen adsorption current is maintained; while Sn adatoms can be progressively stripped from Pt(1 1 1) when Eu is increased beyond 0.45 V, as evidenced by the progressive increase of hydrogen adsorption current due to the recovery of surface sites for hydrogen adsorption. This result demonstrates that the surface concentration of Snad can be con-
6083
veniently controlled under electrochemical conditions. When Snad has been completely stripped out by applying continuously potential cycling between −0.20 and 0.75 V, a pair of peaks close to −0.12 V appears in the voltammogram, corresponding to hydrogen adsorption on surface sites lying on steps of (1 1 1) symmetry. This result indicates that the Pt(1 1 1) surface has been disturbed from well-defined order structure. This perturbation is also evidenced by the disappearance of the sharp spike in the ‘butterfly’ peak at 0.21 V, which is a stamp character of sulfate adsorption on well-defined order structure of Pt(1 1 1). The CV features of bare Pt(1 0 0) and Sn modified Pt(1 0 0) electrode are shown in Fig. 1b. The current peaks of hydrogen adsorption–desorption at 0.02 and 0.12 V observed on bare Pt(1 0 0) have been inhibited almost completely when Sn adsorbed in saturation on Pt(1 0 0) electrode. Along with increasing progressively Eu , the blockage of current of hydrogen adsorption–desorption by Snad is maintained stable till Eu is higher than 0.45 V, illustrating that the Sn adatoms are also stable on Pt(1 0 0) below 0.45 V. When Eu is increased to above 0.45 V, we observe only in the PGPS an oxidation current peak near 0.58 V that is ascribed to Snad oxidation. The reduction of oxidized Snad occurs in a reduction current peak near 0.43 V in the NGPS. It is interesting to see that the current peak around 0.33 V due to hydroxyl species adsorption observed on Pt(1 1 1)/Sn electrode is missing on Pt(1 0 0)/Sn, which may yield different impact on electrocatalytic properties of these electrodes, and it will be described thereafter. It is evident that Snad can be desorbed progressively from Pt(1 0 0) surface when Eu is above 0.45 V in 0.5 M H2 SO4 solution. As a consequence, the current of hydrogen adsorption–desorption is increasing with the recovery of surface sites for hydrogen adsorption. It is worthwhile to note that the current peaks appeared in the voltammograms recorded during stripping out Sn adatoms are quite different from the sharp current peaks observed in the voltammogram of bare Pt(1 0 0) surface. A pair of broad peak lying on from 0.0 to 0.2 V centered at 0.05 V appears in the voltammogram of Pt(1 0 0)/Sn when part Snad has been stripped out. When Sn adatoms have been completely removed by continuously cycling electrode potential between −0.20 and 0.75 V, two pairs of hydrogen adsorption–desorption current peaks appear again. However, these two peaks are much smaller than those in the voltammogram of the bare Pt(1 0 0), and only 68% of surface sites for hydrogen adsorption has been recovered. The results suggest that Pt(1 0 0) surface has been also disturbed from welldefined atomic arrangement through the adsorption of Snad . In the case of Pt(1 1 0) electrode (Fig. 1c), we can hardly observe neither significant inhibition of the current peak of hydrogen adsorption–desorption after the electrode subjected to the same procedure of IRA Sn adatoms adsorption, nor the remarkable Snad oxidation current peak near 0.60 V as observed on Pt(1 1 1)/Sn and Pt(1 0 0)/Sn. These results indicate that Sn may not be irreversibly adsorbed on Pt(1 1 0) surface under present conditions. It is well-known [22–24] that the surface of Pt(1 1 0) is readily reconstructed into a (1 × 2) structure under electrochemical conditions. Two models for the Pt(1 1 0)-(1 × 2) structure are generally accepted. One is the so called missing row model [25], on which the surface sites are of (1 1 1) sym-
6084
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
metry; the other is the Bronzel–Ferrer model [26] that give rise to some small 100 facets of short-range order. In both models, it does not exist hollow sites of long-range order on the Pt(1 1 0)(1 × 2) surface. In contrast, the long-range ordered hollow sites of threefold or fourfold could be found on Pt(1 1 1) or on Pt(1 0 0) surfaces, respectively, which may suggest that the irreversible adsorption of Sn adatoms has been conducted on the hollow sites of long-range order on these two single crystal planes. The surface site occupancy by Snad (or the coverage of Snad , θ Sn ) and the coverage of hydrogen (θ H ) can be calculated from the comparison of hydrogen adsorption charge (QSn H ) with the charge of Snad oxidation (Qox Sn ) using the following equations, Qox /4 NSn = SnS NPt QH
(1.1)
NH QSn = HS NPt QH
(1.2)
θSn = and θH =
NH , NSn and NPt are, respectively, the number of H adatoms, the number of Sn adatoms and the number of surface Pt sites. QSH and QSn H are charges corresponding to hydrogen adsorption on electrodes without and with Snad modification, respectively. The ox QSH , QSn H and QSn are measured by integration of corresponding CV curves. According the previous work of UPD Sn adsorbed on Pt electrode studied through XPS characterization [27], two kinds of tin species were identified on Pt surface, namely, metallic tin and little tin oxy-compounds. Metallic tin was deposited spontaneously on Pt surface due to disproportionation of Sn(II) to Sn(IV), while tin oxide/hydroxide species were derived as a result of Sn(II) complex hydrolysis. It is also known, from cyclic votammetry and electrochemical quartz crystal microbalance (EQCM) studies [28], that four electrons are involved in Sn adatoms oxidation on Pt electrodes, i.e. Snad → Snad 4+ . On account of the experiment procedures used in the present study, the irreversible adsorption of Sn2+ may produce mostly metallic tin and little tin oxide/hydroxide species on Pt(h k l) surfaces. At low potentials all IRA Sn adatoms are reduced into metallic tin, and are oxidized to Sn(IV) species at 0.75 V under present CV conditions. As a consequence, the term Qox Sn /4 in Eq. (1.1) is in direct proportion to NSn . Table 1 lists the electrochemical parameters measured from CVs recorded on Pt(1 1 1), Pt(1 0 0) and Pt(1 1 0) electrodes under saturation adsorption of IRA adatoms. The Pt single crystal electrodes were subjected initially to different pretreatments, Table 1 Parameters of IRA Sn on Pt(h k l) electrodes subjected to different treatments QH /C cm−2
−2 QSn H /μC cm
−2 Qox Sn /μC cm
S θSn
Pt(1 1 1)
Air Ar + H2
233.2 246.9
29.32 42.25
282.4 336.7
0.31 0.34
Pt(1 0 0)
Air Ar + H2
265.8 227.2
50.3 71.7
224.5 251.4
0.21 0.27
Pt(1 1 0)
Air Ar + H2
189.4 219.2
177.6 173.1
2.08 22.36
0.003 0.02
i.e. after annealing in hydrogen–oxygen flame they were cooled in air (oxidant atmosphere) or in Ar + H2 stream (reductant atmosphere). It is known that cooling in Ar + H2 stream will produce a well-defined Pt single crystal surface with less defects [29,30], therefore more hollow sites in long-range order are expected on this surface. As a consequence, it will favor the irreversible adsorption of Sn. As demonstrated in Table 1, the saturation coverage of Sn adatoms on any one of the three electrodes cooling down in Ar + H2 stream is significantly larger than that measured on the same electrode cooling down in air. The results have demonstrated clearly that the long-range order of hollow sites is necessary for IRA Sn adatoms. It is worthwhile to note that, the CV features of Sn adatoms are much pronounced in perchloric acid than in sulphuric acid, and that without the competition adsorption of sulphuric acid anions the Sn adatoms are much stale in perchloric acid. Actually, the Sn adatoms can be easily removed by oxidation at potentials above 0.45 V in H2 SO4 solution, while they could not be stripped in HClO4 solution even at potential as high as 0.75 V. In order to avoid the interference of sulfate adsorption and to study further the surface processes of the IRA Sn adatoms, 0.535 M HClO4 solution that has similar pH value of 0.5 M H2 SO4 solution was used, and the cyclic voltammetric results are illustrated in Fig. 2. As can be seen in Fig. 2a, the hydroxyl species adsorption on Pt(1 1 1)/Sn gives rise to a sharp peak at 0.34 V in the PGPS for a Snad coverage of 0.19, and the desorption yields also a sharp current peak at 0.32 V in the NGPS when the Eu sets at 0.42 V. The oxidation of Snad for this coverage produces a broad peak around 0.57 V in the PGPS, and the reduction of the oxidized Snad gives rise to two broad peaks around 0.34 and 0.27 V in the NGPS. It is interesting to note that the Sn adatoms on Pt(1 1 1) are more stable in perchloric acid solution, which is evidenced by the stable hydrogen adsorption–desorption current recorded with continuously cycling electrode potential between −0.20 and 0.75 V. As the Snad is difficult to remove from Pt(1 1 1) surface (it is also difficult to remove from Pt(1 0 0) surface) in perchloric acid, in the experiments thereafter to prepare different Snad coverage starting from the saturation adsorption of IRA Sn adatoms, the partly stripping out Snad was done in 0.5 M H2 SO4 , the electrode was then transferred into perchloric acid solution to study the surface process or into a solution containing 0.1 M ethanol to investigate the electrocatalytic properties of a defined Snad coverage. At a high Snad coverage (0.24), the current peak of hydroxyl species adsorption is positively shifted to 0.39 V with significant decrease in the amplitude, the current peak of Snad oxidation in the PGPS remains broad and locates around 0.57 V, and the reduction of the oxidized Snad occurs again in broad peaks lying on between 0.60 and 0.18 V in the NGPS. When the Snad coverage is larger than 0.27, the hydroxyl adsorption current peak is disappeared completely, as shown in the voltammogram for a Snad coverage of 0.29, in which only a much broad current peak of Snad oxidation from 0.28 to 0.75 V centered around 0.52 V is observed in the PGPS. In this case the reduction of the oxidized Snad can be seen from 0.60 to 0.0 V in the NGPS and appears a peak near 0.36 V. It is obvious that the hydroxyl species could not
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
6085
tion current of adsorbed Sn adatoms is observed around 0.4 V (Fig. 2c). The saturation coverage of Sn adatoms on Pt(1 1 0) in perchloric acid is determined at 0.02, which is larger than the value measured in sulphuric acid solutions (as listed in Table 1). S measured on Pt(1 1 1) and on Pt(1 0 0) in However, the values θSn perchloric acid are respectively 0.31 and 0.21, which are close to the values reported in Table 1 measured in sulphuric acid solutions. The coadsorption properties of Sn with H on Pt(1 1 1) surface may be characterized by the relationships between 1 − θ H and θ Sn . Fig. 3a plots the variation of 1 − θ H and θ Sn obtained in 0.1 M HClO4 solution, since the study of electrocatalytic properties of Sn adatoms modified Pt single crystal electrode was conducted in 0.1 M HClO4 + 0.1 M CH3 CH2 OH solution, and it has tested that the results are the same as those obtained in 0.535 M HClO4 solution. As stated above, when the θ Sn is below 0.27 the hydroxyl species adsorption will take place, which will cause complexity in analysis of the coadsorption of Sn and H. Therefore, only experimental data for θ Sn > 0.27 were plotted in Fig. 3a. A straight line is observed, and the slope is measured at 2.88 that can be served to estimate the number (n) of
Fig. 2. Cyclic voltammograms of Pt(1 1 1), Pt(1 0 0), Pt(1 1 0) (dashed lines) and Pt(1 1 1)/Sn, Pt(1 0 0)/Sn, Pt(1 1 0)/Sn (solid lines) electrodes in 0.535 M HClO4 solution, sweep rate 50 mV s−1 .
adsorb on Pt(1 1 1)/Sn electrodes with Snad coverage higher than 0.27. Similar CV features are observed on Pt(1 0 0)/Sn electrode in the sulfuric acid and perchloric acid solutions, i.e. only the oxidation current of Snad and the reduction current of the oxidized Snad appear in the voltammogram (Fig. 2b). The oxidation current peak is near 0.58 V in the PGPS and the reduction current peak is around 0.42 V in the NGPS. This CV feature does not depend on the Snad coverage, confirming clearly that it does not exist hydroxyl species adsorption on Pt(1 0 0)/Sn electrode. In the case of Pt(1 1 0) electrode, a small, but not negligible oxida-
Fig. 3. Relationship between 1 − θ H and θ Sn on Pt(1 1 1)/Sn (a) and Pt(1 0 0)/Sn (b). Data obtained in 0.1 M HClO4 solution.
6086
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
Pt(1 1 1) surface sites for hydrogen adsorption blocked per Sn adatom. The slope value of 2.88 suggests that, at high Snad coverage or in saturation adsorption of Sn adatoms on Pt(1 1 1), one Sn adatom may inhibit approximately three surfaces sites for hydrogen adsorption. This result confirms once again that the Sn adatoms are adsorbed on the threefold hollow sites of the Pt(1 1 1) surface, which is in accordance with the coverage of IRA Snad in saturation adsorption determined at 0.31. As it does not exist the adsorption of hydroxyl species on Pt(1 0 0)/Sn electrode, it is convenient to plot all data of (1 − θ H ) versus θ Sn . As illustrated in Fig. 3b, a linear relationship can be maintained for θ Sn varying from 0.022 to 0.182, and the slope of the straight line is measured at 3.10. This result indicates that in most θ Sn studied, by averaging, one Sn adatom may inhibit approximately three surfaces sites for hydrogen adsorption. Such value may suggests that most Sn adatoms adsorb on the fourfold hollow sites of Pt(1 0 0) for which one Sn adatom occupies 4 surface sites, and that some Sn adatoms may also adsorb on the bridge and atop sites of Pt(1 0 0) surface. The coverage of IRA Sn adatoms in saturation adsorption on Pt(1 0 0) has been determined at 0.21, which is close to the value of 0.25 for an adlayer of Sn occupying the fourfold hollow sites on Pt(1 0 0) surface. 3.2. Electrocatalytic properties of Pt(1 1 1) and Pt(1 0 0) modified with IRA Sn adatoms towards ethanol oxidation The oxidation of ethanol is employed as a probe reaction to test the electrocatalytic activity of platinum single crystal electrodes modified with IRA Sn adatoms. As Sn does not adsorb significantly on Pt(1 1 0) through IRA procedure under present conditions, only Pt(1 1 1) and Pt(1 0 0) were studied in this section. In this study, the θ H was calibrated at first in the cell containing 0.1 M HClO4 solution, the electrode was then transferred into another electrochemical cell containing 0.1 M HClO4 + 0.1 M C2 H5 OH solution for investigating ethanol oxidation. Fig. 4a compares the voltammograms of ethanol oxidation on a Pt(1 1 1)/Sn (θ Sn = 0.15) and on the bare Pt(1 1 1) electrodes. The oxidation of ethanol on bare Pt(1 1 1) starts at 0.06 V and gives rise to a current peak near 0.25 V. When the Pt(1 1 1) electrode is covered by Snad at high coverage, the current peak is mostly inhibited. Along with partly desorption of Snad from the electrode, i.e. decreasing θ H , the oxidation current peak is recovered and even larger than that on the bare Pt(1 1 1) electrode as shown in Fig. 4a for a Snad coverage of 0.15. The significant enhanced catalytic activity of Pt(1 1 1)/Sn for Snad coverage of 0.15 in Fig. 4a consists in that the oxidation of ethanol is shifted negatively about 60 mV. The onset potential (Eonset ) of ethanol oxidation is shifted negatively to 0.0 V on the Pt(1 1 1)/Sn; the potential of the current peak in the PGPS (Ep ) is also shifted negatively to 0.2 V on Pt(1 1 1)/Sn. Moreover, the current intensity has been also enhanced to some extent. The intensity current peak (jp ) in the PGPS is measured at 809 A cm−2 on bare Pt(1 1 1), and it is increased to 1200 A cm−2 on the Pt(1 1 1)/Sn electrode. The CV feature that the profile of the j-E curve in the NGPS is almost the same as
Fig. 4. j–E curves of electrocatalytic oxidation of ethanol on Pt(1 1 1) and Pt(1 0 0) (dashed lines); Pt(1 1 1)/Sn and Pt(1 0 0)/Sn (solid lines) in 0.1 M C2 H5 OH + 0.1 M HClO4 .
that in the PGPS suggests that less self-poisoning phenomenon is encountered on the Pt(1 1 1) electrode, which has been also evidenced by studies of in situ FTIR spectroscopy [31]. The enhancement of catalytic activity upon the modification of Sn adatoms on Pt(1 1 1) has been suggested mainly coming from the adsorption of hydroxyl species on Snad modified Pt(1 1 1) surface [32–34]. In contrast to the enhancement of catalytic activity, the Pt(1 0 0)/Sn illustrates a declined activity in comparison with that of the bare Pt(1 0 0). We observe that the current of ethanol oxidation is almost zero when electrode potential is below 0.2 V on bare Pt(1 0 0) electrode, due to the surface poisoning by poison intermediates [31]. When electrode potential is increased to above 0.2 V, visible oxidation current is observed and a sharp current peak appears near 0.47 V with intensity of 6.01 mA cm−2 in the PGPS, which is ascribed to the oxidation of poison species together with ethanol on Pt(1 0 0). A broad oxidation current peak in the NGPS appears at 0.44 V of intensity 1.75 mA cm−2 . However, when the Pt(1 0 0) is modified with Sn adatoms, the intensities of the current peaks in both the PGPS and the NGPS have been declined respectively to 1.84 and 0.87 mA cm−2 , although the Ep in the PGPS is shifted negatively to 0.45 V. The fact that the oxidation current in the potential region from
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
6087
Fig. 6. j–E curves of electrocatalytic oxidation of ethanol on Pt(1 0 0)/Sn at different coverage of Snad in 0.1 M C2 H5 OH + 0.1 M HClO4 .
for which the jp reaches the largest value of 1245 A cm−2 , and the Ep is at the lowest position of 0.20 V. In comparison with the bare Pt(1 1 1) electrode, the Ep has been shifted negatively about 65 mV, and the jp has been enhanced to about 1.7 times of that measured on a bare Pt(1 1 1) surface. When Snad is fully stripped from Pt(1 1 1) electrode, the j–E profile of ethanol oxidation is quite similar to that recorded on the bare Pt(1 1 1) but with a slightly small current intensity. In Fig. 6, the j–E curves recorded in the PGPS for ethanol oxidation on Pt(1 0 0)/Sn electrodes at various θ H are compared. As illustrated previously in Fig. 4b, the Sn adatoms on Pt(1 0 0) present a decline effect in catalytic activity for ethanol oxidation, we observe in Fig. 6 that the jp is decreased monotony with the increase of θ H . It can be seen that the j–E curve of Pt(1 0 0)/Sn electrode after completely stripping out Snad is much different from that of the initial bare Pt(1 0 0), and the jp is only 0.852 mA cm−2 , i.e. about one seventh of that measured on the initial bare Pt(1 0 0) electrode. 4. Conclusions
Fig. 5. (a) j–E curves of electrocatalytic oxidation of ethanol on Pt(1 1 1)/Sn at different coverage of Snad in 0.1 M C2 H5 OH + 0.1 M HClO4 . The variations of the peak potential (b) and the maximum current intensity (c) of C2 H5 OH oxidation with θ Sn on Pt(1 1 1)/Sn.
0.1 to 0.4 V in the PGPS is always much smaller than that in the NGPS indicates a severe self-poisoning on both bare and Sn adatoms modified Pt(1 0 0) surface [31]. Fig. 5a displays j–E curves of ethanol oxidation in the PGPS on Pt(1 1 1) and Pt(1 0 0) modified with Sn adatoms of different coverage. It can observe on Pt(1 1 1)/Sn electrodes that both Ep and jp vary with θ H . The variations of Ep and jp with θ H are plotted in Fig. 5b and c, respectively. It can be determined that the best θ H that yields the maximum enhancement of catalytic activity is 0.14 for the oxidation of ethanol on Pt(1 1 1)/Sn electrode,
The results obtained in the present study demonstrated that the irreversible adsorption of Sn adatoms on basal planes of Pt single crystal is sensitive to the surface structure of Pt single crystal electrode. The CV features of Pt single crystal electrodes modified with IRA Sn adatoms are varied with surface atomic arrangement. It has revealed that Sn can adsorb irreversibly on Pt(1 0 0) and Pt(1 1 1) but not significantly on Pt(1 1 0) electrode under present conditions. The IRA Sn adatoms on Pt(1 1 1) and Pt(1 0 0) are stable at potentials below 0.45 V in sulfuric acid solutions, while they are more stable in perchloric acid medium till 0.75 V. It has observed solely on Pt(1 1 1)/Sn a specific redox peak in the potential region around 0.35 V, which is attributed to the adsorption of hydroxyl species. Based on quantitative analysis of relationship between 1 − θ H and θ Sn , Sn is proposed to adsorb on the threefold hollow sites of Pt(1 1 1) surface and mainly on the fourfold hollow sites of Pt(1 0 0) plane. The coverages of IRA Sn adatoms in saturation adsorption on Pt(1 1 1) and Pt(1 0 0) were determined respectively 0.31 and 0.21, which are
6088
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
in accordance with that the Sn adatoms adsorb preferentially on hollow sites of these surfaces. It has determined a decline effect of the IRA Sn adatoms for ethanol electrocatalytic oxidation on Pt(1 0 0), while the IRA Sn adatoms on Pt(1 1 1) has enhanced significantly the electrocatalytic activity. The oxidation peak potential Ep and the current density jp of ethanol oxidation on Pt(1 1 1)/Sn were varied with Sn coverage θ Sn , and the highest jp as well as the lowest Ep were simultaneously determined at θ Sn of 0.14. The fact that hydroxyl species could adsorb at relative low potentials on Pt(1 1 1)/Sn electrode presents an interpretation of the enhanced electrocatalytic activity of Pt(1 1 1)/Sn for ethanol oxidation. Acknowledgement The study was supported by Natural Science Foundation of China (grants 20673091, 20433060). References [1] E. Herrero, L.J. Buller, H.D. Abruna, Chem. Rev. 101 (2001) 1897. [2] R.R. Adzic, A.V. Tripkovic, N.M. Markovic, J. Electroanal. Chem. 150 (1983) 79. [3] J. Clavilier, A. Fernandezvega, J.M. Feliu, A. Aldaz, J. Electroanal. Chem. 258 (1989) 89. [4] J.M. Feliu, A. Fernandezvega, A. Aldaz, J. Clavilier, J. Electroanal. Chem. 256 (1988) 149. [5] A.N. Haner, P.N. Ross, J. Phys. Chem. 95 (1991) 3740. [6] R. Gomez, A. Fernandezvega, J.M. Feliu, A. Aldaz, J. Phys. Chem. 97 (1993) 4769. [7] B. Beden, F. Kadirgan, C. Lamy, J.M. Leger, J. Electroanal. Chem. 127 (1981) 75. [8] M. Shibata, S. Motoo, J. Electroanal. Chem. 193 (1985) 217. [9] Y.Y. Yang, S.G. Sun, J. Phys. Chem. B 106 (2002) 12499. [10] W.F. Lin, M.S. Zei, M. Eiswirth, G. Ertl, T. Iwasita, W. Vielstich, J. Phys. Chem. B 103 (1999) 6968. [11] J. Clavilier, J.M. Feliu, A. Aldaz, J. Electroanal. Chem. 243 (1988) 419.
[12] E. Herrero, A. Rodes, J.M. Perez, J.M. Feliu, A. Aldaz, J. Electroanal. Chem. 393 (1995) 87. [13] B. Bittinscattaneo, T. Iwasita, J. Electroanal. Chem. 238 (1987) 151. [14] J. Sobkowski, K. Franaszczuk, A. Piasecki, J. Electroanal. Chem. 196 (1985) 145. [15] V. Stamenkovic, M. Arenz, B.B. Blizanac, K.J.J. Mayrhofer, P.N. Ross, N.M. Markovic, Surf. Sci. 576 (2005) 145. [16] H. Massong, S. Tillmann, T. Langkau, E.A.A. El Meguid, H. Baltruschat, Electrochim. Acta 44 (1998) 1379. [17] H. Massong, H.S. Wang, G. Samjeske, H. Baltruschat, Electrochim. Acta 46 (2000) 701. [18] X.Y. Xiao, S. Tillmann, H. Baltruschat, Phys. Chem. Chem. Phys. 4 (2002) 4044. [19] L.H. Jiang, G.Q. Sun, S.G. Sun, J.G. Liu, S.H. Tang, H.Q. Li, B. Zhou, Q. Xin, Electrochim. Acta 50 (2005) 5384. [20] S.A. Campbell, R. Parsons, J. Chem. Soc. -Faraday Trans. 88 (1992) 833. [21] S.G. Sun, A.C. Chen, T.S. Huang, J.B. Li, Z.W. Tian, J. Electroanal. Chem. 340 (1992) 213. [22] J. Clavilier, D. Armand, S.G. Sun, M. Petit, J. Electroanal. Chem. 205 (1986) 267. [23] D. Armand, J. Clavilier, J. Electroanal. Chem. 233 (1987) 251. [24] Y.Y. Yang, S.G. Sun, Y.J. Gu, Z.Y. Zhou, C.H. Zhen, Electrochim. Acta 46 (2001) 4339. [25] G.L. Kellogg, Phys. Rev. Lett. 55 (1985) 2168. [26] H. Niehus, Surf. Sci. 145 (1984) 407. [27] G. Stalnionis, L. Tamasauskaite-Tamasiunaite, V. Pautieniene, A. Sudavicius, Z. Jusys, J. Solid State Electrochem. 8 (2004) 892. [28] M.C. Santos, L.O.S. Bulhoes, Electrochim. Acta 48 (2003) 2607. [29] L.A. Kibler, A. Cuesta, M. Kleinert, D.M. Kolb, J. Electroanal. Chem. 484 (2000) 73. [30] I.M. Tidswell, N.M. Markovic, P.N. Ross, J. Electroanal. Chem. 376 (1994) 119. [31] X.H. Xia, H.D. Liess, T. Iwasita, J. Electroanal. Chem. 437 (1997) 233. [32] C. Lamy, S. Rousseau, E.M. Belgsir, C. Coutanceau, J.M. Leger, Electrochim. Acta 49 (2004) 3901. [33] J.M. Leger, S. Rousseau, C. Coutanceau, F. Hahn, C. Lamy, Electrochim. Acta 50 (2005) 5118. [34] F. Vigier, C. Coutanceau, F. Hahn, E.M. Belgsir, C. Lamy, J. Electroanal. Chem. 563 (2004) 81.
Electrochimica Acta 53 (2008) 6081–6088
Irreversible adsorption of Sn adatoms on basal planes of Pt single crystal and its impact on electrooxidation of ethanol Qing-Wei Zheng, Chun-Jie Fan, Chun-Hua Zhen, Zhi-You Zhou, Shi-Gang Sun ∗,1 State Key Laboratory of Physical Chemistry of Solid Surfaces, Department of Chemistry, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China Received 12 November 2007; received in revised form 18 January 2008; accepted 25 January 2008 Available online 5 February 2008
Abstract The behaviours of irreversible adsorption (IRA) of Sn adatoms on Pt(1 0 0), Pt(1 1 1) and Pt(1 1 0) electrodes were characterized using cyclic voltammetry. It has revealed that Sn can adsorb irreversibly on Pt(1 0 0) and Pt(1 1 1), while not significantly on Pt(1 1 0) electrode. Quantitative analysis of the relationship between 1 − θ H and θ Sn suggests that Sn adatoms may adsorb preferably on hollow sites of Pt(1 1 1) (threefold) and Pt(1 0 0) (fourfold) planes, which is in accordance respectively with the values 0.31 and 0.21 of coverage of IRA Sn adatoms in saturation adsorption determined on these electrodes. The IRA Sn adatoms on different basal planes of Pt single crystal yield different impact on the electrocatalytic oxidation of ethanol. It has revealed that the IRA Sn adatoms on Pt(1 0 0) electrode have declined the activity for ethanol oxidation, while IRA Sn adatoms on Pt(1 1 1) have enhanced remarkably the electrocatalytic activity with Sn coverage θ Sn between 0.09 and 0.18. The oxidation peak potential Ep and the current density jp of ethanol oxidation on Pt(1 1 1)/Sn were varied with θ Sn , and the highest jp (1258 A cm−2 ) as well as the lowest Ep (0.20 V) were measured simultaneously at θ Sn around 0.14. In comparison with the data obtained on a bare Pt(1 1 1), the Ep was shifted negatively by 65 mV and the jp has been enhanced to about 1.7 times on the Pt(1 1 1)/Sn (θ Sn = 0.14), which is ascribed to hydroxyl species adsorption at relatively low potentials on Pt(1 1 1)/Sn surfaces. The current study is of importance in revealing the fundamental aspects of modification of the basal planes of Pt single crystal using Sn adatoms, and the impact of this modification on electrocatalytic activity towards ethanol oxidation. © 2008 Elsevier Ltd. All rights reserved. Keywords: Irreversible adsorption; Sn adatoms; Pt(1 1 1); Pt(1 0 0); Pt(1 1 0); Ethanol electrooxidation
1. Introduction The study of the surface structure of Pt electrodes modified by metal adatoms and the role of adatoms in electrooxidation of small organic molecules is of both fundamental and technological importance in electrocatalysis and fuel cell applications [1]. It is well known that the modification of the electrode surface with adatoms such as Ru, Sn, Pb, As, Sb, Bi, etc. can enhance the catalytic properties of Pt electrode [2–6]. Different techniques to modify electrode surface using adatoms were developed [7], among them the under potential deposition (UPD) [1] and the irreversible adsorption (IRA) [4,6] are most convenient and were frequently employed. The presence of adatoms formed from
∗ 1
Corresponding author. Tel.: +86 592 2180181; fax: +86 592 2183047. E-mail address: [email protected] (S.-G. Sun). ISE member.
0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2008.01.078
UPD or IRA procedure [2,4,5] on electrode surface can modify the electronic and chemical properties of substrates, and as a consequence improve the catalytic properties of electrodes [8–12]. The electrochemical properties of Sn adlayers on Pt have been studied so far mostly on polycrystalline surfaces [13,14]. For a better understanding of the metallic adlayer structures, single crystal electrodes with well-defined atomic arrangement have been introduced. Stamenkovic et al. [15] have characterized bimetallic single crystals of Pt3 Sn(1 1 0) and Pt3 Sn(1 1 1) in ultra high vacuum by using Auger electron spectroscopy (AES), low energy ion scattering spectroscopy (LEISS) and low energy electron diffraction (LEED). Following characterization in the UHV, the Pt3 Sn(1 1 0) and Pt3 Sn(1 1 1) were transferred into an electrochemical cell to study their electrocatalytic properties for CO adsorption and oxidation by using in situ infrared spectroscopy. They demonstrated that continuous oxidative removal of adsorbed CO starts as low as E < 0.1 V (vs. RHE), which is an
6082
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
important property for CO-tolerant catalysts. Electronic effects, surface structure and intermolecular repulsion between adsorbed species are ascribed for the high catalytic activity of Pt3 Sn(h k l) alloys. To the best of our knowledge, the studies so far concerning modification of Pt single crystal electrodes using Sn adatoms were mainly focused on Pt (1 1 1) plane and employing UPD method [16–18]. Massong et al. [16] reported the electrochemical properties of UPD Sn on Pt(1 1 1) at potentials below 0.8 V and with low Sn coverage. A specific reversible redox current peak around 0.6 V (vs. RHE) was observed, ascribing to a surface redox processes of Sn adatoms on Pt(1 1 1) similar to those previously observed for As, Bi and other adatoms. The eps number (electrons transferred per surface site as obtained from the corresponding suppression of hydrogen adsorption) was determined at 1, which was interpreted to a one-electron process of hydroxyl species adsorption. Recently, the PtSn alloy nanomaterials were attractive and used as catalysts in electrooxidation of ethanol [17–19]. In the UPD experiments, the metal ions are always present in solutions, which will certainly influence the precise determination of adatom coverage and introduce complexity in studying electrochemical properties of the adatom modified electrodes [1]. In contrast with UPD, the irreversible adsorption presents advantages consisting in that the adsorption of adatoms can be done outside the electrochemical cell and does not require applying a reduction potential [6]. After the adsorption, the electrode is transferred into an electrochemical cell that contains a study solution without metal ions, so that the adatom coverage can be determined precisely, and the surface processes as well as the electrocatalytic properties of the adatom modified electrode may be finely investigated. Besides the UPD process on Pt single crystal electrode, Sn also shows irreversible adsorption properties [1]. Haner et al. [5] studied the geometric and electronic effects of tin adatoms by depositing spontaneously tin on Pt single crystal surfaces, but they did not observe any enhancement of activity for methanol oxidation. Similar results of irreversibly adsorbed tin on Pt single crystal electrodes were also reported by Campbell et al. [20]. In order to investigate further the surface modification process of Sn adatoms on Pt single crystal electrodes, three basal planes of Pt single crystal, Pt(1 1 1), Pt(1 0 0) and Pt(1 1 0) were prepared and modified by IRA Sn adatoms in the present study. The oxidation of ethanol was employed as a probe reaction to test the electrocatalytic properties of the Pt(h k l)/Sn electrodes. The study has put emphasis upon the surface processes of Pt single crystal basal planes towards Sn adatom modification, and the surface structure effects in electrocatalysis as well. 2. Experimental Pt(1 0 0), Pt(1 1 0) and Pt(1 1 1) were prepared in our laboratory through the method described previously [21]. Before each measurement the Pt single crystal electrodes were treated using Clavilier’s method [22], i.e. they were annealed in a hydrogen–oxygen flame, quenched with super pure water and transferred into electrochemical cell under the protection of a droplet of pure water. After a well-defined cyclic voltammo-
gram that characterizes the surface structure of each electrode has been recorded in cell 1 that contains 0.5 M H2 SO4 solutions, the Pt single crystal electrode was immersed in 1 M H2 SO4 solution containing 10−3 M Sn2+ ions for 1 min, then rinsed with super pure water and transferred back to cell 1 or cell 2 (perchloric acid solutions) at 0.0 V (vs. SCE). Coverage of IRA Sn adatoms (Snad ) in saturation adsorption was thus formed and determined. Starting from the saturation coverage of Snad , different sub-coverage of Snad was obtained by partially stripping Snad in cell 1 by applying a votammetry in controlling the upper limit of potential scan (Eu ) and the number of potential cycling. The calibration of the designated coverage of Snad was then carried out in cell 2, and the electrode was transferred finally into cell 3 that contains ethanol solution for further electrocatalytic study. The solutions were prepared using Millipore water (18.0 M cm) provided from a Milli-Q Lab apparatus (Nihon, Millipore Ltd.), super pure H2 SO4 or HClO4 , ethanol and SnSO4 of analytical grade. A saturated calomel electrode (SCE) was served as reference electrode. Potentials in this paper were reported versus the SCE scale. The solution was deaerated by bubbling pure N2 gas before experiment, and kept a flux of N2 over it during measurements to prevent possible interference of oxygen and impurities from the atmosphere. All experiments were carried out at room temperature around 20 ◦ C. 3. Results and discussion 3.1. Electrochemical characterization of surface modification of Pt(1 1 1), Pt(1 1 0) and Pt(1 0 0) with IRA Sn adatoms The adsorption and desorption of hydrogen is highly sensitive to the atomic arrangement of Pt single crystal electrodes, and has been conveniently employed to characterize the surface structure of the electrode. It is known that the surface structure of Pt single crystal electrodes remains stable unless a significant amount of oxygen is adsorbed on the surface, and a well-defined structure of Pt single crystal surface can be maintained in 0.5 M H2 SO4 solution at electrode potentials below 0.75 V. Fig. 1a shows cyclic voltammograms of Pt(1 1 1) (dot lines, Eu = 0.75 V) and the Snad modified electrode Pt(1 1 1)/Sn (solid lines, varying Eu ) recorded in 0.5 M H2 SO4 solution. It can observe that the electrochemical adsorption of hydrogen yields a current plateau between −0.20 and 0.0 V, and the sulfate adsorption gives a ‘butterfly’ peak in the potential region from 0.07 to 0.30 V on bare Pt(1 1 1), signifying a well-defined order structure of Pt(1 1 1). On the Pt(1 1 1)/Sn of saturation adsorption IRA Snad , the sulfate adsorption current is completely suppressed, and the hydrogen adsorption is inhibited to a large extent. Along with increasing progressively Eu , two oxidation current peaks appear respectively around 0.33 and 0.60 V in the positive going potential scan (PGPS), attributing respectively to adsorption of hydroxyl species [16,18] and Snad oxidation. The desorption of hydroxyl species in the negative-going potential scan (NGPS) gives rise to a sharp current peak near 0.30 V with Eu at 0.45 V, while the reduction of oxidized Snad adatoms occurs in sev-
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
Fig. 1. Cyclic voltammograms of Pt(1 1 1), Pt(1 0 0), Pt(1 1 0) (dashed lines) and Pt(1 1 1)/Sn, Pt(1 0 0)/Sn, Pt(1 1 0)/Sn (solid lines) electrodes in 0.5 M H2 SO4 solution. Initial voltammogram (- - -); recovery voltammogram after stripping Snad (· · ·), sweep rate 50 mV s−1 .
eral broad current peaks lying on potential region between 0.6 and 0.1 V when Eu is above 0.5 V. It can be seen that the Sn adatoms are stable on Pt(1 1 1) in 0.5 M H2 SO4 at potentials below 0.45 V, for which the hydrogen adsorption current is maintained; while Sn adatoms can be progressively stripped from Pt(1 1 1) when Eu is increased beyond 0.45 V, as evidenced by the progressive increase of hydrogen adsorption current due to the recovery of surface sites for hydrogen adsorption. This result demonstrates that the surface concentration of Snad can be con-
6083
veniently controlled under electrochemical conditions. When Snad has been completely stripped out by applying continuously potential cycling between −0.20 and 0.75 V, a pair of peaks close to −0.12 V appears in the voltammogram, corresponding to hydrogen adsorption on surface sites lying on steps of (1 1 1) symmetry. This result indicates that the Pt(1 1 1) surface has been disturbed from well-defined order structure. This perturbation is also evidenced by the disappearance of the sharp spike in the ‘butterfly’ peak at 0.21 V, which is a stamp character of sulfate adsorption on well-defined order structure of Pt(1 1 1). The CV features of bare Pt(1 0 0) and Sn modified Pt(1 0 0) electrode are shown in Fig. 1b. The current peaks of hydrogen adsorption–desorption at 0.02 and 0.12 V observed on bare Pt(1 0 0) have been inhibited almost completely when Sn adsorbed in saturation on Pt(1 0 0) electrode. Along with increasing progressively Eu , the blockage of current of hydrogen adsorption–desorption by Snad is maintained stable till Eu is higher than 0.45 V, illustrating that the Sn adatoms are also stable on Pt(1 0 0) below 0.45 V. When Eu is increased to above 0.45 V, we observe only in the PGPS an oxidation current peak near 0.58 V that is ascribed to Snad oxidation. The reduction of oxidized Snad occurs in a reduction current peak near 0.43 V in the NGPS. It is interesting to see that the current peak around 0.33 V due to hydroxyl species adsorption observed on Pt(1 1 1)/Sn electrode is missing on Pt(1 0 0)/Sn, which may yield different impact on electrocatalytic properties of these electrodes, and it will be described thereafter. It is evident that Snad can be desorbed progressively from Pt(1 0 0) surface when Eu is above 0.45 V in 0.5 M H2 SO4 solution. As a consequence, the current of hydrogen adsorption–desorption is increasing with the recovery of surface sites for hydrogen adsorption. It is worthwhile to note that the current peaks appeared in the voltammograms recorded during stripping out Sn adatoms are quite different from the sharp current peaks observed in the voltammogram of bare Pt(1 0 0) surface. A pair of broad peak lying on from 0.0 to 0.2 V centered at 0.05 V appears in the voltammogram of Pt(1 0 0)/Sn when part Snad has been stripped out. When Sn adatoms have been completely removed by continuously cycling electrode potential between −0.20 and 0.75 V, two pairs of hydrogen adsorption–desorption current peaks appear again. However, these two peaks are much smaller than those in the voltammogram of the bare Pt(1 0 0), and only 68% of surface sites for hydrogen adsorption has been recovered. The results suggest that Pt(1 0 0) surface has been also disturbed from welldefined atomic arrangement through the adsorption of Snad . In the case of Pt(1 1 0) electrode (Fig. 1c), we can hardly observe neither significant inhibition of the current peak of hydrogen adsorption–desorption after the electrode subjected to the same procedure of IRA Sn adatoms adsorption, nor the remarkable Snad oxidation current peak near 0.60 V as observed on Pt(1 1 1)/Sn and Pt(1 0 0)/Sn. These results indicate that Sn may not be irreversibly adsorbed on Pt(1 1 0) surface under present conditions. It is well-known [22–24] that the surface of Pt(1 1 0) is readily reconstructed into a (1 × 2) structure under electrochemical conditions. Two models for the Pt(1 1 0)-(1 × 2) structure are generally accepted. One is the so called missing row model [25], on which the surface sites are of (1 1 1) sym-
6084
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
metry; the other is the Bronzel–Ferrer model [26] that give rise to some small 100 facets of short-range order. In both models, it does not exist hollow sites of long-range order on the Pt(1 1 0)(1 × 2) surface. In contrast, the long-range ordered hollow sites of threefold or fourfold could be found on Pt(1 1 1) or on Pt(1 0 0) surfaces, respectively, which may suggest that the irreversible adsorption of Sn adatoms has been conducted on the hollow sites of long-range order on these two single crystal planes. The surface site occupancy by Snad (or the coverage of Snad , θ Sn ) and the coverage of hydrogen (θ H ) can be calculated from the comparison of hydrogen adsorption charge (QSn H ) with the charge of Snad oxidation (Qox Sn ) using the following equations, Qox /4 NSn = SnS NPt QH
(1.1)
NH QSn = HS NPt QH
(1.2)
θSn = and θH =
NH , NSn and NPt are, respectively, the number of H adatoms, the number of Sn adatoms and the number of surface Pt sites. QSH and QSn H are charges corresponding to hydrogen adsorption on electrodes without and with Snad modification, respectively. The ox QSH , QSn H and QSn are measured by integration of corresponding CV curves. According the previous work of UPD Sn adsorbed on Pt electrode studied through XPS characterization [27], two kinds of tin species were identified on Pt surface, namely, metallic tin and little tin oxy-compounds. Metallic tin was deposited spontaneously on Pt surface due to disproportionation of Sn(II) to Sn(IV), while tin oxide/hydroxide species were derived as a result of Sn(II) complex hydrolysis. It is also known, from cyclic votammetry and electrochemical quartz crystal microbalance (EQCM) studies [28], that four electrons are involved in Sn adatoms oxidation on Pt electrodes, i.e. Snad → Snad 4+ . On account of the experiment procedures used in the present study, the irreversible adsorption of Sn2+ may produce mostly metallic tin and little tin oxide/hydroxide species on Pt(h k l) surfaces. At low potentials all IRA Sn adatoms are reduced into metallic tin, and are oxidized to Sn(IV) species at 0.75 V under present CV conditions. As a consequence, the term Qox Sn /4 in Eq. (1.1) is in direct proportion to NSn . Table 1 lists the electrochemical parameters measured from CVs recorded on Pt(1 1 1), Pt(1 0 0) and Pt(1 1 0) electrodes under saturation adsorption of IRA adatoms. The Pt single crystal electrodes were subjected initially to different pretreatments, Table 1 Parameters of IRA Sn on Pt(h k l) electrodes subjected to different treatments QH /C cm−2
−2 QSn H /μC cm
−2 Qox Sn /μC cm
S θSn
Pt(1 1 1)
Air Ar + H2
233.2 246.9
29.32 42.25
282.4 336.7
0.31 0.34
Pt(1 0 0)
Air Ar + H2
265.8 227.2
50.3 71.7
224.5 251.4
0.21 0.27
Pt(1 1 0)
Air Ar + H2
189.4 219.2
177.6 173.1
2.08 22.36
0.003 0.02
i.e. after annealing in hydrogen–oxygen flame they were cooled in air (oxidant atmosphere) or in Ar + H2 stream (reductant atmosphere). It is known that cooling in Ar + H2 stream will produce a well-defined Pt single crystal surface with less defects [29,30], therefore more hollow sites in long-range order are expected on this surface. As a consequence, it will favor the irreversible adsorption of Sn. As demonstrated in Table 1, the saturation coverage of Sn adatoms on any one of the three electrodes cooling down in Ar + H2 stream is significantly larger than that measured on the same electrode cooling down in air. The results have demonstrated clearly that the long-range order of hollow sites is necessary for IRA Sn adatoms. It is worthwhile to note that, the CV features of Sn adatoms are much pronounced in perchloric acid than in sulphuric acid, and that without the competition adsorption of sulphuric acid anions the Sn adatoms are much stale in perchloric acid. Actually, the Sn adatoms can be easily removed by oxidation at potentials above 0.45 V in H2 SO4 solution, while they could not be stripped in HClO4 solution even at potential as high as 0.75 V. In order to avoid the interference of sulfate adsorption and to study further the surface processes of the IRA Sn adatoms, 0.535 M HClO4 solution that has similar pH value of 0.5 M H2 SO4 solution was used, and the cyclic voltammetric results are illustrated in Fig. 2. As can be seen in Fig. 2a, the hydroxyl species adsorption on Pt(1 1 1)/Sn gives rise to a sharp peak at 0.34 V in the PGPS for a Snad coverage of 0.19, and the desorption yields also a sharp current peak at 0.32 V in the NGPS when the Eu sets at 0.42 V. The oxidation of Snad for this coverage produces a broad peak around 0.57 V in the PGPS, and the reduction of the oxidized Snad gives rise to two broad peaks around 0.34 and 0.27 V in the NGPS. It is interesting to note that the Sn adatoms on Pt(1 1 1) are more stable in perchloric acid solution, which is evidenced by the stable hydrogen adsorption–desorption current recorded with continuously cycling electrode potential between −0.20 and 0.75 V. As the Snad is difficult to remove from Pt(1 1 1) surface (it is also difficult to remove from Pt(1 0 0) surface) in perchloric acid, in the experiments thereafter to prepare different Snad coverage starting from the saturation adsorption of IRA Sn adatoms, the partly stripping out Snad was done in 0.5 M H2 SO4 , the electrode was then transferred into perchloric acid solution to study the surface process or into a solution containing 0.1 M ethanol to investigate the electrocatalytic properties of a defined Snad coverage. At a high Snad coverage (0.24), the current peak of hydroxyl species adsorption is positively shifted to 0.39 V with significant decrease in the amplitude, the current peak of Snad oxidation in the PGPS remains broad and locates around 0.57 V, and the reduction of the oxidized Snad occurs again in broad peaks lying on between 0.60 and 0.18 V in the NGPS. When the Snad coverage is larger than 0.27, the hydroxyl adsorption current peak is disappeared completely, as shown in the voltammogram for a Snad coverage of 0.29, in which only a much broad current peak of Snad oxidation from 0.28 to 0.75 V centered around 0.52 V is observed in the PGPS. In this case the reduction of the oxidized Snad can be seen from 0.60 to 0.0 V in the NGPS and appears a peak near 0.36 V. It is obvious that the hydroxyl species could not
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
6085
tion current of adsorbed Sn adatoms is observed around 0.4 V (Fig. 2c). The saturation coverage of Sn adatoms on Pt(1 1 0) in perchloric acid is determined at 0.02, which is larger than the value measured in sulphuric acid solutions (as listed in Table 1). S measured on Pt(1 1 1) and on Pt(1 0 0) in However, the values θSn perchloric acid are respectively 0.31 and 0.21, which are close to the values reported in Table 1 measured in sulphuric acid solutions. The coadsorption properties of Sn with H on Pt(1 1 1) surface may be characterized by the relationships between 1 − θ H and θ Sn . Fig. 3a plots the variation of 1 − θ H and θ Sn obtained in 0.1 M HClO4 solution, since the study of electrocatalytic properties of Sn adatoms modified Pt single crystal electrode was conducted in 0.1 M HClO4 + 0.1 M CH3 CH2 OH solution, and it has tested that the results are the same as those obtained in 0.535 M HClO4 solution. As stated above, when the θ Sn is below 0.27 the hydroxyl species adsorption will take place, which will cause complexity in analysis of the coadsorption of Sn and H. Therefore, only experimental data for θ Sn > 0.27 were plotted in Fig. 3a. A straight line is observed, and the slope is measured at 2.88 that can be served to estimate the number (n) of
Fig. 2. Cyclic voltammograms of Pt(1 1 1), Pt(1 0 0), Pt(1 1 0) (dashed lines) and Pt(1 1 1)/Sn, Pt(1 0 0)/Sn, Pt(1 1 0)/Sn (solid lines) electrodes in 0.535 M HClO4 solution, sweep rate 50 mV s−1 .
adsorb on Pt(1 1 1)/Sn electrodes with Snad coverage higher than 0.27. Similar CV features are observed on Pt(1 0 0)/Sn electrode in the sulfuric acid and perchloric acid solutions, i.e. only the oxidation current of Snad and the reduction current of the oxidized Snad appear in the voltammogram (Fig. 2b). The oxidation current peak is near 0.58 V in the PGPS and the reduction current peak is around 0.42 V in the NGPS. This CV feature does not depend on the Snad coverage, confirming clearly that it does not exist hydroxyl species adsorption on Pt(1 0 0)/Sn electrode. In the case of Pt(1 1 0) electrode, a small, but not negligible oxida-
Fig. 3. Relationship between 1 − θ H and θ Sn on Pt(1 1 1)/Sn (a) and Pt(1 0 0)/Sn (b). Data obtained in 0.1 M HClO4 solution.
6086
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
Pt(1 1 1) surface sites for hydrogen adsorption blocked per Sn adatom. The slope value of 2.88 suggests that, at high Snad coverage or in saturation adsorption of Sn adatoms on Pt(1 1 1), one Sn adatom may inhibit approximately three surfaces sites for hydrogen adsorption. This result confirms once again that the Sn adatoms are adsorbed on the threefold hollow sites of the Pt(1 1 1) surface, which is in accordance with the coverage of IRA Snad in saturation adsorption determined at 0.31. As it does not exist the adsorption of hydroxyl species on Pt(1 0 0)/Sn electrode, it is convenient to plot all data of (1 − θ H ) versus θ Sn . As illustrated in Fig. 3b, a linear relationship can be maintained for θ Sn varying from 0.022 to 0.182, and the slope of the straight line is measured at 3.10. This result indicates that in most θ Sn studied, by averaging, one Sn adatom may inhibit approximately three surfaces sites for hydrogen adsorption. Such value may suggests that most Sn adatoms adsorb on the fourfold hollow sites of Pt(1 0 0) for which one Sn adatom occupies 4 surface sites, and that some Sn adatoms may also adsorb on the bridge and atop sites of Pt(1 0 0) surface. The coverage of IRA Sn adatoms in saturation adsorption on Pt(1 0 0) has been determined at 0.21, which is close to the value of 0.25 for an adlayer of Sn occupying the fourfold hollow sites on Pt(1 0 0) surface. 3.2. Electrocatalytic properties of Pt(1 1 1) and Pt(1 0 0) modified with IRA Sn adatoms towards ethanol oxidation The oxidation of ethanol is employed as a probe reaction to test the electrocatalytic activity of platinum single crystal electrodes modified with IRA Sn adatoms. As Sn does not adsorb significantly on Pt(1 1 0) through IRA procedure under present conditions, only Pt(1 1 1) and Pt(1 0 0) were studied in this section. In this study, the θ H was calibrated at first in the cell containing 0.1 M HClO4 solution, the electrode was then transferred into another electrochemical cell containing 0.1 M HClO4 + 0.1 M C2 H5 OH solution for investigating ethanol oxidation. Fig. 4a compares the voltammograms of ethanol oxidation on a Pt(1 1 1)/Sn (θ Sn = 0.15) and on the bare Pt(1 1 1) electrodes. The oxidation of ethanol on bare Pt(1 1 1) starts at 0.06 V and gives rise to a current peak near 0.25 V. When the Pt(1 1 1) electrode is covered by Snad at high coverage, the current peak is mostly inhibited. Along with partly desorption of Snad from the electrode, i.e. decreasing θ H , the oxidation current peak is recovered and even larger than that on the bare Pt(1 1 1) electrode as shown in Fig. 4a for a Snad coverage of 0.15. The significant enhanced catalytic activity of Pt(1 1 1)/Sn for Snad coverage of 0.15 in Fig. 4a consists in that the oxidation of ethanol is shifted negatively about 60 mV. The onset potential (Eonset ) of ethanol oxidation is shifted negatively to 0.0 V on the Pt(1 1 1)/Sn; the potential of the current peak in the PGPS (Ep ) is also shifted negatively to 0.2 V on Pt(1 1 1)/Sn. Moreover, the current intensity has been also enhanced to some extent. The intensity current peak (jp ) in the PGPS is measured at 809 A cm−2 on bare Pt(1 1 1), and it is increased to 1200 A cm−2 on the Pt(1 1 1)/Sn electrode. The CV feature that the profile of the j-E curve in the NGPS is almost the same as
Fig. 4. j–E curves of electrocatalytic oxidation of ethanol on Pt(1 1 1) and Pt(1 0 0) (dashed lines); Pt(1 1 1)/Sn and Pt(1 0 0)/Sn (solid lines) in 0.1 M C2 H5 OH + 0.1 M HClO4 .
that in the PGPS suggests that less self-poisoning phenomenon is encountered on the Pt(1 1 1) electrode, which has been also evidenced by studies of in situ FTIR spectroscopy [31]. The enhancement of catalytic activity upon the modification of Sn adatoms on Pt(1 1 1) has been suggested mainly coming from the adsorption of hydroxyl species on Snad modified Pt(1 1 1) surface [32–34]. In contrast to the enhancement of catalytic activity, the Pt(1 0 0)/Sn illustrates a declined activity in comparison with that of the bare Pt(1 0 0). We observe that the current of ethanol oxidation is almost zero when electrode potential is below 0.2 V on bare Pt(1 0 0) electrode, due to the surface poisoning by poison intermediates [31]. When electrode potential is increased to above 0.2 V, visible oxidation current is observed and a sharp current peak appears near 0.47 V with intensity of 6.01 mA cm−2 in the PGPS, which is ascribed to the oxidation of poison species together with ethanol on Pt(1 0 0). A broad oxidation current peak in the NGPS appears at 0.44 V of intensity 1.75 mA cm−2 . However, when the Pt(1 0 0) is modified with Sn adatoms, the intensities of the current peaks in both the PGPS and the NGPS have been declined respectively to 1.84 and 0.87 mA cm−2 , although the Ep in the PGPS is shifted negatively to 0.45 V. The fact that the oxidation current in the potential region from
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
6087
Fig. 6. j–E curves of electrocatalytic oxidation of ethanol on Pt(1 0 0)/Sn at different coverage of Snad in 0.1 M C2 H5 OH + 0.1 M HClO4 .
for which the jp reaches the largest value of 1245 A cm−2 , and the Ep is at the lowest position of 0.20 V. In comparison with the bare Pt(1 1 1) electrode, the Ep has been shifted negatively about 65 mV, and the jp has been enhanced to about 1.7 times of that measured on a bare Pt(1 1 1) surface. When Snad is fully stripped from Pt(1 1 1) electrode, the j–E profile of ethanol oxidation is quite similar to that recorded on the bare Pt(1 1 1) but with a slightly small current intensity. In Fig. 6, the j–E curves recorded in the PGPS for ethanol oxidation on Pt(1 0 0)/Sn electrodes at various θ H are compared. As illustrated previously in Fig. 4b, the Sn adatoms on Pt(1 0 0) present a decline effect in catalytic activity for ethanol oxidation, we observe in Fig. 6 that the jp is decreased monotony with the increase of θ H . It can be seen that the j–E curve of Pt(1 0 0)/Sn electrode after completely stripping out Snad is much different from that of the initial bare Pt(1 0 0), and the jp is only 0.852 mA cm−2 , i.e. about one seventh of that measured on the initial bare Pt(1 0 0) electrode. 4. Conclusions
Fig. 5. (a) j–E curves of electrocatalytic oxidation of ethanol on Pt(1 1 1)/Sn at different coverage of Snad in 0.1 M C2 H5 OH + 0.1 M HClO4 . The variations of the peak potential (b) and the maximum current intensity (c) of C2 H5 OH oxidation with θ Sn on Pt(1 1 1)/Sn.
0.1 to 0.4 V in the PGPS is always much smaller than that in the NGPS indicates a severe self-poisoning on both bare and Sn adatoms modified Pt(1 0 0) surface [31]. Fig. 5a displays j–E curves of ethanol oxidation in the PGPS on Pt(1 1 1) and Pt(1 0 0) modified with Sn adatoms of different coverage. It can observe on Pt(1 1 1)/Sn electrodes that both Ep and jp vary with θ H . The variations of Ep and jp with θ H are plotted in Fig. 5b and c, respectively. It can be determined that the best θ H that yields the maximum enhancement of catalytic activity is 0.14 for the oxidation of ethanol on Pt(1 1 1)/Sn electrode,
The results obtained in the present study demonstrated that the irreversible adsorption of Sn adatoms on basal planes of Pt single crystal is sensitive to the surface structure of Pt single crystal electrode. The CV features of Pt single crystal electrodes modified with IRA Sn adatoms are varied with surface atomic arrangement. It has revealed that Sn can adsorb irreversibly on Pt(1 0 0) and Pt(1 1 1) but not significantly on Pt(1 1 0) electrode under present conditions. The IRA Sn adatoms on Pt(1 1 1) and Pt(1 0 0) are stable at potentials below 0.45 V in sulfuric acid solutions, while they are more stable in perchloric acid medium till 0.75 V. It has observed solely on Pt(1 1 1)/Sn a specific redox peak in the potential region around 0.35 V, which is attributed to the adsorption of hydroxyl species. Based on quantitative analysis of relationship between 1 − θ H and θ Sn , Sn is proposed to adsorb on the threefold hollow sites of Pt(1 1 1) surface and mainly on the fourfold hollow sites of Pt(1 0 0) plane. The coverages of IRA Sn adatoms in saturation adsorption on Pt(1 1 1) and Pt(1 0 0) were determined respectively 0.31 and 0.21, which are
6088
Q.-W. Zheng et al. / Electrochimica Acta 53 (2008) 6081–6088
in accordance with that the Sn adatoms adsorb preferentially on hollow sites of these surfaces. It has determined a decline effect of the IRA Sn adatoms for ethanol electrocatalytic oxidation on Pt(1 0 0), while the IRA Sn adatoms on Pt(1 1 1) has enhanced significantly the electrocatalytic activity. The oxidation peak potential Ep and the current density jp of ethanol oxidation on Pt(1 1 1)/Sn were varied with Sn coverage θ Sn , and the highest jp as well as the lowest Ep were simultaneously determined at θ Sn of 0.14. The fact that hydroxyl species could adsorb at relative low potentials on Pt(1 1 1)/Sn electrode presents an interpretation of the enhanced electrocatalytic activity of Pt(1 1 1)/Sn for ethanol oxidation. Acknowledgement The study was supported by Natural Science Foundation of China (grants 20673091, 20433060). References [1] E. Herrero, L.J. Buller, H.D. Abruna, Chem. Rev. 101 (2001) 1897. [2] R.R. Adzic, A.V. Tripkovic, N.M. Markovic, J. Electroanal. Chem. 150 (1983) 79. [3] J. Clavilier, A. Fernandezvega, J.M. Feliu, A. Aldaz, J. Electroanal. Chem. 258 (1989) 89. [4] J.M. Feliu, A. Fernandezvega, A. Aldaz, J. Clavilier, J. Electroanal. Chem. 256 (1988) 149. [5] A.N. Haner, P.N. Ross, J. Phys. Chem. 95 (1991) 3740. [6] R. Gomez, A. Fernandezvega, J.M. Feliu, A. Aldaz, J. Phys. Chem. 97 (1993) 4769. [7] B. Beden, F. Kadirgan, C. Lamy, J.M. Leger, J. Electroanal. Chem. 127 (1981) 75. [8] M. Shibata, S. Motoo, J. Electroanal. Chem. 193 (1985) 217. [9] Y.Y. Yang, S.G. Sun, J. Phys. Chem. B 106 (2002) 12499. [10] W.F. Lin, M.S. Zei, M. Eiswirth, G. Ertl, T. Iwasita, W. Vielstich, J. Phys. Chem. B 103 (1999) 6968. [11] J. Clavilier, J.M. Feliu, A. Aldaz, J. Electroanal. Chem. 243 (1988) 419.
[12] E. Herrero, A. Rodes, J.M. Perez, J.M. Feliu, A. Aldaz, J. Electroanal. Chem. 393 (1995) 87. [13] B. Bittinscattaneo, T. Iwasita, J. Electroanal. Chem. 238 (1987) 151. [14] J. Sobkowski, K. Franaszczuk, A. Piasecki, J. Electroanal. Chem. 196 (1985) 145. [15] V. Stamenkovic, M. Arenz, B.B. Blizanac, K.J.J. Mayrhofer, P.N. Ross, N.M. Markovic, Surf. Sci. 576 (2005) 145. [16] H. Massong, S. Tillmann, T. Langkau, E.A.A. El Meguid, H. Baltruschat, Electrochim. Acta 44 (1998) 1379. [17] H. Massong, H.S. Wang, G. Samjeske, H. Baltruschat, Electrochim. Acta 46 (2000) 701. [18] X.Y. Xiao, S. Tillmann, H. Baltruschat, Phys. Chem. Chem. Phys. 4 (2002) 4044. [19] L.H. Jiang, G.Q. Sun, S.G. Sun, J.G. Liu, S.H. Tang, H.Q. Li, B. Zhou, Q. Xin, Electrochim. Acta 50 (2005) 5384. [20] S.A. Campbell, R. Parsons, J. Chem. Soc. -Faraday Trans. 88 (1992) 833. [21] S.G. Sun, A.C. Chen, T.S. Huang, J.B. Li, Z.W. Tian, J. Electroanal. Chem. 340 (1992) 213. [22] J. Clavilier, D. Armand, S.G. Sun, M. Petit, J. Electroanal. Chem. 205 (1986) 267. [23] D. Armand, J. Clavilier, J. Electroanal. Chem. 233 (1987) 251. [24] Y.Y. Yang, S.G. Sun, Y.J. Gu, Z.Y. Zhou, C.H. Zhen, Electrochim. Acta 46 (2001) 4339. [25] G.L. Kellogg, Phys. Rev. Lett. 55 (1985) 2168. [26] H. Niehus, Surf. Sci. 145 (1984) 407. [27] G. Stalnionis, L. Tamasauskaite-Tamasiunaite, V. Pautieniene, A. Sudavicius, Z. Jusys, J. Solid State Electrochem. 8 (2004) 892. [28] M.C. Santos, L.O.S. Bulhoes, Electrochim. Acta 48 (2003) 2607. [29] L.A. Kibler, A. Cuesta, M. Kleinert, D.M. Kolb, J. Electroanal. Chem. 484 (2000) 73. [30] I.M. Tidswell, N.M. Markovic, P.N. Ross, J. Electroanal. Chem. 376 (1994) 119. [31] X.H. Xia, H.D. Liess, T. Iwasita, J. Electroanal. Chem. 437 (1997) 233. [32] C. Lamy, S. Rousseau, E.M. Belgsir, C. Coutanceau, J.M. Leger, Electrochim. Acta 49 (2004) 3901. [33] J.M. Leger, S. Rousseau, C. Coutanceau, F. Hahn, C. Lamy, Electrochim. Acta 50 (2005) 5118. [34] F. Vigier, C. Coutanceau, F. Hahn, E.M. Belgsir, C. Lamy, J. Electroanal. Chem. 563 (2004) 81.