Electrocatalytic properties of Au electrodes decorated with Pt submonolayers by galvanic displacement of copper adatoms

Electrochimica Acta 130 (2014) 351–360

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Electrocatalytic properties of Au electrodes decorated with Pt submonolayers by galvanic displacement of copper adatoms B.I. Podlovchenko ∗ , Yu. M. Maksimov, K.I. Maslakov Department of Chemistry, Moscow State University, Moscow 119992, Leninskie Gory, Russia

a r t i c l e

i n f o

Article history: Received 14 October 2013 Received in revised form 26 February 2014 Accepted 27 February 2014 Available online 11 March 2014 Keywords: Au electrode Galvanic displacement of Cu adatoms Pt submonolayers MOR FAOR

a b s t r a c t Submonolayer Pt0 coatings are synthesized by the displacement of copper adatoms (Cuad ) on polycrystalline (pc) Au in PtCl4 2− solutions (supporting electrolyte 0.5 М Н2 SO4 ). Cyclic voltammograms (CVA) measured on Pt0 Au demonstrate that for initial surface coverages ␪Cu < 0.5, the Pt0 adatoms that have displaced Cuad are mainly deposited in the first layer. At large surface coverages, the agglomeration of Pt0 and its partial deposition into the second layer take place; ∼30% of Au surface remains free. SEM, XPS and Auger-spectroscopic data confirm the absence of multilayer Pt0 agglomerates. The specific rates of CН3 OН electrooxidation (per cm2 of Pt) on Pt0 Au turn out to be much lower than on pc Pt0 . This is explained by the fact that dehydrogenation of CН3 OН requires the presence of «areas» formed by a large number of platinum atoms (> 3). For НCOOН, the strong increase (more than by one order of magnitude) in both non-steady-state and steady-state specific electrooxidation currents is observed. It is assumed that new active sites are formed for the current-determining reaction that proceeds through the one-site adsorption of НCOOН molecules. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Platinum is the best catalyst for many chemical and electrochemical reactions including reactions taking place in fuel cells. However, the high cost and shortage of platinum make scientists seek new ways for reducing the platinum content in catalysts (electrocatalysts) and this problem was always challenging [1–3]. One of the ways to solve it is to synthesize bi- and multicomponent catalysts that in addition to platinum include the cheaper and/or less deficient metals. Such bi- and polymetallic structures make it possible to reduce the consumption of platinum and often also exhibit the higher catalytic activity in reactions important for fuel cells such as, e.g., electrooxidation of HCO compounds (including CO). To date, the main regularities of electrooxidation of CН3 OН and НCOOН in acidic solutions on Pt electrodes are already elucidated [4–15]. The НCOOН electrooxidation in the low potential range which is of interest for fuel cells (<0.5 V vs. reversible hydrogen electrode in the same solution) is mainly limited by the rate of dehydrogenation HCOOH molecules. This process is inhibited by accumulation of strongly chemisorbed species (first of all, COads ) on the electrode surface. The current-determining reactions of CH3 OH electrooxidation are as follows: (i) at low potentials, the

∗ Corresponding author. Tel.: +7 495 939 4027; fax: +7 495 932 8846. E-mail address: [email protected] (B.I. Podlovchenko). http://dx.doi.org/10.1016/j.electacta.2014.02.148 0013-4686/© 2014 Elsevier Ltd. All rights reserved.

interaction of relatively weakly surface-bound intermediate species with adsorbed oxygen (probably in the form of OНads ), (ii) at high potentials, dehydrogenation of CH3 OH molecules. Methanol oxidation is also characterized by self-inhibition which is associated with the formation of almost near-monolayer coverages with strongly chemisorbed species such as COads and/or НCOads . According to the literature data [15–22] on binary systems, certain acceleration of electrooxidation of HCO compounds can be achieved due to bifunctional catalysts, the «ensemble effect», and also due to changes in the electronic structure of Pt, i.e., the «electronic effect». The system of gold (single-crystal and polycrystalline) decorated with Pt submono- and monolayers and Pt clusters appears to be convenient as a model system for developing further the principles of synthesis of optimal platinum-containing electrocatalysts [22–35]. Different decoration methods were proposed, both chemical [26–28,31,32] and electrochemical [22–25,29,30,33–35]. A lot of such systems demonstrated high activity in reactions of oxygen electroreduction [24,25,33] and НCOOН electrooxidation [22,26–32]. The pronounced (by a factor of several tens) acceleration of HCOOH electrooxidation was observed [22,28] for submonolayer Pt coatings synthesized by pulsed electrodeposition on polycrystalline (pc) Au [22] and chemical deposition on Au nanoparticles [28]. Adzic et al. [23,24] proposed a highly efficient method for deposition of Pt submono and monolayers on gold. The method is based

352

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

on galvanic displacement of adsorbed atoms of a non-noble metal (М1 ) by Pt atoms due to a redox reaction with solution ions containing Pt(IV) or Pt(II) (in the literature several names are used for this process: galvanic displacement (GD), galvanic replacement (GR), spontaneous deposition (SD), surface limited redox replacement (SLRR)). Later, this method of gold decoration was employed in several studies [25,30,33–35]. Virtually all these studies used PtCl6 2− anions as the displacing agent. The higher specific activity (calculated per Pt surface or Pt mass) of thus obtained electrodes in oxygen reduction and HCOOH oxidation reactions as compared with individual platinum electrodes was mentioned. However, certain problems associated with the structure of formed Pt deposits, its dependence on the formation conditions and also with the stability of both the structure and the activity of the Pt–Au system synthesized by the GD method require additional studies to be carried out. The goal of this study was (i) to prepare Pt deposits on polycrystalline Au by the galvanic displacement of submono- and monolayers of copper adatoms (Cuad ); (ii) elucidate the structure of thus prepared Pt coatings and its dependence on the preparation conditions; (iii) assess their electrocatalytic activity in CН3 OН and НCOOН electrooxidation reactions as a function of the surface coverage with platinum. When solving problem (i) and carrying out the GD process, the primary attention was drawn to eliminating the contact with air when passing from Cuad /Au to Pt/Au [33]. The PtCl4 2− anions were used as the displacing agent because, in contrast to PtCl6 2- , they provide displacement of copper by platinum in the 1: 1 ratio and rule out the possibility of partial ionization of Cuad due to the reaction PtCl6 2- → PtCl4 2− . The comparison of data on the electrooxidation of CН3 OН and НCOOН carried out under the same conditions on platinum-decorated gold is of interest for elucidation the effect of the second component in binary Pt–М systems (where М is the catalytically inactive metal) on the catalytic activity of platinum.

2. Experimental Section. 2.1. Materials and Apparatus All electrochemical measurements were carried out in a threeelectrode cell with separate anodic and cathodic compartments. Working solution was stirred with a magnetic stirrer. Electrochemical experiments were carried out by means of an IPC-pro potentiostate-galvanostat developed at the A. N. Frumkin Institute of Physical Chemistry and Electrochemistry (Moscow). All potentials refer to the reversible hydrogen electrode in the same solution. We used Au (99.99%); all results were obtained in solutions prepared from sulfuric acid (Merck Suprapur), twice crystallized copper sulfate (reagent grade), K2 PtCl4 (Aldrich), methanol and formic acid (both of analytical-grade purity (Fluka)). Water was cleaned in the Milli-Q unit (Millipore, USA). For solution deareation, argon (special grade) was used. The working electrodes represented plates of pc Au and pc Pt (Sgeom = 1.0 cm2 ). Counter electrodes represented gold or platinum wires. The surface morphology of samples was examined on a scanning electron microscope JEOL JSM-6490 LV at the accelerating voltage of 30 kV. Samples surface was analyzed by X-ray photoelectron (XPS) and Auger (AES) spectroscopy using Axis Ultra DLD spectrometer (Kratos Analytical Ltd, UK) with hemispherical analyzer. XPS and Auger spectra were excited with monochromatic Al K␣ (1486.6 eV) radiation source and 10 kV electron gun, respectively. The binding energy scale of the instrument was calibrated with Au 4f7/2 (83.96 eV) and Cu 2p3/2 (932.62 eV) peaks of sputter cleaned gold and copper metals. XPS spectra were acquired in the constant analyzer energy mode with 160 eV pass energy for

survey spectra and 40 eV for narrow scans. Electron gun excited Auger spectra were recorded in the constant retard ratio mode. 2.2. Preparation of Pt0 Au Electrodes and Electrochemical Measurements Prior to experiments, the potentials of Au and Pt electrodes were cycled at v = 50 mV/s in 0.5 М H2 SO4 in intervals of 50–1700 mV (Au) and 50–1450 mV (Pt) until steady-state voltammograms were obtained. Figure 1a shows typical CVA for Pt and Au electrodes. The true surface of Au electrodes (Str ) was determined from the charge consumed in the oxygen electrodesorption under an assumption that at E = 1.70 V, the layer of Oads corresponds to the charge of 420 ␮C cm−2 [36,37]. The surface of pc Pt was assessed from the hydrogen adsorption (210 ␮C cm−2 ) [38]. The electrode roughness factors were 4.6 ± 0.2 for Au and 4.5 ± 0.2 for Pt. The procedure of measuring transients of open-circuit potential observed upon bringing PtCl4 2− anions into contact with the Pt electrode both in the absence and in the presence of Cuad on its surface was described in sufficient detail elsewhere [39]. The measurements on Au electrodes were carried out by the similar procedure. Here, we describe the main experimental stages. As for pc Pt [39], the Cuad monolayer (ML) on pc Au was formed in the solution of 2 mM CuSO4 + 0.5 M H2 SO4 at E= 290 mV. The circuit was opened and immediately after this (in < 2 s) the deaerated solution of 0.5 M H2 SO4 + 10−3 M PtCl4 was added to the solution in contact with the working electrode under the pressure of argon. After its addition, the PtCl4 2− concentration became 10−4 М. A transient of open-circuit potential was recorded until the establishment of the steady-state potential Est (the steady-state criterion was dE/d␶ < 4 mV/min). The working part of the cell was washed with background solution 0.5 M H2 SO4 .The charge corresponding to MLCuad (QMLCu ) was 400 ± 10 ␮C/cm2 (when Str was determined from Oads ). The resulting QMLCu values were close to that calculated in [37] based on the crystallographic structure of pcAu, which pointed to sufficient reliability of the Str determination based on oxygen adsorption. The QMLCu value was later used in determination of the gold surface coverage by Cuad (␪Cu ). Figure 1b shows CVA measured on pc Au after the copper adsorption (adsorption time 60 s) in 2 mM CuSO4 + 0.5 M H2 SO4 (curve 2). By setting the potential of copper adsorption in this solution, the ␪Cu vs. E dependence was obtained (Fig. 1b, insert). By changing the initial potentials Ein at which PtCl4 2− anions were added to solution (i.e., at different ␪Cu ), the amount of deposited Pt was varied. The stationarity criterion for the potentials established in solutions 0.5 M H2 SO4 + 10−4 М PtCl4 and 0.5 M H2 SO4 + 2 mM CuSO4 +10−4 M PtCl4 (Est ) was their variation by less that 0.4 mV min−1 . After the attainment of a steady-state potential, the Au electrode was thoroughly washed with supporting electrolyte solution (0.5 M H2 SO4 ) after which CVA were measured in this solution in order to determine the adsorption characteristics of the Pt0 Au surface formed and the degree of occupation of the Au electrode surface with the Pt deposit. Figure 1c shows a CVA measured of the Pt0 Au electrode immediately after the displacement of a copper monolayer (curve 1) with the upper potential scan limit of 1700 mV. The curve demonstrates the hydrogen adsorption and desorption waves on platinum particles (E ∼ 50–300 mV) and also the waves of oxygen adsorption and desorption on both platinum particles (desorption peak at ∼ 780 mV) and the free gold surface (desorption peak at ∼1200 mV). The repeated cycling of electrode potential up to such high potentials leads to the reconstruction of Pt islets and/or partial dissolution of deposited platinum. According to Fig. 1c (curve 2), after 10 cycles, the adsorption of hydrogen and oxygen on the platinum deposit falls down but the amount of oxygen adsorbed on the gold surface increases. This is why, to preserve the characteristics of Pt0 Au

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

353

Fig. 1. CVA measured (a) in 0.5М H2 SO4 on (1) pc Pt, (2) pc Au; (b) CVA measured on pc Au in (1) 0.5М H2 SO4 , (2) 2 mM CuSO4 + 0.5 М H2 SO4 , insert: dependence of ␪Cu on E; (c) CVA measured in 0.5 М H2 SO4 on Pt0 Au electrode (after the displacement of MLCuad by platinum): (1) 1st cycle, (2) 10th cycle. Potential scan from 0.050 V to 1.700 V. v = 50 mV/s. For explanations, see text.

electrodes, the upper polarization limit in later measurements did not exceed 1.45 V. In CН3 OН and НCOOН solutions (supporting electrolyte 0.5 M H2 SO4 ), CVA were measured with 10-s exposure at the lower potential scan limit. Here, we show steady-state CVA (both complete and their anodic branches). In experiments with Pt deposits on Au, a serious problem is associated with cleaning the Au electrode surface from Pt before new experiments. As a rule, the first stage of cleaning was the mechanical removal of the surface layer and/or chemical etching. We have shown that to remove Pt nanodeposits, it was suffice to cycle the electrode potential in the range from 0.05 to 1.90 V in 0.5 M H2 SO4 solution with v = 0.50 Vs−1 (∼500 cycles). Complete removal of platinum from Au electrodes was indicated by coincidence of CVA for cleaned Au with CVA for original Au and also by the XPS data (see Section 3.3). 3. Results and discussion 3.1. Transients of open-circuit potential Curve 1 in Fig. 2 corresponds to the open circuit potential shift of MLCuad Au electrode upon its contact with PtCl4 2− anions. An arrest is clearly seen in the potential range of ∼350-700 mV where the reaction Cuad + PtCl4 2− → Cu2+ + Pt0 + 4Cl−

(1)

Fig. 2. Open-circuit potential transients (OPT) corresponding to displacement of MLCuad from (1) pc Au and (2) pc Pt by Pt atoms in solution of 0.1 mM K2 PtCl4 + 2 mM CuSO4 + 0.5 М H2 SO4 and also OPT on (3) pc Au and (4) pc Pt in solution 0.1 mM K2 PtCl4 + 0.5 М H2 SO4 .

takes place. A comparison with the analogous curve for pc Pt (curve 2) makes it possible to conclude that the average rate of Cuad removal is higher on pc Au than on Pt. The established steadystate potential Est substantially exceeds the potential of complete

354

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

removal of Cuad as a result of reaction (1), as follows from the adsorption isotherms of copper atoms on the Au electrode surface (Fig. 1b, insert). The potential increase at E > ∼750 mV (i.e., after the removal of Cuad ) is associated with the increase in the total electrode charge Q [38,40] due to the reaction PtCl4 2− + 2e → Pt0 + 4Cl−

(2)

Est values are established in the potential region where oxygen does not adsorb on the Au surface (according to the anodic branch of curve 2 in Fig. 1a) but does adsorb in small amounts on pc Pt. The adsorption of oxygen on platinum is reversible up to ∼ 0.9 V, i.e., the Pt0 Au surface formed should be considered as perfectly polarizable [38,40]. Then, it can be assumed that the potential of this system is determined by the equilibrium Pt0 + H2 O + e ↔ Pt0 OHads + H+

(3)

It should be noted that as regards its origin, Est is the mixed potential, because the steady-state concentration of OHads is formed as a result of compensation of reaction rates of the reduction of PtCl4 2− and traces of dissolved oxygen and the oxidation of traces of organic impurities. It is seen that the potential established on the Pt0 Au surface (curve 1 in Fig. 2) is substantially lower (by ∼60 mV) than the potential established after the contact of a pc Pt electrode with PtCl4 2− (curve 2). The main reason for this is apparently the slower rates of PtCl4 2− reduction on gold electrodes compared to platinum electrodes (as demonstrated by the analysis of curves 3 and 4 in Fig. 2). Moreover, the chemical potential of OНads (␮OН ) on Pt0 can considerably differ from that on the surface of pc Pt. The difference in the properties of monolayer platinum-metal coatings on foreign supports was theoretically predicted [20,21]. This also follows from the difference in curves 1 and 2 in Fig. 2 at E > 800 mV. In contrast to curve 2, curve 1 passes through a maximum. This is probably associated with certain restructuring of the Pt0 layer initially formed on Au by the displacement of Cuad . According to [39], the substitution of Pt for Cuad carried out from PtCl4 2− solutions results in formation of an epitaxial Pt0 layer, which is the reason for the smoothness of curve 2 at high potentials. On the whole, the substantial difference between steady-state potentials established after the displacement of MLCuad Au (curve 1) and MLCuad Pt (curve 2) points to the incomplete blocking of the Au surface by platinum. 3.2. Cyclic voltammetry of Pt0 Au at different surface coverages with platinum Based on the adsorption isotherm ␪Cu (E) (insert in Fig. 1b), different amounts of Pt were deposited (see Section 2.2). Figure 3 shows typical CVA recorded after different Cuad submonolayers (from 0.25MLCuad (curve 4) to 1MLCuad (curve 1)) were displaced by platinum. On gold electrodes, potential cycling in the E range of 0.05 -1.0 V revealed no peaks that can be associated with the adsorption of hydrogen or oxygen (Fig. 1a). The presence of peaks in this potentials range in Fig. 3 (curves 1-4) is apparently associated with hydrogen adsorption-desorption (peaks C and C’) and the removal of adsorbed oxygen (peak B). The quantitative assessment of the surface coverage with platinum was carried under the assumption that (i) in the first approximation, the adsorption of hydrogen and oxygen on Pt0 substituted for Cuad does not differ from their adsorption on bulk polycrystalline platinum; (ii) hydrogen and oxygen are adsorbed on platinum atoms in the ratio H, O: Ptsurf = 1 [38,39]. Then, the gold surface coverage with platinum can be equated to the ratio of charge consumed in the removal of Нads from Pt0 Au (QPtAu H ) to the half-charge consumed in the removal of the Cu monolayer from the original Au electrode surface (QAu Cu ), i.e., 2QPtAu H /QAu Cu . Table 1 compares the latter quantity with the initial  Cu .

Fig. 3. CVA measured on Pt0 Au prepared by displacement of (1) 1MLCuad , (2) 0.75MLCuad , (3) 0.5MLCuad and (4) 0.25 MLCuad . 0.5М H2 SO4 , v = 50 mV/s.

Apparently, in the coverage range  Cu < ∼0.5, these two quantities are close, whereas for  Cu > 0.5, 2QPtAu Н /QAu Cu <  Cu . This allows us to assume that at low initial  Cu , platinum is deposited mainly as a monolayer in the ratio of 1 Pt atom per 1 Au surface atom. It can be assumed that considerable amounts of individual Pt atoms are present on the surface in the beginning of displacement and that hydrogen does not adsorb on them. However, for the coverages under consideration ( Pt ≥0.25), the fraction of such Pt0 is obviously insignificant. Otherwise, for low  Cu ( Pt ), we could expect that  Cu > 2QH PtAu /QCu Au ; however, as follows from Table 1, for  Cu < 0.5 the ratio  Cu : 2QH PtAu /QCu Au is close to 1. This is yet another evidence that in the considered interval of primary  Cu values the forming Pt0 coating consists mainly of islets. As the initial  Cu increases, a certain conglomeration of Pt atoms obviously occurs as well as their partial deposition on Pt0 rather than on Au sites free from Cuad . Earlier [39], the displacement of MLCuad from pc Pt resulted in formation of epitaxial Pt0 layer. Hence, it seems to be most probable that Pt «islets» formed on Au contain fragments no thicker than 2 layers. The fact that gold surface sites not blocked by platinum are left after the displacement of MLCuad in a PtCl4 2− solution is evidenced by the presence of the peak A in curve 1 (Fig. 3) at potentials close to those of Oads electrodesorption from pc Au (Fig. 1). The gold surface fraction left unoccupied after the displacement of MLCuad was quantitatively assessed by the Oads electrodesorption peak at 1.17 V in the first CVA measured up to 1.70 V. Based on three experiments, it was found that the average fraction of unoccupied gold surface fraction left after displacement of MLCuad was 0.3 (± 0.05). This adequately agrees with the surface fraction free of Pt estimated above based on the hydrogen adsorption on Pt0 (the more so, if we take into account the aforementioned possibility of slight dissolution of Pt at high anodic potentials even during one cycle). Insofar as the displacement of a Cuad monolayer does not produce a Pt0 monolayer, then in order to identify gold samples with different surface coverages with Pt0 , we used their original coverages with Cuad so that samples corresponding to  Cu = 0.25, 0.5, etc. are designated as Pt0.25 0 Au, Pt0.5 0 Au, etc. Table 1 Adsorption parameters of Pt0 Au electrodes (for explanations, see text). Cu Eads , mV

 Cu (±0.05)

Cu QAu , mC

Cu QPtAu , mC

Cu H 2QPtAu /QAu

290 340 450 500

1 0.75 0.5 0.25

1.9 1.9 1.9 2.0

0.55 0.47 0.35 0.27

0.63 0.52 0.42 0.25

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

355

Fig. 4. SEM images of (a) pc Au and (b) pt01 Au.

The above estimates of the Au surface coverage with platinum were made under the assumption that the main adsorption characteristics of hydrogen and oxygen on Pt0 islets on Au are close to those on pc Pt. A more careful comparison of energy spectra of hydrogen and oxygen adsorption on Pt0 and pc Pt revealed certain differences. Thus, in place of three peaks observed in the Нads desorption region in curve 1, Fig. 1a (pc Pt), only one peak is observed in curves 1–4, Fig. 3. Moreover, the potential of the peak corresponding to Нads removal from Pt0 Au somewhat depends on the degree of gold surface occupation with platinum. On Pt0.5 0 Au samples, the potential of Oads electrodesorption from Pt0 (peak B in curve 2, Fig. 3) is ∼ 40 mV more negative as compared with the corresponding peak on pc Pt (curve 1, Fig. 1a). These results point to the presence of a certain electronic effect associated with the Pt contact with the support; however, the latter effect was assessed to be relatively small. 3.3. Analysis of SEM, XPS and Auger spectroscopy data A comparison of SEM images (Fig. 4) obtained on the clean Au surface (a) and Au surface covered with MLPt0 prepared by GD (b) points to the presence of Pt islets. It should be borne in mind that the electrochemical measurements were carried out in situ immediately after the displacement whereas SEM images were obtained ex situ, i.e., after the Pt1 0 Au sample was exposed to air for a relatively long period. Thus, the shape of particles observed by SEM may be strongly deformed by the «adhesion» of impurities to Pt0 upon contact with air. This is why the data in Fig. 4 merely make it possible to presume of a certain tendency towards conglomeration for Pt0 atoms that have displaced Cuad which agrees with the electrochemical data shown above. The tendency of Pt0 atoms to conglomeration was also noted in earlier publications [23,24,33], in which MLCuad was displaced by

Fig. 5. (a) Overall XPS spectra of (1) pt01 Au and (2) pt00.5 Au collected at the emission angle of 900 . (b) Comparison of XPS spectrum lines of Pt4f electrons for (1) pt01 Au and (2) pt00.5 Au samples collected at angles of 900 and 300 . Spectra are normalized as regards intensity.

Pt from Pt(111) in PtCl6 2− solutions. In these studies, the similar changes of SEM images were observed after the displacement of MLCuad , although Pt0 was deposited in the half amount as compared with our studies (because Pt(IV) was used in place of Pt(II)). X-ray photoelectron spectra recorded for Au samples after their electrochemical cleaning from Pt0 (see Experimental) showed the absence of platinum, which confirms the efficiency of procedures used for regeneration of Au surface (the spectra are not shown here). Figure 5a shows the overall X-ray photoelectron spectra of samples Pt1 0 Au and Pt0.5 0 Au (for emission angle of 90◦ ). Figure 5b demonstrates photoelectron spectra of Pt4f lines for the same samples. The thickness of the layer analyzed was ∼4 nm, i.e., more than an order of magnitude larger as compared with the Pt monolayer. When samples were washed or stayed in contact with atmosphere in the evacuation chamber, considerable amounts of impurities could be accumulated on their surfaces. This made impossible the reliable quantitative analysis. For the qualitative analysis, we used the Pt/Au ratio calculated based on the 4f spectra for gold and platinum collected at two different photoelectron emission angles with respect to the sample surface:

356

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

Sample

0

Pt1 Au Pt0.5 0 Au

Photoelectron emission angle 900

300

0.26 0.18

0.35 0.23

X-ray photoelectron spectra showed, as expected, the lower Pt content in the Pt0.5 0 Au sample as compared with Pt1 0 Au. When photoelectrons were collected at angle 30◦ with respect to the surface, the analysis depth was half the depth at angle 90◦ . It is seen that for the smaller angle, the relative contribution of the 4f Pt line increases, which at least confirms the predominant presence of platinum on the surface. To find the degree of uniformity of the platinum distribution over the surface of Pt1 0 Au sample, Auger spectra were obtained in 9 points (Fig. 6). Table 2 shows the results of calculations of the Pt/Au ratio based on these spectra. On the whole, except for one point (4), the scatter of Pt/Au ratio is small, which points to the sufficiently uniform distribution of Pt0 over the Au surface. Point (4) is probably situated on some surface defect (depression), which is apparently the reason for such strong difference in the surface concentration of Pt0 in this point and in the other points. The conclusion on the sufficiently uniform Pt0 distribution over the gold surface does not contradict the SEM image (Fig. 4b), because the average distance between Pt0 conglomerates of different size can be assessed as no larger than ∼20 nm, whereas the region probed by Auger spectroscopy was ∼1 ␮m х 1 ␮m. The smaller Pt/Au ratios for Pt01 Au samples calculated based on Auger spectra (Table 2) as compared with XPS data are associated with the greater analysis depth for the Auger spectroscopy (5–6.5 nm in our case). It deserves mention that the binding energy of Pt4f7/2 electrons for Pt1 0 Au and Pt0.5 0 Au samples (Fig. 5b) equal to 70.8 eV (the binding energy of Au4f7/2 electrons is 83.96 eV) is 0.4 eV lower than the binding energy for bulk platinum (71.2 eV [41]). Such a decrease in binding energy was observed for both the surface platinum atoms that differ from volume platinum atoms only in their electronic

Fig. 6. Secondary electron images of a surface region on the pt01 Au sample on which Auger spectra were measured in different points (designated as Pt-Au).

Fig. 7. Positive going scans of CVAs in 0.5 M CH3 OH + 0.5 М H2 SO4 solution on (1) pc Pt and pt0x Au electrodes for x: (2) 1.0, (3) 0.75, (4) 0.5, (5) 0.25. Insert: complete CVAs on (1) Pt and (2) pt00.5 Au electrodes. v = 50 mV/s.

structure [42] and also for Pt–Au systems [43,44] in which the lower binding energy is associated with the charge transfer at the platinum interaction with gold atoms. The found binding energy of Pt4f7/2 electrons testifies that the majority of Pt atoms are not in the multilayer Pt0 conglomerates otherwise the binding energy for Pt4f7/2 electrons would be closer to that for bulk platinum. Thus, the results of characterization of the surface layer of Pt0 Au samples by SEM, XPS and Auger spectroscopy adequately agree with the data of electrochemical measurements shown above (bearing in mind that the former were obtained ex situ and the latter–in situ). 3.4. Methanol electrooxidation Figure 7 shows potentiodynamic curves for the methanol oxidation reaction (MOR) for different surface coverages with platinum. The currents are shown per cm2 of the true platinum surface. The presence of Pt on the gold surface results in appearance of a methanol oxidation peak at ∼ 840 mV (no such peak is observed on gold) which becomes higher with the increase in the amount of deposited Pt0 . However, for Pt0 1 Au (curve 2) the specific activity of Pt0 remains much lower than for smooth pc Pt electrode (curve 1). For the surface estimation method used (Hads : Ptsurf = 1), the shown specific currents characterize in fact the activity of a single Pt surface atom. Thus, the «neighborhood» with gold reduces the specific activity of Pt0 so that this effect is the most pronounced for Pt0 0.25 Au (the activity of Pt0 is ∼ 6-times lower than the activity of Ptsurf for bulk Pt). At high potentials, the methanol electrooxidation is mainly limited by the stage of dehydrogenation [4,8]. From our point of view, the decrease in the activity of Pt0 atoms on the gold surface with respect to «destruction» of CН3 OН molecules should be associated with the geometrical factor, i.e., the absence of a sufficient number of large «platinum areas» required for the destructive chemisorption of CН3 OН molecules. According to [4,15], more than three Pt surface atoms are required for the chemisorption of one CН3 OН molecule. As the amount of deposited Pt0 and its degree of conglomeration increase, the geometrical effect should weaken, which agrees with experimental data. The fact that chemisorption of CН3 OН requires a large number of surface sites was used in many studies for explaining the decrease in the MOR specific rate with

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

357

Table 2 The Pt/Au ratio on the surface of Pt1 0 Au sample in nine points (calculated based on Auger spectra). Point

1

2

3

4

5

6

7

8

9

Pt/Au

0.18

0.16

0.16

0.07

0.17

0.17

0.18

0.15

0.19

the decreasing size of Pt nanoparticles (the negative size effect) [45–47]. The comparison of CVA for Pt and Pt0 0.5 Au electrodes (insert in Fig. 7, curve 1 and 2, respectively) has shown that the differences in the peak potentials in the anodic (Ep f ) and cathodic (Ep b ) scans were almost the same for both electrodes. At the same time, the ratios of peaks (Ip f /Ip b ) noticeably differed being 1.15 for Pt and 1.60 for Pt0 0.5 Au. The ratio Ip f /Ip b > 1 suggests that the removal of CН3 OН chemisorption products in the end of the anodic scan is incomplete. This ratio is far larger for Pt0 0.5 Au as compared with Pt which points to the lower rates of oxidation of strongly chemisorbed species (SCS) on Pt. To a certain extent, the latter result can be associated with the «electronic effect». In catalysis, according to the theory [20,21], the «electronic effect» is mainly associated with the changes in d-bond centers. The absence of correlation between the activity of platinum-gold alloys with respect to MOR and the character d-bonds was mentioned in [48]. However, the behavior of Pt0 Au can strongly differ from that of the AuPt alloy. On the whole, the problem of the «electronic effect» contribution in catalysis of HCO-substances electrooxidation on Pt0 Au requires further studies, in particular, the studies of CO chemisorption and electrooxidation. Of high practical interest is the degree of Pt utilization in MOR and hence its specific activity expressed per Pt mass unit (the mass specific activity, MSA). The Pt MSA in Pt0.25 Au was assessed based on the maximum current in curve 5 Fig. 7 to be ∼700 mA/mg, which exceeds MSA of platinum in disperse PtRu catalysts for MOR (e.g., see [49]). Thus, the effect of the high degree of dispersion (close to its limiting value) prevails over the geometric effect. At the same time it should be noted that the CН3 OН electrooxidation currents under consideration pertain to the region of sufficiently high potentials (0.7–1.0 V), which lowered down their feasibility for MOR. These results are important for understanding the mechanism of the multistage MOR process and the nature of activity of mixed catalysts. We failed to carry out the correct comparison of steady-state MOR currents at potentials E < 0.6 V (important for practice) on the Pt0 Au catalysts under study due to their smallness and poor reproducibility.

3.5. Formic acid electrooxidation Fundamental changes occur as we pass from MOR to formic acid electrooxidation (FAOR) (Fig. 8). On pc Au, the FAOR currents are virtually absent up to ∼ 1.2 V (the corresponding curve is not shown). On pc Pt, the currents remain low up to 750 mV (curve 1) and a relatively small peak is observed at ∼ 900 mV. On Pt0 Au electrodes, the beginning of FAOR can be discerned even at ∼ 150 mV and a high current peak is observed at 630 mV. The maximum current somewhat increases when going from Pt0.25 Au to Pt0.75 Au and slightly falls dawn for Pt1 Au. However, these changes fit the ∼2-fold gain and can be ascribed to secondary effects such as the effect of the perimeter of the Pt0 islet/free Au boundary, the effect of the partial transition of Pt0 to the second layer, the effects of structural defects, etc. On the whole, for E < 700 mV, we observed the unusually high rise of the Pt0 activity on the Au surface with respect to this reaction as compared with Pt atoms on the surface of pc Pt (by a factor of 20-25 at ∼ 600 mV). This

Fig. 8. Positive going scans of CVAs in 0.5 M HCOOH + 0.5 M H2 SO4 solution on (1) Pt electrode and pt0x Au electrodes for x: (2)1.0, (3)0.75, (4)0.5, (5)0.25. v = 50 mV/s.

result is qualitatively similar to the data found in [22,28] on how submonolayer platinum coatings on the surface of pc Au and Au nanoparticles affect the FAOR. Certain quantitative deviations appear to be sufficiently reasonable due to substantial difference in the methods of preparation of submonolayer Pt coatings, electrode pretreatment, surface roughness of Au electrodes and other factors. Based on the difference in FAOR currents on Pt0 0.5 Au observed in anodic and cathodic scans of CVA (Fig. 9), it can be concluded that this non-steady-state process proceeds under conditions of severe surface blocking by strongly chemisorbed species (SPS), where COads is the main species. Calculations have shown that the maximum currents in CVA recorded at v = 50 mV/s are lower by more than one order of magnitude as compared with the processes limited by mass transfer, i.e., the process occurs in the kinetic mode. Figure 9b compares CVA on Pt0 0.5 Au with analogous curves measured on pc Pt. The ratio of currents in peaks Р2 and Р1 can be taken as the qualitative characteristic of FAOR inhibition observed during the electrode potential cycling. For Pt0 0.5 Au, the ratio IP2 /IP1 is 1.7, according to Fig. 9b; for pc Pt, this ratio is 2.8, i.e., the weaker inhibition of the reaction under study for Pt deposited on Au can be concluded based only on CVA measurements. Chronoamperograms measured at certain fixed potential values provide more information on the FAOR inhibition by accumulated products. As mentioned above, insofar as the voltammetric characteristics measured at E< 0.6 V are more interesting for practice, Fig. 10 shows the I vs. ␶ curves at E =0.4 V (in the anodic branch of steady-state CVA, the scanning was stopped at 0.4 V). As seen, the currents on Pt0 0.5 Au drop down sufficiently strongly with time. However, quasi-steady-state currents established on Pt0.5 Au in ∼ 7 min were higher than the currents on pc Pt by approximately one order of magnitude. Based on the literature [10,15,50] and our [5,45,51] data on the FAOR mechanism, the scheme of НCOOН electrooxidation can assume to be as follows:

358

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

Fig. 9. CVA measured in 0.5 M HCOOH + 0.5 M H2 SO4 solution: (a) on pt00.5 Au at (1) v = 50 mV/s, (2) v = 5 mV/s; (b) on (1) Pt, (2) pt00.5 Au at v = 50 mV/s.

i1 HCOOHads

HCOOads + Hads -e CO2 + H+

HCOOHsol

-e H+

i2 SCS

CO2

The upper reaction route is current-determining («direct path» [15,22]). It proceeds by the dehydrogenation mechanism where one adsorption site is sufficient for the adsorption of an intermediate species [15,22,45,51]. The НCOOН oxidation by the parallel route proceeds through strongly chemisorbed species (SCS) and is much slower as compared with the main route (i2 « i1 ). At low potentials (<∼600 mV), SCS are mainly represented by COads formed by the dehydration mechanism. In [15], it was assumed that at the higher potentials the species are also accumulated. The «harmful» effect of the second route (despite the latter also leads to CO2 ) is that

the active sites (designated by dark circles) on which adsorption and dehydrogenation of HCOOH proceed are blocked by SCS. Thus, the promoting effect of Pt deposited on Au is apparently associated with the appearance of new active sites as we pass from the pc Pt surface to Pt0 Au. The lattice parameters of Pt and Au considerably differ (a is 0.3920 nm for Pt and 0.4078 nm for Au), which make inevitable the different arrangement of Pt0 atoms on Au as compared with Pt atoms on the surface of pc Pt. The new active sites can represent, e.g., individual Pt atoms on the perimeter of Pt0 islets or Pt0 atoms that occupy defective sites on the pc Au surface. The considerable contribution of the electronic effect seems to be improbable. The weakening of the blocking effect can be associated with the fact that the formation of SCS requires no less than 2 adsorption sites [15]. New active sites may be more selective with respect to processes that proceed through one-site adsorption. This explanation is fundamentally close to the explanation of the phenomenon under consideration given in [22,28] in which the FAOR acceleration on Pt submonolayers on Au was associated with the ensemble effect. The fact that such acceleration of FAOR on Pt can be associated with difficulties in the SCS formation was also assumed in studies in which the promoting action of lead and bismuth on the FAOR was noted (e.g., see [52–54]). At present, it is impossible to separate the effects of the «decrease» in blocking from those of formation of new active sites for НCOOН dehydrogenation. Quasisteady-state current in FAOR (Fig. 10) assessed per mg of Pt is ∼0.45 A/mg for Pt0.5 Au/GC, which is ∼10-time higher than the mass specific activity of Pt in highly disperse Pt/C (according to data of [53]). This indicates that a very high degree of Pt utilization in Ptx 0 Au is reached in FAOR.

4. Conclusion

Fig. 10. Chronoamperograms of (1) pt00.5 Au and (2) Pt electrodes at the potential of 400 mV in 0.5 M HCOOH + 0.5 M H2 SO4 solution.

• It was shown that transients of open-circuit potential observed after bringing Cuad Au into contact with the PtCl4 2− solution (0.5 М Н2 SO4 as the supporting electrolyte) provide important information on the process of displacement of Cuad by Pt atoms. • As follows from voltammetric studies, the displacement of (1-х) MLCuad on the surface of pc Au in PtCl4 2− solution leads to the formation of submonolayer Pt0 coatings. For х< 0.5, the conglomeration of Pt atoms and the partial formation of the second layer occur.

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

• SEM, XPS and Auger spectroscopic studies confirmed the absence of multilayer conglomeration of Pt0 on the surface of Au. • The specific activity of Pt0 on Au in the reaction of CН3 OН electrooxidation (per Pt atom) is substantially lower than the activity of Pt atoms on the pc Pt surface, which was explained by the absence of sufficient number of platinum «areas» required for the multisite adsorption of CН3 OН molecules. • The strong increase in the specific rates of НCOOН electrooxidation on Pt0 Au as compared with pc Pt (by an order of magnitude and more) is observed. A scheme of formic acid electrooxidation is put forward. It is assumed that new active sites are formed on Pt0 Au which selectively accelerate the НCOOН oxidation route via one-site adsorption of its molecules or intermediate species. Acknowledgments This work was supported by the Russian Foundation for Fundamental Research, Project Nos. 12-03-00998a and in part by M.V. Lomonosov Moscow State University Program of Development. References [1] W. Vielstich, A. Lamm, H. Gasteiger (Eds.), Handbook of Fuel Cells, Fundamentals, Technology, Applications (Hardcover), J. Willy and Sons, London, 2003. [2] A.S. Arico, S. Srinivasan, V. Antonucci, DMFCs From Fundamental Aspects to Technology Development, Fuel Cells 1 (2001) 133. [3] C. Conntanceau, S., Brimand, C., Lamy, J.-M. Leger, L., Dubau, S., Rousseau, F. Vigier, Review of different methods for developing nanoelectrocatalysts for the oxidation of organic compounds, Electrochimica Acta 53 (2008) 6865. [4] O.A. Petry, B.I. Podlovchenko, A.N. Frumkin, Hira Lal, The behaviour of platinized-platinum and platinum-ruthenium electrodes in methanol solutions, Journal of Electroanalytical Chemistry 10 (1965) 253. [5] B.I. Podlovchenko, O.A. Petry, A.N. Frumkin, Hira Lal, The behaviour of a platinized-platinum electrode in solutions of alcohols containing more than one carbon atom, aldehydes and formic acid, Journal of Electroanalytical Chemistry 11 (1966) 12. [6] M.W. Breiter, Electrochemical Processes in Feul Cells, Berlin: Springer-Verlag, 1969. [7] B.I. Podlovchenko, R.P. Petukhova, On methanol electrooxidation at a platinized platinum electrode in the range medium surface coverages with chemisorbed substances, Elektrochimiya 10 (1974) 489. [8] T. Iwasita, Electrocatalysis of methanol oxidation, Electrochimica Acta 47 (2002) 3663. [9] H. Wang, C. Wingender, H. Bultrushat, M. Lopez, M.T. Rutz, Methanol oxidation on Pt, PtRu, and colloidal Pt electrocatalysts: a DEMS study of product formation, Journal of Electroanalytical Chemistry 509 (2001) 163. [10] R. Parsons, T. VanderNood, The oxidation of small organic molecules: A survey of recent fuel cell related research, Journal of Electroanalytical Chemistry 257 (1988) 9. [11] S. Park, Y. Xie, M.J. Weaver, Electrocatalytic Pathways on Carbon-Supported Platinum Nanoparticles: Comparison of Particle-Size-Dependent Rates of Methanol, Formic Acid, and Formaldehyde Electrooxidation, Langmur 18 (2002) 5792. [12] M. Leger, Mechanistic aspects of methanol oxidation on platinum-based electrocatalysts, Journal Applied Electrochemistry 31 (2001) 767. [13] A. Capon, R. Parsons, The oxidation of formic acid at noble metal electrodes Part III. Intermediates and mechanism on platinum electrodes, Journal of Electroanalytical Chemistry 45 (1973) 205. [14] T.J. Schmidt, B.N. Grgur, N.M. Marcovic, P.N. Ross, Oscillatory behavior in the electrochemical oxidation of formic acid on Pt(100): rotation and temperature effects, Journal of Electroanalytical Chemistry 500 (2001) 36. [15] M. Neurock, M. Janik, A. Wieckowski, A first principles comparison of the mechanism and site requirements for the electrocatalytic oxidation of methanol and formic acid over Pt, Faraday Disscussion 140 (2009) 363. [16] S. Motoo, M. Watanabe, Electrocatalysis by ad-atoms: Part I. Enhancement of the oxidation of methanol on platinum and palladium by gold ad-atoms, Journal of Electroanalytical Chemistry 60 (1975) 259. [17] T.D. Jarvi, E.M. Stuve, Electrocatalysis, in: J. Lipkovski, P.N. Ross (Eds.), Electrocatalysis., Wiley-VCH, 1998, p. 72. [18] H. Schibata, N. Furuya, M. Watanabe, Electrocatalysis by ad-atoms: Part XXV. Electrocatalytic effects on the elementary steps in ethanol oxidation by nonoxygen-adsorbing ad-atoms, Journal of Electroanalytical Chemistry 267 (1989) 163. [19] A. Cuesta, M. Escudero, B. Lanova, H. Baltruschat, Cyclic Voltammetry, FTIRS, and DEMS Study of the Electrooxidation of Carbon Monoxide, Formic Acid, and Methanol on Cyanide-Modified Pt(111) Electrodes, Langmur 25 (2009) 6500.

359

[20] T. Bligaard, J.K. Norskov, Ligand effects in heterogeneous catalysis and electrochemistry, Electrochimca Acta 52 (2007) 5512. [21] E. Santos, W. Schmickler, Electrocatalysis of Hydrogen Oxidation—Theoretical Foundations, Angewandte Chemie International Edition. 46 (2007) 8262. [22] M.D. Obradoviˇc, A.V. Tripkoviˇc, S.L. Gojkoviˇc, The origin of high activity of Pt–Au surfaces in the formic acid oxidation, Electrochimica Acta 55 (2009) 204. [23] S.R. Brankovic, J.X. Wang, R.R. Adzic, Metal monolayer deposition by replacement of metal adlayers on electrode surfaces, Surface Science 474 (2001) L173. [24] K. Sasaki, Y. Mo, J.X. Wang, M. Balasubramanian, F. Uribe, J. McBreen, R.R. Adzic, Pt submonolayers on metal nanoparticles—novel electrocatalysts for H2 oxidation and O2 reduction, Electrochimica Acta 48 (2003) 3841. [25] M. Van Brussel, G. Kokkinidis, I. Vandendael, C. Buess-Herman, High performance gold-supported platinum electrocatalyst for oxygen reduction, Electrochemistry Communications 4 (2002) 808. [26] S. Papadimitriou, S. Armyanov, E. Valova, A. Hubin, O. Steenhaut, E. Pavlidou, G. Kokkinidis, S. Sotiropoulos, Methanol Oxidation at Pt-Cu, Pt-Ni, and Pt-Co Electrode Coatings Prepared by a Galvanic Replacement Process, Journal of Physical Chemistry C114 (2010) 5217. [27] S. Wang, N. Kristian, S. Jiang, X. Wang, Controlled deposition of Pt on Au nanorods and their catalytic activity towards formic acid oxidation, Electrochemistry Communications 10 (2008) 961. [28] N. Kristian, Y. Yan, X. Wang, Highly efficient submonolayer Pt-decorated Au nano-catalysts for formic acid oxidation, Chemical Communications (2008) 353. [29] R. Zhou, R. Yue, F. Jiang, Y. Du, P. Yang, C. Wang, J. Xu, Electrocatalytic Oxidation of Formic Acid at Pt Modified Electrodes: Substrate Effect of Unsintered Au Nano-Structure, Fuel cells 12 (2012) 971. [30] Yaolun Yu, Yueping Hu, Xuewei Liu, Weiqiao Deng, Xin Wang, The study of Pt@Au electrocatalyst based on Cu underpotential deposition and Pt redox replacement, Electrochimca Acta 54 (2009) 3092. [31] N. Kristian, Y. Yu, P. Gunawan, R. Xu, W. Deng, X. Liu, X. Wang, Controlled synthesis of Pt-decorated Au nanostructure and its promoted activity toward formic acid electro-oxidation, Electrochimica Acta 54 (2009) 4916. [32] S. Patra, J. Das, H. Yang, Selective deposition of Pt on Au nanoparticles using hydrogen presorbed into Au nanoparticles during NaBH4 treatment, Electrochimica Acta 54 (2009) 3441. [33] D. Misak, T.C. Ruthenburg, W.R. Fawcett, Copper deposition and its replacement by platinum on a gold electrode, Electrochimica Acta 55 (2010) 7610. [34] A. Rincón, M.C. Peréz, C. Gutierrez, Dependence of low-potential CO electrooxidation on the number of Pt monolayers on gold, Electrochimica Acta 55 (2010) 3152. [35] D. Gokcen, S.-E. Bae, S.R. Brankovic, Reaction kinetics of metal deposition via surface limited red-ox replacement of underpotentially deposited metal monolayers, Electrochimica Acta 56 (2011) 5545. [36] A.A. Michri, A.G. Pshchenichnikov, R.K. Burstein, Determination of the surface of smoth gold electrodes, Soviet Electrochemistry 8 (1972) 351. [37] M.C. Santos, L.H. Mascaro, S.A.S. Machado, Voltammetric and rotating ringdisk studies of underpotential deposition of Ag and Cu on polycrystalline Au electrodes in aqueous H2 SO4 , Electrochimica Acta 43 (1998) 2263. [38] A.N. Frumkin, Potentsialy Nulevogo Zaryada (Zero-Charge Potentials), Nauka, Moscow, 1981. [39] B.I. Podlovchenko, U.E. Zhumaev, Yu.M. Maksimov, Galvanic displacement of copper adatoms on platinum in PtCl4 2- solutions, Journal of Electroanalytical Chemistry 651 (2011) 30. [40] B.I. Podlovchenko, E.A. Kolyadko, A correlation between the changes in total charge and open circuit potential of a hydrogen electrode during the adsorption of neutral particles and the formation of adatoms from ions, Journal of Electroanalytical Chemistry 506 (2001) 11. [41] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photoelectron Spectroscopy, Chigasaki: ULVAC-PHI, Inc, 1995. [42] K. Dückers, K.C. Prince, H.P. Bonzel, V. Cháb, K. Horn, Adsorption-induced surface core-level shifts of Pt(110), Physical Review B 36 (1987) 6292. [43] S. Uhm, H. Jeon, J. Lee, Electrocatalytic Oxidation of HCOOH on an Electrodeposited AuPt Electrode: its Possible Application in Fuel Cells, J. Electrochem. Science and Technology 1 (2010) 10. [44] R.A.P. Smith, Surface Characterisation of Heterogeneous Catalysts by XPS: Part II, Platinum Metals Review 53 (2009) 109. [45] Yu.M. Maksimov, B.I. Podlovchenko, T.L. Azarchenko, Preparation and electrocatalytic properties of platinum microparticles incorporated into polyvinylpyridine and Nafion films, Electrochimica Acta 43 (1998) 1053. [46] S.H. Joo, K. Kwon, D.J. You, Ch. Pak, H. Chang, J.M. Kim, Preparation of high loading Pt nanoparticles on ordered mesoporous carbon with a controlled Pt size and its effects on oxygen reduction and methanol oxidation reactions, Electrochimica Acta 54 (2009) 5746. [47] B.I. Podlovchenko, V.A. Krivchenko, Yu.M. Maksimov, T.D. Gladysheva, L.V. Yashina, S.A. Evlashin, A.A. Pilevsky, Specific features of the formation of Pt(Cu) catalysts by galvanic displacement with carbon nanowalls used as support, Electrochimica Acta 76 (2012) 137. [48] M.W. Breiter, Reactivity and Surface Composition Anodic Methanol Oxidation on Platinum—Gold Alloys, Journal of Physical Chemistry 69 (1965) 3377. [49] C.-F. Chi, M.-C. Yang, H.-S. Weng, A proper amount of carbon nanotubes for improving the performance of Pt–Ru/C catalysts for methanol electrooxidation, Journal of Power Sources 193 (2009) 462.

360

B.I. Podlovchenko et al. / Electrochimica Acta 130 (2014) 351–360

[50] Y.X. Chen, S. Ye, M. Heinen, Z. Jusus, R.J. Behm, Bridge-Bonded Formate: Active Intermediate or Spectator Species in Formic Acid Oxidation on a Pt Film Electrode? Langmuir 22 (2006) 10399. [51] A.V. Smolin, B.I. Podlovchenko, Y.M. Maksimov, Electrooxidation of formic acid on palladium electrodeposits in the sulfuric acid electrolytes, Russian Journal of Electrochemistry 33 (1997) 440. [52] E. Casado-Rivera, D.J. Volpe, L. Alden, C. Lind, C. Downie, T. Vázquez-Alvarez, ˜ Electrocatalytic Activity of Ordered A.C.D. Angelo, F.J. DiSalvo, H.D. Abruna,

Intermetallic Phases for Fuel Cell Applications, Journal of American Chemical Society 126 (2004) 4043. [53] B.-W. Zhang, Ch.-L. He, Y.-X. Jiang, M.-H. Chen, Y.-Y. Li, L. Rao, Sh.-G. Sun, High activity of PtBi intermetallics supported on mesoporous carbon towards HCOOH electro-oxidation, Electrochemistry Communications 25 (2012) 105. [54] A.S. Bauskar, C.A. Rice, Spontaneously Bi decorated carbon supported Pt nanoparticles for formic acid electro-oxidation, Electrochimica Acta 93 (2013) 152.