Peculiarities in the electrocatalytic behavior of ultralow platinum deposits on gold synthesized by galvanic displacement

    Peculiarities in the electrocatalytic behavior of ultralow platinum deposits on gold synthesized by galvanic displacement B.I. Podlovchenko, Yu.M. Maksimov PII: DOI: Reference:

S1572-6657(15)30061-8 doi: 10.1016/j.jelechem.2015.08.004 JEAC 2228

To appear in:

Journal of Electroanalytical Chemistry

Received date: Revised date: Accepted date:

14 May 2015 29 July 2015 2 August 2015

Please cite this article as: B.I. Podlovchenko, Yu.M. Maksimov, Peculiarities in the electrocatalytic behavior of ultralow platinum deposits on gold synthesized by galvanic displacement, Journal of Electroanalytical Chemistry (2015), doi: 10.1016/j.jelechem.2015.08.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

Peculiarities in the electrocatalytic behavior of ultralow platinum deposits on gold synthesized by galvanic displacement

IP

T

B. I. Podlovchenko*, Yu. M. Maksimov.

Department of Chemistry, Moscow State University, Moscow 119992, Leninskie Gory, Russian

SC R

Federation

Abstract

NU

Composites Ptn0Au are synthesized in PtCl42- solutions by galvanic displacement of Cuad monolayers (MLCuad) from polycrystalline (pc) Au (to afford Pt10Au), and also from Pt10Au

MA

(Pt20Au) and Pt20Au (Pt30Au). The data of open-circuit potential transients, CVA, SEM, and XPS studies indicate that the displacement of MLCuad proceeds not layer-by-layer but to form Pt clusters. The degree of blocking of the Au surface by Pt is approximately (%): 65 (Pt10Au), 80

D

(Pt20Au) and 90 (Pt30Au). These Ptn0Au deposits simulate the gradual transition of Pt coatings

TE

close to monolayer to those formed by small Pt particles. The specific rates of methanol oxidation reaction (MOR) (per cm2 of EASAPt) increase in the row: Pt10Au < Pt20Au < Pt30Au <

CE P

pc Pt. For the formic acid oxidation (FAOR), the inverse dependence is observed: pc Pt < Pt30Au < Pt20Au < Pt10Au. The difference is explained by the fact that the chemisorption of СН3ОН requires a larger area (≥ 3 Pt surface atoms), whereas its single-site adsorption results in formic acid oxidation by the direct path. It is assumed that the Pt/Au interface of particles plays the

AC

important role in the formation of active sites (“ensembles”) for FAOR. The high specific mass activities (mA/mg Pt) of Ptn0Au composites is observed not only in FAOR, but also in MOR, which is associated with the high degree of dispersion of these Pt deposites 2

m EASA/g Pt).

Keywords: Au electrode, galvanic displacement of Cu adatoms, Pt multilayers, MOR, FAOR

1.Introduction

*

Corresponding author. Tel.: +7 495 939 4027; fax: +7 495 932 8846.E-mail address:

[email protected] (B.I. Podlovchenko). ISE member.

(~ 90-170

ACCEPTED MANUSCRIPT

2

Since the 60s of the last century up to nowadays, the electrooxidation of HCO compounds (small organic molecules) was always in the focus of attention of scientists [1-8]. First of all this is associated with the fuel cell problem. Fuels such as methanol and formic acid are the best alternative to hydrogen. Platinum is the best single-component catalyst for the methanol

T

oxidation reaction (MOR) and Pd is the best catalyst for formic acid oxidation (FAOR).

IP

However, at present it is the nanostructurized catalysts built not of “pure” platinum metals but of

SC R

their composites with the other d-metals that are the most popular. Their use allows one to decrease the consumption of the more expensive component and/or to enhance its activity. The promising methods of enhancing the activity of a Pt catalyst consist in modifying its

NU

surface with ultralow amounts of another noble metal (Pd [9-11] and Au [12,13]) and also dispersing Pt microamounts over the surface of some other noble metal (Pd [14-16], Au [13,17-

MA

22]). Among numerous chemical and electrochemical methods proposed for modifying the Pt catalyst surface, the method of galvanic displacement (GD) (also called galvanic replacement, spontaneous deposition, surface-limited redox deposition) [17, 21-25] shows the best promise. In

D

this method, adatoms of the less electropositive non-noble metal (most often, Cu or Pb) serve as

formal equations

TE

the mediators for the noble metal deposition. The modification of platinum can be described by

CE P

M1ad/Pt + (n/z) M2z+→ (n/z)M20/Pt +M1n+, M1ad/M2 + (n/z)Ptz+ → nPt0/M2 + M1n+, where M1 is the non-noble metal (M1ad is its adatom) and M2 is the noble metal.

AC

The GP method is characterized by simplicity and easiness of dosing the deposited metals, because the deposit amount is determined by the amount of displaced metal M1. MOR and FAOR are the complex reactions which can proceed by several parallel routes [1-8, 13, 22, 27-31]. Their common feature is that the current-determining route (the direct path) proceeds via adsorbed intermediate species weakly bound with the surface. The second route is the oxidation through strongly bound chemisorbed species (SCS), which is characterized by relatively low rates and inhibition of the direct-path oxidation by SCS. The main difference between the MOR and FAOR mechanisms falls to the region of low potentials (≤ 0.6 В vs. RHE), which is of greatest interest for fuel cells. The oxidation via the current-determining path is limited by the dehydrogenation stage for FAOR [3,6-8,28] and by the interaction by adsorbed species with adsorbed oxygen (probably, in the form of ОНads) for MOR [1,2,5]. There are also differences in the composition of adsorbed species both weakly bound and SCS [2, 5-7, 22]. According to the literature data, the acceleration of СН3ОН and НСООН oxidation can be achieved by increasing their amount and/or by changing the binding energy of OHads (the

ACCEPTED MANUSCRIPT

3

bifunctional mechanism) [32, 33], the geometric factor [6,22,34,35], the ensemble effect [36-38] and also by modifying the electronic state of the platinum surface [13, 39,40]. The Pt-Au system turned out to be suitable for developing scientific concepts on the nature of electrocatalytic activity of Pt and the ways of its enhancement and also on the oxidation

T

mechanisms of СН3ОН и НСООН [12,13,17-22]. The studies on gold electrodes decorated with

IP

small amounts of Pt gained in importance for practice when it was shown that the specific

SC R

activity of FAOR on Pt(Au) electrodes much exceeds the activity of monoplatinum catalysts. At the same time, no promoting effect of Au on the activity of Pt surface in MOR was observed [12,22,38].

NU

In our previous study [22], we considered the electrocatalytic behavior of electrodes of polycrystalline (pc) gold decorated with submonolayers of Pt by the galvanic displacement of copper adatoms (Ptx0Au, x≤1). It was shown that the displacement of a monolayer (ML)Cuad

MA

failed to afford a monolayer Pt coating. On Ptx0Au, the specific rates of MOR (per cm2 of the electrochemically active Pt surface, i.e., EASAPt) were found to be lower than on compact pc Pt. At the same time, for FAOR the strong acceleration was observed. In [22], we also discussed the

D

possible reasons for such a different effect of the Pt-Au contact on the platinum activity in MOR

TE

and FAOR.

In the present study, Pt was deposited on pc Au in amounts corresponding to the

CE P

displacement of from one to three MLCuad (the resulting deposit did not exceed 1.8 µg/cm2 EASAAu). We followed the changes in the adsorption properties and electrocatalytic activity of Pt (MOR and FAOR) in Ptn0Au composites (n=1, 2, 3) with the increase in n. These composites

AC

simulate the transition from Pt coatings close to a monolayer to the layers of Pt clusters. Analyzing the changes in the properties of Ptn0Au observed with the increase in n is important as regards both the further development of fundamental concepts on the electrocatalysis by Pt-Au composites and the assessment of prospects for their practical application.

2. Experimental The materials, the cell, the apparatus used and the procedure of Au electrode preparation to measurements were the same as in [22]. The roughness factors of electrodes were ~3.3 for Au and ~ 5.2 for Pt. The electrochemically active surface of gold EASAAu (the geometric surface was 1 cm2) was determined based on the oxygen adsorption [41,42]; the EASAPt was determined based on hydrogen adsorption [43]. The measurements were carried out at 19±10С. The working electrode potentials are shown with respect to the reversible hydrogen electrode (RHE). The monolayer of copper atoms (MLCuad) on the Au electrode was formed in solution of 2mM CuSO4 + 0.5M H2SO4 at E = 290 mV [22]. Immediately after opening the circuit, a portion

ACCEPTED MANUSCRIPT

4

of deaerated solution of 10-3M K2PtCl4 + 0.5M H2SO4 was added under argon pressure to the working compartment in order to reach the 10-4 M PtCl42- concentration. The transient of opencircuit potential was recorded to the point of establishment of its stationary value Est (the stationarity criterion dE/dτ <0.4 mV/min). The working compartment of the cell was repeatedly

T

washed with the supporting electrolyte solution, after which the measurements were carried out

IP

on the resulting electrode. The electrode corresponding to the first MLCuad displacement was

SC R

designated as Pt10Au. Note that the subscript indicates not the coverage of the Au surface with platinum but the amount of displaced copper expressed in ML [22]. The platinum amount in the Pt0Au composite was increased as follows. On the freshly synthesized Pt10Au, CVA were recorded first in 0.5 M H2SO4 (in the E range of 0.05 – 1.45 V) and then in 2 mM CuSO4 + 0.5M

NU

H2SO4 (in the E range of 0.29 – 1.45 V). In the latter solution, a new MLCuad was formed on Pt10Au. After this the same operations as those undertaken when preparing Pt10Au on Au were the same procedures from Pt20Au.

MA

performed. As a result, the Pt20Au electrode was prepared. The Pt30Au electrode was formed by The electrocatalytic activity of Pt0nAu electrodes was tested in solutions of 0.5M CH3OH

D

+ 0.5M H2SO4 and 0.5M HCOOH + 0.5M H2SO4. The CVA were recorded with 20-s exposure

TE

at the lower potential scan limit. In this paper, we show the steady-state CVA. Voltammetric curves of MOR and FAOR were recorded on Ptn0Au electrodes formed from Pt n-10Au and never

CE P

before subjected to polarization measurements in CH3OH and HCOOH solutions.

3.Results and discussion

AC

3.1. Transients of open-circuit potential Curves in Fig.1 show the potential variation on MLCuadAu (1), MLCuadPt10 (2), MLCuadPt20Au (3) and MLCuadPt (4) electrodes after bringing them in contact with PtCl42anions under open-circuit conditions. The delays observed in the potential region from 300 to ~750 mV correspond to the removal of Cuad according to the reaction Cuad + PtCl42- → Cu2+ + Pt0 + 4Cl-.

(1)

The time required for the MLCuad removal increased with the increase in the gold coverage with platinum (curves 1 - 3), which agrees with the high rates of Cuad displacement from pc Au as compared with pc Pt [44]. However, for the formation of Pt30Au too, the time of MLCuad displacement (curve 3) was still substantially lower than the time of MLCuad displacement from pc Pt (curve 4). This allows us to assume the presence of Au areas free of Pt on the Pt30Au surface and/or the acceleration of reaction (1) on the Pt atoms in contact with Au. After the removal of MLCuad, the potential rise in curves of Fig. 1 was determined by the changes in the electrode total charge Q [43-45] as a result of reaction

ACCEPTED MANUSCRIPT PtCl42- + 2е → Pt0 + 4Cl-.

5

(2)

The stationary potential (Еst) on Ptn0Au electrodes was established the higher the larger n, but did not exceed 900 mV. At these potentials, oxygen is not yet adsorbed on Au but is adsorbed in small amounts on Pt and its adsorption is reversible [43]. It can be assumed [44] that the reaction

T

Pt0 + H2O +e ↔ Pt0OHads + H+is potential determining. However, Est is of the mixed nature,

IP

because the surface concentration of OHads is determined by compensation of rates of the

SC R

reduction of PtCl42- and the oxidation of traces of organic substances and also of PtCl42- to PtCl62. As was demonstrated in [44], the rate of reaction (2) is substantially lower on Au than on Pt. Insofar as on Pt10Au a considerable part of surface is formed by Au [22], Est was established on

NU

this electrode at a value lower (by ~60 mV) (curve 1) as compared with Pt (curve 4). As the amount of deposited Pt increased (with the increase in n) and, correspondingly, the free gold surface fraction decreased, the potential Est shifted in the positive direction (curve 2 and 3).

MA

The peculiar shape of the E vs. τ curve at the formation of Pt10Au (curve 1 Fig.1), namely, the presence of a maximum, was already discussed in [44]. According to Fig.1, this effect is less pronounced at the displacement of the 2nd MLCuad (curve 2). At the displacement of

D

the 3rd MLCuad (curve 3), the maximum was virtually absent as in the curve of MLCuad

TE

displacement from Pt (curve 4). It seems reasonable that as the amount of Pt deposit increased, the effect of Au support on the process of MLCuad displacement weakened. At the same time, for

CE P

Pt30Au also, Est (curve 3) was considerably lower as compared with the displacement of MLCuad by platinum from a Pt support (curve 4). This can be largely associated with the differences in the surface structure of Pt deposits on Au and on smooth pc Pt. The reactions that determine Еst,

AC

are obviously structure-depending.

3.2. CVA

Figure 2 shows CVA for electrodes prepared by the displacement of MLCuad by platinum from Au (Pt10Au), Pt10Au (Pt20Au) and Pt20Au (Pt30Au). All these curves demonstrate the well pronounced regions of Нads ionization-adsorption in the low potential range and also the regions of oxygen adsorption (at Е≥0.75 V in the anodic scan) and its removal (peak at Е ≈ 0.65 V in the cathodic scan) from platinum. The anodic scan at Е ~ 1.38 V demonstrated a bend corresponding to the beginning of adsorption of Оads on Au. This bend was also observed in CVA for Pt30Au (curve 3), which points to incomplete blocking of Au atoms on the surface even after the 3-fold displacement of Cuad monolayers. To determine the oxygen adsorption on the part of gold surface that remained unoccupied by Pt, we recorded CVA up to 1.7 V [22,41] (these curves for Au, Pt10Au and Pt30Au are shown in the insert to Fig. 2). The fraction of gold surface non-blocked by platinum was determined as the

ACCEPTED MANUSCRIPT

6

ratio of the cathodic peak area at Е = 1.15 V on Ptn0Au electrodes to the area of the similar peak on Au. These fractions were (rough estimates) 0.35, 0.20, and 0.10 for Pt10Au, Pt20Au, and Pt30Au, respectively. This alone suggests that the Cuad replacement by platinum from Ptn0Au the displacement of a MLCuad from Au observed earlier [22].

T

electrodes proceeds by the mechanism different from the layer-by-layer mechanism, similarly to

IP

The complicated 3D mechanism of formation of Pt deposits at the successive displacement of

SC R

copper monolayers follows from the unexpectedly strong increase in the hydrogen adsorption observed in the row Pt10Au → Pt20Au → Pt30Au. According to the simplest mechanism of displacement, a Pt atom takes the place of the displaced Cuad; and, hence, the surface gain should

NU

be relatively small. The gain in the surface accessible to Нads suggests that the processes of Cuad ionization and Pt adsorption are sufficiently separated spatially and the electrons produced in the reaction Cuad – 2е → Cu2+ are consumed in the formation of new Pt clusters. How much this

MA

affects the variation in the Pt surface area can be seen from two schematic versions of Cuad

AC

CE P

TE

D

displacement:

A simple calculation shows that for the layer-by-layer displacement (version Ι), the number of sites for hydrogen adsorption should increase by a factor of ~1.2, whereas at the formation of a new cluster (version ΙΙ) the number of sites will increase by a factor of ~1.7. Moreover, the unexpectedly large increase in the hydrogen adsorption upon the transition Pt10Au → Pt30Au can be associated with the lower H coverage on the Pt monolayer on Au. The results of [22] make it possible to compare the number of Pt atoms deposited by displacing a submonolayer Cuad coverage from Au (AQAuCu/2F, where QAuCu is the charge consumed in the removal of Cuad; A, F are the constants of Avogadro and Faraday, respectively) with the number of H atoms adsorbed on this Pt (AQPtH/F). Their ratio at low and medium Cuad coverage is close to 1. Thus, no substantial difference is observed between the hydrogen adsorption on submonolayer Pt coatings and on the surface atoms of compact Pt.

ACCEPTED MANUSCRIPT

7

Figure 3 shows the anodic potentiodynamic curves of the removal of Cuad monolayers from PtnAu (curves 1-3). Their comparison with the analogous curve of electrodesorption of MLCuad from Au (insert in Fig.3) points to dramatic changes in the energy spectrum of Cuad. The

T

potentials of the total removal of Cuad become much more positive and overlap the potentials of

IP

the beginning of adsorption of Оads(ОНads) on Pt. The double layer region is absent. For the

SC R

potential scan rate used (v = 50 mV/s), the superposition of regions of Cuad desorption and Оads adsorption is typical of Pt electrode (insert in Fig. 3). With the transition Pt10Au → Pt20Au → Pt30Au, the position and shape of Cuad desorption peaks change, which points to modifications in

NU

the structure of Pt0 clusters formed and also to the changes in the nature of their contact with Au. Based on the curves of MLCuad electrodesorption from Pt (insert in Fig. 3), Pt10Au and Pt20Au

MA

electrodes (curves 1 and 2, Fig.3), the amounts of Cuad (in charges QCu) displaced upon the formation of Pt10Au, Pt20Au and Pt30Au were estimated and then the amounts of deposited Pt were calculated for each step of displacement. Table shows the corresponding values (in

D

parentheses, the total amounts of Pt deposit are shown).

2

CE P

1

EASAPt, cm2

QCu, µC

mPt, µg

Specific EASAPt, m2/g

2.2

1320

1.3

169

3.4

1960

1.9 (3.2)

105

5.1

2150

2.6 (5.8)

88

AC

n

TE

Characteristics of Ptn0Au electrodes

3

Based on the hydrogen regions of CVA anodic branches (Fig. 2), the EASAPt values were determined (using the value of 210 µC/cm2) and then their specific values (in m2/g) were found. It is evident that the degree of dispersion of Pt remained high, albeit decreased with the increase in n. It should be noted that the mPt values in table are slightly overestimated because we ignored the additional deposition of Pt due to the changes in the total electrode surface charge [43-45]. It is impossible to precisely determine this error; however, it can be approximately assessed as not exceeding 10%.

ACCEPTED MANUSCRIPT

8

3.3. SEM and XPS. Comparing the SEM images (Fig.4) for Au (а) and Pt10Au (b) shows that Pt is relatively uniformly distributed over the surface during the 1st displacement of the MLCuad. At the same time, in image b, one can distinguish the Au surface area free of Pt and also the Pt «aggregates»

T

resembling ridges. This confirms the conclusion drawn above based on electrochemical

IP

measurements (see Section 3.2 and [22]) that Pt does not form a monolayer when displaces a

SC R

MLCuad from the Au surface.

As the amount of deposited platinum increases during its gradual substitution for Cuad monolayers, the clusters grow and their conglomeration intensifies, which is clearly seen in the

NU

image for Pt30Au (c). The size of conglomerates was 25-50 nm, ignoring the cohered particles. Were these conglomerates crystals, then assuming their average diameter (d) to be 30 nm, the specific surface of the Pt deposit can be assessed based on the sphere model as 9.5 m2/g.

MA

According to electrochemical data (table) for the Pt30Au sample, its SPtH is almost one order of magnitude higher; and, hence, the visible particles consist of much finer particles-clusters with d≈ 3.0 nm (according to the sphere model). This agrees with version II of the scheme of Cuad

D

displacement.

TE

Figure 5 shows the overall XPS spectra (for emission angle 900) for Pt10Au and Pt30Au. By and large, the spectra are similar; however, the ratios of heights of Pt and Au peaks are

CE P

different. Insofar as before being studied in the vacuum chamber, the samples were in contact with air, their surface could contain considerable amounts of impurities. Moreover, the analysis depth was ~ 4 nm, which is many times larger than the thickness of the first atomic layer on

AC

samples. This is why, similarly to [22], the quantitative estimates were drawn based on the atomic ratio Pt/Au. The latter was 0.25 for Pt10Au and 2.1 for Pt30Au. Such a large difference confirms that a loose Pt layer is formed from agglomerates of clusters and cohered particles. Such a layer should be rich of structural defects which can considerably affect the electrocatalytic activity of Pt. The binding energies of Pt4f7/2 electrons for Pt10Au and Pt30Au were found to be 70.8 and 71.1 eV, respectively. For Pt10Au, this value was considerably lower (by 0.4 eV) than for bulk platinum (71.2 eV [46]), which is typical Pt surface atoms and also for Pt atoms in contact with gold atoms [47-49]. The fact that Pt4f7/2 binding energies in the Pt30Au sample are close to those of bulk Pt suggests that the majority of Pt atoms are present in the form of relatively coarse particles. 3.4. MOR According to Fig. 6, the specific activity of Pt deposited on Au in MOR calculated per cm2 of EASA increases in the series: Pt10Au < Pt20Au < Pt30Au, approaching the activity of smooth pc

ACCEPTED MANUSCRIPT

9

Pt (curve 4). According to [22], for submonolayer Pt coatings on Au (Ptx0Au), the activity changed in the series Pt0.25 0Au < Pt0.5 0Au
T

number of studies (e.g., see [34,35,50]). At the high potentials corresponding to the MOR peak

IP

(~0.8 V), the rate of this reaction is largely limited by the dehydrogenation stage [2, 5]. Insofar

SC R

as the chemisorption of a СН3ОН molecule requires no less than 3 Pt surface atoms [2,5,6,22], the MOR rate is largely determined by the «geometric factor», i.e., the necessity to have sufficiently large surface regions suitable for the destruction of СН3ОН. The effect of this factor weakens with the growth of Pt particles. No substantial effect of the «electronic factor»

NU

on the activity of Ptx0Au particles in MOR was observed for submonolayer Pt coatings [22]; the more so, this effect can hardly be expected for Ptn0Au composites.

MA

As was noted earlier [22], a substantial difference in the ratio of MOR peaks in the anodic and cathodic scans was observed between Pt0.50Au and pc Pt (1.6 and 1.15, respectively). For Pt30Au and pc Pt, this difference is virtually absent (1.15 and 1.2, according to curves in the insert to

D

Fig.6), which shows that the effect of the Au support on the activity of Pt becomes weaker.

TE

If we multiply the MOR peak currents (Fig. 6) by the specific EASAPt values (table), we obtain the specific mass activity of Pt in Ptn0Au (A/g): 729 (in Pt10Au), 580 (in Pt20Au) and 640

CE P

(in Pt30Au). Taking into account the possible errors in assessing the specific EASAPt values and imperfect reproducibility of currents from one sample to another, the differences in the specific mass activity of Pt can be considered in the first approximation as insignificant. For deposits

AC

under consideration, the degree of utilization of Pt in MOR is very high: the specific mass activity of Ptn0Au is 6-7 times higher than the activity of commercial Pt/C catalysts [51]. The high degree of dispersion of Pt is the key factor. Above, as in the majority of cited studies by other authors, we tested the non-stationary MOR currents in CVA in the region of high Е (~ 0.8 V). It should be noted that from the point of view of using MOR in fuel cells, it is the stationary MOR currents in the region of low potentials (< 0.6 V) that are of the greatest interest. Unfortunately, on our samples, these currents turned out to be very low (due to accumulation of chemisorption products, first of all, СОads) and, the more so, insufficiently reproducible to be used for correct comparison of the activity of Ptn0Au at different n.

3.5. FAOR Figure 7 shows anodic CVA branches for Ptn0Au (curves 1-3) and pc Pt electrodes in solution of 0.5 М НСООН + 0.5 М Н2SO4. It is known [6,7,29,30,52] that currents below ~ 0.60 V are

ACCEPTED MANUSCRIPT

10

mainly determined by НСООН electrooxidation via the direct path in the presence on the surface of strongly chemisorbed species (SCS). At the higher Е, these currents are supplemented by currents of electrodesorption of strongly chemisorbed species (the peak at ~ 0.9 V), and then the FAOR is suppressed by the oxygen adsorption. The specific activity of Pt

T

in the Ptn0Au composites with respect to FAOR is much higher than the activity of pc Pt, i.e.,

IP

the effect of the Au support is opposite to that observed for MOR. For Е = 0.6 V, the specific

SC R

activity varies in the row Pt10Au > Pt20Au > Pt30Au > pc Pt. The simplest explanation can be reduced to the size factor which for FAOR in positive, in contrast to MOR [34,52,53]. In the course of linear potential scanning, the FAOR currents being nonstationary strongly

NU

depend of quite a number of factors, namely, the electrode surface pretreatment, the lower and upper potential scan limits, the scan rate, the time of exposure at the potential limits, etc. Nonstationary currents are poorly reproducible. The factors mentioned above are among the

MA

main reasons for the large difference in the quantitative estimates of the degree of promotion of FAOR in catalytic systems. Such difference can reach a factor of 10 and more [13,17,19,21,22]. According to Fig.7, at 0.60 V, the specific surface activity of Pt10Au in FAOR exceeds the

TE

factor of ~ 1.5.

D

activity of pc Pt by a factor of ~ 5, whereas the activity of Pt30Au exceeds that of pc Pt by a As was mentioned above, the stationary currents at Е < 0.6 V are of the greater interest.

CE P

According to chronoamperograms in Fig. 8, the FAOR currents dramatically decrease in time after the interruption of the anodic scan at a given Е. This suggests that the process of НСООН electrooxidation is inhibited due to accumulation of SCS (COads, :O2CHads [6,30,31]). The

AC

inhibition on Ptn0Au electrodes is less pronounced as compared with pc Pt. Quasistationary specific currents observed in 450 s on Pt10Au were 10 times higher than on pc Pt, whereas on Pt30Au these currents were ~ 4 times higher than on pc Pt. It is evident that the ratios of activities of Ptn0Au to pc Pt, estimated based on quasistationary currents, strongly differ from the corresponding ratios for nonstationary currents discussed above. This confirms the statement that CVA are insufficient for estimating the activities of Pt-based catalysts in reactions of electrooxidation of small molecules. The explanations of the promoting effect of Au on the specific rate of FAOR on Pt for the case of monolayer Ptx0Au coatings (х < 1) and Pt10Au were discussed in detail in [22]. There, the general scheme of FAOR based on our and reference data was also shown. The FAOR acceleration on Pt «islets» on gold as compared with bulk Pt was associated with blocking of sites for SCS formation and also with the formation of new active centers (“ensembles”) for the FAOR direct path which proceeds via single-site adsorption of НСООН. It was assumed that the Pt/Au interface of particles and also the Pt atoms occupying defective sites on the Au

ACCEPTED MANUSCRIPT

11

support play the important role in the formation of new active sites. The «electronic factor» can also play a certain role. The found results for Ptn0Au agree with this interpretation. As Pt clusters are formed and grow with the increase in n, the effect of the Au support should weaken (which is the case, see Figs 7 and 8), particularly, due to the strong decrease in the fraction of Pt

T

atoms that are in contact with gold. For Pt30Au, the retention of the much higher activity as

IP

compared with bulk Pt can be associated to a certain extent with the high defectiveness of Pt

formation of centers for the FAOR direct path.

SC R

nanoparticles, which should complicate the multi-site adsorption of SCS and favor the The overall CVA curves for FAOR on Pt30Au and pc Pt are close to one another as regards both their shape and the peak potentials (insert to Fig.7). The acceleration of FAOR due

NU

to the presence of the Pt–Au contact is also observed in the cathodic scan for Pt30Au; however, this effect is weak, namely, the ratio of peak currents for Pt30Au and pc Pt is only ~ 1.4. It is

MA

interesting that for Pt0.50Au, according to [22], the maximum specific FAOR current in the cathodic scan exceeds only by a factor of ~1.1 the current on pc Pt. In the cathodic scan, the current is mainly determined by the HCOOH dehydrogenation rate on the platinum surface freed

D

from SCS and oxides [6, 54]. Hence, these data suggest that the promoting effect of Au on

TE

FAOR is mainly associated with the reduction of the blocking effect of strongly chemisorbed species due to acceleration of their oxidation.

CE P

According to quasistationary currents (Fig. 8), the specific mass activity of Pt in FAOR was (mA/mg): Pt10Au – 150, Pt20Au - 38, Pt30Au – 28. Such a strong change in the specific activity in FAOR was associated with the decrease in both the specific surface of Pt and its

AC

activity with the increase in n. Nonetheless, even for Pt30Au, the specific mass activity of Pt in FAOR exceeds its stationary activity in a Pt electrodeposit by a factor of almost 20 (~ 1.5 mA/mg of Pt at 0.4 V [11]).

4. Conclusions 

The electrocatalytic properties of three Pt deposits, namely, Pt10Au, Pt20Au and Pt30Au, synthesized by the substitution of Pt for MLCuad on Au, Pt10Au and Pt20Au, respectively, are compared.



The MLCuad monolayers are shown to be displaced by platinum to form Pt clusters and their conglomerates rather than by the layer-by-layer mechanism. Even after the displacement of three MLCuad, about 10% of Au surface still remains unblocked by Pt.

ACCEPTED MANUSCRIPT 

12

The MOR specific rate (per cm2 of EASAPt) increases in the row Pt10Au < Pt20Au < Pt30Au < pc Pt. This effect was explained by the increase in the number of «areas» suitable for the multi-site adsorption of СН3ОН molecules.



For FAOR, the inverse order (as compared with MOR) of variation of the specific

T

activity of tested composites is observed: Pt10Au > Pt20Au > Pt30Au > pc Pt. Such an

IP

order was explained by the fact that the number of active sites («ensembles»)

SC R

decreased with n for the FAOR direct path which proceeds via the single-site adsorption of НСООН. 

It was assumed that the Pt/Au interface plays an important role in the formation of

NU

new active centers for FAOR. As n increased, the specific ratio of Pt atoms in contact with Au decreased, which correlated with the changes in the activity of Ptn0Au. 

Composites Ptn0Au demonstrated the very high specific mass activity of Pt (in A/g)

MA

not only in FAOR but also in MOR, which was associated with the high degree of

TE

FIGURE CAPTIONS

D

dispersion of Pt (~ 90–170 м2 EASA/g Pt).

Fig.1. Transients of open-circuit potential corresponding to displacement of MLCuad from (1)

CE P

Au, (2) Pt10Au, (3) Pt20Au and (4) Pt in 0.1mM K2PtCl4 + 0.5M H2SO4. Fig.2. CVA measured in 0.5M H2SO4 on (1) Pt10Au, (2) Pt20Au, (3) Pt30Au. Insert: CVA for (1) Au, (2) Pt10Au and (3) Pt30Au with the upper scan limit of 1.7 V. v = 50 mV/s.

AC

Fig.3. Anodic potentiodynamic curves measured in 2mM CuSO4 + 0.5M H2SO4 on (1) Pt10Au, (2) Pt20Au, (3) Pt30Au. Insert: analogous curves for Au and Pt. v = 50 mV/s. Fig.4. SEM images of (a) pc Au, (b) Pt10Au, (c) Pt30Au. Fig.5. Overall XPS spectra of (1) Pt10Au and (2) Pt30Au collected at the emission angle of 900. Fig.6. Positive going CVA scans in 0.5M СН3ОН +0.5M H2SO4 solutions on (1) Pt10Au, (2) Pt20Au, (3) Pt30Au and (4) pc Pt. v = 50 mV/s. Insert: overall CVA for (1) Pt30Au and (2) pc Pt. Рис.7. Positive going CVA scans in 0.5M HCOOH +0.5M H2SO4 solutions on (1) Pt10Au, (2) Pt20Au, (3) Pt30Au and (4) pc Pt. v = 50 mV/s. Insert: overall CVA for (1) Pt30Au and (2) pc Pt. Рис.8.Chronoamperograms of (1) Pt10Au, (2) Pt20Au, (3) Pt30Au and (4) Pt electrodes at E=400 mV in 0.5M HCOOH +0.5M H2SO4 solution.

References [1] M.W. Breiter, Electrochemical Processes in Fuel Cells. Berlin: Springer-Verlag,1969.

ACCEPTED MANUSCRIPT

13

AC

CE P

TE

D

MA

NU

SC R

IP

T

[2] O.A. Petry, B.I. Podlovchenko, A.N. Frumkin, Hira Lal, J. Electroanal. Chem. 10 (1965) 253-269. [3] R. Parsons, T. Vander Nood, J. Electroanal. Chem. 257 (1988) 9-45. [4] W. Vielstich, A. Lamm, H.A. Gasteiger, Handbook of Fuel Cells: Fundamentals, Technology, Applications, John Wiley and Sons, 2003. [5] T. Iwasita, Electrochim. Acta 47(2002) 3663-3674. [6] M. Neurock, M. Janik, A. Wieckowski, Faraday Disscussion 140 (2009) 363. [7] J.V. Perales-Rondon, A. Ferre-Vilaplana, J. M. Feliu, E. Herrero, J. Am. Chem. Soc. 136 (2014) 13110−13113. [8] B.I. Podlovchenko, Yu.M. Maksimov, S.A. Evlashin, T.D. Gladysheva, K.I. Maslakov, V.A. Krivchenko, J. Electroanal.Chem. 743 (2015) 93-98. [9] N. V. Long, M. Ohtaki, T. D. Hien, J. Randy, M. Nogami, Electrochim. Acta 56 (2011) 9133-9143. [10] R. Huang, Y.-H. Wen, Z.-Z. Zhu, S.-G. Sun, J. Phys. Chem. C 116 (2012) 8664-8671. [11] T.D. Gladysheva, A.Yu. Filatov, B.I. Podlovchenko, Mendeleev Commun. 25 (2015)56-58. [12] S. Motoo, M. Watanabe, J.Electroanal. Chem. 60 (1975) 259-266. [13] M.D. Obradović, A.V. Tripković, S.L. Gojković, Electrochim. Acta 55 (2009) 204-209. [14] K. Sasaki, J.X. Wang, H. Naohara, N. Marinkowic, K. More, H. Inada, R.R. Adžić, Electrochim. Acta 55 (2010) 2645-2652. [15] Y.-Y. Chu, Z.-B. Wang, Z.-Z. Jiang, D.-M. Gu, G.-P. Yin, J. Power Sources 203 (2012) 1725. 16. M. Liao, Y. Wang, G. Chen, H. Zhou, Y. Li, C.-J. Zhong, B.H. Chen, J. Power Sources 257 (2014) 45-51. [17] Y. Xia, J. Liu, W. Huang, Z. Li, Electrochim. Acta 70 (2012) 304–312. [18] S. Wang, N. Kristian, S. Jiang, X. Wang, Electrochem. Commun. 10 (2008) 961-964. [19] N. Kristian, Y. Yan, X. Wang, Chem. Commun. (2008) 353-355. [20] R. Zhou, R. Yue, F. Jiang, Y. Du, P. Yang, C. Wang, J. Xu, Fuel cells 12 (2012) 971-977. [21] N. Kristian, Y. Yu, P. Gunawan, R. Xu, W. Deng, X. Liu, X. Wang, Electrochim. Acta 54 (2009) 4916-4924. [22] B.I. Podlovchenko, Yu. M. Maksimov, K.I. Maslakov, Electrochim. Acta 130 (2014) 351– 360 . [23] S.R. Branković, J.X. Wang, R.R. Adžić, Surf. Sci. 474 (2001) L173-L179. [24] B.I. Podlovchenko, U.E. Zhumaev, Yu.M. Maksimov, J. Electroanal. Chem. 651 (2011) 2930. [25] D. Gokcen, S.-E. Bae, S.R. Brankovic, Electrochim. Acta 56 (2011) 5545-5553. [26] S.Ambrosik, B.Rawlings, N.Vasiljević, N.Dimitrov, Electrochem. Commun. 44 (2014) 1922. [27] B.I. Podlovchenko, O.A. Petry, A.N. Frumkin, Hira Lal, J. Electroanal. Chem. 11 (1966) 12-25. [28] A.V. Smolin, B.I. Podlovchenko, Y.M. Maksimov, Russ. J. Electrochem. 33 (1997) 440446. [29] Y.X. Chen, S. Ye, M. Heinen, Z. Jusus, R.J. Behm, Langmuir 22 (2006) 10399-10408. [30] H. Okamoto, Y. Numata, T. Gojuki, Y. Mukouyama, Electrochim. Acta 116 (2014) 263270.

ACCEPTED MANUSCRIPT

14

AC

CE P

TE

D

MA

NU

SC R

IP

T

[31] K. Jiang, H.-X. Zhang, Sh. Zou ,W.-B. Cai, Phys. Chem. Chem. Phys. 16 (2014) 2036020376. [32] M. Watanabe, S. Motoo, J. Electroanal. Chem. 60 (1975) 267-273. [34] Yu.M. Maksimov, B.I. Podlovchenko, T.L. Azarchenko, Electrochim. Acta 43 (1998) 10531059. [35] S.H. Joo, K. Kwon, D.J. You, Ch. Pak, H. Chang, J.M. Kim, Electrochim. Acta 54 (2009) 5746-5753. [36] E. Casado-Rivera, Z. Gál, A.C.D. Angelo, C. Lind, F.J. DiSalvo, H.D. Abruña, Chem. Phys. Chem.4 (2003) 193-199. [37] N.M. Marković, H.A. Gasteiger, P.N. Ross Jr., X. Jiang, I. Villegas, M.J. Weaver, Electrochim. Acta 40 (1995) 91-98. [38] J. V. Perales-Rondón, E. Herrero, J. M. Feliu, J. Electroanal. Chem. 742 (2015) 90-96. [39] T. Bligaard, J.K. Norskov, Electrochim. Acta 52 (2007) 5512-5516. [40] E. Santos, W. Schmickler, Angew. Chem. Int. Ed.46 (2007) 8262-8265. [41] A.A. Michri, A.G. Pshchenichnikov, R.K. Burstein, Sov. Electrochem. 8 (1972) 351-353. [42] M.C. Santos, L.H. Mascaro, S.A.S. Machado, Electrochim. Acta 43 (1998) 2263-2273. 43. A.N. Frumkin, Potentsialy Nulevogo Zaryada (Zero-Charge Potentials), Nauka, Moscow, 1981. [44] B. I. Podlovchenko, Yu. M. Maksimov, Mendeleev Commun. 23 (2013) 157-159. [45] B.I. Podlovchenko, E.A. Kolyadko, J. Electroanal. Chem. 506 ( 2001) 11-21. [46] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photoelectron Spectroscopy, Chigasaki: ULVAC-PHI, Inc, 1995. [47] K. Duckers, K.C. Prince, H.P. Bonzel, V. Chab, K. Horn, Phys. Rev. B 36 (1987) 62926301. [48] S. Uhm, H. Jeon, J. Lee, J. Electrochem. Science and Technology 1 (2010) 10-18. [49] R.A.P. Smith, Platinum Metals Review 53 (2009) 109-110. [50] B.I. Podlovchenko, V.A. Krivchenko, Yu.M. Maksimov, T.D. Gladysheva, L.V. Yashina, S.A. Evlashin, A.A. Pilevsky, Electrochim. Acta 76 (2012) 137– 144. [51] Y. Tong, J. Pu, H. Wang, Sh. Wang, Ch. Liu, Zh. Wang, J. Electroanal. Chem. 728 (2014) 66–71. [52] R.P. Petukhova, V.G. Garzon, B.I. Podlovchenko, Vestn. MSU ser. Khim. 28 (1987) 6266. [53] F. J. E. Scheijen, G. L. Beltramo, S. Hoeppener, T. H. M. Housmans, M. T. M. Koper, J. Solid State Electrochem. 12 (2008) 483–495. [54] R.R.Adžić, A.V. Tripcović, J. Electroanal. Chem. 99 (1979) 43–53.

Acknowledgments This work was supported by the Russian Foundation for Fundamental Research, Project Nos. 15-03-03436 А and in part by M.V. Lomonosov Moscow State University Program of Development.

ACCEPTED MANUSCRIPT

15

Highlights. 

Ptn0Au composites (n =1÷3) were obtained by the successive displacement of three MLCuad in the PtCl42- solution. The displacement of MLCuad proceeded to afford Pt clusters rather than layer-by-layer.



In the series Pt10Au – Pt20Au – Pt30Au – pc Pt, the specific surface activity of Pt

IP

T





SC R

increased as regards MOR and decreased as regards FAOR.

The possible reasons for the different behavior of Ptn0Au activity with the increase in n in FAOR and MOR are discussed.

NU

Composites Ptn0Au demonstrated the high degree of Pt utilization not only in FAOR but

CE P

TE

D

MA

also in MOR.

AC



TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 1

16

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 2

17

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 3

18

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 4

19

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 5

20

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 6

21

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 7

22

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 8

23