Journal of Electroanalytical Chemistry 819 (2018) 136–144

Contents lists available at ScienceDirect

Journal of Electroanalytical Chemistry

journal homepage: www.elsevier.com/locate/jelechem

In situ determination of the nanostructure effects on the activity, stability T

and selectivity of Pt-Sn ethanol oxidation catalysts
Laura Calvilloa,b,⁎, Lucila Mendez De Leoc, Stephen J. Thompsona, Stephen W.T. Pricea,d,
Ernesto J. Calvoc, Andrea E. Russella,⁎⁎
a
Chemistry, University of Southampton, Highfield, Southampton SO17 1BJ, UK
b
Dipartimento di Scienze Chimiche, Università di Padova, Via Marzolo 1, 35131 Padova, Italy
c
INQUIMAE, Departamento de Química Inorgánica, Analítica y Química Física, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Argentina
d
Diamond Light Source Ltd., Harwell Science and Innovation Campus, Didcot OX11 0DE, UK

A R T I C L E I N F O A B S T R A C T

Keywords: Nanoparticle catalysts comprising two PtSn alloys with different Pt:Sn atomic ratios and a Sn modified Pt cat-
Pt-Sn nanostructures alyst were prepared in order to study the effect of the particle nanostructures on the activity towards the ethanol
Ethanol electrooxidation electrooxidation and the selectivity to CO2. An accurate model of the electronic and structural properties, ob-
In situ FTIRS tained by ex situ analysis, was established. Alloying of Sn with Pt causes the expansion of the lattice parameter of
In situ XAFS
Pt and modifies its electronic structure. In contrast, the deposition of Sn on the Pt surface has neither effect. The
Structural stability
activity of the catalysts towards ethanol oxidation was established voltammetrically and the CO2 selectivity via in
situ Fourier transform infrared spectroscopy (FTIRS). Results indicated that the modification of the electronic
environment of Pt in Pt-Sn alloys results in a weaker adsorption of the intermediates (acetaldehyde and acetic
acid), which desorb easily from the surface of the catalyst resulting in incomplete oxidation to CO2. In contrast,
when the electronic structure is not perturbed (Sn modified Pt sample), the amount of CO2 produced increases.
The stability of the different nanostructures under working conditions was investigated by in situ X-ray ab-
sorption spectroscopy (XAS) measurements, which show that initially both the Sn modified Pt and Pt-Sn alloy
nanostructures are stable under applied potential in the potential window studied and in presence of ethanol.
Accelerated aging studies showed that the Sn modified Pt nanostructure remained stable, whereas a significant
structural change was observed for the Pt-Sn alloys.

1. Introduction systems operating with these catalysts is still insufficient for more de-
manding practical applications, especially at low temperatures. The
The design of catalysts that are able to completely oxidize ethanol to overall efficiency of Pt-based electrocatalysts suffers the consequences
CO2 is an ongoing research challenge. Ethanol contains a CeC bond of effective catalyst poisoning by CO that is an intermediate of the
that is necessary to break in order to achieve its complete oxidation. The oxidation of any alcohol [2,3]. In order to minimize the poisoning ef-
interest in the ethanol oxidation reaction (EOR) comes from the fact fects and to increase the electrocatalytic activity, they are usually
that the use of ethanol as fuel in direct ethanol fuel cells (DEFCs) makes modified with oxophilic metals such as Ru, Sn, Mo, or Rh [4,5,6]. Using
these devices very promising for energy conversion, having the ad- these catalysts, however, results in high yields of partial oxidation
vantages of easier handling, storage and transport of the fuel than products like acetaldehyde and acetic acid, releasing two and four
gaseous hydrogen [1]. Their use becomes even more advantageous electrons per ethanol molecule, respectively, and hence generating
when bioethanol produced from waste biomass is used as a sustainable lower energy conversion efficiency.
source of the fuel. The binary Pt-Sn and ternary Pt-Rh-Sn catalysts are generally con-
Ethanol is difficult to oxidize on platinum or platinum alloys sup- sidered as the most effective ethanol oxidation reaction (EOR) catalysts
ported on carbon materials, which constitute the most common cata- [6,7,8]. Nevertheless, the origin of the promotion effect due to the
lysts for DEFCs. Binary and ternary platinum-based catalysts have been presence of Sn in the ethanol oxidation in acidic media is still under
tested as electrode materials, but the efficiency and/or durability of debate and contradictions are found in the literature. There seems to be

⁎
Correspondence to: L. Calvillo, Dipartimento di Scienze Chimiche, Università di Padova, Via Marzolo 1, 35131 Padova, Italy.
⁎⁎
Corresponding author.
E-mail addresses: laura.calvillolamana@unipd.it (L. Calvillo), a.e.russell@soton.ac.uk (A.E. Russell).

http://dx.doi.org/10.1016/j.jelechem.2017.09.060
Received 29 June 2017; Received in revised form 26 September 2017; Accepted 27 September 2017
Available online 28 September 2017
1572-6657/ © 2017 Elsevier B.V. All rights reserved.
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

an agreement about the positive effect of Sn on the electrooxidation of with ultrapure water and dried overnight. The materials obtained are
CO due to its capacity to activate water at lower potentials than Pt, labelled as Pt3Sn1/C and Pt1Sn1/C.
providing OHads species that help to oxidize the COads [9,10,11].
However, contradictory results are reported about the ethanol electro- 2.1.3. Electrodes
oxidation, which may be explained by the different Pt:Sn ratio and Pt- Electrodes of the Pt-Sn catalyst materials were manufactured by
Sn crystal structures used in these studies and/or by the extent of Pt:Sn painting Nafion-based inks on to carbon paper (TGP-H-60) with a final
alloying/mixing. For example, López-Suarez et al. state that the oxi- Pt loading of 0.6 mg Pt cm− 2. The catalyst inks were prepared by dis-
dation state of Sn, the Pt-Sn alloy formation and the Pt:Sn atomic ratio persing the catalyst in a mixture of water and Nafion® (10.85 wt% so-
have a minimal influence on the catalytic activity of PtSn/C materials lids in water, Johnson Matthey). The electrodes were then pressed at
[12]; whereas Du et al. affirm that the Pt:Sn atomic ratio and the 180 °C and 1 bar for 3 min. Circular button electrodes of 1.25 cm2 area
crystalline structure of Sn play an important role in the selectivity to were cut and hydrated in ultra-pure (Purite, 18 MΩ cm) boiling water
CO2 [13]. before their use.
Studies have shown that shape, structure, composition, and archi-
tecture are extremely important parameters [14,15], which affect cat- 2.2. Physicochemical characterization
alytic activities of bimetallic nanostructured materials. Therefore, in
order to design more efficient electrocatalysts, further understanding of The bulk composition analysis was performed by an energy dis-
the effects that the addition of other metals has on the Pt properties as persive spectroscopy (EDX) probe attached to a Philips XL-30 ESEM
well as relating these effects to catalytic activity, selectivity, and sta- microscope.
bility is needed. This requires the use of in situ techniques to investigate Transmission electron microscopy (TEM) was carried out on a JEOL
the dependence of the product distribution with the applied potential JEM 2100 LaB6 microscope operating at 200 kV, conducted at the
and to determine in situ structural changes induced by the applied Research Complex at the Harwell Campus, UK. In addition, STEM-
electrochemical conditions [16]. HAADF images for the Sn/Pt/C and Pt3Sn/C catalysts were obtained
In this work, we have prepared carbon supported Pt-Sn catalysts using a Titan FEI with Cs probe-corrected (1 Å) at the “Laboratorio de
with different nanostructures, establishing an accurate model of the Microscopias Avanzadas” at “Instituto de Nanociencia de Aragón -
electronic and structural properties of each nanostructure using ex situ Universidad de Zaragoza”. Energy dispersive X-ray spectroscopy (EDS)
characterization techniques and verifying that the structures are re- measurements with a resolution of 5 eV/channel were recorded in order
tained in the electrochemical environment and in the presence of to obtain the composition profiles for Pt and Sn along the metal na-
ethanol using in situ characterization techniques. The EOR activity of noparticles.
each catalyst has been measured using both thin film electrode studies X-ray diffraction (XRD) patterns were recorded using a SuperNova
and in situ FTIRS in order to determine the products obtained. This Dual Wavelength diffractometer with θ-θ configuration and operating
combined approach enables structure/property relationships to be de- with Cu Kα radiation (λ = 0.15406 nm) generated at 45 kV and 40 nm.
fined. Finally, the structures of the catalysts, fresh and after potential The mean crystallite size along the c direction (Lc) of the metal crys-
cycling, have been studied using in situ XAS, providing the first such tallites was calculated using the Scherrer equation with a value of
comprehensive characterization of both the initial state of the PtSn k = 0.9 using the Pt(220) width peak around 2θ = 70° to avoid the
nanostructures as well as their long-term structural stability. influence of the broad band of the carbon support at 2θ ~25°
[18,19,20,21]. The (220) peak was also used to determine the lattice
2. Experimental parameter, assuming the Pt fcc structure. From the values of the lattice
parameter, the atomic fraction of Sn in the Pt-Sn alloys were calculated,
2.1. Synthesis of catalysts and preparation of electrodes assuming that the Pt-Sn lattice parameter follows Vergard's law and
that the dependence of the lattice parameter on Sn content is the same
2.1.1. Sn modified Pt supported on carbon for supported and unsupported Pt: χSn = (lPtSn − lPt)/0.352, where
The commercial 20 wt% Pt/C catalyst from Johnson Matthey, la- lPtSn and lPt are the lattice parameters for the Pt-Sn/C and Pt/C, re-
belled as Pt/C-JM, was modified with one monolayer of Sn. 27.7 mg of spectively. Using these values, the amount of Sn alloyed (Snal) was
SnCl2·2H2O were dissolved in 30 mL of ultrapure water and subse- calculated from the relation: Snal = χSn / [(1 − χSn)(Sn/Pt)at], being
quently added dropwise under sonication to a Pt/C-JM dispersion, (Sn/Pt)at the nominal Sn/Pt atomic ratio.
which was prepared by the dissolution of 300 mg of Pt/C-JM catalyst in X-ray photoelectron spectra (XPS) were acquired with an ESCAPlus
30 mL of ultrapure water. After 30 min, an aqueous solution of NaBH4 Omicron spectrometer fitted with a Mg Kα (1253.6 eV) 150 W X-ray
(Sigma Aldrich, 9.3 mg of NaBH4 in 40 mL of ultrapure water) was source. Data processing was performed with XPSPEAK software to de-
added to the mixture dropwise under sonication. Once the addition of termine the chemical states of Pt and Sn and the surface composition of
NaBH4 was completed, the mixture was stirred for 1 h at room tem- Pt-Sn nanoparticles. The background was corrected using the Shirley
perature and, finally, the catalyst was filtered, washed with ultrapure method, and the binding energy of C 1s peak from the support at
water and dried overnight. This Sn modified Pt/C catalyst is labelled as 284.5 eV was taken as an internal standard. The Pt:Sn surface atomic
Sn/Pt/C. ratio was calculated from the Pt 4f and Sn 3d peak areas taking into
account the corresponding sensitivity factors.
2.1.2. Pt-Sn alloy supported on carbon
Two Pt-Sn alloys supported on carbon were prepared by the polyol 2.3. Electrochemical characterization
method [17], using H2PtCl6 (8 wt%, Sigma-Aldrich) and SnCl2·2H2O
(Sigma Aldrich) as metal precursors. The synthesis process involved the A systematic study applying conventional electrochemical techni-
dissolution of both metal precursors in 60 mL of ethylene glycol (EG) ques (cyclic voltammetry and CO-stripping voltammetry) was per-
under sonication conditions. The amount of each of the metal pre- formed in order to determine the efficiency of the catalysts in the
cursors was that necessary to obtain Pt:Sn atomic ratios of 3:1 and 1:1 ethanol oxidation reaction (EOR). The electrochemical measurements
and a total metal loading of 20 wt%. Then, the pH was adjusted to 11 were conducted in a standard three-electrode electrochemical cell. Pt
using a 1 M NaOH solution (in EG) and 240 mg of carbon (Vulcan XC- gauze was used as counter electrode and a Hg/Hg2SO4 electrode (MMS,
72R) was added and maintained under sonication for 15 min. Subse- calibrated as +0.695 V vs. RHE) placed inside a Luggin capillary was
quently, the system was refluxed at 195 °C for 2 h and then quickly used as reference electrode. All the potentials in the text are referred to
cooled in a cold water bath. Finally, the catalysts were filtered, washed the reversible hydrogen electrode (RHE). A 0.5 M H2SO4 solution,

137
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

prepared from high purity reagents (Fisher Scientific) and purged with to acquire in situ IR spectra. The cell was connected to a Jaissle IMP88
nitrogen gas, was used as supporting electrolyte. All the electro- Potentiostat (Germany) controlled by the computer via a digital to
chemical experiments were carried out at room temperature. The oxi- analog converter (Agilent USB AD/DA converter). The IR window was a
dation of ethanol was studied by cyclic voltammetry. 1 M CH3CH2OH in 1 in. (~ 25 mm) CaF2 equilateral prism (Harrick Scientific Technology,
0.5 M H2SO4 solution was used to carry out the experiments. Cyclic PleasantVille, NY). Prior to the assembly of the spectroelectrochemical
voltammetry experiments between 0.05 and 0.7 V vs. RHE were per- cell, the gold electrode was modified with the catalysts by depositing
formed. Ten conditioning scans at 0.1 V s− 1 were initially performed in 30 μL of catalyst ink prepared with 2 mg catalyst, 15 μL Nafion solution
order to get a good contact between the electrode and electrolyte and, (10.85 wt%) and 500 μL water. A Pt wire and an Ag/AgCl (3 M KCl)
subsequently, three more cyclic voltammograms were measured at were used as counter and reference electrodes, respectively. The cell
0.02 V s− 1. The third CV was used for comparison of the performance was filled with 0.1 M HClO4 solution and fifty conditioning CVs were
of the different catalysts towards the ethanol electrooxidation. The run between 0.05 and 0.6 V vs. RHE at 0.2 V s− 1 to clean the surface.
currents shown in the text are normalized by the electrochemical active Three further cycles were recorded at 0.010 V s− 1 to ensure that the
surface area (specific activity) determined by monitoring the complete electrode was stable. Then, the base electrolyte was exchanged with a
elimination of a CO adsorbed monolayer (CO stripping), assuming a 0.5 M CH3CH2OH + 0.1 M HClO4 solution for the IR measurements.
charge of 420 μC cm− 2 involved in the oxidation of a monolayer of Prior to the IR measurements, the thickness of the thin layer of elec-
linearly adsorbed CO. The procedure for the CO stripping measure- trolyte between the optical window and the gold electrode was set to
ments is detailed in the Supplementary Information (SI). 7.5–8.2 μm, with the value determined by comparing the experimental
The samples were subjected to an accelerated aging treatment reflectivity spectrum of the thin layer cell attenuated by the layer of the
(AAT) in order to study the stability of the nanostructures with time. solvent to the reflectivity curve calculated from the optical constants of
They consisted of 1000 cycles between 0.05 and 0.6 V in 0.5 M H2SO4 the cell constituents [27,28]. Potential loops were performed for each
at 60 °C using a scan rate of 0.05 V s− 1. experiment, varying the potential between 0.05 and 1 V vs. RHE with a
0.05 V step. Each potential step comprises an equilibration time of
2.4. In situ X-ray absorption measurements 1 min followed by the acquisition of the spectrum by averaging 200
scans at 4 cm− 1 resolution. For each sample, a spectrum was measured
An in situ electrochemical cell was used to collect EXAFS data of the at 0.05 V and taken as reference (R0), since no ethanol oxidation occurs
Pt-Sn/C catalysts as a function of potential [22]. The working electrode at this potential. The spectra presented in the text correspond to the
was held in place by an Au wire contact, a Pt wire served as the counter ratio between the spectrum at the specific potential and the reference
electrode, and the reference electrode was a mercury mercurous sulfate one. In SNIFTIRS spectra usually there are bands in both up and down
(Hg/Hg2SO4) electrode that was connected to the cell via a short length direction with respect to the baseline. Taking R0 as the reflectance in
of tubing containing the electrolyte. The cell was controlled by an the reference spectrum and R the sample spectrum reflectance, positive
Autolab potentiostat running with NOVA 1.8 Software. The electrolyte bands (R0 > R) correspond to consumption of species and negative
was purged with N2 and then pumped through the cell using a peri- bands (R0 < R) correspond to formation of new species.
staltic pump. The measurements were carried out at three different
fixed potentials (0.2, 0.4 and 0.6 V vs. RHE). The measurements were 3. Results
run both in 0.5 M H2SO4 and 0.5 M H2SO4 + 0.5 M CH3CH2OH. Details
of the procedure can be found in the SI. X-ray adsorption (XAS) mea- Three nanoparticle Pt-Sn catalysts with different structure and
surements for the fresh samples were recorded on beamline B18 at composition were prepared: two Pt-Sn alloys with Pt:Sn atomic ratio of
Diamond Light Source (UK) with ring energy of 3 GeV and a current of 3:1 (Pt3Sn1/C) and 1:1 (Pt1Sn1/C), and a one monolayer Sn modified Pt
300 mA. The monochromator used was Si(311) crystals operating in catalyst (Sn/Pt/C). The commercial Pt/C catalyst from Johnson
Quick EXAFS (QEXAFS) mode. The measurements were carried out in Matthey (JM) was also included in this study as reference.
fluorescence mode at the Pt L3 (11,564 eV) and Sn K (29,200 eV) ab- The bulk composition of the catalysts was determined by using SEM-
sorption edges at 298 K using a 9-element Ge detector. XAS measure- EDX and the results are reported in Table S1. The metal loading was
ments for the aged samples were recorded on beamline CLAESS at ALBA around 19.5–25.9 wt%, depending on the type of material. In all cases,
synchrotron light source (Spain). Calibration of the monochromator except for the Sn/Pt/C, the amount of Sn was slightly lower than the
was carried out using a Pt foil and Sn foil previously to the nanoparticle nominal one.
measurements at the respective absorption edge. The distribution and size of the metal particles deposited on the
Ex-situ XAS measurements were also recorded in transmission mode carbon support were studied by TEM and representative images for all
on beamline B18 at Diamond Light Source (UK) and details can be the Pt-Sn catalysts are exhibited in Fig. S1. All catalysts show a narrow
found in the SI. particle size distribution. TEM images of the Sn/Pt/C sample show well
The acquired data was processed and analyzed using the programs dispersed nanoparticles on the support. In contrast, the Pt-Sn alloy
Athena and Artemis [23], respectively, which implement the FEFF6 and samples show different degrees of agglomeration, which is most evident
IFEFFIT codes [24]. Fits were carried out using a k range of 3–16 Å− 1 for the Pt3Sn1/C material (Fig. S1b).
and an R range of 1.4–4.2 Å at the Pt L3 edge, and 3–10 Å− 1 and The local composition and architecture of the Sn/Pt/C and Pt3Sn1/C
1.1–3.2 Å at the Sn K edge, with multiple k weightings of 1, 2 and 3. samples were investigated by using an EDS probe in the STEM mode
Different FEFF inputs were used in these fits depending on the mate- (Fig. 1). Multiple single-particle EDS analyses of both samples con-
rials. An alloy 3:1 Pt:Sn model was created based on [25,26] using the firmed the bimetallic nature of the particles. In both samples, only a few
lattice parameters obtained from the XRD analysis. monometallic Pt particles were observed and no monometallic Sn
particles were detected. In order to confirm the core-shell and alloy
2.5. In situ IR measurements nanostructures, STEM-EDS line spectra were recorded. The composition
profile obtained for the Sn/Pt/C sample show that the distribution of Sn
Electrochemical in situ subtractive normalized interfacial Fourier reaches a maximum at the edge of the particles (shell) and a minimum
transform infrared spectroscopy (SNIFTIRS) experiments were carried at the center of the particle, whereas Pt reaches a maximum con-
out on a Thermo Nicolet 8700 (Nicolet, Madison, WI) spectrometer centration at the center of the particles (core). This indicates that Sn
equipped with a custom-made external tabletop optical mount and an was deposited on the surface of the Pt nanoparticles. On the other hand,
MCT-A detector. A custom made glass-filled Teflon® electrochemical the composition profile for the Pt3Sn1/C sample indicates that the dis-
cell with a polycrystalline gold disc electrode was coupled to the set-up tribution profile of both metals is the same, confirming the formation of

138
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

Fig. 1. HAADF atomic-resolution images (a,c)

and corresponding STEM-EDS spectra (b,d) along
the line in the image for the Sn/Pt/C (a,b) and
Pt3Sn/C (c,d) samples. The insets in (b) and (d)
show the schema of the Pt-Sn nanostructures
where blue and red circles represent Pt and Sn
atoms, respectively. (For interpretation of the
references to color in this figure legend, the
reader is referred to the web version of this ar-
ticle.)

the alloy. is greater than that of Pt/C-JM (3.9175 Å) reference and increases with
The effect of Sn addition on the structural and crystallographic the amount of Sn to 3.9776 Å and 4.0260 Å for Pt3Sn1/C and Pt1Sn1/C,
properties of Pt was studied by XRD (Fig. 2a). The broad diffraction respectively. The crystallite size was calculated using the Scherrer
peak observed at 2θ = 22.5° in all the samples is attributed to the equation. Pt3Sn1/C and Pt1Sn1/C materials have an average crystallite
sample holder, which overlaps the peak of the (002) plane of the carbon size of 3.4 and 3.3 nm, respectively. These values are in good agreement
support at 2θ = 22.7°. The two alloys show the typical features of the with those observed by TEM (Fig. S1).
fcc structure of Pt. However, the diffraction peaks are slightly shifted to Sn/Pt/C shows the characteristic peaks of the fcc structure of Pt. In
lower 2θ values, confirming the incorporation of Sn into the fcc struc- this case, no shifts in the peaks are observed, but the peaks associated to
ture of Pt. In the literature, a shift of the Pt fcc structure peaks to lower the (110), (101), (211), (220) and (202) planes of SnO2 are observed,
2θ values has been observed with the increase of the Sn content which is in agreement with the fact that Sn was not incorporated into
[11,29,30]. However, in this case the peaks of the Pt fcc structure for the structure of Pt but deposited on its surface as tin oxide, as seen by
the Pt1Sn1/C sample are broad and have low intensity, which is at- STEM-EDS. The fact that Sn was not incorporated into the Pt structure is
tributed to a diminishment of the Pt crystallinity due to the large Sn also deduced from the lattice parameter (3.9196 Å), which is very si-
content – this complicates accurate determination of the peak position milar to that of Pt/C-JM (3.9175 Å). The average crystallite size for this
shift. In fact, for this sample, additional peaks associated with the material is 4.0 nm, the same as that for the commercial 20 wt% Pt/C-
(110), (101), (211), (220) and (202) planes of SnO2 can be observed, JM catalyst used in its synthesis, and the amount of Sn alloyed is only
indicating the poor grade of alloying between Pt and Sn, which is only 8%.
66.8% versus 92.6% for the Pt3Sn1/C sample. Pt1Sn1 does not show the The effect of Sn on the electronic properties of Pt, as well as the
peaks associated with the hcp phase expected for a Pt1Sn1 alloy, in- surface composition, was determined by XPS. Fig. 2b shows the Pt 4f
dicating that this sample might be a mixture of the fcc phase with a and Sn 3d XPS spectra for the Pt-Sn catalysts and the commercial Pt/C
considerable amount of Sn not alloyed to Pt and segregated to the from JM; the corresponding analysis of both regions is reported in Table
surface forming SnO2. In contrast, no peaks related to Sn or its oxides S2. The Pt 4f lines contain two peaks, which correspond to Pt 4f7/2 and
are observed for the Pt3Sn1/C sample. Due to the incorporation of Sn 4f5/2 states from the spin-orbital splitting. Each peak was deconvoluted
into the fcc structure of Pt, an expansion of the lattice cell is expected using three different chemically shifted components (Pt0, Pt2 + and
since the atomic radius of Sn (1.61 Å) is larger than that of Pt (1.39 Å). Pt4 +) to identify the predominant oxidation state of Pt. As seen in Table
In agreement with this, the lattice parameter of the Pt-Sn alloy samples S2, Pt0 is the predominant oxidation state in all the Pt-Sn catalysts

Fig. 2. XRD diffraction patterns (left) and pho-

toemission spectra (right) for the commercial Pt/
C-JM, Sn/Pt/C, Pt1Sn1/C, and Pt3Sn1/C catalysts.
In the right panel, the Pt 4f and Sn 3d regions as
well as the corresponding chemical shifted com-
ponents are shown.

139
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

(around 68–78%). However, all samples present a variable amount of the fitted models that will be described below for each sample.
oxidized species, deduced from the Pt 4f7/2 peaks at 72.7–74.5 eV. The The commercial Pt/C-JM sample was measured under the same
presence of Pt oxide phases in all the samples can be attributed to air conditions as a reference. As mentioned above, the Pt oxide in the
exposure. This result is explained in more detail with the ex situ EXAFS surface of the Pt NPs due to contact with air is reduced under the
analysis (see Ex situ XAS measurements Section in SI). electrochemical conditions (Table S4). The reduction of Pt results in an
A shift in the binding energy of the Pt core level to lower energies, increase of the Pt atoms in the first coordination shell from 5.1 to 9.07
compared to those for Pt/C, is observed for the Pt3Sn1/C alloy catalyst (at 0.4 V vs. RHE, double layer region). The applied potential seems to
(Table S2). This effect may be caused by electronegativity differences of not have any significant influence on the Pt structure in the 0.2–0.6 V
Pt and Sn, and could indicate a charge transfer from Sn, which is less potential range, since there is no variation of the coordination number,
electronegative, to Pt, which is more electronegative, indicating that the PtePt bond length or the disorder (σ2). The slight variations ob-
there is an electronic modification of Pt. For the Pt1Sn1/C sample, served are within the error of the fits. The adsorption of ethanol does
however, this effect is not observed, which could be attributed to the not cause changes in the coordination distance or the disorder, and
poor extent of allowing between Pt and Sn. For the Pt sample modified therefore it can be said that the presence of ethanol does not alter the Pt
with Sn, the binding energies are the same as those for the commercial structure.
Pt/C, indicating that, in this case, Sn does not modify the electronic The analysis of the Sn/Pt/C at the Pt L3 edge indicates that Pt is
environment of Pt. The electronic effect of the addition of Sn in the reduced under these electrochemical conditions, as observed for the Pt/
different nanostructures was also confirmed by X-ray absorption near C-JM sample, which is reflected in the increase of the Pt atoms in the
edge spectroscopy (XANES) data analysis, and it is explained in the SI first coordination shell. It was not possible to fit the Pt-Sn shell because
(Fig. S3). the amount of Pt atoms with a Sn neighbour is too small to be reliably
The Sn 3d XPS spectra also show two peaks, which correspond to Sn fitted, as described in the SI for the ex situ measurements. The analysis
3d5/2 and 3d3/2 states. In this case, the peaks were deconvoluted into at the Sn K edge suggests that the SnO2 shell is very stable and it is not
two components, one around 487.5 eV attributed to Sn2 + and Sn4 + reduced under electrochemical conditions due to the high affinity of Sn
species and the other one around 485.6 eV corresponding to Sn0 (con- towards oxygen (see also Fig. S6b). Varying the potential does not result
sidered only for the alloy materials). It should be noted that it is not in changes of the coordination number (N) or coordination distances
possible to distinguish between Sn2 + and Sn4 + by this technique, as (R), indicating that the Sn/Pt nanostructure is stable under operating
their binding energies are too close. As seen in Fig. 2b, Sn is pre- conditions. The introduction of ethanol in the cell does not cause var-
dominantly oxidized in all the samples, which is in agreement with iation of the structural parameters at 0.2 and 0.4 V at the Pt edge.
other reports in the literature [10,11,29,31] and can be attributed to However, an increase of the SneSn bond distance from 3.223 to
the strong affinity of Sn towards oxygen. In the Pt3Sn1/C sample, a well- 3.239 Å and the decrease of the SneO coordination number from 5.28
defined shoulder at 485.6 eV is observed, indicating that the presence to 4.81 are observed at 0.6 V in the presence of ethanol. This could
of Sn0 is significant in this material, which is attributed to the formation indicate a slight reduction of the SnO2 layer due to the adsorption of
of the Pt-Sn alloy. In the fit of the Pt1Sn1/C sample, it was necessary ethanol.
also to include the peak corresponding to Sn0 in order to obtain the best For the Pt3Sn/C sample, a slight increase of the Pt-Sn interactions at
fit. the Pt L3 edge is observed when increasing the potential, which has
The surface Pt/Sn atomic ratio was calculated from the Pt 4f and Sn been previously attributed to a change in the morphology, from more
3d peak areas, taking into account the corresponding sensitivity factors spherical at low potentials to a more raft-like at higher potentials [32].
(Table S2). In all the Pt-Sn/C samples, the Pt/Sn ratio in the surface In that work, the samples were only analyzed at the Pt L3 edge. Ana-
(XPS) is lower than that in the bulk (EDX). For the Pt sample modified lysing the samples at the Sn K edge, a slight increase of the Sn-Pt in-
with Sn, this is justified by the presence of Sn only on the surface of the teractions, as well as an increase of the Sn-O ones, is also observed. In
Pt nanoparticles. However, in the case of the alloy samples, it indicates presence of ethanol, the migration of the Sn atoms into the NPs could
a surface Sn enrichment. This behaviour has already been described in take place, which may be attributed to a higher affinity of Pt towards
the literature and it is explained by the lower surface free energy of Sn, ethanol. These morphological changes are evidenced by a decrease of
respect to Pt, causing the migration of Sn from bulk to the surface [29]. the NPt-Pt and increase of NPt-Sn with an increase of the RPt-Pt and a
The stability of these materials under electrochemical conditions decrease of RPt-Sn. In the presence of ethanol, a slight decrease in the
over the applied potential (potential range = 0.2–0.6 V vs. RHE) and number of Sn-O interactions is also observed for this sample, which is
the effects of ethanol on the structures were determined by using in situ interpreted as a slight reduction of the SnO2 layer due to the adsorption
EXAFS measurements. The parameters extracted from the EXAFS of ethanol, as for the Sn/Pt/C sample. However, despite the small
spectra fits for the fresh samples are reported in Tables 1, 2 and S4, changes, it can be concluded that both nanostructures are stable in the
whilst the Fourier transformed EXAFS spectra and the corresponding k- potential range studied and in presence of ethanol.
space spectra are depicted in Figs. S4 and S5. Once the stability of the Pt-Sn nanostructures under EOR conditions
Comparing the structural parameters determined from the ex situ was confirmed, the EOR activities were established using cyclic vol-
and in situ EXAFS measurements, attention is drawn to the dis- tammetry. Fig. 3 shows the activity of the different Pt-Sn materials and
appearance of the Pt oxide (Pt-O1 path) under electrochemical condi- the commercial Pt/C. It can be clearly seen that the addition of Sn re-
tions. This result is expected since the potentials investigated are below sults in an approximately 0.2 V shift of the onset potential to more
the potential at which platinum oxide starts to be formed. The reduc- negative potentials and in the increase of the current densities, com-
tion of Pt oxide under working conditions is evident from the XANES pared to those obtained with the commercial Pt/C-JM. This effect is
spectra recorded at 0.4 V (double layer region) at the Pt L3 edge (Fig. attributed to the fact that Sn promotes the water activation at lower
S6a). A significant reduction of the white line (WL) intensity is ob- potentials than Pt (see CO-stripping CVs in Fig. S7), incorporating
served, compared with the ex situ XANES data reported in Fig. S3a. In oxidant species (OHads) to the surface that helps to oxidize ethanol. This
the case of Pt/C-JM and Sn/Pt/C, the intensity of the WL is the same as result is consistent with the fact that the ethanol oxidation starts at
that of the Pt foil, indicating that Pt mainly exists as metallic phase. The around 0.6 V on Pt NPs, a potential at which OHads species are present
decrease of the WL intensity observed for the Pt-Sn alloys compared to on the NP surface [33]. In the literature, the better performance of the
the Pt foil, which becomes more significant with the increase of the Pt-Sn alloys in the electrooxidation of ethanol, both in terms of current
alloy degree, confirms that the modification of the electronic structure density at a given potential and onset potential, has also been attributed
of Pt due to the electron donation from Sn is maintained under working both to the expansion of the lattice parameter of Pt (seen by XRD),
conditions. The reduction of Pt is also accompanied by other changes in which could catalyse the cleavage of the CeC bond, and to the

140
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

Table 1
Structural parameters obtained for the Sn/Pt/C catalyst from fitting the Pt L3 and Sn K edges EXAFS data acquired under potential control in 0.5 M H2SO4 and in 0.5 M CH3CH2OH
+ 0.5 M H2SO4.

Electrolyte Potential vs. RHE Shell N R/Å σ2 × 104/Å2 ΔE0/eV Rf

0.5 M H2SO4 0.2 V Pt-Pt1 8.92 ± 0.29 2.743 ± 0.001 55 ± 1 7.4 ± 0.3 0.005
Sn-Sn 1.65 ± 0.91 3.213 ± 0.019 56 ± 41 5.0 ± 0.9 0.011
Sn-O 5.36 ± 0.38 2.053 ± 0.007 32 ± 9
0.4 V Pt-Pt1 9.10 ± 0.27 2.744 ± 0.001 57 ± 1 7.3 ± 0.3 0.003
Sn-Sn 1.47 ± 0.74 3.214 ± 0.015 27 ± 31 4.3 ± 0.9 0.012
Sn-O 5.43 ± 0.40 2.056 ± 0.007 35 ± 9
0.6 V Pt-Pt1 8.90 ± 0.33 2.744 ± 0.002 56 ± 2 7.2 ± 0.3 0.003
Sn-Sn 1.69 ± 1.00 3.222 ± 0.021 58 ± 46 4.3 ± 0.9 0.012
Sn-O 5.39 ± 0.40 2.053 ± 0.007 29 ± 9
0.5 M H2SO4 + 0.5 M EtOH 0.2 V Pt-Pt1 9.07 ± 0.30 2.747 ± 0.002 57 ± 2 7.0 ± 0.5 0.003
Sn-Sn 1.72 ± 0.82 3.226 ± 0.017 54 ± 35 5.2 ± 0.9 0.010
Sn-O 5.32 ± 0.36 2.053 ± 0.007 37 ± 9
0.4 V Pt-Pt1 9.28 ± 0.27 2.747 ± 0.001 58 ± 1 7.0 ± 0.6 0.005
Sn-Sn 1.83 ± 0.85 3.223 ± 0.017 58 ± 35 5.1 ± 0.8 0.010
Sn-O 5.28 ± 0.36 2.053 ± 0.006 34 ± 8
0.6 V Pt-Pt1 9.15 ± 0.43 2.747 ± 0.002 58 ± 2 6.9 ± 0.4 0.006
Sn-Sn 1.73 ± 1.28 3.239 ± 0.008 79 ± 66 5.0 ± 1.0 0.019
Sn-O 4.81 ± 0.40 2.052 ± 0.009 21 ± 10

electronic effect that could result in a weaker bond between Pt and corresponds to the CeO stretch of the carbonyl group in both acet-
carbon atoms, preventing and/or reducing the poisoning by inter- aldehyde and acetic acid [37], whereas the band at 1280 cm− 1 corre-
mediate species [33,34]. Therefore, both the structural and electronic sponds to the CeOH stretching in the acetic acid [38]. The latter band
effect could explain the higher current densities obtained with the two will be used to quantify the acetic acid formed, since it can be attributed
Pt-Sn alloys (Pt3Sn/C and Pt1Sn1/C) than with the Sn modified Pt sur- unequivocally to this product. However, no bands can be exclusively
face. Recently, it has been demonstrated that Pt-Sn materials promote ascribed to the formation of acetaldehyde. The broad band centred at
the partial oxidation of ethanol to acetaldehyde and acetic acid [7,35], ca. 1623 cm− 1 is associated with the consumption of water during the
suggesting that the higher currents obtained during the CV measure- electrooxidation of ethanol, and the positive bands at ~2900 cm− 1 to
ments are due to the production of these partial oxidation products and the loss of ethanol in the thin layer [37]. It is noteworthy that the bands
not the complete oxidation of ethanol to CO2. associated with linear (2050 cm− 1, COL) and bridge (1836 cm− 1, COB)
In order to determine the origin of variation in EOR activity ob- bonded CO were not observed for most of the catalysts [39]. Only the
served in Fig. 3 and any differences in selectivity of the catalysts, the band characteristic of linear bonded CO was observed on the Pt1Sn1/C
EOR at the different nanostructures was followed in situ by FTIR, as sample at potentials as low as 0.17 V (Fig. 4d).
shown in Fig. 4. Similar adsorbed species can be observed during the Although the species adsorbed on the different Pt-Sn materials were
electrooxidation of ethanol on all the Pt-Sn materials investigated and similar, the product distributions varied significantly with the different
on the commercial Pt/C (see Fig. 4a–d). The band observed at ca. Pt-Sn nanoarchitectures as illustrated in Fig. 4e–h, which show the
2342 cm− 1 is attributed to the formation of CO2 and, therefore, to the amount of CO2 and acetic acid formed as a function of the applied
complete oxidation of ethanol [36,37]. The band at ca. 1715 cm− 1 potential. In agreement with the literature, both the Pt-Sn alloys

Table 2
Structural parameters obtained for the Pt3Sn1/C catalyst from fitting simultaneously the Pt L3 and Sn K edges EXAFS data acquired under potential control in 0.5 M H2SO4 and in 0.5 M
CH3CH2OH + 0.5 M H2SO4.

Electrolyte Potential vs. RHE Shell N R/Å σ2 × 104/Å2 ΔE0/eV Rf

0.5 M H2SO4 0.2 V Pt-Pt 8.96 ± 0.38 2.757 ± 0.003 67 ± 2 5.5 ± 0.5 0.007
Pt-Sn 1.14 ± 0.31 2.796 ± 0.012 102 ± 24
Sn-Pt 5.22 ± 1.37 2.796 ± 0.012 102 ± 24 3.5 ± 1.5
Sn-O 3.55 ± 0.42 2.052 ± 0.015 47 ± 17
0.4 V Pt-Pt 8.86 ± 0.39 2.762 ± 0.003 67 ± 3 6.2 ± 0.5 0.007
Pt-Sn 1.27 ± 0.30 2.800 ± 0.009 96 ± 20
Sn-Pt 5.25 ± 1.23 2.800 ± 0.009 96 ± 20 3.4 ± 1.3
Sn-O 3.73 ± 0.43 2.051 ± 0.011 59 ± 17
0.6 V Pt-Pt 8.98 ± 0.39 2.758 ± 0.003 70 ± 3 6.0 ± 0.5 0.008
Pt-Sn 1.52 ± 0.41 2.805 ± 0.011 115 ± 29
Sn-Pt 5.42 ± 1.82 2.805 ± 0.011 115 ± 29 3.0 ± 1.7
Sn-O 3.78 ± 0.61 2.047 ± 0.015 49 ± 23
0.5 M H2SO4 + 0.5 M EtOH 0.2 V Pt-Pt 8.84 ± 0.28 2.756 ± 0.002 65 ± 2 5.2 ± 0.4 0.005
Pt-Sn 1.15 ± 0.41 2.797 ± 0.014 147 ± 40
Sn-Pt 8.30 ± 3.17 2.797 ± 0.014 147 ± 40 2.3 ± 2.0
Sn-O 3.23 ± 0.75 2.033 ± 0.017 42 ± 30
0.4 V Pt-Pt 8.74 ± 0.31 2.761 ± 0.002 65 ± 2 5.6 ± 0.4 0.009
Pt-Sn 1.22 ± 0.44 2.793 ± 0.015 129 ± 41
Sn-Pt 6.80 ± 2.98 2.793 ± 0.015 129 ± 41 3.1 ± 2.4
Sn-O 3.67 ± 0.93 2.041 ± 0.021 49 ± 35
0.6 V Pt-Pt 8.81 ± 0.65 2.764 ± 0.005 68 ± 5 6.1 ± 0.9 0.013
Pt-Sn 1.35 ± 0.78 2.784 ± 0.024 119 ± 63
Sn-Pt 6.98 ± 4.90 2.784 ± 0.024 119 ± 63 3.8 ± 1.0
Sn-O 3.17 ± 1.46 2.030 ± 0.033 39 ± 6

141
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

desorb easily from the surface of the catalyst not allowing their further
oxidation to CO2. This effect is most significant for the Pt3Sn1/C (92.6%
alloying) compared to Pt1Sn1/C (67% alloying). In the case of the Sn/
Pt/C catalyst, the deposition of Sn on the surface of Pt NPs did not cause
any electronic effect (as observed in the XPS and XANES), implying a
stronger adsorption of the intermediates, which allows them to be
further oxidized to CO2. Our results contradict those of Du et al. who
report the differences between alloy and non-alloyed Pt-Sn catalyst in
the conversion to CO2 and attributed these to the composition of the
catalysts and not to the crystalline structure [13].
The long-term stability of the nanostructures is also of importance in
designing more effective EOR electrocatalysts. The stability was
therefore investigated by using in situ XAS and post-mortem XPS mea-
surements performed on the samples subjected to an accelerated aging
treatment (AAT). The structural and electronic results extracted from
the XAS and XPS analysis are reported in Tables 3 and S5 and Figs.
S8–S10, whilst the electrochemical data can be found in Fig. S11.
The EXAFS analysis of the Sn modified Pt structure demonstrates the
Fig. 3. Cyclic voltammograms in 0.5M H2SO4 + 0.5M CH3CH2OH obtained at room stability of this nanostructure in the potential window studied and
temperature and at a scan rate of 0.02 V s− 1 for the Pt-Sn/C catalysts and the commercial precludes significant dissolution of Sn, since no changes in the EXAFS
Pt/C from JM. Inset: enlargement of the onset potential for the ethanol electroxidation
parameters were observed. The stability of Sn was also confirmed by
during the forward scan after double layer removal.
XPS, since the atomic Pt:Sn ratio did not change with the AAT. This
result is also supported by the electrochemical data that shows only a
produced more acetic acid than CO2 [20]. Additionally, the acetic acid/ slight decrease of the electrochemical surface area, calculated both
CO2 ratio was found to be greater for Pt3Sn1/C than Pt1Sn1/C, thus from the hydrogen UPD region and the CO-stripping peak.
increasing with the extent of Pt-Sn alloying. In contrast, the Sn modified In the case of the Pt3Sn1/C sample, however, significant changes in
Pt surface nanostructure was more selective to CO2. the atomic structure are deduced from the EXAFS fit. A significant
Expansion of the lattice parameter of Pt by alloying has been pre- decrease of the Pt-Sn interactions, accompanied by an increase of the
viously suggested to facilitate the CeC bond breaking, improving the PteSn bond length, suggests the segregation of Sn to the surface with
selectivity to CO2 [29,40]. However, such trend in the selectivity to CO2 cycling, resulting in the de-alloying of the Pt-Sn nanoparticles. In ad-
with lattice constant was not found for the catalysts in our study. Al- dition, the increase of the Sn-O coordination number confirms the
though, the lattice expansion may contribute to the overall increased segregation of the Sn atoms and the formation of a SnO2 layer at the
current density as shown in Fig. 3, this is not reflected in the in situ surface of the nanoparticles (SnO2/PtxSn/C). The XPS results corrobo-
FTIRS results. rate the de-alloying of the Pt-Sn nanoparticles with the subsequent in-
The selectivity to acetic acid formation (reduced selectivity to CO2) crease of SnO2. The Sn 3d photoemission line shows a decrease of the
is attributed to the electronic effect of Sn on the Pt, resulting in weaker metal Sn component respect to the as-prepared sample (Figs. 2 and S10,
adsorption of the intermediates (acetaldehyde and acetic acid), which Tables S2 and S5). However, the electronic effect on Pt was preserved

Fig. 4. In situ FTIR spectra taken at different potentials in 0.5 M CH3CH2OH + 0.1 M HClO4 (upper panels); and normalized areas of the CO2 (2343 cm− 1) and acetic acid (1280 cm− 1)
bands as function of the applied potential (bottom panels) for Pt/C-JM (a,e), Sn/Pt/C (b,f), Pt3Sn1/C (c,g) and Pt1Sn1/C (d,h) catalysts. The insets in (e)–(h) show the Pt-Sn nanostructures
where blue and red circles represent Pt and Sn atoms, respectively.

142
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

Table 3
Structural parameters obtained for the aged Pt-Sn materials supported on carbon from fitting the Pt L3 and Sn K edge EXAFS data acquired at 0.2 V in 0.5 M H2SO4. For the Pt3Sn1/C
sample, both edges were fitted simultaneously.

Sample Shell N R/Å σ2 × 104/Å2 ΔE0/eV Rf

Pt/C-JM Pt-Pt 9.00 ± 0.86 2.753 ± 0.004 53 ± 5 8.0 ± 0.7 0.020

Sn/Pt/C Pt-Pt 8.82 ± 0.46 2.748 ± 0.005 65 ± 5 6.6 ± 0.5 0.003
Sn-Sn 1.38 ± 0.82 3.213 ± 0.019 41 ± 41 5.6 ± 0.9 0.011
Sn-O 5.33 ± 0.40 2.048 ± 0.007 37 ± 10
Pt3Sn1/C Pt-Pt 7.97 ± 0.79 2.759 ± 0.006 57 ± 5 6.2 ± 1.1 0.016
Pt-Sn 0.57 ± 0.61 2.808 ± 0.012 103 ± 33
Sn-Pt 4.61 ± 1.66 2.808 ± 0.012 103 ± 33 3.7 ± 1.4
Sn-O 4.14 ± 0.36 2.049 ± 0.010 41 ± 12

after the AAT, which is evidenced by the shift in the binding energy of structural changes suffered by the Pt3Sn1/C sample, its overall activity
the Pt core level to lower energies compared to that of Pt/C and Sn/Pt/ outperformed that of the Sn/Pt/C sample, suggesting that the electronic
C. The electrochemical data shows a significant decrease of the HUPD effects of alloying persisted and have an important role in describing
area and the CO-stripping peak in agreement with the segregation of Sn the activity of this material.
to the surface and the formation of a SnO2 layer. In addition, during the
AAT a redox couple at 0.55/0.8 V associated with Sn appears [7], Acknowledgments
whereas the reduction peak of Pt at 0.8 V disappears.
However, even after the structural changes suffered by the Pt3Sn1/C L.C. acknowledges support via an EU Marie Curie Intra-European
during the AAT, its overall activity (not necessarily CO2 selectivity as Fellowship under contract no. FP7-PEOPLE-2010-IEF-272632. EJC and
discussed above) is still higher than that of the Sn/Pt/C sample. This LMD acknowledge financial support from CONICET and ANPCyT
suggests that the electronic effects retained after the AAT, confirmed by (Argentina). The authors acknowledge the LMA-INA, UK Catalysis Hub,
XANES and XPS, continue to have an important role in describing the Diamond Light Source and the ALBA synchrotron light source for of-
activity of this material. fering access to their instruments and expertise and for provision of
beamtime (SP8071 and 2012100428). Dr. Giannantonio Cibin, Dr.
Andy Dent and Dr. Stephen Parry are acknowledged for the excellent
4. Conclusions beamline support at the Diamond Light Source and Dr. Laura Simonelli
at ALBA Synchrotron Light Source. The data used in this paper can be
Three different Pt-Sn nanostructures, two Pt-Sn alloys with varying found in the Southampton University Deposit with DOI: https://doi.
extent of alloying and a Sn modified Pt catalyst, were investigated to org/10.5258/SOTON/D0294.
study the effect of the addition of Sn on the structural and electronic
properties of Pt and their relative effects on the overall ethanol elec- Appendix A. Supplementary data
trooxidation activity and selectivity to CO2 formation. The Pt3Sn1/C
and Pt1Sn1/C samples are alloys with a heterogeneous composition, Supplementary data to this article can be found online at https://
with more Sn on the surface than in the bulk of the nanoparticles. doi.org/10.1016/j.jelechem.2017.09.060.
Pt3Sn1/C presented a high grade of crystallinity with a 92.6% of al-
loying. In contrast, Pt1Sn1/C showed a more amorphous character that References
was attributed to the excess of Sn present as SnO2 at the surface or near
to the surface of the alloy particles. The Sn/Pt/C sample consisted of a [1] C. Lamy, A. Lima, V. LeRhun, F. Delime, C. Coutanceau, J.-M. Léger, Recent ad-
monolayer of Sn deposited on the Pt nanoparticles forming a core-shell vances in the development of direct alcohol fuel cells (DAFC), J. Power Sources 105
structure. Alloying of Sn with Pt caused an increase in the lattice (2002) 283–296.
[2] J.S. Spendelow, A. Wieckowski, Electrocatalysis of oxygen reduction and small al-
parameter of Pt and modified the electronic structure, whilst the de- cohol oxidation in alkaline media, Phys. Chem. Chem. Phys. 9 (2007) 2654–2675.
position of Sn on the Pt surface did not modify the Pt lattice parameter [3] H. Igarashi, T. Fujino, Y. Zhu, H. Uchida, M. Watanabe, CO tolerance of Pt alloy
nor perturbed its electronic environment. electrocatalysts for polymer electrolyte fuel cells and the detoxification mechanism,
Phys. Chem. Chem. Phys. 3 (2001) 306–314.
All the Pt-Sn materials showed higher EOR activity than the com- [4] L. Colmenares, H. Wang, Z. Jusys, L. Jiang, S. Yan, G.Q. Sun, R.J. Behm, Ethanol
mercial Pt/C reference catalyst, both in terms of onset potential and oxidation on novel, carbon supported Pt alloy catalysts-model studies under defined
current density, with the Pt-Sn alloys giving the highest yield of ethanol diffusion conditions, Electrochim. Acta 52 (2006) 221–233.
[5] H.D. Herrera-Méndez, P. Roquero, M.A. Smit, L.C. Ordóñez, Carbon-supported
oxidized. In situ FTIR was used to determine any effect of the Pt-Sn
platinum molybdenum electro-catalysts and their electro-activity toward ethanol
nanostructure on the products distribution. Our study advanced the oxidation, Int. J. Electrochem. Sci. 6 (2011) 4454–4469.
detailed understanding of the oxidation of ethanol on PtSn catalysts by [6] G. Yang, A.I. Frenkel, D. Su, X. Teng, Enhanced electrokinetics of CeC bond split-
showing that the architecture of the nanoparticles affects the selectivity ting during ethanol oxidation by using a Pt/Rh/Sn catalyst with a partially oxidized
Pt and Rh core and a SnO2 shell, ChemCatChem 8 (2016) 1–6.
towards CO2. On the Pt-Sn alloys, the formation of acetaldehyde and [7] J.-M. Jin, T. Sheng, X. Lin, R. Kavanagh, P. Hamer, P. Hu, C. Hardacre, A. Martinez-
acetic acid was confirmed and related to the weaker adsorption of such Bonastre, J. Sharman, D. Thompsett, W.-F. Lin, The origin of high activity but low
products due to the modification of the electronic environment of Pt, CO2 selectivity on binary PtSn in the direct ethanol fuel cell, Phys. Chem. Chem.
Phys. 16 (2014) 9432–9440.
which desorb easily from the surface of the catalyst and thus are not [8] A. Kowal, M. Li, M. Shao, K. Sasaki, M.B. Vukmirovic, J. Zhang, N.S. Marinkovic,
completely oxidized to CO2. In contrast, on the Sn modified Pt sample, P. Liu, A.I. Frenkel, R.R. Adzic, Ternary Pt/Rh/SnO2 electrocatalysts for oxidizing
the amount of ethanol completely oxidized to CO2 increased, indicating ethanol to CO2, Nat. Mater. 8 (2009) 325–330.
[9] B.E. Hayden, M.E. Rendall, O. South, The stability and electro-oxidation of carbon
that the product distribution can be tuned by the Pt-Sn nanostructure. monoxide on model electrocatalysts: Pt(111)–Sn(2 × 2) and Pt(1 1 1)–Sn
Both nanoarchitectures were shown to be retained under working (√ 3 × √ 3)R30°, J. Mol. Catal. A Chem. 228 (2005) 55–65.
conditions and in presence of ethanol in the 0.2–0.6 V potential [10] E.M. Crabb, R. Marshall, D. Thompsett, Carbon monoxide electro-oxidation prop-
erties of carbon-supported PtSn catalysts prepared using surface organometallic
window. Following AAT cycles, a segregation of Sn from the bulk to the chemistry, J. Electrochem. Soc. 147 (2000) 4440–4447.
surface of the Pt-Sn alloy nanoparticles was found, causing Pt-Sn de- [11] S. García-Rodríguez, F. Somodi, I. Borbáth, J.L. Margitfalvi, M.A. Peña,
alloying and formation of a SnO2 layer on the surface. In contrast, the J.L.G. Fierro, S. Rojas, Controlled synthesis of Pt-Sn/C fuel cell catalysts with ex-
clusive Sn-Pt interaction. Application in CO and ethanol electrooxidation reactions,
Sn modified Pt nanostructure was very stable. However, even after the

143
L. Calvillo et al. Journal of Electroanalytical Chemistry 819 (2018) 136–144

Appl. Catal. B Environ. 91 (2009) 83–91. colloidal synthesis and structural control, Chem. Mater. 25 (2013) 1400–1407.
[12] F.E. Lopez-Suarez, C.T. Carvalho-Filho, A. Bueno-Lopez, J. Arboleda, A. Echavarrıa, [27] V. Zamlynny, J. Lipkowski, In Advances in Electrochemical Science and
K.I.B. Eguiluz, G.R. Salazar-Banda, Platinum–tin/carbon catalysts for ethanol oxi- Engineering, Vol. 9 WILEY, Weinheim, 2006.
dation: influence of Sn content on the electroactivity and structural characteristics, [28] V. Zamlynny, I. Zawisza, J. Lipkowski, PM FTIRRAS studies of potential-controlled
Int. J. Hydrog. Energy 40 (2015) 12674–12686. transformations of a monolayer and a bilayer of 4-pentadecylpyridine, a model
[13] W. Du, G. Yang, E. Wong, N.A. Deskings, A.I. Frenkel, D. Su, X. Teng, Platinum-tin surfactant, adsorbed on a Au(111) electrode surface, Langmuir 19 (2003) 132–145.
oxide core–shell catalysts for efficient electro-oxidation of ethanol, J. Am. Chem. [29] J.H. Kim, S.M. Choi, S.H. Nam, M.H. Seo, S.H. Choi, W.B. Kim, Influence of Sn
Soc. 136 (2014) 10862–10865. content on PtSn/C catalysts for electrooxidation of C1–C3 alcohols: synthesis,
[14] S. Alayoglu, A.U. Nilekar, M. Mavrikakis, B. Eichhorn, Ru–Pt core–shell nano- characterization, and electrocatalytic activity, Appl. Catal. B Environ. 82 (2008)
particles for preferential oxidation of carbon monoxide in hydrogen, Nat. Mater. 7 89–102.
(2008) 333–338. [30] J.M. Sieben, M.M.E. Duarte, Nanostructured Pt and Pt–Sn catalysts supported on
[15] S. Alayoglu, B. Eichhorn, Rh–Pt bimetallic catalysts: synthesis, characterization, and oxidized carbon nanotubes for ethanol and ethylene glycol electro-oxidation, Int. J.
catalysis of core–shell, alloy, and monometallic nanoparticles, J. Am. Chem. Soc. Hydrog. Energy 36 (2001) 3313–3321.
130 (2008) 17479–17486. [31] G. Li, P.G. Pickup, Decoration of carbon-supported Pt catalysts with Sn to promote
[16] R. Parsons, T. VanderNoot, The oxidation of small organic molecules: a survey of electro-oxidation of ethanol, J. Power Sources 173 (2007) 121–129.
recent fuel cell related research, J. Electroanal. Chem. 257 (1988) 9–45. [32] S. Mukerjee, J. Mcbreen, An in situ X-ray absorption spectroscopy investigation of
[17] F. Fievet, J.P. Lagier, B. Blin, Homogeneous and heterogeneous nucleations in the the effect of Sn additions to carbon-supported Pt electrocatalysts: part I, J.
polyol process for the preparation of micron and submicron size metal particles, Electrochem. Soc. 146 (1999) 600–606.
Solid State lonics 32–33 (1989) 198–205. [33] S.C. Zignani, E.R. Gonzalez, V. Baglio, S. Siracusano, A.S. Aricò, Investigation of a
[18] J.-H. Zhou, J.-P. He, Y.-J. Ji, W.-J. Dang, X.-L. Liu, G.-W. Zhao, C.-X. Zhang, J.- Pt3Sn/C electro-catalyst in a direct ethanol fuel cell operating at low temperatures
S. Zhao, Q.-B. Fu, H.-P. Hu, CTAB assisted microwave synthesis of ordered meso- for portable applications, Int. J. Electrochem. Sci. 7 (2012) 3155–3166.
porous carbon supported Pt nanoparticles for hydrogen electro-oxidation, [34] T.S. Almeida, K.B. Kokoh, A.R. De Andrade, Effect of Ni on Pt/C and PtSn/C pre-
Electrochim. Acta 52 (2007) 4691–4695. pared by the Pechini method, Int. J. Hydrog. Energy 36 (2011) 3803–3810.
[19] D.L. Boxall, C.M. Lukehart, Rapid synthesis of Pt or Pd/Carbon nanocomposites [35] J.M.S. Ayoub, R.F.B. De Souza, J.C.M. Silva, R.M. Piasentin, E.V. Spinacé,
using microwave irradiation, Chem. Mater. 13 (2001) 806–810. M.C. Santos, A.O. Neto, Ethanol electro-oxidation on PtSn/C-ATO electrocatalysts,
[20] W. Li, W. Zhou, H. Li, Z. Zhou, B. Zhou, G. Sun, Q. Xin, Nano-stuctured Pt–Fe/C as Int. J. Electrochem. Sci. 7 (2012) 11351–11362.
cathode catalyst in direct methanol fuel cell, Electrochim. Acta 49 (2004) [36] V. Del Colle, J. Souza-Garcia, G. Tremiliosi-Filho, E. Herrero, J.M. Feliu,
1045–1055. Electrochemical and spectroscopic studies of ethanoloxidation on Pt stepped sur-
[21] S. Kim, S.-J. Park, Effect of acid/base treatment to carbon blacks on preparation of faces modified by tin adatoms, Phys. Chem. Chem. Phys. 13 (2011) 12163–12172.
carbon-supported platinum nanoclusters, Electrochim. Acta 52 (2007) 3013–3021. [37] F.C. Simoes, D.M. dos Anjos, F. Vigier, J.-M. Léger, F. Hahn, C. Coutanceau,
[22] A.M. Wise, P.W. Richardson, S.W.T. Price, G. Chouchelamane, L. Calvillo, P.J. E.R. Gonzalez, G. Tremiliosi-Filho, A.R. de Andrade, P. Olivi, K.B. Kokoh,
Hendra, M.F. Toney, A.E. Russell, Inhibitive effect of Pt on Pd-hydride formation of Electroactivity of tin modified platinum electrodes for ethanol electrooxidation, J.
Pd@Pt core-shell electrocatalysts: an in situ EXAFS and XRD study, Electrochim. Power Sources 167 (2007) 1–10.
Acta, (under revision). [38] R.F.B. De Souza, L.S. Parreira, J.C.M. Silva, F.C. Simoes, M.L. Calegaro, M.J. Giz,
[23] B. Ravel, M. Newville, ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray C.A. Camara, A.O. Neto, M.C. Santos, PtSnCe/C electrocatalysts for ethanol oxi-
absorption spectroscopy using IFEFFIT, J. Synchrotron Radiat. 12 (2005) 537–541. dation: DEFC and FTIR “in-situ” studies, Int. J. Hydrog. Energy 36 (2011)
[24] M. Newville, EXAFS analysis using FEFF and FEFFIT, J. Synchrotron Radiat. 8 11519–11527.
(2001) 96–100. [39] M.H. Shao, R.R. Adzic, Electrooxidation of ethanol on a Pt electrode in acid solu-
[25] W. Zhou, L. Liu, B. Li, P. Wu, Q. Song, Structural, elastic and electronic properties of tions: in situ ATR-SEIRAS study, Electrochim. Acta 50 (2005) 2415–2422.
intermetallics in the Pt–Sn system: a density functional investigation, Comput. [40] W.J. Zhou, S.Q. Song, W.Z. Li, Z.H. Zhou, G.Q. Sun, Q. Xina, S. Douvartzides,
Mater. Sci. 46 (2009) 921–931. P. Tsiakaras, Direct ethanol fuel cells based on PtSn anodes: the effect of Sn content
[26] X. Wang, L. Altmann, J. Stöver, V. Zielasek, M. Bäumer, K. Al-Shamery, H. Borchert, on the fuel cell performance, J. Power Sources 140 (2008) 50–58.
J. Parisi, J. Kolny-Olesiak, Pt/Sn intermetallic, core/shell and alloy nanoparticles:

144