Morphology-dependent activity of Pt nanocatalysts for ethanol oxidation in acidic media: Nanowires versus nanoparticles

Article abstract of DOI:10.1016/j.electacta.2011.08.055

The morphology of nanostructured Pt catalysts is known to affect significantly the kinetics of various reactions. Herein, we report on a pronounced morphology effect in the electrooxidation of ethanol and carbon monoxide (CO) on Pt nanowires and nanoparti

Full text of DOI:10.1016/j.electacta.2011.08.055

Electrochimica Acta 56 (2011) 9824–9830
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Morphology-dependent activity of Pt nanocatalysts for ethanol oxidation in
acidic media: Nanowires versus nanoparticles
Wei-Ping Zhoua,∗, Meng Lia, Christopher Koenigsmannb, Chao Mac, Stanislaus S. Wongb,c
,
Radoslav R. Adzica
a Chemistry Department, Brookhaven National Laboratory, Upton, NY 11973, United States
b Department of Chemistry, State University of New York at Stony Brook, Stony Brook, NY 11794, United States
c Condensed Matter Physics and Materials Sciences Department, Brookhaven National Laboratory, Building 480, Upton, NY 11973, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 12 May 2011
Received in revised form 16 August 2011
Accepted 16 August 2011
Available online 25 August 2011
The morphology of nanostructured Pt catalysts is known to affect significantly the kinetics of various
reactions. Herein, we report on a pronounced morphology effect in the electrooxidation of ethanol and
carbon monoxide (CO) on Pt nanowires and nanoparticles in an acidic solution. The high resolution trans-
mission electron microscopy analysis showed the inherent morphology difference between these two
nanostructured catalysts. Voltammetric and chronoamperometric studies of the ethanol electrooxida-
tion revealed that these nanowires had a higher catalytic activity by a factor of two relative to these
nanoparticles. The rate for CO monolayer oxidation exhibits similar morphology-dependent behavior
with a markedly enhanced rate on the Pt nanowires. In situ infrared reflection–absorption spectroscopy
measurements revealed a different trend for chemisorbed CO formation and CO2-to-acetic acid reaction
product ratios on these two nanostructures. The morphology-induced change in catalytic activity and
selectivity in ethanol electrocatalysis is discussed in detail.
Keywords:
Fuel cell
Heterogeneous catalysis
Platinum
Nanowires
Ethanol oxidation
© 2011 Elsevier Ltd. All rights reserved.
1. Introduction
enhanced catalytic performance in the oxygen reduction reaction
(ORR) and methanol oxidation reaction (MOR), as demonstrated
Nanostructured Pt is one of the most widely studied electro-
catalytic materials in energy related technologies, particularly in
low-temperature fuel cells [1]. A fundamentally interesting feature
of these nanostructured Pt catalysts is their size- and morphology-
dependent electrocatalytic activity [1–15]. The conventional Pt
electrocatalysts with zero-dimensional (0-D) nanoparticle mor-
phologies generally have large numbers of low coordination atoms
(LCAs) and defects on their surfaces. By contrast, Koenigsmann et al.
[14,16] have shown that the Pt nanostructures with ultrathin one-
dimensional (1-D) nanowire morphologies maintain elongated
single crystalline segments with smooth crystal planes that are con-
nected by grain boundaries forming a nanowire structural motif.
Hence, these structures possess proportionally less surface LCAs
as compared with their associated 0-D analogues [13–18]. More-
over, the electronic property of noble metal nanowires (Au, Pt, Pd)
could be largely altered due to a surface contraction effect and a
surface-stress-induced phase transformation, when their diame-
ter decreased below a critical value of 2 nm [19–22]. The unique
structural and electronic properties of Pt catalysts linked to 1-D
morphologies could be largely responsible for their significantly
from our previous studies as well as those from other groups
[13–15,17,18].
The structural effect on the Pt catalytic activity in the ethanol
oxidation reaction (EOR) has been demonstrated from a num-
ber of studies on well-ordered [23–27] as well as nanostructured
[2,8] surfaces. The role of the LCAs has been highlighted in these
studies and these sites are generally regarded to favor C–C bond
splitting [2,8,24–26]. However, ethanol oxidation is a complicated
reaction and comprises multiple reaction steps that may main-
tain different structural sensitivity in each step. Although C–C
bond splitting and OH adsorption for surface CO oxidation have
been reported to be step site-favored reaction steps in the EOR
[23–25,27], C-dehydrogenation is believed to be a predominantly
terrace-favored reaction, as evidenced from the methanol oxida-
tion reaction [25,28,29]. Therefore, the Pt nanostructures with a
high density of surface LCAs may not necessarily lead to a high
activity, given different site preferences associated with different
reaction steps in the EOR. For example, Colmati et al. [25] sug-
gested that the optimized performance of the Pt catalyst for the EOR
can only be achieved with a balance between surface terrace and
step sites. Additionally, it has also been demonstrated that the step
sites with different symmetry and electronic property may have
different reactivity and selectivity in the EOR [23–25,27]. Further-
more, nanostructured materials may exhibit catalytic behaviors
∗
Corresponding author. Tel.: +1 631 3447298; fax: +1 631 3445815.
E-mail address: wpzhou@bnl.gov (W.-P. Zhou).
0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2011.08.055

W.-P. Zhou et al. / Electrochimica Acta 56 (2011) 9824–9830
9825
significantly different from that of well-ordered surface studies. For
example, CO monolayer oxidation is generally considered as a step
site-favored reaction from well-ordered surface studies [23,29,30].
In fact, it has been reported by several other studies that smaller
nanoparticles are less active than larger particles for CO monolayer
oxidation, although the number of the LCAs is expected to increase
progressively as the particle size is decreased [3,5,6]. It has been
suggested that this size-dependent activity is due to stronger bond-
ing of CO to smaller particles and hence a decrease in the surface
diffusion rate [5,6]. To further explore the nanostructure-induced
change in electrocatalytic activity, we have carried out ethanol and
CO electrooxidation on Pt nanocatalysts with distinctively different
morphologies.
In the present work, we used a combination of electrochem-
ical methods, in situ infrared reflection–absorption spectroscopy
(IRRAS), and high resolution transmission electron microscopy
(HRTEM) to study ethanol oxidation on two nanostructured Pt
catalysts, namely nanowires (NWs) and nanoparticles (NPs). Our
ultra-thin Pt NWs (d: 1.6 0.4 nm) had larger crystalline facets and
fewer surface LCAs as compared with commercial Pt catalysts com-
posed of nanoparticles (d: 2.4 0.4 nm) [14]. We found that the
ultrathin Pt NWs exhibited a significantly lowered onset potential
and also an increased faradic current with a factor of two to three
as compared with the Pt nanoparticles in the EOR. In fact, the ini-
tial onset potential for the EOR on our ultrathin nanowires was
approaching the one obtained on a state-of-the-art Pt/Sn bimetal-
lic catalyst [31]. Our in situ IRRAS study provided mechanistic
insight into this morphology-dependent activity and selectivity of
the EOR. We demonstrated that the morphology of the nanostruc-
tured materials may have a more profound effect than anticipated
from well-ordered surface studies on ethanol electrocatalysis.
Fig. 1. Voltammetric curves for (a) Pt nanowires and (b) Pt nanoparticles in 0.1 M
HClO4 solution. Scan rate: 50 mV s−1. Currents are normalized to the measured Pt
surface area (Hads charge after double layer correction). Typical HRTEM images are
highlighted of Pt nanowires (c) and Pt nanoparticles (d).
with 5 ␮L of 4 ␮g/10 ␮L Nafion® solution (diluted with water from
5% Nafion solution by Aldrich) and dried again before the EOR and
CO oxidation measurements.
For the CO (Matheson, 99.8%) adsorption and electrooxidation
experiments, the saturated CO adlayer was formed in a CO satu-
rated solution (holding at 0.05 V for 5 min) followed by purging
of the electrolyte with Ar (20 min). Thus, the CO electrooxidation
process was performed in a CO-free solution.
Ethanol (anhydrous, Reagent Grade) was purchased from
Sigma–Aldrich and all solutions were made with Milli-Q water
(Millipore, Bedford, MA). All potentials were measured using an
Ag/AgCl (3 M Cl−) electrode (Bio), but reported with respect to a
reversible hydrogen electrode (RHE). All electrochemical measure-
ments were performed in a 0.1 M HClO4 (Optima, Fisher) solution at
room temperature. The selection of HClO4 solution as the support-
ing electrolyte discards the possible anion effect in the catalytic
process. All the electrochemical data are reported with the sur-
face area normalized in terms of the electrochemical active surface
area (ECSA) that was calculated using the charge density obtained
from an integrated H adsorption peak after correcting for the dou-
2. Experimental
2.1. Preparation of nanostructured Pt nanowires
A Pt/C sample (46.4 wt.%) from Tanaka Kikinzoku (TKK) Inter-
national Inc. was used as the Pt nanoparticle reference without
further treatment. Details of ultra-thin Pt nanowire (NW) syn-
thesis and characterization have been reported elsewhere [14].
In brief, a 2.5 mL aliquot of a hexachloroplatinic (IV) acid hydrate
(H2PtCl6 · xH2O, >99.9%, Aldrich) solution (10.0 mM, aqueous) was
dissolved in a solvent system comprised of 20 mL dimethylfor-
mamide (EMD, anhydrous), 12.5 mL toluene (Acros, reagent grade),
and 2.5 mL triethylamine (Fisher, reagent grade). The Pt precur-
sor was reduced by the addition of 20 mg of sodium borohydride
while stirring, and the reaction was allowed to proceed for 3 h.
The resulting black product was precipitated by centrifugation and
was further washed several times with either absolute methanol
or ethanol. The isolated product was subsequently dispersed and
treated in a solution of 6 M HCl, sonicated, and washed several
times with aliquots of water and ethanol. The morphology and
dimensions of as-prepared Pt NWs were characterized using a
transmission electron microscopy instrument (JEOL 3000, BNL).
The average diameter of wires was 1.6 0.4 nm with an average
length of 100 25 nm, as shown in Fig. 1c.
ble layer contribution and assuming a relationship of 210 ␮C/cmP2t
.
All of the current densities reported here are normalized to the elec-
trochemical active surface area of Pt so that the activity of different
Pt catalysts tested can be directly compared.
2.3. In situ infrared reflection–absorption spectroscopy
In situ IRRAS studies were performed with
let Nexus 670 FT-IR spectrometer equipped with
a
Nico-
a MCT
(Mercury–Cadmium–Telluride) detector cooled with liquid
nitrogen. An unpolarized light beam was used. The spectral res-
olution was set to 8 cm−1 and 128 interferograms were together
added to each spectrum. Spectra are given in absorbance units
defined as A = −log(R/R0), where R and R0 represent the reflected
IR intensities corresponding to the sample- and reference-single
beam spectrum, respectively. The reference spectrum was col-
lected at 0.05 V in the same solution with 0.1 M ethanol and 0.1 M
HClO4. Band intensity and position were analyzed with the Omnic
program. A ZnSe hemisphere was used as the IR window, and the
working electrodes used in this IR study, including Pt NW and Pt
NP electrocatalysts deposited on a polycrystalline Au disk, were
pressed against the IR window to create a thin solution layer with
a thickness of a few micrometers. Pure Ar and dry air were used to
2.2. Electrochemical measurements
Preparing a thin film of catalysts on a glassy carbon electrode
(5 mm diameter, Pine instrument) involved dispersing a certain
amount of catalysts in water (in our case, 1 mg catalysts dispersed
into 1 mL water) and sonicating for ca. 10 min to create a uniform
suspension [32]. Then, 10 ␮L of this suspension was placed onto a
glassy carbon disk and dried in the air. The electrode was covered

9826
W.-P. Zhou et al. / Electrochimica Acta 56 (2011) 9824–9830
(b) second scan
0.02
(a) first scan
0.02
0.8
0.4
0.0
0.8
0.4
0.0
0.01
0.00
0.01
0.00
0.0
0.2
0.4
0.6
E/V vs. RHE
0.0
0.2
0.4
E/V vs. RHE
0.0
0.2
0.4
0.6
0.8
0.0
0.2
0.4
0.6
0.8
E/V vs. RHE
E/V vs. RHE
Fig. 2. Comparison of voltammetric curves for the first scan (a) and the second scan (b) taken in the positive direction (solid line) and negative direction (dashed line) for
Pt nanowires (black line) and nanoparticles (red line) after the sample was immersed in 0.2 M ethanol in 0.1 M HClO4 solution at 0.07 V under potential control. Scan rate:
10 mV s−1. Currents are normalized to the measured Pt surface area. Inset shows the comparison of the curves at the low current density region. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)
purge the electrolyte and spectrometer, respectively, in order to
remove the spectral interference from CO2 and water vapor in air.
and b show and compare the first and second CV scans, respec-
tively, of the Pt NWs and Pt NPs. Significant voltammetric changes
were not observed after the second scan for both of the samples.
However, there are several important features noted between these
two samples during the first and second scans. In the first positive-
going scan, the Pt NPs evince a low activity with an onset potential
at about 0.45 V (measured at j = 0.01 mA cm−2). However, the Pt
NWs display significantly higher activity with an onset potential
occurring at about ∼0.23 V measured at the same current density,
as well as a greater shift of ∼0.22 V toward the negative direc-
tion as compared with the Pt NPs. This onset potential shift on the
ethanol oxidation is approaching that obtained for state-of-the-art
bimetallic catalysts. In fact, a 0.25 V negatively shifted onset poten-
tial has been reported on Pt3Sn/C when compared with Pt/C [31].
Moreover, our nanowires also demonstrated an increased current
density, about 2–3 times higher relative to that of nanoparticles in
the potential range of 0.20–0.92 V in both the positive and negative
scan directions. A similar activity enhancement for ethanol oxida-
tion on Pd 1-D nanostructures as compared with their associated
0-D counterparts has also been reported [17].
Subsequent CV scans recorded on both nanoscale electrocata-
lysts show noteworthy differences, including a positively shifted
onset potential and a decreased current density as compared with
the first CV scan. The Pt NWs still display a considerably lower onset
potential (∼0.48 V at j = 0.01 mA cm−2) as compared with the Pt NPs
(∼0.59 V) in the second positive-going scan. Moreover, the Pt NWs
demonstrated a lower current decrease relative to the Pt NPs, e.g.
∼5% on NWs versus ∼15% on NPs at peak potential (0.88 V here). A
similar activity loss in consecutive scans of Pt catalysts in ethanol
solution has been reported elsewhere [33,34]. Previous in situ FTIR
experiments [33,34] as well as recent sum-frequency generation
(SFG) studies [35] show that the CO are accumulated on the sur-
face during the first negative-going scan and attained the highest
coverage when the electrode reached the negative most potential
of the scan window (0.07 V in this case). In addition, results from
those spectroscopy studies as well as differential electrochemical
mass spectrometry (DEMS) [26,36] and surface enhanced Raman
3. Results and discussion
Fig. 1 shows typical cyclic voltammetry (CV) scans of the Pt
NWs and Pt NPs recorded in an argon-saturated 0.1 M HClO4 solu-
tion at room temperature, together with representative, associated
HRTEM images. The HRTEM image in Fig. 1c highlights the presence
of well-defined individual nanowires with an average diameter of
1.6 0.4 nm. Specifically, the nanowires consist of multiple crys-
talline segments with an average length of 6 2 nm that extends
along the axis of the nanowire (see Fig. 6) [14]. The average par-
ticle size of the Pt NPs is 2.4 0.4 nm (Fig. 1d), which is slightly
larger than the Pt NWs. For a close comparison of the electro-
chemical features of two Pt nanostructures, the catalyst surface
area has been normalized to the Pt specific surface area. Current
features at ∼0.14 V associated with (1 1 0)-type of defect sites are
clearly observed on the Pt NPs but are almost indiscernible on the
Pt NWs (Fig. 1a and b), suggesting considerably less LCAs on the
nanowires [14]. In the surface hydroxyl formation region, the Pt
NWs display a positively shifted onset of oxidation and also a sup-
pressed intensity as compared with the Pt NPs, a clear indication of
a weaker binding of OH species on the former. Our previous work
[14] demonstrated that decreasing the binding of oxygen to Pt NWs
leads to a significantly enhanced ORR activity as compared with the
Pt NPs.
Fig. 2 shows CV scans of ethanol oxidation on the Pt NWs and Pt
NPs in 0.2 M ethanol in 0.1 M HClO4 solution at room temperature.
The positive potential limit set at 0.92 V prevents possible perturba-
tion of the Pt morphology during potential cycling to higher values.
The ethanol oxidation experiments were always performed with
immersion of the catalysts at 0.07 V under potential control where
ethanol oxidation is negligible [33] and with subsequent scans in
a positive-going direction. This method is very helpful for under-
standing the catalytic behavior of a Pt surface covered either with
or without surface intermediates formed during CV scans. Fig. 2a

W.-P. Zhou et al. / Electrochimica Acta 56 (2011) 9824–9830
9827
(a) Pt NPs
Pt NW
Pt NW
Pt NP
Pt NP
0.2
0.86 V
0.2
0.1
0.0
0.1
0.0
0.77 V
0.68 V
0.59 V
0.50 V
0.55
0.60
E/V vs. RHE
0.65
0.42 V
0.24 V
2400
2100
1800
1500
1200
(b) Pt NWs
CO2
CH3COOH
0
1000
2000
time /s
3000
COads
0.86 V
0.77 V
Fig. 3. Comparison of current–time plots for ethanol oxidation activity on Pt
nanowires (black) and nanoparticles (red) in 0.2 M ethanol in 0.1 M HClO4 solution
at 0.65 V for 3600 s reaction time. Inset shows comparison of current densities mea-
sured at three different potentials (0.55 V, 0.60 V and 0.65 V) at 500 s reaction time.
All current data were recorded after the catalysts were immersed at 0.07 V under
potential control and then stepped to reaction potentials. Currents are normalized
to the measured Pt surface area. (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)
0.68 V
0.59 V
0.50 V
0.42 V
0.24 V
2400
2100
1800
1500
-1
1200
spectroscopy (SERS) [37] measurements indicate the difficulty in
the electrooxidation of –CHx species at low potentials, which may
also lead to the accumulation of –CHx adsorbates on the surface
at potentials near the onset of ethanol oxidation. Therefore, the
decreased activity of Pt catalysts in a subsequent positive-going CV
scan can be ascribed to the poisoning effect of both adsorbed CO and
–CHx surface species mutually formed in the preceding negative-
going scan. It is unlikely that the adsorption of acetic acid played
a significant role in the onset potential shift, given its particularly
low concentration at the interface at low potentials near the onset
of ethanol oxidation and the similar adsorption strength to that of
the (bi)sulfate that does not show such a significant effect upon the
onset potential shift [25,35]. Regardless, the Pt NWs still exhibit
better catalytic performance, i.e. higher currents and lower onset
potentials, as compared with the Pt NPs, despite the partially cov-
ered surface by adsorbed CO and –CHx species. This significantly
enhanced catalytic performance noted with nanowires highlights
the inherent advantages of using 1-D nanostructures in ethanol
electrocatalysis.
Chronoamperometric (CA) measurements have been performed
to test the activity and stability of these two Pt catalysts, as shown
in Fig. 3. The CA data were recorded after the catalysts had been
immersed at 0.07 V under potential control and then stepped to
0.6 V. Clearly, the Pt NWs display higher current density, that is,
higher activity, both at the experimental beginning and also, after
3600 s of reaction at 0.6 V. The activity loss is significantly different
for both catalysts. That is, it takes about 1000 and 60 s of reac-
tion time for the Pt NWs and Pt NPs, respectively, to lose 50% of
their initial activity. After 3600 s of reaction, the Pt NWs still dis-
played an outstanding activity with a measured current density of
0.076 mA cm−2, which was about 2-fold higher than that of Pt NPs
(0.036 mA cm−2). The ethanol oxidation at three different poten-
tials (e.g., 0.55 V, 0.60 V, and 0.65 V) at 500 s of reaction shows a
consistently enhanced activity for the Pt NWs, i.e. slightly varied
at about 3-fold enhancement, as compared with the Pt NPs (inset
Wavenumber /cm
Fig. 4. In situ IRRAS spectra recorded during ethanol electrooxidation on Pt nanopar-
ticles (a) and Pt nanowires (b) in a 0.2 M ethanol in 0.1 M HClO4 solution. 128
interferograms (resolution of 8 cm−1) were collected and added into each spectrum.
in Fig. 3). The CA data again confirmed the significantly enhanced
catalytic performance on 1-D nanostructures seen in our comple-
mentary potentiodynamic measurements. However, mechanisms
for the current decay still remain unclear and further investigation
is expected to give insight into the enhanced stability of the 1D Pt.
To shed light upon this morphology-induced activity change in
ethanol oxidation, in situ IRRAS studies were carried out to verify
surface intermediate formation and product distribution on both
Pt catalysts. Fig. 4 displays the recorded spectra and Table 1 lists
the associated frequencies and band assignments. Carbon monox-
ide (CO), carbon dioxide (CO2), and acetic acid (CH3COOH) are the
main species observed on both Pt catalysts. The band at around
2030 cm−1 can be assigned to a linear bound COads, an indication of
C–C bond cleavage in ethanol molecules. However, chemisorbed
surface COads is also a catalytically poisoning intermediate that
Table 1
In situ IRRAs spectra band assignments.
Wavenumber/cm−1 Assignment
2343
2030
1705
CO2 asymmetric stretching
CO linear bonding
C
C
O stretching of CH3CHO and CH3COOH in solution
O stretching of adsorbed acetaldehyde and acetyl
1620–1635
∼1598
1396–1410
∼1350
1280
H–O–H deformation of adsorbed water
O–C–O stretching of adsorbed acetate
CH plane bending of adsorbed acetate
C–O stretching of CH3COOH in solution
CH3 symmetric deformation and C–H wagging in CH3CHO
1368

9828
W.-P. Zhou et al. / Electrochimica Acta 56 (2011) 9824–9830
Fig. 5. (a) Integrated band intensities of linear band COads (2030 cm−1) as a function of electrode potential in IRRAS spectra recorded in a positive scan-direction during
ethanol oxidation on Pt nanowires (bottom panel) and nanoparticles (upper panel). (b) The ratio of the integrated band intensity of CO2 (2343 cm−1) with respect to
CH3COOH (1280 cm−1) plotted as a function of applied electrode potential.
impedes further ethanol oxidation on Pt. The positive-going peak
near 2343 cm−1 is attributed to the asymmetric stretch vibration of
CO2, which is the product of the ethanol total oxidation pathway.
The band located around 1705 cm−1 can be ascribed to the stretch-
ing vibration of the C O bond in both acetaldehyde and acetic
acid, and the appearance of this band indicates partial oxidation of
ethanol. The band at 1280 cm−1 is the characteristic C–O stretching
vibration of acetic acid, which can be used for quantitative analysis
of acetic acid formation [27,38].
Since oxidation of the surface COads intermediate is a crucial step
in a complete oxidation of ethanol to CO2 [8,23,25,27,33,34,38],
we further performed CO stripping experiments to explore this
morphology-dependent effect in electrocatalysis. Fig. 6 shows the
oxidation of an adsorbed CO monolayer on the Pt NWs and Pt NPs
in 0.1 M HClO4 solution. While our ultrathin NWs had extended
crystalline facets and a small portion of surface LCAs, they dis-
played a markedly negative-potential shift for the CO stripping as
compared with commercial NPs. In fact, the corresponding main
peak potentials are 0.73 V and 0.90 V, respectively. That is, the Pt
NWs exhibited a significantly higher CO oxidation activity (based
on the main stripping peak) than the Pt NPs. A similar observation
of enhanced CO oxidation activity on carbon supported Pt NWs
as compared with their 0-D counterparts has also been reported
[15]. In addition to the negatively shifted peak potential of CO
stripping, a small but clearly discernable CO oxidation feature in
the so-called pre-ignition potential region, well below the peak
Evolution of the COads band and CO2-to-CH3COOH ratios on two
nanostructures has been checked. Fig. 5a shows integrated band
intensities of the linear COads band at 2030 cm−1, plotted against
electrode potentials, formed in a positive scan-direction on the Pt
NWs and NPs. In Fig. 5b, we plotted as a function of electrode poten-
tial the ratio of integrated band intensities between the CO2 band
(2343 cm−1) and the CH3COOH band (1280 cm−1). The high ratio
represents the catalysts’ high efficiency in the complete oxidation
of ethanol to CO2. The potential dependence of COads evolution on
the Pt NPs is consistent with previous studies on carbon-supported
Pt catalysts with similar particle sizes [39]. While the Pt NWs
exhibit a very similar COads evolution profile to that of Pt NPs, there
are several distinctive differences between the two nanostructures.
First, we observed that the onset potential of the COads band on the
Pt NPs is slightly below 0.3 V and is lower than that on Pt NWs
(0.3 V), probably because of the high density of step and edge sites
on nanoparticles that are favored in C–C bond splitting [2,8,25,26].
Second, we found that the COads band begins to decrease at 0.45 V
on the Pt NWs but at 0.62 V on the Pt NPs, coincident with the
appearance of CO oxidation features in the pre-ignition potential
region shown in the voltammetric curve (see discussion below in
Fig. 6). The significant negative-potential shift (∼0.17 V) in COads
oxidation suggests an efficient removal of surface chemisorbed CO
formed on the Pt nanowires. In addition, the Pt NWs displayed a
slightly increased CO2-to-CH3COOH reaction product ratio relative
to that of the Pt NPs at potentials below 0.60 V where the presence
of COads was predominantly observed for both catalysts. A similar
ratio was seen in the potential range of 0.6–0.8 V where CO oxida-
tion can occur on nanoparticles. However, when the surface CO was
significantly lower at high potentials (>0.8 V), the Pt NWs again dis-
played a greatly increased CO2-to-CH3COOH ratio. This observation
may have been caused by a significant oxidation of Pt nanoparti-
cles in this particular potential range, thereby preventing effective
ethanol oxidation.
Fig. 6. Voltammetric curves for CO monolayer stripping on Pt nanowires (a) and
nanoparticles (b) in 0.1 M HClO4 solution. Scan rate: 50 mV s−1. Shown are the first
scan (solid line) starting at 0.07 V and the second scan (dash-dot line). The electrode
was exposed to dissolved CO for 5 min at 0.07 V followed by Ar bubbling for 20 min.
Typical HRTEM images of a Pt nanowire (a) and nanoparticle (b) are also shown. The
scale bar is 2 nm for both images.

W.-P. Zhou et al. / Electrochimica Acta 56 (2011) 9824–9830
9829
potential (E < 0.6 V), was also observed on the Pt NWs but was
absent on the Pt NPs. A similar feature for CO oxidation in the
pre-ignition potential region has been previously reported on well-
ordered Pt surfaces with so-called “defect sites”, which play a
critical role in CO oxidation at low potentials [29,40,41]. The exis-
tence of these active sites is a very important indicator for a high
CO poisoning resistance catalyst, because CO oxidation at this low
potential region can free more surface sites for reactant adsorption
and reaction [29,40,41]. Note that the surface coverage of COads
formed during the EOR begins to decrease at a significantly lower
potential on the Pt NWs, in line with the characteristic voltammet-
ric feature of CO oxidation at the pre-ignition potential region. The
enhanced activity for CO monolayer oxidation may possibly be a
consequence of the unique structural and electronic properties of
our ultrathin nanowires.
their proportionally fewer LCA sites, since CO binds stronger on
the LCAs [45–47]. In addition, it has been reported that a signifi-
cant surface contraction and phase transformation may occur for
the thin Pt nanowires when its diameters decreased below 2 nm
[14,15,19,21]. Based on Nørskov and Hammer theory [48,49], a
down-shift of the d-band center due to a surface contracting strain
effect will decrease the adsorption strength of reactants. Recently,
Wang et al. [15] reported a negative shift of the Pt 4f core-level
binding energy, an experimental measure of the relative d-band
center [48], for carbon supported Pt NWs as compared with NP
analogues for the first time. This electronic change was accompa-
nied by a weakening of the interaction with surface adsorbates. Not
surprisingly, the unsupported ultrathin Pt nanowires in this report
show a corresponding shift in the oxide reduction feature to higher
potentials, which suggests a similar structure dependent electronic
effect.
On the other hand, recent studies have suggested that the
agglomeration of Pt nanostructures that generates large amounts
of defect sites also increases the CO oxidation activity [3,5,6,42,43].
However, based on a preponderance of previous results, we believe
that the enhanced CO stripping arises from an intrinsic difference
in the electrocatalytic properties of ultrathin Pt nanowire morphol-
ogy. Specifically, our CV data and previous ORR work [14] suggested
that the Pt NWs possess fewer surface defects when compared with
associated 0-D analogues. Furthermore, we did not observe any
significant effect of particle dispersion on the ORR activity of these
nanowires in our previous ORR study [14], highlighting the intrinsic
relationship between the structure of 1-D nanostructure and their
electrocatalytic activity, as opposed to agglomeration effects. How-
ever, the effects of the aggregation may not be ruled out entirely
because of the complexity of this hierarchical nanowire network
structure. Regardless, the data presented in Fig. 6 provides evi-
dence for the presence of unique sites that are active toward CO
oxidation, which is further justification for the high activity and
poisoning resistance in ethanol oxidation observed in the case of
the ultrathin Pt nanowires.
It has been previously reported that Pt nanoparticles with a high
density of surface LCAs usually generated a correspondingly high
activity for the EOR [2,8]. However, our Pt 1-D nanostructure has a
morphology that is clearly different from this category. Therefore,
the highly enhanced catalytic activity and selectivity in ethanol
oxidation and CO monolayer stripping on Pt catalysts with 1-D
nanostructures as compared with associated 0-D analogues raises
very interesting questions. Herein, we believe that the enhanced
catalytic performance of nanowires may be assigned to its unique
structural property, that is, a specific ensemble of surface terrace-
defect sites that are favored in different reaction steps in ethanol
oxidation. A series of reactions involving C-dehydrogenation, C–C
bond splitting, and –CHx oxidation that generates the chemisorbed
CO intermediate at low potentials in the EOR requires the presence
of multiple contiguous sites (terrace sites) [25,29,44]. At potentials
where the CO and –CHx oxidation rate is smaller than the formation
rate, the chemisorbed CO and –CHx accumulate on the surface and
act as a catalytically poisoning adsorbate, blocking Pt sites and also
affecting nearby sites for further ethanol adsorption and oxidation.
Considering the small portion of available terrace sites on 2–3 nm
particles, their smaller electrocatalytic activity as compared with
nanowires, therefore, could be rationalized due to quickly dimin-
ished terrace sites that are required for ethanol adsorption and
dehydrogenation. Consistent with this hypothesis of a COads and
–CHx inhibiting effect is the onset shift of ethanol oxidation on a
surface partially covered with COads and –CHx intermediate (Fig. 2).
Therefore, a delicate balance of surface terrace and defect sites on
1-D nanostructures may be conducive to a high ethanol oxidation
activity.
In the case of ethanol oxidation, this structure dependent elec-
tronic change of the Pt NWs may lead to a weaker binding of CO
and thereby improve its surface diffusion rate, which may lead to
an enhanced oxidation rate. Heinen et al. [34] have reported that
the rate determining step is the removal of surface COads poisoning
rather than the C–C bond splitting at low potentials for the EOR
on the polycrystalline Pt surface. Therefore, it is hypothesized that
the high EOR activity on 1-D nanostructures may also possibly be
due to their unique structural and electronic properties that lead
to efficient removal of surface COads. It is critical to highlight, how-
ever, that we cannot conclusively comment on the effects of the 1D
morphology and structure with respect to the oxidation of adsorbed
–CHx intermediates on the basis of the existing experimental data.
Regardless, this interpretation is consistent with results from IRRAS
and CO stripping experiments that surface COads can be removed
at lower potentials on nanowires than on nanoparticles. At high
potentials where O/OH adsorption is facile and COads coverage is
small, a competition between O/OH adsorption and ethanol oxi-
dation will occur on the surface. In a similar manner to the CO
poisoning effect, the presence of stabilized O/OH species with large
coverage could also impede the EOR. Therefore, the surface that has
a weaker OH binding strength can provide for more sites for ethanol
oxidation that will lead to high activity. Admittedly, an accurate
quantitative evaluation of CO, CO2, and acetic acid formation is
difficult to accomplish by IRRAS measurements alone, due to the
thin layer configuration in nanostructured material studies [38].
Therefore, the present interpretation based on IRRAS data is qual-
itative in nature. Nevertheless, analysis of the CO formation trend
and also the ratio between CO2 and acetic acid production with
applied potentials provide for very useful mechanistic insights into
this morphology-dependent activity in the EOR.
4. Conclusions
Our work clearly demonstrates the morphology effect on the
reaction rates of ethanol oxidation and CO monolayer oxidation on
Pt. Namely, the Pt catalyst with 1-D morphology evinces a struc-
ture comprised of a specific ensemble of large crystal facets and a
small density of defect sites distinctly different from the commer-
cial Pt catalysts with 0-D nanostructures. Electrocatalytic tests have
demonstrated that 1-D nanostructures with this structure exhib-
ited greatly enhanced activity and efficiency in the EOR, a negatively
shifted onset potential for ∼0.22 V, and a high current density
with at least a two-fold enhancement in the potential region
of 0.2–0.9 V as compared with associated 0-D counterparts. The
markedly high activity of CO monolayer oxidation seen also on the
1-D nanostructures, coupled with the observation of chemisorbed
COads intermediates in the EOR, provides evidence that the unique
structural and electronic properties of 1-D nanostructures can lead
to an efficient removal of surface COads. Therefore, the enhanced
Furthermore, the unique structure property of 1-D morphology
has also been advantageous in enhancing COads diffusion due to

9830
W.-P. Zhou et al. / Electrochimica Acta 56 (2011) 9824–9830
catalytic performance of 1-D nanostructures in the EOR may be
attributed to the likely interplay between balanced terrace-step
sites for ethanol adsorption and reaction as well as a facilitated
removal of chemisorbed CO on the nanowires. These results are
clearly different from previous conclusions in which it was advo-
cated that Pt nanostructures with a high density of low coordination
atoms are preferred in the EOR [2,8]. Although the detailed mecha-
nism for this morphology-dependent activity still remains elusive,
our results evidently show that the morphology of nanostructured
catalysts could have a far more profound effect than anticipated
in a complex surface reaction such as the ethanol oxidation reac-
tion. These results therefore are of importance in providing needed
structural insight into the rational design of and search for novel
nanostructured materials in ethanol electrocatalysis.
[13] S. Sun, F. Jaouen, J.-P. Dodelet, Adv. Mater. 20 (2008) 3900.
[14] C. Koenigsmann, W.-P. Zhou, R.R. Adzic, E. Sutter, S.S. Wong, Nano Lett. 10
(2010) 2806.
[15] S. Wang, S.P. Jiang, X. Wang, J. Guo, Electrochim. Acta 56 (2011) 1563.
[16] C. Koenigsmann, S.S. Wong, Energy Environ. Sci. 4 (2011) 1161.
[17] H. Wang, C.W. Xu, F.L. Cheng, S.P. Jiang, Electrochem. Commun. 9 (2007)
1212.
[18] H. Zhou, W.-P. Zhou, R.R. Adzic, S.S. Wong, J. Phys. Chem. C 113 (2009) 5460.
[19] M.I. Haftel, K. Gall, Phys. Rev. B 74 (2006) 035420.
[20] Y. Kondo, K. Takayanagi, Science 289 (2000) 606.
[21] E. Santos, P. Quaino, G. Soldano, W. Schmickler, Electrochem. Commun. 11
(2009) 1764.
[22] C. Koenigsmann, A.C. Santulli, K. Gong, M.B. Vukmirovic, W.-P. Zhou, E. Sutter,
S.S. Wong, R.R. Adzic, J. Am. Chem. Soc. 133 (2011) 9783.
[23] M.T.M. Koper, S.C.S. Lai, E. Herrero, in: M.T.M. Koper (Ed.), Fuel Cell Catalysis: A
Surface Science Approach, John Wiley & Sons, Inc., Hoboken, New Jersy, 2009.
[24] D.J. Tarnowski, C. Korzeniewski, J. Phys. Chem. B 101 (1997) 253.
[25] F. Colmati, G. Tremiliosi-Filho, E.R. Gonzalez, A. Berna, E. Herrero, J.M. Feliu,
Phys. Chem. Chem. Phys. 11 (2009) 9114.
[26] A.A. Abd-El-Latif, E. Mostafa, S. Huxter, G. Attard, H. Baltruschat, Electrochim.
Acta 55 (2010) 7951.
Acknowledgements
[27] S.C. Chang, L.W.H. Leung, M.J. Weaver, J. Phys. Chem. 94 (1990) 6013.
[28] H.A. Gasteiger, N. Markovic, J.P.N. Ross, E.J. Cairns, J. Phys. Chem. 97 (1993)
12020.
[29] J.S. Spendelow, Q. Xu, J.D. Goodpaster, P.J.A. Kenis, A. Wieckowski, J. Elec-
trochem. Soc. 154 (2007) F238.
[30] N.P. Lebedeva, M.T.M. Koper, J.M. Feliu, R.A. van Santen, J. Phys. Chem. B 106
(2002) 12938.
[31] A.V. Tripkovic, K.D. Popovic, J.D. Lovic, V.M. Jovanovic, S.I. Stevanovic, D.V. Trip-
kovic, A. Kowal, Electrochem. Commun. 11 (2009) 1030.
[32] W.-P. Zhou, K. Sasaki, D. Su, Y. Zhu, J.X. Wang, R.R. Adzic, J. Phys. Chem. C 114
(2010) 8950.
This work was supported in part by U.S. Department of Energy,
Divisions of Chemical and Material Sciences, under the Contract
No. DE-AC02-98CH10886. Synthesis and characterization work (CK
and SSW) on nanowires was supported by the U.S. Department of
Energy, Basic Energy Sciences, Materials Sciences and Engineer-
ing Division. WPZ thanks for the financial support from the LDRD
program at Brookhaven National Laboratory.
[33] M.H. Shao, R.R. Adzic, Electrochim. Acta 50 (2005) 2415.
[34] M. Heinen, Z. Jusys, R.J. Behm, J. Phys. Chem. C 114 (2010) 9850.
[35] R.B. Kutz, B. Braunschweig, P. Mukherjee, R.L. Behrens, D.D. Dlott, A. Wieck-
owski, J. Catal. 278 (2011) 181.
[36] T. Iwasita, E. Pastor, Electrochim. Acta 39 (1994) 531.
[37] S.C.S. Lai, S.E.F. Kleyn, V. Rosca, M.T.M. Koper, J. Phys. Chem. C 112 (2008) 19080.
[38] M. Li, A. Kowal, K. Sasaki, N. Marinkovic, D. Su, E. Korach, P. Liu, R.R. Adzic,
Electrochim. Acta 55 (2010) 4331.
References
[1] A. Wieckowski, E.R. Savinova, C.G. Vayenas (Eds.), Catalysis and Electrocatalysis
at Nanoparticle Surfaces, Marcel Dekker, Inc., New York, 2003.
[2] N. Tian, Z.-Y. Zhou, S.-G. Sun, Y. Ding, Z.L. Wang, Science 316 (2007) 732.
[3] M. Arenz, K.J.J. Mayrhofer, V. Stamenkovic, B.B. Blizanac, T. Tomoyuki, P.N. Ross,
N.M. Markovic, J. Am. Chem. Soc. 127 (2005) 6819.
[4] S. Park, Y. Xie, M.J. Weaver, Langmuir 18 (2002) 5792.
[5] F. Maillard, M. Eikerling, O.V. Cherstiouk, S. Schreier, E. Savinova, U. Stimming,
Faraday Discuss. 125 (2004) 357.
[6] F. Maillard, E.R. Savinova, U. Stimming, J. Electroanal. Chem. 599 (2007) 221.
[7] J.X. Wang, H. Inada, L. Wu, Y. Zhu, Y. Choi, P. Liu, W.-P. Zhou, R.R. Adzic, J. Am.
Chem. Soc. 131 (2009) 17298.
[39] Q. Wang, G.Q. Sun, L.H. Jiang, Q. Xin, S.G. Sun, Y.X. Jiang, S.P. Chen, Z. Jusys, R.J.
Behm, Phys. Chem. Chem. Phys. 9 (2007) 2686.
[40] A. Wieckowski, M. Rubel, C. Gutierrez, J. Electroanal. Chem. 382 (1995) 97.
[41] N.M. Markovic, J.P.N. Ross, Surf. Sci. Rep. 45 (2002) 117.
[42] A. López-Cudero, J. Solla-Gullón, E. Herrero, A. Aldaz, J.M. Feliu, J. Electroanal.
Chem. 644 (2010) 117.
[8] Z.-Y. Zhou, Z.-Z. Huang, D.-J. Chen, Q. Wang, N. Tian, S.-G. Sun, Angew. Chem.
Int. Ed. 49 (2010) 411.
[9] M. Subhramannia, V.K. Pillai, J. Mater. Chem. 18 (2008) 5858.
[10] S.W. Lee, S. Chen, W. Sheng, N. Yabuuchi, Y.-T. Kim, T. Mitani, E. Vescovo, Y.
Shao-Horn, J. Am. Chem. Soc. 131 (2009) 15669.
[11] J.Y. Chen, B. Lim, E.P. Lee, Y.N. Xia, Nano Today 4 (2009) 81.
[12] Y. Takasu, T. Iwazaki, W. Sugimoto, Y. Murakami, Electrochem. Commun. 2
(2000) 671.
[43] E.G. Ciapina, S.F. Santos, E.R. Gonzalez, J. Electroanal. Chem. 644 (2010) 132.
[44] S.-C. Chang, Y. Ho, M.J. Weaver, Surf. Sci. 265 (1992) 81.
[45] B. Hammer, Top. Catal. 37 (2006) 3.
[46] A.D. Karmazyn, V. Fiorin, S.J. Jenkins, D.A. King, Surf. Sci. 538 (2003) 171.
[47] B.E. Hayden, K. Kretzschmar, A.M. Bradshaw, R.G. Greenler, Surf. Sci. 149 (1985)
394.
[48] B. Hammer, Y. Morikawa, J.K. Nørskov, Phys. Rev. Lett. 76 (1996) 2141.
[49] B. Hammer, J.K. Nørskov, Adv. Catal. 45 (2000) 71.