Surface composition tuning of Au-Pt bimetallic nanoparticles for enhanced carbon monoxide and methanol electro-oxidation

Article abstract of DOI:10.1021/ja402072r

The ability to direct bimetallic nanoparticles to express desirable surface composition is a crucial step toward effective heterogeneous catalysis, sensing, and bionanotechnology applications. Here we report surface composition tuning of bimetallic Au-Pt electrocatalysts for carbon monoxide and methanol oxidation reactions. We establish a direct correlation between the surface composition of Au-Pt nanoparticles and their catalytic activities. We find that the intrinsic activities of Au-Pt nanoparticles with the same bulk composition of Au_0.5Pt_0.5 can be enhanced by orders of magnitude by simply controlling the surface composition. We attribute this enhancement to the weakened CO binding on Pt in discrete Pt or Pt-rich clusters surrounded by surface Au atoms. Our finding demonstrates the importance of surface composition control at the nanoscale in harnessing the true electrocatalytic potential of bimetallic nanoparticles and opens up strategies for the development of highly active bimetallic nanoparticles for electrochemical energy conversion.

Full text of DOI:10.1021/ja402072r

Article
pubs.acs.org/JACS
Surface Composition Tuning of Au−Pt Bimetallic Nanoparticles for
Enhanced Carbon Monoxide and Methanol Electro-oxidation
Jin Suntivich,†,▽ Zhichuan Xu,‡,▽ Christopher E. Carlton,‡ Junhyung Kim,‡ Binghong Han,†
Seung Woo Lee,‡ Nicephore Bonnet,† Nicola Marzari,†,¶ Lawrence F. Allard,⊥ Hubert A. Gasteiger,*,‡,∥
́
Kimberly Hamad-Schifferli,*,‡,§ and Yang Shao-Horn*,†,‡
‡
§
†Department of Materials Science and Engineering, Department of Mechanical Engineering, and Department of Biological
Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States
⊥Microscopy Group, Oak Ridge National Laboratory, 1 Bethel Valley Road, Building 4515, Oak Ridge, Tennessee 37831,
United States
S
* Supporting Information
ABSTRACT: The ability to direct bimetallic nanoparticles to express
desirable surface composition is a crucial step toward effective heterogeneous
catalysis, sensing, and bionanotechnology applications. Here we report
surface composition tuning of bimetallic Au−Pt electrocatalysts for carbon
monoxide and methanol oxidation reactions. We establish a direct correlation
between the surface composition of Au−Pt nanoparticles and their catalytic
activities. We find that the intrinsic activities of Au−Pt nanoparticles with the
same bulk composition of Au0.5Pt0.5 can be enhanced by orders of magnitude
by simply controlling the surface composition. We attribute this enhance-
ment to the weakened CO binding on Pt in discrete Pt or Pt-rich clusters
surrounded by surface Au atoms. Our finding demonstrates the importance of surface composition control at the nanoscale in
harnessing the true electrocatalytic potential of bimetallic nanoparticles and opens up strategies for the development of highly
active bimetallic nanoparticles for electrochemical energy conversion.
(vide infra). By tuning the surface composition of Au−Pt
nanoparticles, we establish a direct correlation between surface
compositions and catalytic activities of Au−Pt nanoparticles for
the CO electro-oxidation reaction (COR: CO + H2O → CO2 +
2H+ + 2e−) and methanol electro-oxidation reaction (MOR:
CH3OH + H2O → CO2 + 6H+ + 6e−). We find that the
intrinsic activities of Au−Pt nanoparticles with a bulk
nanoparticle composition of Au0.5Pt0.5 can be enhanced by at
least an order of magnitude by simply controlling the surface
composition, thereby demonstrating the importance of surface
composition control for energy conversion nanomaterials.
Au−Pt nanoparticles have been the subject of many
electrocatalytic studies such as O2 reduction reaction,15,16
MOR,17,18 and COR.13 In bulk, Au and Pt are immiscible, but
at nanoscale, the two elements can alloy and undergo solid
solutioning.19 In this study, we utilize this phenomenon to
synthesize single-phase Au0.5Pt0.5 nanoparticles. We then
control the Au0.5Pt0.5 nanoparticles to have surface Pt or Au
enrichment by taking advantage of the difference in the surface
free20 and binding energies with various adsorbates between Pt
and Au.21 The degree of surface Pt or Au enrichment is
quantified by element-specific surface electro-adsorption/
desorption via cyclic voltammetry and energy-dispersive
INTRODUCTION
■
Fundamental studies on extended surfaces have shown that the
activity of an electrocatalyst can be described by its surface
electronic structure, which, if properly controlled, can radically
enhance the electrocatalytic activity.1−5 However, realization of
this enhancement in nanomaterials has been challenging
because of the difficulty in controlling surface composition at
the nanoscale. As precise control over surface composition is
critical to obtain the most optimum surface electronic structure
and catalytic activity,1,3,6 mastering the ability to control surface
composition for nanoparticles is a critical step toward
applications in fuel cells based on small molecules of energy
consequence such as methanol and natural gas derivatives.
Toward this end, methods to precisely control surface
chemistry at the nanoscale including selective leaching,2,7
adsorbate-driven segregation,8−12 and multistep synthesis13,14
have been developed. However, a direct correlation between
tailored surface compositions of nanoparticles and their
catalytic activities, particularly in a low pH environment
mimicking a proton-exchanged membrane electrolyte, still
remains elusive due to the fact that only a few metals are
stable in acid and that a very limited number of techniques can
allow quantification of surface compositions of nanoparticles.2,7
In this contribution, we accomplish the surface composition
control by strategically selecting Au−Pt chemistry as both Au
and Pt are stable in acid and can be quantified electrochemically
Received: February 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja402072r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
spectroscopy (EDS) in an aberration-corrected scanning
transmission electron microscope (STEM). We show that Au
enrichment on the surface greatly enhances the intrinsic
activities of the Au−Pt nanoparticles for COR by orders of
magnitude in comparison to commercial Pt and Au nano-
particles. For MOR, Au−Pt nanoparticles with a surface Pt
composition of ∼70 atom % have the maximum MOR activity,
comparable to state-of-the-art Pt-alloy nanoparticle catalysts.
The activity enhancement of Au−Pt nanoparticles in catalyzing
COR and MOR is attributed to the weakening of the CO
binding on Pt following formation of a discrete Pt or Pt-rich
cluster surrounded by Au atoms.
EXPERIMENTAL SECTION
■
Nanoparticle Synthesis and Catalyst Preparation. Amounts of
0.25 mmol of HAuCl4 (Sigma-Alrich) and 0.25 mmol of H2PtCl6
(Sigma-Alrich) were dissolved in 20 mL of oleylamine (Sigma-Alrich)
at 40 °C under an Ar blanket. The solution was evaporated and heat-
treated at 160 °C for 2 h. Afterward, 100 mL of ethanol was added,
and the mixture was centrifuged at 6000 rpm for 10 min. The
sediments were redispersed and ultrasonicated in a mixture of 600 mL
of hexane and ∼320 mg of Vulcan XC-72 (Premetek, USA).
Afterward, the solution-containing supported Au0.5Pt0.5 nanoparticles
were purged with Ar at room temperature and then dried in vacuum
for 24 h. Procedures for sequential loading Au−Pt nanoparticles on
Vulcan carbon (21 wt %) and thermal treatments are detailed in the
Supporting Information.
Figure 1. Schematic models of Au0.5Pt0.5 nanoparticles undergoing
surface-energy-driven restructuring upon different thermal treatments.
Differential surface energies of Au and Pt in different environments
enable adsorbate-driven segregation, which leads to different surface
compositions despite all having the same bulk composition.
500 °C (Figure 1a and 1b), during which an average particle
size grew from ∼6 to ∼8 nm (Supporting Information, Figure
S1) and no phase separation was detected by XRD (Supporting
Information, Figure S2). The surface composition change from
Pt-rich to Au-rich was evaluated using element-specific surface
electro-adsorption/desorption via cyclic voltammetry (CV) and
aberration-corrected EDS in STEM. Using the charge
associated with the hydrogen adsorption/desorption and the
Au oxide reduction in the CV data, the Pt and Au surface
compositions were extracted, respectively (Supporting In-
formation, Figure S3). Specifically, upon heating at 500 °C in
air (Figure 1a and 1b), Pt-rich (90 3 atom % Pt) surfaces of
colloidal Au0.5Pt0.5 nanoparticles were changed to a Au-rich
Scanning Transmission Electron Microscopy-Energy-Disper-
sive X-Ray Spectroscopy (STEM-EDS). A JEOL JEMS2200 STEM
outfitted with a CEOS aberration corrector and a Bruker XFlash 5030
T detector at the Advanced Microscopy Laboratory at the Oak Ridge
National Laboratory was used to form conventional bright and dark
field images, as well as EDX maps as part of the SHaRE user program.
All imaging and spectroscopy was done at 200 KeV. The EDS maps
were formed with acquisition times of ≥20 min utilizing drift
correction and were postprocessed with a smooth algorithm (Bruker
Espirit). The effect of the smoothing algorithm is shown in Supporting
Information Figure S11. We note that the surface composition as
evaluated by STEM-EDS does not precisely match with the result from
CV; this is likely a result of the resolution limit of STEM-EDS, which
was estimated to be ∼0.5 nm, approximately twice the distance
between two (111) planes.22 An additional STEM-EDS image is
included in the Supporting Information.
Electrochemical Characterization. The thin-film rotating disk
electrode for the Au0.5Pt0.5 nanoparticle study was a drop cast from a
solution containing the catalyst and an appropriate amount of Nafion
(ion power) with a final composition of 18 μgAu+Pt cm−2disk, 68 μgVulcan
cm−2disk, and 38 μgNafion cm−2disk. Electrochemical measurements were
conducted with a rotating-disk electrode using a bipotentiostat (Pine).
The electrolyte was prepared using a dilution of either H2SO4 (Fluka)
or HClO4 (GFS Chemicals) using Milli-Q water (18 MΩ·cm). The
potential range for both COR and MOR studies was restricted
between 0.05 and 1.1 V vs RHE. Error bars represent standard
deviations from at least three independent measurements.
surface (88
1 atom % Au). This change in the surface
composition was further confirmed by STEM-EDS analysis of
Pt and Au distributions within individual Au0.5Pt0.5 nano-
particles. Specifically, a Pt shell (blue) and Au core (yellow)
were found for colloidal Au0.5Pt0.5 nanoparticles (Figure 3a),
while a Au shell and Pt were observed for Au0.5Pt0.5
nanoparticles after the 500 °C heat treatment in Ar in Figure
3b. Additional representative TEM images are included in the
Supporting Information.
Au0.5Pt0.5 nanoparticles with ∼70 and ∼30 atom % Pt on the
surface were also obtained by heat-treating as-synthesized
Au0.5Pt0.5 nanoparticles supported on carbon (Figure 1c and
1d) at temperature lower than 500 °C. A treatment at 250 °C
in air gave rise to a 66
4 atom % Pt surface from CV
RESULTS AND DISCUSSION
measurements (Figure 2c), in good agreement with the STEM-
EDS analysis of Pt and Au distributions within individual
particles (Figure 3c). We note that this condition (250 °C in
air) represents the lower limit of the heat-treating temperatures
for surface Au enrichment, where the majority of surfactant
molecules that stabilize Pt on the surface must be removed to
allow Au to migrate and enrich the surface. In comparison, the
heat treatment at 250 °C in Ar did not result in any measurable
change in the surface composition (Supporting Information,
Figure S4a) because fewer surfactant molecules were removed
during heating in Ar than in air. This hypothesis is supported by
the thermal gravimetric analysis data, where more weight loss
■
Tuning the Surface Composition of Au0.5Pt0.5 Nano-
particles. We synthesized four different Au0.5Pt0.5 nano-
particles with surface atomic compositions ranging from ∼90
to ∼10 atom % Pt (Figure 1). First, Au0.5Pt0.5 nanoparticles
were synthesized by a colloidal route as demonstrated
previously,23 which yielded AuPt particles with Pt-enriched
surface and a nominal bulk composition of Au0.5Pt0.5, as
revealed by inductive plasma coupling. We then induced surface
Au enrichment using the fact that Au has a lower surface energy
than Pt20,24 by heat-treating as-synthesized Au0.5Pt0.5 nano-
particles supported on carbon (21 wt. % Au0.5Pt0.5/C) in Ar at
B
dx.doi.org/10.1021/ja402072r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. Cyclic voltammograms of Au0.5Pt0.5 nanoparticles in 0.5 M H2SO4 at 50 mV/s scan rate. (a) As-synthesized Au0.5Pt0.5 (90 3% surface Pt),
(b) Ar treatment at 500 °C for 30 min (13 5% surface Pt), (c) air treatment at 250 °C for 30 min (68 3% surface Pt), and (d) air treatment at
250 °C for 30 min, followed by Ar treatment at 350 °C for 30 min (30
independent measurements.
1% surface Pt). Standard deviations were calculated form three
was found for Au0.5Pt0.5 nanoparticles following the heat
treatment in Ar than in air at 250 °C (Supporting Information,
Figure S4b). Additionally, a subsequent heat treatment in Ar at
350 °C gave rise to further surface Au enrichment, yielding a 28
1 atom % Pt surface concentration (Figure 2d) as measured
using CV and confirmed by spectral Pt and Au maps from the
EDS analysis (Figure 3d). We point out that, because the 30
atom % Pt particle preparation requires a two-step heat
treatment, dealloying of Pt and Au can begin to occur, as can be
seen via TEM (Figure 3d). It also should be mentioned that
some differences may exist in the surface Pt and Au
concentrations as quantified via CV and EDS measurements
as EDS in the STEM with spatial resolution of 0.5 nm is less
sensitive to the topmost layer of Au0.5Pt0.5 nanoparticles than
CV measurements.22
CO Electro-Oxidation on Au0.5Pt0.5 Nanoparticles in
Acid. The CO oxidation current was obtained from scanning a
thin film of Au0.5Pt0.5 nanoparticles deposited on a rotating disk
electrode (RDE) in a CO-saturated HClO4 solution as a
function of potential. Geometric current densities (current
normalized to the RDE area) of the Au0.5Pt0.5 nanoparticles
with different surface compositions (Figure 4a) showed that
increasing the surface Au concentration enhanced the intrinsic
activity of COR with Au0.5Pt0.5 nanoparticles with ∼10 and ∼30
atom % Pt surface exhibiting specific activities (current
normalized to the combined electrochemical active surface
area of Pt and Au) markedly higher than those with ∼70 and
∼90 atom % Pt by 1 order of magnitude and commercial Au/C
and Pt/C (Tanaka Kikinzoku, “TKK”) by 2 orders of
magnitude (see Figure 4b and Supporting Information, Figure
S5). The enhancement in the specific activity was also observed
at the mass level, where the mass activities for Au0.5Pt0.5
nanoparticles with ∼10 and ∼30 atom % Pt on the surface
were found to be significantly higher than those with ∼70 and
∼90 atom % Pt and commercial Au/C and Pt/C (Supporting
Information, Figure S6).
We hypothesize the enhancement in the CO electro-
oxidation kinetics of Au0.5Pt0.5/C to stem from the modified
electronic structure of surface Pt that is surrounded by Au
neighbors,25 which results in a weakened binding strength of
COad. This hypothesis is supported by the recent single-crystal
work, whereby isolated surface Pt clusters surrounded by Au
neighbors (see Supporting Information, Scheme S1) were
shown to have weaker CO binding than Pt due to lateral
interactions and the absence of binding contribution from the
next nearest Pt atoms.25,26 We believe that this configuration,
where the Pt or Pt-rich clusters are surrounded by Au or Au-
rich atoms, well represents the surface structure of our
Au0.5Pt0.5 nanoparticles with surface Au enrichment as Pt
prefers to have Pt as its nearest neighbor due to favorable
enthalpy of the Pt−Pt bond.27 Further evidence on the
importance of weakened COad binding came from a positive
reaction order of CO concentration on the COR activity
observed for Au0.5Pt0.5 nanoparticles with 30% surface Pt and a
C
dx.doi.org/10.1021/ja402072r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. Aberration-corrected scanning transmission electron microscopy-energy-dispersive X-ray spectroscopy (STEM-EDS) of Au0.5Pt0.5
nanoparticles. (a) As-synthesized Au0.5Pt0.5 (90% surface Pt), (b) Ar treatment at 500 °C for 30 min (13% surface Pt), (c) air treatment at 250
°C for 30 min (68% surface Pt), (d) air treatment at 250 °C for 30 min, followed by Ar treatment at 350 °C for 30 min (30% surface Pt). The signal
count from the trace line is shown in the top inset, showing Pt and Au Lα and, in the bottom inset, their calculated relative concentrations. The right
panel shows Pt, Au, and their combined Lα mappings. The intensity plots are based on a raw EDS signal that was smoothed in a manner detailed by
Figure S10 (Supporting Information). The colors of the maps were universally scaled so that the color scale of 1 represents 10% less than the
maximum concentration of either Pt or Au in any of the nanoparticles. Signal processing procedures and additional representative TEM images are
included in the Supporting Information.
zero reaction order (invariant activity with CO concentration)
for ∼70 atom % Pt (Supporting Information, Figure S7). This
finding is in contrast to the case of Pt/C, which has a negative
reaction order, indicative of the poisonous nature of CO for
COR.28
assess the catalytic trend, and we compare the COR activity
normalized by the Pt surface area (Figure 4b, blue, right axis).
We find that the COR activity monotonically increases with Au
content, consistent with the continuing decrease in the CO
adsorption energy with increasing Au concentration, in
agreement with the literature.26 The critical role of CO binding
strength in the COR kinetics can be understood using the
framework of the well-known Langmuir−Hinshelwood mech-
anism, where COad + OHad → COOHad is the rate-limiting
step.31,32 Specifically, weakened CO binding on Pt reduces Pt
poisoning, which decreases CO coverage and allows OH to
nucleate more efficiently, and the activation energy, which is
prohibitively high as a result of strongly adsorbed COad on Pt,
consequently enhancing the COR kinetics.
We rule out the possibility that CO may adsorb on Au
because the CO stripping charge scales with Pt instead of the
total surface area (see Supporting Information, Figure S8) and
that Au is known to undergo CO desorption below room
temperature in the form of either extended29 or Au−Pt alloy
surfaces.26,30 As we did not observe significant differences in the
surface atomic structure after different heat treatments (see
Supporting Information Figure S9 and Table S1), we also rule
out the possibility that the enhanced COR kinetics stem from
Au0.5Pt0.5 nanoparticles with different surface facets, in
particular, high-index planes. We therefore propose that the
enhancement comes from the intrinsic electronic structure
modification on Pt by Au. In this hypothesis, Au serves to
modify Pt electronic structure and thereby to more accurately
Methanol Electro-Oxidation on Au0.5Pt0.5 Nanopar-
ticles in Acid. We further applied the thin-film RDE
methodology to examine the MOR activities of Au0.5Pt0.5
nanoparticles (Figures 5a,b). Methanol oxidation currents
normalized to the combined electrochemically active surface
D
dx.doi.org/10.1021/ja402072r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. Activities of Au0.5Pt0.5 NPs at various surface compositions for carbon monoxide electro-oxidation reaction (COR). (a) As-measured COR
current densities of Au0.5Pt0.5 NPs at various surface compositions in 0.1 M HClO4 (measured at 1600 rpm, 5 mV s−1 scan rate, 0.018 mgmetal cm−2
disk
loading). (b) The specific COR activities, is, Au0.5Pt0.5 NPs (capacitance-corrected) normalized to either both Pt and Au surface area (red, left axis)
or Pt surface area (blue, right axis) from a positive-going scan direction as a function of surface composition at 0.7 V vs RHE potential. The activity
for 100% Pt was obtained from commercial Pt/C catalyst (Tanaka Kikinzoku). Error bars represent standard deviations of at least three independent
measurements.
Figure 5. Activities of Au0.5Pt0.5 NPs at various surface compositions for the methanol electro-oxidation reaction (MOR). (a) The methanol electro-
oxidation reaction (MOR) geometric current densities of Au0.5Pt0.5 NPs at various surface compositions in 0.1 M HClO4 (at 50 mV s−1 scan rate,
0.018 mgmetal cm−2 loading). (b) The specific MOR activities, is, of Au0.5Pt0.5 NPs (capacitance-corrected) normalized to either both Pt and Au
disk
surface area (red, left axis) or Pt surface area (blue, right axis) from the positive-going scan direction as a function of surface composition at 0.7 V vs
RHE potential. The activity for 100% Pt was obtained from commercial Pt/C catalyst (Tanaka Kikinzoku). Error bars represent standard deviations
of at least three independent measurements.
area of Pt and Au were measured as shown in Figure 5b and
Supporting Information Figure S10. Like COR, the specific
activities for MOR were found to greatly depend on the surface
composition, exhibiting an increasing catalytic activity with
increasing surface Au composition from the surface Pt-rich side,
until reaching a maximum at the surface Pt concentration of
∼70%, where the activity began to subsequently decrease with
more surface Au on the surface Au-rich side (known as the
“volcano relationship” in catalysis). Interestingly, Au0.5Pt0.5
nanoparticles at this optimum surface composition had higher
specific activities than commercial Pt/C (both at ∼3 and ∼7
nm diameters, supplied by TKK and Johnson Matthey,
respectively). The enhanced activity of these Au0.5Pt0.5
nanoparticles relative to Pt/C was also realized when
normalized to the catalyst mass (Supporting Information
Figure S11, compared only against ∼7 nm commercial catalyst
to avoid specific area difference), demonstrating the robustness
of the catalytic enhancement in these nanomaterials. The MOR
activity of the champion Au0.5Pt0.5 nanoparticles was found to
be comparable to state-of-the-art Pt−Ru and Pt3Sn nano-
particles with respect to both mass and surface area
(Supporting Information, Table S2), although this was still
lower than the result from the Pt−Ru extended surface.
To explain the volcano trend in the MOR activity, we first
focus on the Au0.5Pt0.5 nanoparticles with Pt-enriched surface
(the right side of the volcano), where MOR activity increases
with Au addition. We attribute the enhancement on this branch
to the modified Pt electronic structure25 with weakened CO
binding strength as proposed in the previous section.
Specifically, the modified Pt electronic structure can more
efficiently catalyze the COR, which is the rate-limiting step in
the indirect mechanism of the MOR.31,32 This hypothesis is
congruent with previous reports that the electrochemical
oxidation of methanol on Pt-based surfaces proceeds
preferentially through the indirect pathway, where methanol
undergoes dissociative adsorption and electrochemical dehy-
drogenation to form COad (CH3OH → COad + 4H+ + 4e−)
before proceeding to the final step with the oxidation of the
COad,1,33 the latter of which is considered to be the rate-
limiting step. This proposal agrees well with the observation
E
dx.doi.org/10.1021/ja402072r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
that the increase in the MOR activity with surface Pt
enrichment (Figure 5) is concomitant with COR enhancement
(Figure 4). We therefore propose that the origin of MOR
enhancement can be attributed largely to the weakened CO
binding on Pt, which increases the COR reactivity on Au0.5Pt0.5
nanoparticles. Within the framework of this hypothesis, Au
again modifies the electronic structure of surface Pt, and
therefore we normalize the MOR activity by the Pt surface area
to more accurately assess the catalytic trend (Figure 5b, blue,
right axis). With this normalization, the MOR activity
monotonically increases with Au content on the volcano’s
right side (>50 atom % Pt), the same trend as what was
observed in COR. This supports our hypothesis for the catalytic
enhancement in the case of the Pt-rich surface Au0.5Pt0.5
nanoparticles.
To explain the activity trend of the Au0.5Pt0.5 nanoparticles
with Au surface enrichment (the left side of the volcano), which
has reduced MOR activity with increasing surface Au
composition (both per total and Pt surface areas), we propose
the rate-limiting step on this branch to be the dissociative
adsorption/dehydrogenation of the methanol. With increasing
Au surface enrichment, the probability of forming 3-fold Pt
assembly on the surface1,34 decreases, and thus the ability for
the surface to dissociatively adsorb/dehydrogenate methanol
decreases, resulting in the negative contribution to the MOR
activity.
nanoparticle electrocatalyst. Further studies that combine the
advantages from both the electronic effect (such as Au), as
demonstrated in this work, and the classical bifunctional effect
(such Ru or Sn) can be an interesting area for the design of
highly active catalysts that promote the oxidation of small
organic molecules.
CONCLUSION
■
We present a correlation between electrocatalytic activities and
surface compositions of bimetallic nanoparticles. By manipulat-
ing and controlling the heat treatment conditions, we utilize
surface energy as a means of selectively enriching either Pt or
Au on the Au0.5Pt0.5/C nanoparticle catalysts. Our finding
demonstrates that tuning the surface composition is critical to
enhance the catalytic activity and that, without it, the as-
measured catalytic activity may not be representative of the true
catalytic potential of nanoparticles. The enhanced COR and
MOR activities of Au0.5Pt0.5 can be attributed to the weakened
CO binding on Pt in the presence of Au, where Au serves as an
electronic modifier to reduce CO poisoning on Pt. Our finding
reflects the critical importance of controlling surface
composition for nanoparticle electrocatalysis, where we
demonstrate the same bulk Au0.5Pt0.5 nanoparticles can have
enhanced or reduced activities depending on the surface
composition. This avenue of optimization must be integrated in
the design of future highly active electrocatalysts.
The optimum MOR activity occurs on Au0.5Pt0.5 nano-
particles with surface Pt composition of ∼70 atom %, while the
optimum MOR activity occurs at ∼90 atom % Pt surface
concentration1,34 for Pt−Ru surfaces. This difference can be
attributed to the dissimilar mechanisms responsible for the
enhancement in the COR activity on Au0.5Pt0.5 nanoparticles
and Pt−Ru and Pt3Sn catalysts reported previously.1,34,35
Unlike Au, Ru or Sn can promote water oxidation36,37 to
increase the availability of OHad at lower overpotential and
consequently enhance the activity of COR via the bifunctional
Langmuir−Hinshelwood mechanism (COad-Pt + OHad-Ru →
CO2 + H+ + e−). Au, however, is much more inert, and
therefore the catalytic reactions on the Au0.5Pt0.5 nanoparticles
likely proceed only through Pt sites that have been modified by
nearby and subsurface Au atoms, where the maximum COR
activity might be obtained when Pt sites have attained the
optimum CO binding. This is in contrast to in the case of Pt−
Ru, where the optimum composition is reached when
appropriate populations of COPt and OHRu are formed.
We show that controlling the surface atomic concentration of
Au on Au0.5Pt0.5 nanoparticles is immensely critical for the
COR and MOR activities. The COR activity of the champion
Au0.5Pt0.5 nanoparticles is also comparable to state-of-the-art
Pt3Sn nanoparticles with respect to both mass and surface area
(Supporting Information Table S3), although it is important to
recognize that this value is still lower than the result from the
Pt3Sn extended surface. In addition, our champion Au0.5Pt0.5
nanoparticle catalyst has significantly higher MOR activity than
other Au−Pt catalysts reported previously (Supporting
Information Table S2).38−40 Even though Au alone is not
active in water oxidation,36 unlike Ru or Sn, it is interesting to
note that the optimum MOR activities of Au0.5Pt0.5 nano-
particles are comparable to those of state-of-the-art MOR
catalysts such as Pt−Ru and Pt3Sn nanoparticles (Supporting
Information Table S2). Our work demonstrates that electronic
structure control by way of surface composition manipulation
represents another avenue for designing a future generation of
ASSOCIATED CONTENT
* Supporting Information
■
S
Details on specific heat treatment condition, surface area
determination, and reaction order analysis. Particle size
distribution of all Au−Pt particles, X-ray diffraction on all
Au−Pt particles, details on surface area extraction from cyclic
voltammograms, thermogravimetry of Au−Pt colloids, CO
reaction order data for Au−Pt particles, CO stripping
voltammetry of Au−Pt particles, additional STEM-EDS images,
and activity comparison. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
hubert.gasteiger@tum.de; schiffer@mit.edu; shaohorn@mit.
■
edu
Present Addresses
∥Department of Chemistry, Technische Universitat Munchen,
̈
̈
Lichtenbergstrasse 4, D 85747 Garching, Germany.
#MIT Lincoln Laboratory, 244 Wood St., Lexington, MA
02420.
¶
́
Theory and Simulation of Materials, Ecole Polytechnique
erale de Lausanne, 1015 Lausanne, Switzerland.
Fed
́ ́
Author Contributions
▽These authors contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported in part by the MRSEC Program of
the National Science Foundation under award number DMR
0819762, the DOE Hydrogen Initiative program under award
number DE-FG02-05ER15728, and the DOE EERE award DE-
EE0000458 via subcontract through General Motors. Micros-
copy work at Oak Ridge National Laboratory’s High Temper-
F
dx.doi.org/10.1021/ja402072r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(31) Lee, S. W.; Chen, S. O.; Sheng, W. C.; Yabuuchi, N.; Kim, Y. T.;
Mitani, T.; Vescovo, E.; Shao-Horn, Y. J. Am. Chem. Soc. 2009, 131,
15669.
(32) Lebedeva, N. P.; Koper, M. T. M.; Feliu, J. M.; van Santen, R. A.
J. Phys. Chem. B 2002, 106, 12938.
ature Materials Laboratory was sponsored by the U.S.
Department of Energy, Office of Energy Efficiency and
Renewable Energy, Vehicle Technologies Program. J.S. was
supported by the Chesonis Foundation Fellowship. The
authors would like to thank Dr. Eva Mutoro for the help
with the illustration.
(33) Lin, W. F.; Iwasita, T.; Vielstich, W. J. Phys. Chem. B 1999, 103,
3250.
(34) Gasteiger, H. A.; Markovic, N.; Ross, P. N.; Cairns, E. J. J.
Electrochem. Soc. 1994, 141, 1795.
(35) Gasteiger, H. A.; Markovic, N. M.; Ross, P. N. J. Phys. Chem.
1995, 99, 8945.
(36) Borkowska, Z.; Tymosiak-Zielinska, A.; Shul, G. Electrochim.
Acta 2004, 49, 1209.
(37) House, C. I.; Kelsall, G. H. Electrochim. Acta 1984, 29, 1459.
(38) Lin, Z. H.; Shih, Z. Y.; Tsai, H. Y.; Chang, H. T. Green Chem.
2011, 13, 1029.
REFERENCES
■
(1) Gasteiger, H. A.; Markovic, N.; Ross, P. N.; Cairns, E. J. J. Phys.
Chem. 1993, 97, 12020.
(2) Stamenkovic, V. R.; Mun, B. S.; Arenz, M.; Mayrhofer, K. J. J.;
Lucas, C. A.; Wang, G. F.; Ross, P. N.; Markovic, N. M. Nat. Mater.
2007, 6, 241.
(3) Gasteiger, H. A.; Markovic, N.; Ross, P. N.; Cairns, E. J. J. Phys.
(39) Ding, Y.; Ge, X. B.; Wang, R. Y.; Liu, P. P. Chem. Mater. 2007,
19, 5827.
(40) Luo, J.; Maye, M. M.; Kariuki, N. N.; Wang, L. Y.; Njoki, P.; Lin,
Y.; Schadt, M.; Naslund, H. R.; Zhong, C. J. Catal. Today 2005, 99,
291.
Chem. 1994, 98, 617.
(4) Norskov, J. K.; Bligaard, T.; Rossmeisl, J.; Christensen, C. H.
Nature Chem. 2009, 1, 37.
(5) Yu, W.; Porosoff, M. D.; Chen, J. G. Chem. Rev. 2012, 112, 5780.
(6) Tian, N.; Zhou, Z. Y.; Sun, S. G.; Ding, Y.; Wang, Z. L. Science
2007, 316, 732.
(7) Chen, S.; Sheng, W. C.; Yabuuchi, N.; Ferreira, P. J.; Allard, L. F.;
Shao-Horn, Y. J. Phys. Chem. C 2009, 113, 1109.
(8) Mayrhofer, K. J. J.; Juhart, V.; Hartl, K.; Hanzlik, M.; Arenz, M.
Angew. Chem., Int. Ed. 2009, 48, 3529.
(9) Tao, F.; Grass, M. E.; Zhang, Y. W.; Butcher, D. R.; Renzas, J. R.;
Liu, Z.; Chung, J. Y.; Mun, B. S.; Salmeron, M.; Somorjai, G. A. Science
2008, 322, 932.
(10) Kitagawa, H.; Kobayashi, H.; Yamauchi, M.; Kubota, Y.; Kato,
K.; Takata, M. J. Am. Chem. Soc. 2010, 132, 5576.
(11) Yang, H. Angew. Chem., Int. Ed. 2011, 50, 2674.
(12) Wang, D. L.; Xin, H. L.; Wang, H. S.; Yu, Y. C.; Rus, E.; Muller,
D. A.; DiSalvo, F. J.; Abruna, H. D. Chem. Mater. 2012, 24, 2274.
(13) Peng, Z. M.; Yang, H. Nano Res. 2009, 2, 406.
(14) Mazumder, V.; Chi, M. F.; More, K. L.; Sun, S. H. Angew. Chem.,
Int. Ed. 2010, 49, 9368.
(15) Hartl, K.; Mayrhofer, K. J. J.; Lopez, M.; Goia, D.; Arenz, M.
Electrochem. Commun. 2010, 12, 1487.
(16) Zhang, J.; Sasaki, K.; Sutter, E.; Adzic, R. R. Science 2007, 315,
220.
(17) Luo, J.; Njoki, P. N.; Lin, Y.; Mott, D.; Wang, L. Y.; Zhong, C. J.
Langmuir 2006, 22, 2892.
(18) Luo, J.; Wang, L.; Mott, D.; Njoki, P. N.; Lin, Y.; He, T.; Xu, Z.;
Wanjana, B. N.; Lim, I. I. S.; Zhong, C. J. Adv. Mater. 2008, 20, 4342.
(19) Braidy, N.; Purdy, G. R.; Botton, G. A. Acta Mater. 2008, 56,
5972.
(20) Needs, R. J.; Mansfield, M. J. Phys. Condens. Matter 1989, 1,
7555.
(21) Hammer, B.; Norskov, J. K. Adv. Catal. 2000, 45, 71.
(22) Carlton, C. E.; Chen, S.; Ferreira, P. J.; Allard, L. F.; Shao-Horn,
Y. J. Phys. Chem. Lett. 2012, 3, 161.
(23) Xu, Z. C.; Carlton, C. E.; Allard, L. F.; Shao-Horn, Y.; Hamad-
Schifferli, K. J. Phys. Chem. Lett. 2010, 1, 2514.
(24) Greeley, J.; Norskov, J. K. Electrochim. Acta 2007, 52, 5829.
(25) Pedersen, M. O.; Helveg, S.; Ruban, A.; Stensgaard, I.;
Laegsgaard, E.; Norskov, J. K.; Besenbacher, F. Surf. Sci. 1999, 426,
395.
(26) Eyrich, M.; Diemant, T.; Hartmann, H.; Bansmann, J.; Behm, R.
J. J. Phys. Chem. C 2012, 116, 11154.
(27) Vandertoorn, L. J.; Tiedema, T. J. Acta Metallurgica 1960, 8,
711.
(28) Schmidt, T. J.; Gasteiger, H. A.; Behm, R. J. J. Electrochem. Soc.
1999, 146, 1296.
(29) Kim, J.; Samano, E.; Koel, B. E. J. Phys. Chem. B 2006, 110,
17512.
(30) Ren, H.; Humbert, M. P.; Menning, C. A.; Chen, J. G.; Shu, Y.
Y.; Singh, U. G.; Cheng, W. C. Appl. Catal., A 2010, 375, 303.
G
dx.doi.org/10.1021/ja402072r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX