Precisely-controlled synthesis of Au@Pd core-shell bimetallic catalyst via atomic layer deposition for selective oxidation of benzyl alcohol

Article abstract of DOI:10.1016/j.jcat.2015.01.019

In this study, we report a novel facile strategy for atomically-precise synthesis of supported Au@Pd core-shell bimetallic catalyst via atomic layer deposition (ALD). By choosing a proper deposition condition, we can selectively deposit Pd only on Au nanoparticle surface, but not on SiO₂ support to exclusively form Au@Pd core-shell bimetallic nanoparticles, while avoiding monometallic nanoparticle formation; therein, the Pd shell thickness can be atomically precisely tuned by varying the number of Pd ALD cycles. In solvent-free oxidation of benzyl alcohol, the catalytic activities of the resulted Au@Pd/SiO₂ core-shell bimetallic catalysts showed a clear volcano-like trend with the Pd shell thickness, reaching a maximum at a Pd shell thickness of 0.6-0.8 nm due to the optimized synergistic effect. More importantly, we believe this new strategy of precise synthesis of core-shell structured bimetallic catalyst using ALD can be general to other supported bimetallic catalysts for broad applications.

Full text of DOI:10.1016/j.jcat.2015.01.019

Journal of Catalysis 324 (2015) 59–68
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Precisely-controlled synthesis of Au@Pd core–shell bimetallic catalyst
via atomic layer deposition for selective oxidation of benzyl alcohol
⇑
Hengwei Wang, Chunlei Wang, Huan Yan, Hong Yi, Junling Lu
Department of Chemical Physics, Hefei National Laboratory for Physical Sciences at the Microscale, CAS Key Laboratory of Materials for Energy Conversion,
University of Science and Technology of China, Hefei, Anhui 230026, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, we report a novel facile strategy for atomically-precise synthesis of supported Au@Pd core–
shell bimetallic catalyst via atomic layer deposition (ALD). By choosing a proper deposition condition, we
can selectively deposit Pd only on Au nanoparticle surface, but not on SiO support to exclusively form
2
Au@Pd core–shell bimetallic nanoparticles, while avoiding monometallic nanoparticle formation;
therein, the Pd shell thickness can be atomically precisely tuned by varying the number of Pd ALD cycles.
Received 2 December 2014
Revised 18 January 2015
Accepted 20 January 2015
2
In solvent-free oxidation of benzyl alcohol, the catalytic activities of the resulted Au@Pd/SiO core–shell
Keywords:
bimetallic catalysts showed a clear volcano-like trend with the Pd shell thickness, reaching a maximum at
a Pd shell thickness of 0.6–0.8 nm due to the optimized synergistic effect. More importantly, we believe
this new strategy of precise synthesis of core–shell structured bimetallic catalyst using ALD can be gen-
eral to other supported bimetallic catalysts for broad applications.
AuPd bimetallic catalyst
Core–shell structure
Atomic layer deposition
Precisely controlled synthesis
Benzyl alcohol oxidation
Synergistic effect
Ó 2015 Elsevier Inc. All rights reserved.
1
. Introduction
method often generates undesired monometallic nanoparticles
[
7,11,12]. Other strategic routes in a more controlled manner were
Bimetallic catalysts often show superior catalytic properties
also reported, such as anion coordination protocol [6], sacrificial
hydrogen method [13], surface-specific reductants [14], redox
transmetalation reactions [15], and controlled simultaneous reduc-
tion [16]. Typically, the surface structures, lattice strains, and elec-
tronic properties of core–shell bimetallic nanoparticles largely
depend on the shell thickness, which thus determines their cata-
lytic performance [8,9,16–18]. Nevertheless, a facile and general
strategy to achieve atomically precise control over the shell thick-
ness while avoiding monometallic nanoparticle formation is still
missing.
ALD is a variation on chemical vapor deposition in which met-
als, oxides, and other materials are deposited on surfaces by a
sequence of self-limiting reactions [19–22]. In recent years, ALD
has demonstrated its great potential in advanced catalysts synthe-
sis beyond its original application in microelectronics [23–28]. Due
to its self-limiting features in each deposition cycle, ALD provides a
possible method to design and modify catalysts at the nanoscale
through precise control over the structure and composition of
the underlying support, the catalytic active sites, and the protec-
tive layer [29–34]. Very recently, we reported a general strategy
of low-temperature selective metal ALD for atomically precise syn-
thesis of supported bimetallic catalysts [34], wherein monometal-
lic nanoparticle formation is avoided by selectively growing a
secondary metal on the primary metal nanoparticle but not on
compared to those of their parent monometallic counterparts
due to the synergistic effect of two components [1–3]. This behav-
ior may mainly originate from the ensemble effect, (particular
metal atoms arrangements that are required for facilitating a par-
ticular catalytic process) or electronic effect (electronic modifica-
tion resulting from hetero-nuclear metal–metal bond formation)
[
1,4]. The catalytic properties of bimetallic catalysts often vary
dramatically with their size, structure, and composition. Thus, pre-
cisely tuning these factors of bimetallic nanoparticles will make
them a new class of materials with enhanced properties for a range
of applications.
In particular, core–shell structured bimetallic nanoparticles are
currently of enormous interest. In addition to their unique proper-
ties, another motivation is to decrease precious metal consumption
by coating inexpensive metal cores with a small amount of pre-
cious metal while retaining catalytic activity [5,6]. Lots of efforts
have been devoted to exploring facile methods of synthesis of
core–shell structured bimetallic nanoparticles. The most com-
monly used synthetic method is the reduction of a second metal
onto pre-formed primary metal particles [7–10], while such
⇑
Corresponding author.
E-mail address: junling@ustc.edu.cn (J. Lu).
http://dx.doi.org/10.1016/j.jcat.2015.01.019
021-9517/Ó 2015 Elsevier Inc. All rights reserved.
0

6
0
H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
-WI). All flow rates were kept at 20 ml min 1 with
ꢁ
the support using ALD. Meanwhile, the size, composition, and
structure of the bimetallic nanoparticles can be precisely con-
trolled by tailoring the precursor pulse sequences.
catalyst (Pd/SiO
mass flow controllers.
2
However, Au ALD is still currently not applicable due to a lack of
a proper Au ALD precursor, even though Au monometallic and Au-
based bimetallic catalysts are extremely interesting systems in
catalysis [35–37]. For example, AuPd bimetallic catalysts have
showed superior catalytic performance in many reactions such as
CO oxidation [38,39], direct synthesis of hydrogen peroxide
2.1.3. Au@Pd/SiO
Pd ALD was carried out on a viscous flow reactor (GEMSTAR-6™
Benchtop ALD, Arradiance). Ultrahigh purity N (99.999%) was
used as carrier gas at a flow rate of 200 ml min . Pd ALD was used
to selectively deposit Pd only on the Au nanoparticle surface of the
2
bimetallic catalysts using selective Pd ALD
2
ꢁ1
2 2
Au/SiO catalyst, but not on the SiO support to exclusively form
[
40,41], direct synthesis of vinyl acetate [42], formic acid decompo-
Au@Pd core–shell bimetallic nanoparticles, while avoiding mono-
metallic nanoparticle formation. Such selective Pd ALD process
sition [43,44], and oxidation of carbon–hydrogen bonds [45]. In
particular, AuPd bimetallic nanoparticles can remarkably enhance
the catalytic activity and product selectivity in primary alcohol
oxidations [37,46–56].
Herein we report a novel strategy to precisely synthesize SiO
supported Au@Pd core–shell catalysts using a combined wet-
chemistry and ALD method: We first synthesized Au/SiO catalysts
was executed on the as-prepared Au/SiO
Pd(II) hexafluoroacetylacetonate (Pd(hfac)
as the Pd precursor and ultrahigh purity H
2
the Pd(hfac) precursor was contained in a sealed stainless steel
2
catalysts at 150 °C, using
2
, Sigma–Aldrich, >97%)
2
as reductant [34]. Here
2
bottle at 65 °C to get sufficient vapor pressure, and the inlet lines
were heated to 110 °C to avoid any condensation. The timing
sequence for selective Pd ALD was 300, 180, 25, and 180 s for
2
using the deposition–precipitation (DP) method, then Pd was
selectively deposited only on the surface of Au nanoparticles but
2 2 2 2
Pd(hfac) exposure, N purge, H exposure, and N exposure,
not on the SiO
2
support to exclusively form uniform Au@Pd
respectively. A series of Au@Pd bimetallic catalysts were synthe-
sized by different numbers of Pd ALD cycles, which are denoted
core–shell nanoparticles, while avoiding monometallic nanoparti-
cle formation. By varying the number of Pd ALD cycles, the thick-
ness of Pd shell was precisely tuned. Extensive characterizations
were carried out to confirm this selective deposition and uniform
formation of core–shell structured bimetallic nanoparticles.
Finally, we evaluated the catalytic performance of the resulted
as Au@xPd/SiO
the number of Pd ALD cycles). In order to further confirm the selec-
tive deposition of Pd only on the Au surface, but not on the SiO
support, Pd ALD was also carried out on the bare SiO support for
, here x = 1,
2
(here x = 1, 3, 5, 8, 10, 15, and 20, representing
2
2
different cycles under the same conditions (xc-Pd/SiO
2
Au@Pd/SiO
2
core–shell bimetallic catalysts using solvent-free aer-
2, and 8) as a control experiment.
obic oxidation of benzyl alcohol as a probe reaction, and we found
that the catalytic activities showed a clear volcano-like trend as a
function of Pd ALD cycles (or Pd shell thickness), wherein a Au@Pd
core–shell catalyst with a Pd shell thickness of 0.6–0.8 nm showed
a maximum activity due to the optimized synergistic effect via
both ensemble and electronic promotion.
2.2. Characterizations
2.2.1. Structure and compositions
Transmission electron microscopy (TEM) measurements were
performed on a JEOL-2010 instrument operated at 200 kV to char-
2 2
acterize the morphology of Au/SiO and Pd/SiO catalysts, while
characterizations of the Au@Pd bimetallic catalysts were more
carefully carried out on an aberration-corrected high-angle annu-
lar dark-field scanning TEM (HAADF-STEM) instrument at 200 kV
(JEOL-2010F, University of Science and Technology of China).
Meanwhile, energy-dispersive X-ray (EDX) spectroscopy was also
collected on the same equipment. The compositions and loadings
of catalysts were analyzed by an inductively coupled plasma-
atomic emission spectrometer (ICP-AES); therein, all samples were
dissolved in hot aqua regia.
2
. Experimental
2
2
.1. Catalyst synthesis
.1.1. Au/SiO
A 1 wt% Au/SiO
57]. Typically, 1 g HAuCl
Co., Ltd.) was dissolved into 50 ml deionized water to prepare
2
catalyst
catalyst was first prepared using DP method
ꢀ4H O (Sinopharm Chemical Reagent
2
[
4
2
ꢁ1
0
.0485 mol L
4 4
HAuCl aqueous solution. Next, 1.1 ml HAuCl
aqueous solution, 1.0 g spherical SiO2, and 80 ml deionized water
were co-added into a three-necked bottle and mixed for 30 min
under vigorous stirring at 60 °C, and ammonia was used to adjust
the pH value between 9 and 10. Then, the system was continued
vigorously stirred for another 12 h. The suspension was then cen-
trifuged and washed with deionized water for several times, and
dried at 80 °C overnight. Finally, the resulted sample was calcined
2.2.2. DRIFTS CO chemisorption
The diffuse reflectance infrared Fourier transform spectroscopy
(DRIFTS) CO chemisorption measurements were performed on a
Nicolet iS10 spectrometer equipped with an MCT detector and a
low-temperature reaction chamber (Praying Mantis Harrick).
Before DRIFTS measurements, the samples were calcined in 10%
O in He and followed by reduction in 10% H in He at 150 °C. After
2 2
cooling the sample to room temperature under He, a background
spectrum was collected. Subsequently, the sample was exposed
ꢁ1
at 250 °C under 10% O
obtain the Au/SiO catalyst. Note that mono-dispersed SiO
were synthesized according to the modified Stöber method [58].
2
in He at a flow rate of 20 ml min for 4 h to
2
2
spheres
ꢁ
1
to 10% CO in He at a flow rate of 20 ml min for about 30 min until
saturation. Next, the sample was purged with He at a flow rate of
ꢁ1
2
0 ml min for another 30 min to remove the gas phase CO and
2
.1.2. Pd/SiO
A 1 wt% Pd/SiO
method [59]. Typically, 52 mg Pd(acac)
Reagent Co., Ltd.) was dissolved into 50 ml acetylacetone to pre-
2
catalyst
catalyst was prepared by wet impregnation
(Sinopharm Chemical
weakly bonded CO on Au surface, and then the DRIFT spectrum
was collected with 256 scans at a resolution of 4 cm .
ꢁ1
2
2
2.2.3. UV–vis and XPS studies
ꢁ1
pare a 1.04 mg ml impregnation solution. 30.2 ml Pd solution
and 1 g spherical SiO were co-added into a 100 ml flask and stir-
red at 25 °C for 24 h. The solvent was slowly evaporated under stir-
ring. The obtained solid was dried at 110 °C overnight and further
The UV–vis spectra were measured on a Shimadzu DUV-3700
spectrophotometer. The X-ray photoelectron spectroscopy (XPS)
measurements were taken on a Thermo-VG Scientific Escalab 250
2
spectrometer equipped with an Al anode (Al Ka = 1486.6 eV). The
calcined at 500 °C under 10% O
tion step at 250 °C under 10% H
2
in He for 3 h followed by a reduc-
binding energies were calibrated using the C 1s peak at 284.4 eV
as the internal standard [9]. All samples were pretreated in 10%
2
in Ar for 2 h to obtain the Pd/SiO
2

H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
61
in Ar at a flow rate of 20 ml min 1 for 1 h at 150 °C before UV–
ꢁ
H
2
2 2
the Au/SiO catalyst was 1.01% and the Pd loading in the Pd/SiO
vis and XPS measurements.
catalyst was 0.92%, both close to the calculated values which dem-
onstrate the high efficiencies of DP and impregnation processes.
2.3. Catalytic performance
3
.2. Precise synthesis of Au@Pd/SiO
2
bimetallic catalysts using selective
Solvent-free aerobic oxidation of benzyl alcohol using molecu-
Pd ALD
lar O
tions. 5 ml benzyl alcohol and 20 mg catalyst were added into a
5 ml three-necked glass flask equipped with a reflux condenser.
Prior to reaction, the system was first charged with O by bubbling
at a flow rate of 15 ml min for 20 min to
remove air. Under the continuous flow of O , the reactor was
2
was carried out in a batch-type reactor under mild condi-
Using the strategy of low-temperature selective metal ALD on
metals but not on oxide support, we developed recently, selective
Pd ALD was performed on the Au/SiO
2
2
2
catalyst by alternatively
ꢁ1
ultrahigh purity O
2
exposing to Pd(hfac) and hydrogen at 150 °C [34]. Here, hydrogen
2
2
was chosen as the reducing regent instead of formaldehyde to fur-
ther inhibit the growth on the support [34]. Indeed, we observed
that there was negligible Pd loading (0.04%, close to the detection
immersed into a silicon oil bath at 90 °C to initiate the reaction.
During the reaction, the mixture was vigorously stirred at a rate
of 1250 rpm to exclude any mass transfer limitation [60]. Finally,
the reaction products were analyzed using a Shimadzu GC-2014
gas chromatograph equipped with an Rtx-1 capillary column and
an auto-injector.
limit) even after 8 cycles of Pd ALD on the bare SiO
shown in Fig. 2a, which strongly indicates that Pd does not nucle-
ate on the SiO support under these conditions due to the inert
properties of Si–OH groups, consistent with our previous results
61]. However, the Pd loadings in the Au@xPd/SiO samples
2
support as
2
[
2
3
. Results and discussion
.1. Morphology of Au/SiO
Au/SiO and Pd/SiO catalysts were first prepared using the DP
(x = 1, 3, 5, 8, 10, 15, and 20) considerably increased along with
the number of Pd ALD cycles. Therefore, it is obviously that the
Pd must be deposited on the surface of Au nanoparticles of the
3
2 2
and Pd/SiO catalysts
2 2
Au/SiO catalyst, but not on the SiO support, demonstrating the
2
2
advantage of selective Pd ALD method in bimetallic catalyst syn-
thesis. It is noted that the increase of Pd loading showed a slight
lower rate in the first several ALD cycles than the one in the follow-
ing cycles, which could contribute to the increase in the surface
area of Au@Pd bimetallic nanoparticles via the particle size growth.
and impregnation method, respectively. TEM images and corre-
sponding metal particle size distribution histograms show that
metal nanoparticles were well dispersed in both Au/SiO
SiO samples with a rather narrow size distribution within 3–
nm, and the mean size calculated from TEM images were
.0 ± 0.6 nm and 4.7 ± 0.8 nm in Au/SiO and Pd/SiO , respectively
2
and Pd/
2
6
4
On Au surface, we speculate that the Pd (hfac)
2
precursor very
2
2
likely undergoes a dissociative chemisorption process via Eq. (1)
⁄
⁄
(Fig. 1). ICP-AES measurements showed that the Au loading in
by forming Pdhfac and hfac surface species (the asterisk
Fig. 1. Representative TEM images of (a) Au/SiO
2
, (b) Pd/SiO
2
-WI, and (c), (d) the corresponding particle size distributions.

6
2
H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
2 2
Fig. 2. (a) ICP-AES results of Pd loadings as a function of Pd ALD cycles. Solid circles, Pd ALD on the Au/SiO sample, solid squares, Pd ALD on the SiO support. (b) A linear
fitting of the calculated Pd shell thickness as a function of Pd ALD cycles.
designates a surface species, and h designates a nucleation site),
similar to our previous observation of exposing trimethylalumi-
num on noble metal surfaces [62,63]; next in the hydrogen reduc-
tion step, both hfac ligands were removed through Eq. (2) by
of the Au@Pd bimetallic catalysts characterized by other techniques
as follows.
⁄
forming Pd–H surface species and gaseous Hhfac product. The
3.3. HAADF-STEM characterization and EDX mapping analysis
deposited Pd adatoms on the Au nanoparticles might likely remain
isolated from each other or form very small Pd aggregates due to
the low surface coverage caused by the steric hindrance of hfac
Aberration-corrected HAADF-STEM measurements were further
carried out to investigate the atomic structural information of
these bimetallic samples. As shown in Fig. 3a–c, the as-prepared
Au@3Pd/SiO , Au@8Pd/SiO and Au@20Pd/SiO2 samples clearly
ligand. In the following Pd cycles, the Pd(hfac)
2
precursor can
⁄
either react with the pre-formed Pd–H species by forming Pd–
2
2,
⁄
Pd-hfac and gaseous Hhfac product (Eq. (3)) or react with the
show the gradual particle size growth with Pd ALD cycles, and
the growth rate, determined from the slope of the linear fitting
of measured mean particle size as a function of Pd ALD cycles,
was about 0.14 nm per cycle (Fig. 3d). At higher magnification,
all these samples demonstrated a clear Au core–Pd shell structure,
where the darker contrast is due to the atomically lighter Pd and
the brighter contrast is attributed to the heavier Au (Fig. 3e–g,
and i). Here, Au nanoparticles consisted of different orientations
wherein the lattice distances of 0.235–0.237 nm and 0.205 nm
are assigned to the (111) and (200) planes of Au, respectively.
The Pd shell in a darker contrast rather uniformly coated on the
Au nanoparticles, and the thickness of Pd shells measured from
the HAADF-STEM images at different locations of these samples
re-exposed Au surface via Eq. (1); hydrogen will again remove all
the hfac ligand in the reduction step (Eq. (4)). As a consequence,
the amount of Pd on the Au nanoparticles can be precisely con-
trolled by varying the number of ALD cycles.
ꢂ
ꢂ
2
à þ PdðhfacÞ ðgÞ ! Pdhfac þ hfac
ð1Þ
ð2Þ
ð3Þ
ð4Þ
2
ꢂ
ꢂ
ꢂ
Pdhfac þ hfac þ 3=2H
2
ðgÞ ! Pd—H þ Ã þ 2HhfacðgÞ
ꢂ
ꢂ
Pd—H þ PdðhfacÞ ðgÞ ! Pd—Pd—hfac þ HhfacðgÞ
2
ꢂ
ꢂ
Pd—Pd—hfac þ H
2
ðgÞ ! Pd—Pd—H þ HhfacðgÞ
Given that low-temperature selective deposition of Pd on Au nano-
particles will likely form Au@Pd core–shell structured bimetallic
nanoparticles, we estimated the Pd shell thickness after different
Pd ALD cycles using the Au ‘‘magic clusters’’ model [9,64]; therein,
a Au nanoparticle was treated as a central Au atom surrounded by
closed shells of identical Au atoms. In this model, a 4 nm Au nano-
particle contains seven shells of Au atoms (total number: 1415) [9];
Pd deposited on a 4 nm Au core was approximately considered to
form the other shells of the seven-shell Au core. According to the
molar ratio of Au/Pd determined by the ICP-AES measurements, this
model provides a useful way to estimate the Pd surface coverage in
various Au@Pd bimetallic catalysts with different Pd loadings. Given
that the thickness of one full monolayer Pd is about 0.227 nm (lat-
tice distance of Pd (111) planes) [65], thus the thickness of Pd shells
in these bimetallic catalysts is obtained. The details of the calcula-
tion results are shown in Table 1. To clearly illustrate the trend of
the Pd shell thickness as a function of ALD cycles, we further plotted
the calculated results as shown in Fig. 2b. It is obviously that the
estimated Pd shell thickness rather linearly increased at a rate of
were about 0.33 nm, 0.77 nm, and 1.61 nm for Au@3Pd/SiO
2
,
Au@8Pd/SiO2, and Au@20Pd/SiO , respectively. The deposition rate,
2
determined from the slope of the linear fitting of measured thick-
ness of Pd shells, as a function of Pd ALD cycles, was 0.08 nm per Pd
ALD cycle (or 0.35 Pd monolayer per cycle), as shown in Fig. 3h,
which is very close to the expected value, about half of particle size
growth rate (0.14 nm per cycle, Fig. 3d). Despite the inevitable
errors between the real Au nanoparticles and the ‘‘magic cluster’’
model, or the measuring errors from the STEM images, the deposi-
tion rate determined from the STEM images consists very well with
the one obtained from the ICP-AES results using the ‘‘magic clus-
ter’’ model, implying the rationality of our estimation about the
Pd shell thicknesses. The core–shell structure was further con-
firmed by the elemental mapping of Au La and Pd La on the
Au@8Pd/SiO sample using EDX (Fig. 3i–l). Even though Au has
2
lower surface energy than Pd [66,67], it is not surprising that we
could obtain Au@Pd core–shell nanoparticles by selective Pd ALD
at 150 °C, because structural transformation from Au@Pd core–
shell to Au–Pd alloy often occurs after heat treatment at about
300 °C or even higher [68–72]. As a consequence, low-temperature
selective Pd ALD appears to be possible for synthesis of Au@Pd
core–shell bimetallic nanoparticles with precisely tunable Pd shell
thickness and compositions by varying the number of Pd ALD
cycles.
0
.09 nm per Pd ALD cycle (or 0.39 Pd monolayer per cycle). Even
though one might argue that the variation of size and structures
of Au nanoparticle and the contact area between the particles and
support would induce uncertainties to the estimations, we believe
these approximate results would help us understand the structures

H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
63
Table 1
2
Pd loadings in a series of Au@xPd/SiO bimetallic catalysts with different Pd ALD cycles and the corresponding thicknesses of Pd shells calculated according to the ‘‘magic clusters
model’’ based on a 4 nm Au core, which contains 1415 Au atoms.
Pd shell number (n) Numbers of Pd atoms in shell (natom)a Pd ALD cycles Pd loading (%) Pd/Au molar ratio Pd coverage (ML) Thickness of Pd shell (nm)
1
2
3
4
5
6
7
8
642
812
1
3
5
0.17
0.40
0.57
0.90
1.34
3.0
0.31
0.73
1.05
1.65
2.47
5.52
7.36
0.68
1.5
2.0
2.9
3.8
6.5
7.7
0.16
0.34
0.46
0.66
0.87
1.48
1.75
1002
1212
1442
1692
1962
2252
8
10
15
20
4.0
a
Equation: natom = 10 * (n + 7)2 + 2.
3
.4. DRIFT of CO chemisorption study
well with the ICP-AES and STEM results. Note that the Au SPR
bands demonstrated a slight blue shift as increasing the Pd ALD
cycles, which is contributed to the electron density increase in
Au caused by the Pd shells [8,79].
The detailed surface structures and compositions of Au@xPd/
SiO core–shell bimetallic catalysts were further analyzed by
2
DRIFTS CO chemisorption, because IR studies of CO chemisorption
have been widely used to characterize the surface composition of
bimetallic nanoparticles [65,68,73–76]. On Au catalyst, CO often
weakly bonds to the low-coordinated sites of Au nanoparticles
and can be fairly easily purged away with inert gas at room tem-
perature (not shown here) [77]. Thus, the CO chemisorption peaks
observed on the Au@Pd bimetallic catalysts after 30 min He purg-
ing should attribute to chemisorbed CO on Pd surfaces. Fig. 4 illus-
trates the DRIFT spectra of CO chemisorption on various bimetallic
On the other hand, the disturbance of Pd shell 3d electrons by
the Au core as a function of the Pd shell thickness was investigated
using XPS measurements. As shown in Fig. 6a, the Pd 3d3/2 and
3d5/2 binding energies were 340.4 and 335.2 eV on the Pd/SiO -
2
WI sample, respectively, which are assigned to zero-valence Pd
[9]. Focusing on the Pd 3d5/2 peak, we observed it was 334.7 eV
on the Au@1Pd/SiO
binding energies as increasing Pd ALD cycles. On the Au@20Pd/
SiO sample, the Pd 3d5/2 binding energy was 335.1 eV very close
to the Pd/SiO -WI sample [9,65]. The 3d5/2 binding energy shifts
2
sample, and it gradually shifted to higher
2
catalysts recorded at room temperature. On the Au@1Pd/SiO
2
cat-
2
ꢁ1
alyst, two peaks at 2082 and 1980 cm were observed, which are
assigned to linear and twofold-bridged CO on Pd, respectively. The
strong linear CO together with the much weaker twofold-bridged
CO strongly indicates that Pd can be majorly isolated Pd adatoms
or tiny aggregates [73,76]. With an increase of Pd ALD cycles from
on these bimetallic catalysts from zero-valence Pd (335.2 eV) were
further emphasized as a function of Pd ALD cycles as shown in
Fig. 6b. It is obvious that the thinner of the Pd shell, the more neg-
ative shift of the Pd 3d5/2 binding energy, strongly indicating the
modification of Pd electronic properties by Au tightly depends on
the Pd shells thickness [9,74]. The negative shift implies the Pd
3d level draws electrons from Au in the bimetallic system, even
though electrons intuitively transfer from Pd to Au, suggested by
our UV–vis results, which is due to the higher electronegativity
of Au. This phenomenon could be interpreted with the electron
transfer model that in a AuPd bimetallic system, Au usually gains
s, p electrons and loses d electrons while Pd loses s and p electrons
but gains d electrons [4,8,68,80]. Therefore, Pd gaining d electrons
from Au explains the negative shift of Pd 3d binding energy in
1
to 10, the intensity of linear CO chemisorption band gradually
ꢁ1
decreased along with a blue shift from 2082 cm
to about
ꢁ
1
2
093 cm , while the twofold-bridged CO developed aggressively.
The blue shift of linear CO is due to the increasing dipole–dipole
coupling effect, implying the aggregation of Pd atoms [73]. Mean-
ꢁ1
while, a new broad shoulder peak at around 1930 cm appeared
on the Au@8Pd/SiO and Au@10Pd/SiO samples, which is assigned
2
2
to bridge-bonded CO on Pd(111) facet, implying the formation of
continuous Pd islands or films in (111) orientation [65,73,75]. This
trend in DRIFT spectra of CO chemisorption clearly indicates the
gradual evolution of Pd species on Au nanoparticle surfaces while
increasing the Pd ALD cycles: from isolated Pd adatoms or very
small aggregates to large ensembles, and to continuous islands or
films [73,74]. Thus, DRIFT CO chemisorption results provide
another strong evidence of precise control over the surface struc-
ture of AuPd bimetallic catalysts through the facile low-tempera-
ture selective Pd ALD.
2
Au@xPd/SiO bimetallic samples. Moreover, this electron transfer
model between Au and Pd indicates a stronger interaction of Au–
Pd than Au–Au or Pd–Pd, which favors the formation of isolated
Pd adatoms on Au at low Pd coverages [4,42,73], consistent with
our observation in DRIFT CO chemisorption on Au@1Pd/SiO
(Fig. 4). Such electronic interaction between Au and Pd persists
on the Au@20cPd/SiO sample, showing a small negative shift
2
2
(ꢃ0.1 eV) of Pd 3d5/2 binding energy, even though the Pd shell
thickness on this sample was about 1.61 nm.
3.5. UV–vis and XPS studies
3.6. Catalytic performance
UV–vis and XPS characterizations were employed to further
understand the electronic properties of the well-controlled
Au@xPd/SiO bimetallic catalysts. As shown in Fig. 5, the UV–vis
spectrum of the Au/SiO catalyst showed a characteristic surface
plasma resonance (SPR) band centered at 529 nm [8,74,78], while
the Pd/SiO -WI catalyst does not show any SPR band between
00 and 800 nm. Deposition of Pd onto Au nanoparticles for even
one Pd ALD cycle (Au@1Pd/SiO ) led to significantly decline and
broaden the Au SPR band, indicating that the Au surface was cov-
ered by Pd to form Au@Pd bimetallic nanoparticles [8,78]. The Au
SPR band was gradually obscured while increasing the Pd ALD
cycles, implying the gradual thickening of Pd shells, consistent very
The catalytic performance of monometallic and bimetallic cata-
2
lysts was evaluated in solvent-free selective oxidation of benzyl
alcohol at 90 °C using molecular O as oxidant [60]. As shown in
Fig. 7, the Au/SiO catalyst was totally inactive, which is likely
due to the large size of the Au particles or the inert property of
SiO support [37]. The Pd/SiO -WI catalyst also showed a poor
2
2
2
2
2
2
2
2
activity, and the conversion was only 35% after 6 h reaction, even
though the selectivity to benzaldehyde was about 94%
[60,65,81,82]. As a control experiment, the 8c-Pd/SiO ALD catalyst
2
was also tested and it did not show any catalytic activity as what
we expected, again providing strong evidence that Pd cannot

6
4
H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
Fig. 3. Aberration-corrected HAADF-STEM images of different bimetallic catalysts at low and high magnifications: (a) and (e) Au@3Pd/SiO
2 2
; (b) and (f) Au@8Pd/SiO ; (c) and
(
g) Au@20Pd/SiO . (d) The growth of Au@Pd particle size as a function of Pd ALD cycles. (h) The Pd shell thickness as a function of Pd ALD cycles. (i) A lower magnification
2
2
HAADF-STEM image of Au@8Pd/SiO , and corresponding EDX mapping images: (j) Au La1 and (k) Pd La1 signals, and (l) the reconstructed Au@Pd bimetallic composition
image.
deposit on the SiO
the other hand, remarkable increases of catalytic activities were
observed on all the Au@xPd/SiO (x = 1, 3, 5, 8, 10, 15, 20) bimetal-
lic catalysts. Among them, high conversion of benzyl alcohol about
2
support under current Pd ALD conditions. On
90% after 6 h reaction was obtained on Au@8Pd/SiO
2
, Au@10Pd/
SiO , Au@15Pd/SiO , and Au@20Pd/SiO catalysts.
2
2
2
2
Table 2 summarizes the conversion and the product selectivity
on all these catalysts after the initial 1 h and 6 h reaction, respec-

H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
65
Fig. 7. Conversions of benzyl alcohol versus reaction time over Au/SiO
2 2
, Pd/SiO -
WI, 8c-Pd/SiO , and various Au@xPd/SiO catalysts. Reaction conditions: benzyl
2
2
ꢁ
1
Fig. 4. DRIFT spectra of CO chemisorption at 298 K on various Au@xPd/SiO
2
2
alcohol, 5 ml; catalyst, 20 mg; O , 15 ml min ; stirring speed, 1250 rpm; temper-
catalysts at the CO saturation coverage.
ature, 90 °C.
high as 35%. Similar trend was also observed in the low conversion
range, when we compare the benzaldehyde selectivity at the same
conversion. For example, at about 55% conversion, the benzalde-
hyde selectivity was about 93% and 67% on Au@5Pd/SiO
2
and
Au@15Pd/SiO , respectively. Such observations are consistent with
2
the previous studies that toluene is generated from the dispropor-
tionation of benzyl alcohol via either a hydrogen transfer or an
oxygen transfer mediated by surface-enriched Pd over a Pd-rich
shell/Au-rich core structure [65]. Nevertheless, the benzaldehyde
2
selectivity all increased to near 90% on all the Au@xPd/SiO bime-
tallic catalysts as preceding the reaction for 6 h, consistent with the
previous studies that the benzaldehyde selectivity is independent
of Au:Pd ratio [49,65]. Meanwhile, the toluene formation was sig-
nificantly inhibited to only a few percent. Corma and coworkers
suggested that the Au–H and Pd–H species that formed on sup-
ported Au and Pd catalysts are important in the toluene formation
[
83]. Therefore, we speculate that the accumulation of surface oxy-
2 2 2
Fig. 5. UV–vis spectra of Au/SiO , Pd/SiO -WI, and various Au@xPd/SiO catalysts.
gen on the Pd shells with reaction time facilitates the removal of
Au–H and Pd–H surface species, thus might be the main reason
for the effective inhibition of toluene formation pathway.
To gain the information of the intrinsic reaction rate over the Pd
shell thickness, we calculated both the turnover frequency (TOF)
and specific activity after 1 h reaction based on the number of sur-
face Pd sites estimated using the ‘‘magic cluster’’ model and the
tively. After the first 1 h reaction, the benzaldehyde selectivity was
rather low and toluene was the main by-product for all the bime-
tallic catalysts. The benzaldehyde selectivity decreased as increas-
ing Pd ALD cycles, which was only 64% at a conversion of 45% on
2
the Au@20Pd/SiO catalyst, while the toluene selectivity was as
Fig. 6. (a) XPS spectra in the Pd 3d region for a variety of catalysts. Here, the spectrum of Pd/SiO
energies shift for a variety of bimetallic catalysts compared to the Pd binding energy in the Pd/SiO
2
-WI sample was amplified by 5 times for a better comparison. (b) Binding
-WI as a function of Pd ALD cycles.
2

6
6
H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
Table 2
Catalytic performance comparisons in solvent-free selective oxidation of benzyl alcohol on series catalysts after 1 h and 6 h reaction. Reaction conditions: benzyl alcohol, 5 ml;
ꢁ
1
catalyst, 20 mg; O
2
, 15 ml min ; stirring speed, 1250 rpm; temperature, 90 °C.
Samples
Time (h)
Conversion (%)
Selectivity (%)
Benzaldehyde
Toluene
Benzoic acid
Benzyl benzoate
Au/SiO
Pd/SiO
c-Pd/SiO
2
1
6
1
6
1
6
1
6
1
6
1
6
1
6
1
6
1
6
<0.5
<0.5
3.4
35
<0.5
<0.5
2.6
8.8
15
54
35
91
33
NA
NA
93
94
NA
NA
92
91
80
93
73
87
74
91
67
88
64
88
NA
NA
0.6
0.5
NA
NA
8.5
7.1
18
2.2
26
4.6
24
NA
NA
2.2
1.9
NA
NA
NA
0.8
1.0
2.4
0.8
4.3
0.8
3.2
0.4
1.5
0.2
2.2
NA
NA
4.7
4.1
NA
NA
NA
1.0
1.0
2.0
0.8
3.7
1.2
5.0
1.8
5.5
0.8
3.0
2
-WI
8
2
Au@1Pd/SiO
Au@5Pd/SiO
Au@8Pd/SiO
2
2
2
Au@10Pd/SiO
Au@15Pd/SiO
Au@20Pd/SiO
2
2
2
88
55
90
45
1
31
5.2
35
7.0
94
total amount of Pd, respectively. Here, we did not take Au into
account, since Au is almost inactive and Pd is widely considered
as the active sites [37,73]. Interestingly, the histograms of TOF
and specific activity both demonstrate a volcano-like trend that
they gradually increased as Pd shell thickness (or Pd ALD cycles)
effect [2,37,48]. Moreover, the volcano-like trend of catalytic activ-
ities as a function of the Pd shell thickness clearly indicates that
aerobic oxidation of benzyl alcohol is a structure-sensitive reaction
[2,47,65,82]. A similar volcano-like behavior was also observed on
the Au–Pd alloy catalysts [53]. However, when Au@8Pd/SiO
2
was
ꢁ1
ꢁ1
and reached maximums at 27,600 h
and 9800 h
on the
annealed at 550 °C for 30 min in 10% H in He to form Au–Pd alloy
2
Au@8Pd/SiO
2
bimetallic catalyst (Pd shell thickness, ꢃ0.8 nm),
nanoparticles [85] (indicted by the observation of a large decease
of CO chemisorption peaks, especially the bridge-bonded CO in
the DRIFT spectra, which is not shown here), the specific activity
decreased by about 60–70%, even though the TEM measurements
showed the AuPd particle size did not change after high-tempera-
ture reduction.
A generally accepted mechanism of oxidation of benzyl alcohol
over Pd-contained catalysts is first to form the Pd alcoholate inter-
mediate through the interaction of the O–H bond of the alcohol
respectively; then, they both decreased as further increasing Pd
shell thickness (Fig. 8), which is different from the values reported
previously, wherein the maximum activity was observed on
Au@Pd core–shell catalysts with a Pd shell thickness of one atomic
layer [84] and 2.2 nm [56], respectively. However, we noticed the
leaching of metal nanoparticles to a certain extent during the recy-
cling tests on the Au@8Pd/SiO
interaction between metal particles and the ‘‘inert’’ SiO
2
bimetallic catalyst due to the weak
support
2
and lack of catalyst stability enhancement by high-temperature
pretreatment [37]. Nevertheless, the superior catalytic perfor-
mance of Au@Pd core–shell bimetallic catalysts over the Au and
Pd monometallic catalysts can be addressed to the synergistic
with
a
Pd site, followed by the b-hydride elimination
[2,65,78,86]. Nevertheless, the role of the terrace and low-coordi-
nated (edges and corners) Pd sites is still in the debate. Kaneda
et al. [86] suggested that the low-coordinated Pd atoms are the cat-
alytic active site for the formation of Pd alcoholate and the b-
hydride elimination, wherein the later step is the rate-limiting
step. On the other hand, Baiker et al. [87] proposed that the dehy-
drogenation of alcohol to carbonyl product could occur virtually on
2 3
all the surface Pd atoms of Pd/Al O catalyst. Wang et al. [82] also
observed that TOF depends significantly on the mean size of Pd
particles with a maximum at the medium mean particle size
(
3.6–4.3 nm), thus speculated that benzyl alcohol might first
chemisorb on the Pd terrace sites via the interaction of the delocal-
ized -bond of benzene ring and then probably forms Pd alcoholate
p
species, followed by the b-hydride elimination on the nearby edge
or corner Pd sites.
In our studies, we observed that isolated Pd adatoms or small
aggregates on the Au@1Pd/SiO
ently lower activity than the continuous Pd islands or shells in
111) orientation on the Au@Pd/SiO bimetallic catalyst. This
2
bimetallic catalyst had an appar-
(
2
result is in line with the geometric effect suggested by Wang
et al. [82] that both terrace and low-coordinated Pd atoms play
important role in this reaction, and also explains well the lower
activity of Au–Pd alloy compared to Au@Pd core–shell bimetallic
nanoparticles, since significantly smaller amount of continuous
Pd islands are present on the alloy surface. On the other hand, fur-
ther increasing Pd shell thicker than 0.8 nm (>8 Pd ALD cycles)
resulted in a considerable decrease of TOF and specific activity,
Fig. 8. Initial activities of different catalysts in benzyl alcohol oxidation after 1 h
reaction. Specific activity (black column) normalized to the total Pd content; TOF
(
red column) normalized to the surface Pd content. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of
this article.)

H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
67
which is probably mainly due to the decline of electronic effects,
since the Pd 3d binding energy shift is much less pronounced over
the Au@xPd/SiO (x > 8) bimetallic catalysts (Fig. 6). As a conse-
quence, the activity appears to be optimized by precisely tuning
the Pd shell thickness of Au@Pd core–shell bimetallic catalysts,
through the maximization of the synergistic effect via both ensem-
ble and electronic promotion.
[8] A. Sárkány, O. Geszti, G. Sáfrán, Appl. Catal. A: Gen. 350 (2008) 157–163.
[
9] M.O. Nutt, K.N. Heck, P. Alvarez, M.S. Wong, Appl. Catal. B: Environ. 69 (2006)
115–125.
2
[
10] I. Srnova-Sloufova, B. Vlckova, Z. Bastl, T.L. Hasslett, Langmuir 20 (2004) 3407–
3415.
11] M. Harada, K. Asakura, N. Toshima, J. Phys. Chem. 97 (1993) 5103–5114.
12] H. Yu, P.C. Gibbons, K.F. Kelton, W.E. Buhro, J. Am. Chem. Soc. 123 (2001)
[
[
9
198–9199.
[13] N. Toshima, Y. Shiraishi, A. Shiotsuki, D. Ikenaga, Y. Wang, Eur. Phys. J. D 16
2001) 209–212.
14] J. Turkevic, G. Kim, Science 169 (1970) 873.
[15] W.R. Lee, M.G. Kim, J.R. Choi, J.I. Park, S.J. Ko, S.J. Oh, J. Cheon, J. Am. Chem. Soc.
27 (2005) 16090–16097.
(
[
4
. Conclusions
1
[
[
16] A. Sarkany, P. Hargittai, A. Horvath, Top. Catal. 46 (2007) 121–128.
17] P. Strasser, S. Koh, T. Anniyev, J. Greeley, K. More, C.F. Yu, Z.C. Liu, S. Kaya, D.
Nordlund, H. Ogasawara, M.F. Toney, A. Nilsson, Nat. Chem. 2 (2010) 454–460.
18] S.F. Xie, S.I. Choi, N. Lu, L.T. Roling, J.A. Herron, L. Zhang, J. Park, J.G. Wang, M.J.
Kim, Z.X. Xie, M. Mavrikakis, Y.N. Xia, Nano Lett. 14 (2014) 3570–3576.
19] S.M. George, Chem. Rev. 110 (2010) 111–131.
In conclusion, we have successfully demonstrated that low-
temperature selective Pd ALD only on Au nanoparticles, but not
on the SiO support could be a facile method to exclusively synthe-
size well-dispersed Au@Pd core–shell bimetallic catalysts sup-
ported on SiO , while avoiding the formation of monometallic
[
[
2
2
[20] T. Suntola, J. Hyvarinen, Annu. Rev. Mater. Sci. 15 (1985) 177–195.
[
[
[
21] R.L. Puurunen, J. Appl. Phys. 97 (2005) 121301.
nanoparticles. The thickness of Pd shell can be precisely tuned at
a rate of ꢃ0.08 nm per Pd ALD cycle. ICP-AES and aberration-cor-
rected HAADF-STEM measurements were performed to confirm
the selective Pd deposition, and the formation of Au@Pd core–shell
structured bimetallic catalysts, respectively; DRIFTS CO chemi-
sorption, and UV–vis and XPS measurements were further
employed to reveal the evolution of Pd atoms arrangement on Au
nanoparticle surface and the electronic properties of the bimetallic
nanoparticles as a function of Pd ALD cycles (or Pd shell thickness),
respectively. In solvent-free aerobic oxidation of benzyl alcohol, we
discovered that the catalytic activities of resulted Au@Pd core–
shell bimetallic catalysts showed a clear volcano-like trend as a
function of Pd shell thickness, wherein a Au@Pd core–shell catalyst
with a Pd shell thickness of 0.6–0.8 nm showed a maximum TOF of
22] B.S. Lim, A. Rahtu, R.G. Gordon, Nat. Mater. 2 (2003) 749–754.
23] J.S. King, A. Wittstock, J. Biener, S.O. Kucheyev, Y.M. Wang, T.F. Baumann, S.K.
Giri, A.V. Hamza, M. Baeumer, S.F. Bent, Nano Lett. 8 (2008) 2405–2409.
[24] A.C. Johansson, J.V. Larsen, M.A. Verheijen, K.B. Haugshoj, H.F. Clausen, W.M.M.
Kessels, L.H. Christensen, E.V. Thomsen, J. Catal. 311 (2014) 481–486.
[
25] H. Feng, J.W. Elam, J.A. Libera, W. Setthapun, P.C. Stair, Chem. Mater. 22 (2010)
133–3142.
3
[26] T.D. Gould, A.M. Lubers, B.T. Neltner, J.V. Carrier, A.W. Weimer, J.L. Falconer,
J.W. Medlin, J. Catal. 303 (2013) 9–15.
[
27] C.P. Canlas, J.L. Lu, N.A. Ray, N.A. Grosso-Giordano, S. Lee, J.W. Elam, R.E.
Winans, R.P. Van Duyne, P.C. Stair, J.M. Notestein, Nat. Chem. 4 (2012) 1030–
1036.
28] Y.J. Pagan-Torres, J.M.R. Gallo, D. Wang, H.N. Pham, J.A. Libera, C.L. Marshall,
J.W. Elam, A.K. Datye, J.A. Dumesic, ACS Catal. 1 (2011) 1234–1245.
29] J. Lu, B. Fu, M.C. Kung, G. Xiao, J.W. Elam, H.H. Kung, P.C. Stair, Science 335
(2012) 1205–1208.
[
[
[
30] J.L. Lu, P.C. Stair, Angew. Chem. Int. Ed. 49 (2010) 2547–2551.
ꢁ
1
ꢁ1
[31] J.L. Lu, J.W. Elam, P.C. Stair, Acc. Chem. Res. 46 (2013) 1806–1815.
2
7,600 h (or a specific activity of 9800 h ) due to the synergistic
[
[
[
32] A.C. Alba-Rubio, B.J. O’Neill, F.Y. Shi, C. Akatay, C. Canlas, T. Li, R. Winans, J.W.
Elam, E.A. Stach, P.M. Voyles, J.A. Dumesic, ACS Catal. 4 (2014) 1554–1557.
33] T.J. Schwartz, B.J. O’Neill, B.H. Shanks, J.A. Dumesic, ACS Catal. 4 (2014) 2060–
effect via both ensemble and electronic promotion. Finally, it is
worthy to note that synthesizing core–shell structured bimetallic
nanoparticles using a combined wet-chemistry and gas phase
ALD method could further extend the application of ALD in cataly-
sis. Moreover, this low-temperature selective metal ALD on the
metal particle surface of a monometallic catalyst could be general
for other metals and supports and thus opens great opportunities
for atomically precise synthesis of various supported bimetallic
catalysts with core–shell structures for broad applications.
2069.
34] J. Lu, K.B. Low, Y. Lei, J.A. Libera, A. Nicholls, P.C. Stair, J.W. Elam, Nat. Commun.
5 (2014).
[35] M. Haruta, Faraday Discuss. 152 (2011) 11–32.
[
[
36] M. Valden, X. Lai, D.W. Goodman, Science 281 (1998) 1647–1650.
37] D.I. Enache, J.K. Edwards, P. Landon, B. Solsona-Espriu, A.F. Carley, A.A. Herzing,
M. Watanabe, C.J. Kiely, D.W. Knight, G.J. Hutchings, Science 311 (2006) 362–
365.
38] K. Qian, W. Huang, Catal. Today 164 (2011) 320–324.
39] J. Xu, T. White, P. Li, C.H. He, J.G. Yu, W.K. Yuan, Y.F. Han, J. Am. Chem. Soc. 132
[
[
(
2010) 10398–10406.
[
[
40] J. Edwards, B. Solsona, P. Landon, A. Carley, A. Herzing, C. Kiely, G. Hutchings, J.
Catal. 236 (2005) 69–79.
41] J.K. Edwards, B. Solsona, P. Landon, A.F. Carley, A. Herzing, M. Watanabe, C.J.
Kiely, G.J. Hutchings, J. Mater. Chem. 15 (2005) 4595.
42] M. Chen, D. Kumar, C.W. Yi, D.W. Goodman, Science 310 (2005) 291–293.
43] W.Y. Yu, G.M. Mullen, D.W. Flaherty, C.B. Mullins, J. Am. Chem. Soc. 136 (2014)
11070–11078.
44] Y. Huang, X. Zhou, M. Yin, C. Liu, W. Xing, Chem. Mater. 22 (2010) 5122–5128.
45] L. Kesavan, R. Tiruvalam, M.H. Ab Rahim, M.I. bin Saiman, D.I. Enache, R.L.
Jenkins, N. Dimitratos, J.A. Lopez-Sanchez, S.H. Taylor, D.W. Knight, C.J. Kiely,
G.J. Hutchings, Science 331 (2011) 195–199.
Conflict of interest
The authors declare no competing financial interest.
[
[
Acknowledgments
[
[
This work was supported by National Natural Science Founda-
tion of China (21473169), the National Science Foundation for
Young Scientists of China (51402283), the Fundamental Research
Funds for the Central Universities (WK2060030014), the Recruit-
ment Program of Global Youth Experts, the Scientific Research
Foundation for the Returned Overseas Chinese Scholars, and Uni-
versity of Science and Technology of China for the startup funds.
[46] W. Hou, N. Dehm, R. Scott, J. Catal. 253 (2008) 22–27.
[
[
[
[
[
[
[
47] X. Yang, C. Huang, Z. Fu, H. Song, S. Liao, Y. Su, L. Du, X. Li, Appl. Catal. B:
Environ. 140–141 (2013) 419–425.
48] C.Y. Ma, B.J. Dou, J.J. Li, J. Cheng, Q. Hu, Z.P. Hao, S.Z. Qiao, Appl. Catal. B:
Environ. 92 (2009) 202–208.
49] A. Villa, N. Janjic, P. Spontoni, D. Wang, D.S. Su, L. Prati, Appl. Catal. A: Gen. 364
(
2009) 221–228.
50] Y. Chen, H. Lim, Q. Tang, Y. Gao, T. Sun, Q. Yan, Y. Yang, Appl. Catal. A: Gen. 380
2010) 55–65.
(
References
51] M. Sankar, N. Dimitratos, P.J. Miedziak, P.P. Wells, C.J. Kiely, G.J. Hutchings,
Chem. Soc. Rev. 41 (2012) 8099–8139.
52] R.C. Tiruvalam, J.C. Pritchard, N. Dimitratos, J.A. Lopez-Sanchez, J.K. Edwards,
A.F. Carley, G.J. Hutchings, C.J. Kiely, Faraday Discuss. 152 (2011) 63–86.
53] J. Pritchard, L. Kesavan, M. Piccinini, Q. He, R. Tiruvalam, N. Dimitratos, J.A.
Lopez-Sanchez, A.F. Carley, J.K. Edwards, C.J. Kiely, G.J. Hutchings, Langmuir 26
[
[
[
1] F. Maroun, F. Ozanam, O.M. Magnussen, R.J. Behm, Science 293 (2001) 1811–
814.
2] Z. Guo, B. Liu, Q. Zhang, W. Deng, Y. Wang, Y. Yang, Chem. Soc. Rev. 43 (2014)
480–3524.
3] J. Greeley, T.F. Jaramillo, J. Bonde, I.B. Chorkendorff, J.K. Norskov, Nat. Mater. 5
2006) 909–913.
1
3
(
2010) 16568–16577.
(
[
[
54] S. Meenakshisundaram, E. Nowicka, P.J. Miedziak, G.L. Brett, R.L. Jenkins, N.
Dimitratos, S.H. Taylor, D.W. Knight, D. Bethell, G.J. Hutchings, Faraday
Discuss. 145 (2010) 341–356.
55] N. Dimitratos, J.A. Lopez-Sanchez, J.M. Anthonykutty, G. Brett, A.F. Carley, R.C.
Tiruvalam, A.A. Herzing, C.J. Kiely, D.W. Knight, G.J. Hutchings, Phys. Chem.
Chem. Phys. 11 (2009) 4952–4961.
[
[
4] F. Gao, D.W. Goodman, Chem. Soc. Rev. 41 (2012) 8009–8020.
5] K. Sasaki, H. Naohara, Y. Cai, Y.M. Choi, P. Liu, M.B. Vukmirovic, J.X. Wang, R.R.
Adzic, Angew. Chem. Int. Ed. 49 (2010) 8602–8607.
[
[
6] C.J. Serpell, J. Cookson, D. Ozkaya, P.D. Beer, Nat. Chem. 3 (2011) 478–483.
7] B. Lim, H. Kobayashi, T. Yu, J.G. Wang, M.J. Kim, Z.Y. Li, M. Rycenga, Y.N. Xia, J.
Am. Chem. Soc. 132 (2010) 2506–2507.

6
8
H. Wang et al. / Journal of Catalysis 324 (2015) 59–68
[
[
56] A.M. Henning, J. Watt, P.J. Miedziak, S. Cheong, M. Santonastaso, M.H. Song, Y.
Takeda, A.I. Kirkland, S.H. Taylor, R.D. Tilley, Angew. Chem. Int. Ed. 52 (2013)
[71] D. Jana, A. Dandapat, G. De, J. Phys. Chem. C 113 (2009) 9101–9107.
[72] S.J. Mejia-Rosales, C. Fernandez-Navarro, E. Perez-Tijerina, J.M. Montejano-
Carrizales, M. Jose-Yacaman, J. Phys. Chem. B 110 (2006) 12884–12889.
[73] L. Zhang, A. Wang, J.T. Miller, X. Liu, X. Yang, W. Wang, L. Li, Y. Huang, C.-Y.
Mou, T. Zhang, ACS Catal. 4 (2014) 1546–1553.
1
477–1480.
57] K. Qian, L.F. Luo, H.Z. Bao, Q. Hua, Z.Q. Jiang, W.X. Huang, Catal. Sci. Technol. 3
2013) 679–687.
(
[
[
58] E. Bourgeat-Lami, J. Lang, J. Colloid Interface Sci. 210 (1999) 281–289.
59] B. Li, H. Li, W.-Z. Weng, Q. Zhang, C.-J. Huang, H.-L. Wan, Fuel 103 (2013) 1032–
[74] A. Wang, X.Y. Liu, C.-Y. Mou, T. Zhang, J. Catal. 308 (2013) 258–271.
[75] T. Lear, R. Marshall, J.A. Lopez-Sanchez, S.D. Jackson, T.M. Klapotke, M. Baumer,
G. Rupprechter, H.J. Freund, D. Lennon, J. Chem. Phys. 123 (2005) 174706.
[76] N.E. Kolli, L. Delannoy, C. Louis, J. Catal. 297 (2013) 79–92.
[77] J.D. Grunwaldt, M. Maciejewski, O.S. Becker, P. Fabrizioli, A. Baiker, J. Catal. 186
(1999) 458–469.
[78] W. Zhou, J.Y. Lee, Electrochem. Commun. 9 (2007) 1725–1729.
[79] A.V. Gaikwad, P. Verschuren, E. Eiser, G. Rothenberg, J. Phys. Chem. B 110
(2006) 17437–17443.
1
038.
[
60] D.I. Enache, D. Barker, J.K. Edwards, S.H. Taylor, D.W. Knight, A.F. Carley, G.J.
Hutchings, Catal. Today 122 (2007) 407–411.
61] J. Lu, P.C. Stair, Langmuir 26 (2010) 16486–16495.
62] J. Lu, B. Liu, J.P. Greeley, Z.X. Feng, J.A. Libera, Y. Lei, M.J. Bedzyk, P.C. Stair, J.W.
Elam, Chem. Mater. 24 (2012) 2047–2055.
63] J. Lu, L. Bin, N.P. Guisinger, P.C. Stair, J.P. Greeley, J.W. Elam, Chem. Mater. 26
[
[
[
(
2014) 6752–6761.
[80] J.A. Rodriguez, Surf. Sci. Rep. 24 (1996) 225–287.
[
[
64] L.N. Lewis, Chem. Rev. 93 (1993) 2693–2730.
65] Y. Chen, H. Wang, C.-J. Liu, Z. Zeng, H. Zhang, C. Zhou, X. Jia, Y. Yang, J. Catal.
[81] T. Mallat, A. Baiker, Chem. Rev. 104 (2004) 3037–3058.
[82] F. Li, Q. Zhang, Y. Wang, Appl. Catal. A: Gen. 334 (2008) 217–226.
[83] A. Abad, C. Almela, A. Corma, H. Garcia, Chem. Commun. (2006) 3178–3180
[84] T.A.G. Silva, E. Teixeira-Neto, N. Lopez, L.M. Rossi, Sci. Rep. 4 (2014) 5766.
[85] X. Wei, X.F. Yang, A.Q. Wang, L. Li, X.Y. Liu, T. Zhang, C.Y. Mou, J. Li, J. Phys.
Chem. C 116 (2012) 6222–6232.
2
89 (2012) 105–117.
[
[
[
66] A. Jablonski, S.H. Overbury, G.A. Somorjai, Surf. Sci. 65 (1977) 578–592.
67] L.Z. Mezey, J. Giber, Surf. Sci. 117 (1982) 220–231.
68] C.W. Yi, K. Luo, T. Wei, D.W. Goodman, J. Phys. Chem. B 109 (2005) 18535–
1
8540.
[86] K. Mori, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, J. Am. Chem. Soc. 126
(2004) 10657–10666.
[87] D. Ferri, C. Mondelli, F. Krumeich, A. Baiker, J. Phys. Chem. B 110 (2006) 22982–
22986.
[
[
69] T. Nakagawa, H. Nitani, S. Tanabe, K. Okitsu, S. Seino, Y. Mizukoshi, T.A.
Yamamoto, Ultrason. Sonochem. 12 (2005) 249–254.
70] Y. Ding, F.R. Fan, Z.Q. Tian, Z.L. Wang, J. Am. Chem. Soc. 132 (2010) 12480–
1
2486.