High-performance Pd-Au bimetallic catalyst with mesoporous silica nanoparticles as support and its catalysis of cinnamaldehyde hydrogenation

Article abstract of DOI:10.1016/j.jcat.2012.04.003

A high-performance bimetallic catalyst with mesoporous silica nanoparticles as support, PdAu/MSN, was prepared by an organic impregnation-hydrogen reduction approach. A series of investigations were conducted to assess the effects of (i) the porous nanopa

Full text of DOI:10.1016/j.jcat.2012.04.003

Journal of Catalysis 291 (2012) 36–43
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
High-performance Pd–Au bimetallic catalyst with mesoporous silica
nanoparticles as support and its catalysis of cinnamaldehyde hydrogenation
Xu Yang a,b, Dan Chen a,b, Shijun Liao a,b, , Huiyu Song a,b, Yingwei Li a,b, Zhiyong Fu a,b, Yunlan Su c
⇑
a
School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510641, China
Key Lab for Fuel Cell Technology of Guangdong Province & Key Lab of Enhanced Heat Transfer and Energy Conservation, Ministry of Education, China
Institute of Chemistry, Chinese Academy of Sciences, Beijing 100191, China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
A high-performance bimetallic catalyst with mesoporous silica nanoparticles as support, PdAu/MSN, was
prepared by an organic impregnation–hydrogen reduction approach. A series of investigations were con-
ducted to assess the effects of (i) the porous nanoparticle support on the dispersion of active components
and on the catalyst’s performance, (ii) the addition of gold on the dispersion of active components and the
catalyst’s activity, and (iii) the preparation parameters, such as solvent, pressure, and temperature, on the
catalyst’s activity. The active metallic components were highly dispersed, with particle size 2.5 nm. The
addition of gold to the catalyst favorably promoted the hydrogenation of cinnamaldehyde. The activity of
PdAu0.2/MSN (with Au/Pd molar ratio 0.2:1) was up to four times higher than that of Pd/MSN (without Au
as a promoter) and eight times higher than that of commercial Pd/C catalyst. The enhanced activity of
PdAu0.2/MSN can be attributed to the synergistic effect of Pd with the added Au and the highly dispersed
active components. The ultrahigh activity, as well as its novel structure with controllable compositions,
makes this catalyst very attractive for both fundamental research and practical applications.
Ó 2012 Elsevier Inc. All rights reserved.
Received 3 February 2012
Revised 3 April 2012
Accepted 4 April 2012
Available online 22 May 2012
Keywords:
Bimetallic
Cinnamaldehyde hydrogenation
Mesoporous silica
Pd–Au
Synergistic effect
À1
1
. Introduction
of alcohol (TOFs up to 270,000 h ) [12,13] and the direct synthesis
2 2
of H O [14,15]; recently, they extended the catalyst’s applications
In green chemistry [1,2], achieving high dispersion of
a
to oxidation of the primary C–H bond [16]. Venezia and coworkers
[17–19] prepared PdAu catalysts with varied Au/Pd ratios for the
hydrodesulfurization of aromatic compounds and found that add-
ing Au could modify the substrate’s adsorption model. Goodman’s
[11,20], as well as Prati’s group [9], investigated the ensemble
effect of Pd monomer isolated by Au, which showed unique
properties for the synthesis of vinyl acetate and glycerol oxidation,
supported noble metal catalyst (e.g., Pd, Au, Pt) has been a signifi-
cant issue in the field of heterogeneous catalysis [3–6], due to these
catalysts’ inherent high activity but high cost. However, it is well
known that during the activation process (calcination or hydrogen
reduction), some of the noble metals, especially palladium and
gold, are easily reduced and are apt to form large metal particles,
due to the unique physical/chemical properties of noble metals
2
respectively. Zhao’s group found that physically mixed TiO -
(
autocatalysis, mobility, and so on), resulting in poor metal disper-
supported Pd and Au catalysts can accelerate the hydrogen
spillover process [10].
sion and low activity. Thus, preparing a high-performance, hetero-
geneous noble metal catalyst with high dispersion of the active
components has been an ongoing challenge.
Alloying with a second metal has proved effective for obtaining
a high-activity catalyst, especially for Pd-based catalysts, as
reported in several studies [7–11]. The promotion effect of alloying
not only improves resistance to particle sintering [8] but also
modifies the geometry and electronic properties of the active spe-
Mesoporous silica (MS) has gained intense interest in the field of
nanomaterials and catalysis [21–24]. It is an ideal support candidate
for the preparation of high-dispersion catalysts, based on its unique
advantages over conventional support materials: (i) high surface
area and pore volume facilitate the adsorption and dispersion of me-
tal salt via the impregnation method, avoiding the addition of the
acid/base organic polymer required for precipitation–deposition
or other complicated methods [25]; (ii) intrinsic porosity provides
more access to the active center for the substrate [26]; (iii) mesopor-
ous channels trap active metal particles that can resist deactivation
during reaction [21,27]; and most importantly from another point
of view, (iv) inactive siliceous properties make it an ideal support
for investigating the interaction between active components, with
less interruption by metal–support interaction [28].
2
cies. For example, Hutchings’ group reported a TiO -supported
bimetallic catalyst with a Pd-rich shell and an Au-rich core that
showed superior catalytic performance for the selective oxidation
⇑
Corresponding author at: School of Chemistry and Chemical Engineering, South
China University of Technology, Guangzhou 510641, China.
E-mail address: chsjliao@scut.edu.cn (S. Liao).
0
021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.04.003

X. Yang et al. / Journal of Catalysis 291 (2012) 36–43
37
The cinnamaldehyde (CALD) hydrogenation route, depicted in
Scheme 1, can result in various product distributions, depending
on the active metal, solvent, and promoter [29–32]. Gallezot and
Richard reported [33] that a Pd catalyst favored the production of
hydrocinnamaldehyde (HALD), while a Pt-based catalyst facilitated
the production of cinnamyl alcohol, both of which are important
intermediates in the preparation of pharmaceuticals [34]. To date,
most published research has applied mono-Pt or modified Pt cata-
lysts for selective hydrogenation of CALD to cinnamyl alcohol, and
few reports have involved the utilization of PdAu bimetallic cata-
lysts for producing HALD.
In this paper, we extended our previous work using mesoporous
silica nanoparticles (MSNs) as supports [35] and prepared a series
of high-performance bimetallic catalysts, PdAu/MSN. We found
that under mild conditions, adding Au and using MSN as a support
greatly enhanced Pd dispersion and the catalyst’s performance to-
ward selective hydrogenation of CALD to HALD. The effect of metal
composition on the catalyst’s structure and behavior was investi-
gated to explore the intrinsic alloy synergistic effect.
Atomic absorption analysis (AAS) was performed with a Spec-
trAA-220Z (Varian, USA).
The N adsorption–desorption isotherms were measured with a
2
Tristar 3010 isothermal nitrogen sorption analyzer (Micromeritics,
USA) using a continuous-adsorption procedure.
2
H chemisorption was measured according to a previously re-
ported method [37]. The measurement was performed at 298 K
in a standard volumetric glass apparatus from Belsorp-HP (Anker-
smid, Netherlands). The double isotherm method was used to
determine the amount of irreversibly adsorbed hydrogen, which
allows calculation of the apparent metallic dispersion (H/Pd),
assuming an adsorption stoichiometry of H:Pd = 1:1. The amount
of chemisorbed hydrogen was determined by extrapolating the
leaner part of the isotherms to zero pressure.
Scanning transmission electron microscopy (STEM) and high-
resolution transmission electron microscopy (HRTEM) were car-
ried out with a JEOL JEM2010 microscope (JEOL, Japan) under an
accelerating voltage of 200 kV.
X-ray photoelectron spectroscopy (XPS) with an AXIS Ultra DLD
(
Kratos, Britain) was used to examine the catalysts’ electronic
properties. The binding energies were calibrated using a C1s bind-
ing energy of 284.8 eV. The peaks were fitted by a nonlinear least
square fitting program using a properly weighted sum of Lorentz-
ian and Gaussian component curves after background subtraction,
according to Shirley and Sherwood.
2
. Experimental
2.1. Catalyst preparation
Mesoporous silica nanoparticles were prepared according to a
2
H TPR was performed on an AutoChem II 2920 chemisorption
typical MCM-41 synthesis route [36]: 0.25 g CTABr was dissolved
in 100 ml deionized water under mild stirring. After the addition
of ammonium solution (25 wt.%), TEOS was added dropwise
analyzer (Micromeritics, USA): a sample of 50 mg was heated from
À1
room temperature to 600 °C at a rate of 10 °C min in 5% H
2
/Ar
À1
(
40 ml min , STP).
to obtain final mixture composition of 0.051CTABr : 414-
a
H
2
O : 3.29NH : 1TEOS. After 10 h of stirring, white precipitates
3
were obtained and then filtered, washed, dried, and calcined at
2.3. Catalyst evaluation
6
00 °C for 4 h to remove the residual template.
PdAu bimetallic catalysts were prepared according to the proce-
dures we reported previously [35]. First, the MSN silica support
The hydrogenation of CALD was used for the hydrogenation
model reaction. The reaction mixture, composed of 20–50 mg cat-
alyst, 1.0 g CALD, and 10 ml solvent, was introduced into a 50-ml
autoclave under stirring at 1000 rpm, a value chosen to eliminate
the mass transfer limitation. Air in the autoclave was purged by
hydrogen for 20 min, and then the reaction proceeded at the re-
quired temperature (50–100 °C) and at 0.5 MPa of 99.99% pure
hydrogen.
was pretreated in a muffle oven at 200 °C for 2 h and then was
impregnated with an ethanol solution of PdCl and HAuCl , at
2 4
various Au/Pd molar ratios, at room temperature under stirring.
The solvent was then evaporated in a water bath at 70 °C, with fur-
ther vacuum drying at 40 °C for 12 h. The PdAu/MSN catalyst was
finally prepared by reducing the sample for 2 h under hydrogen
flow at 200 °C in a tubular furnace with programmed temperature
À1
The products were analyzed using a gas chromatograph (Agilent
control, at a rate of 3 °C min
.
6
(
3
840N, USA) equipped with a FID detector. A GC–MS analyzer
Shimadzu GCMS-QP5050A, Japan), equipped with a 0.25 mm Â
0 m DB-WAX capillary column, was used for the identification
Three catalyst samples were prepared, with Au/Pd molar ratios
of 0.1, 0.2, and 0.4, respectively. For comparison, Au/MSN and
Pd/MSN were prepared using the same procedures. The basic
specifications of all five catalyst samples are listed in Table 1.
of products. The column oven temperature was programmed from
À1
1
20 °C (held for 1 min) to 280 °C at a rate of 20 °C min
.
2.2. Catalyst characterization
3
. Results and discussion
X-ray diffraction (XRD) patterns were obtained with a D/max-
IIIA X-ray diffractometer (Rigaku, Japan) using CuK
a radiation
À1
3.1. Catalyst characterizations
(
35 kV, 30 mA) at a rate of 0.5 °C min
.
3.1.1. XRD analysis
Fig. 1 shows the small-angle XRD patterns for the PdAu/MSN
Cinnamal alcohol
(CALC)
catalyst samples and the MSN sample. All of the samples, including
the pristine MSN support and the metal-loaded catalysts, exhibited
a very strong main peak and three other peaks in the small-angle
domain, indexed as (100), (110), (200), and (210), which are
reflections of the typical hexagonal mesophase p6 mm, demon-
strating the high quality of mesopore packing [36]. The ordered
mesopores contributed to the high surface area and pore volume
of the MSN support, qualities that can be beneficial to the adsorp-
tion of metal salt during impregnation and to the improvement of
metal dispersion. The mesoporous properties of the samples are
given in Table 1.
OH
H
c
.
ta
a
O
OH
H
c
.
Cinnamal dehyde
CALD)
O
a
Hydrocinnamal alcohol
(HALC)
t
a
(
Hydrocinnamal dehyde
HALD)
(
Scheme 1. Selective hydrogenation of cinnamaldehyde.

3
8
X. Yang et al. / Journal of Catalysis 291 (2012) 36–43
Table 1
Composition and porosity parameters of MSN and PdAu/MSN with various Au/Pd molar ratios.
Metal loading (wt.%)a
S
BET (m2
g
À1
)b
Pore size (nm)b
Pore volume (cm3 )b
g
À1
Catalyst
Pd/C
Pd/MSN
PdAu0.1/MSN
Pd, 2.0 (1.7)
Pd, 2.0 (1.9)
Pd, 2.0 (1.9)
Au, 0.37 (0.35)
Pd, 2.0 (1.9)
Au, 0.74 (0.73)
Pd, 2.0 (1.9)
Au, 1.48 (1.40)
Au, 2.0 (1.8)
241
624
617
—
3.1
3.2
—
1.20
1.21
PdAu0.2/MSN
PdAu0.4/MSN
Au/MSN
622
616
620
3.1
3.0
3.1
1.19
1.13
1.17
a
The recipe metal content and the content measured by atomic absorption spectroscopy (in parentheses).
Derived from nitrogen adsorption–desorption isotherm.
b
(100)
(
a)
(111)
(200)
(220)
(
v)
(110)
(
200) (220)
MSN support
PdAu0.1/MSN
(
iv)
(iii)
ii)
i)
(
PdAu0.2/MSN
PdAu0.4/MSN
(
10
20
30
40
50
60
70
80
1
2
3
4
5
6
7
8
9
10
o
o
2
Theta ( )
2
Theta ( )
Fig. 1. Small-angle XRD patterns of MSN support and PdAu/MSN catalysts.
(
b)
Au (111) Pd (111)
The metal crystallites of the catalysts were characterized using
wide-angle XRD, as shown in Fig. 2a. Although the reduction tem-
perature was low (200 °C), all of the catalyst samples exhibited
three clear diffraction peaks, indexed as the (111), (110), and
(v)
(
220) planes of a face-centered cubic structure, indicating that
all the metal components were well reduced and supported on
the MSN supports. Compared with Pd/MSN and Au/MSN, all three
PdAu/MSN catalysts showed a broadened diffraction peak for Pd
(
iv)
iii)
ii)
i)
(
(
111) between 38° and 40°, implying that the addition of Au im-
(
proved the dispersion of Pd on the MSN support and achieved
smaller metal particle size. Table 2 presents all the particle size
data, calculated using the Scherrer equation. The particle size de-
creased from 12.8 to 9.3 nm when MSN was used as the support
instead of active carbon. Adding Au also sharply improved the
dispersion of the active components; after just a small addition
of Au (PdAu0.1/MSN), the particle size decreased from 9.3 to
(
30 32 34 36 38 40 42 44 46 48 50
o
2 Theta ( )
Fig. 2. XRD patterns (a) and enlarged regional patterns (b) between 30° and 50° for
the catalysts (i) Pd/MSN, (ii) PdAu0.1/MSN, (iii) PdAu0.2/MSN, (iv) PdAu0.4/MSN, and
3
.2 nm. The optimal Au/Pd molar ratio was 0.2, yielding the small-
(
v) Au/MSN.
est size (3.1 nm). Although the high surface area of MSN is helpful
for getting a high dispersion of palladium, it is clear that the addi-
tion of gold played a crucial role for enhancing the dispersion and
decreasing the particle size of active components.
From the enlarged regional pattern between 30° and 50° (see
Fig. 2b), it can be observed that when the Au/Pd ratio was
increased from 0 to 0.4, the reflection peak of the samples shifted
toward a lower angle, from 40.06° to 38.28° (see Table 2), implying
that the metal crystallite changed from ‘‘Pd-like’’ to ‘‘Au-like,’’ or
indicating PdAu alloy formation [38–40].
Table 2
Specifications of several catalysts prepared with various metal compositions.
Catalyst
Hydrogen chemisorption
d(111)a
DXRDa
DTEMb
V
irre
(l
mol/g)
Dispersion
Pd/C
Pd/MSN
PdAu0.1/MSN
PdAu0.2/MSN
PdAu0.4/MSN
0.015
0.022
0.085
0.100
0.058
0.08
0.12
0.45
0.53
0.31
–
12.8
9.3
3.2
3.1
4.5
–
40.06
39.92
39.51
38.28
6.2
2.8
2.5
4.0
3.1.2. Hydrogen chemisorption
a
The hydrogen chemisorption results are presented in Table 2.
Hydrogen can be dissociatively adsorbed onto the surface or
Calculated from XRD patterns with Jade software.
Average values calculated from 200 individual crystallites in TEM images.
b

X. Yang et al. / Journal of Catalysis 291 (2012) 36–43
39
absorbed into the Pd metal bulk. The dispersion value derived from
the hydrogen chemisorption value, therefore, does not reflect the
actual exposed Pd surface; however, the increasing trend in hydro-
gen adsorption for all five samples is quite consistent with the
decreasing trend in particle size. The relatively low hydrogen
adsorption for Pd/C and Pd/MSN can be attributed to their low me-
tal dispersions, as discussed above. For the bimetallic samples,
their higher irreversible hydrogen adsorption amounts compared
with the mono-Pd catalyst varied with the Au/Pd ratios, in the se-
with particle sizes ranging from 1 to 10 nm (see Fig. 3b); even
some larger Pd aggregations (>10 nm) can be seen, implying the
sintering or aggregation of the active metal component in this cat-
alyst, actually the aggregation of pure palladium catalyst has been
recognized and reported previously by many researchers [35,41].
All of the other bimetallic catalysts showed homogeneous metal
dispersion, with particle sizes of ca. <5 nm (Fig. 3c–d), demonstrat-
ing that particle size can be controlled effectively by the addition of
appropriate amounts of Au. Once the ratio of Au/Pd goes up to 0.4
(PdAu0.4/MSN), the metal particles seem to become larger, which
agrees well with the XRD results. It should be noted that the aver-
age particle sizes observed from the TEM images are consistent
with the values derived from the XRD data with the Scherrer equa-
tion, except for mono-Pd/MSN. This discrepancy can be ascribed to
the polydispersity of Pd/MSN [41]; a comprehensive explanation
can be found in Natter et al. [42].
PdAu0.2/MSN was observed further using STEM and HRTEM, as
presented in Fig. 4. Highly dispersed metal particles trapped in
the mesochannels of the MSN can be observed in Fig. 4a and b.
Some of the metal nanoparticles are arranged in lines (indicated
by the red arrows in Fig. 4b), or stuck in the pore’s end (Fig. 4c), ob-
served from the direction perpendicular or parallel to the meso-
channel, respectively. It was reported previously that active
metal particles confined in the channels of mesopores could result
in some interesting nanoscale effects, such as sintering resistance
quence PdAu0.4 < PdAu0.1 < PdAu0.2
.
Considering the deviation of particle sizes determined by XRD
method, we can say that the adsorption of hydrogen is inversely
proportional to the particle size, although the hydrogen not only
can be adsorbed onto the surface of palladium, but also can be dis-
solved into the lattice of palladium crystal. As we will discuss later,
the sequence of hydrogen adsorption of three catalysts is actually
quite consistent with that of their catalytic activities.
3.1.3. TEM/STEM image analysis
Typical TEM images of the mono- and bimetallic catalysts are
presented in Fig. 3. Fig. 3a shows the MSN support as a monodis-
perse elliptical form (ꢀ100 nm) with honeycomb-like porosity,
characteristic of typical hexagonal packing mesopores for MCM-
4
1 family molecular sieves, which is consistent with the results
from small-angle XRD analysis.
The metal dispersion of the loaded catalysts can be seen clearly
from the TEM images. The mono-Pd/MSN exhibited polydispersity,
[
26] and low metal melting point [43]. The former property can im-
prove the catalyst’s recyclability and the latter, we assume, could
(
a)
(b)
40
80 120 160 200 240
Particle size (nm)
4
5
6
7
8
9
Particle size (nm)
(
c)
(d)
1
2
3
4
5
6
2
3
4
5
6
Particle size (nm)
Particle size (nm)
Fig. 3. HRTEM images of (a) MSN, (b) Pd/MSN, (c) PdAu0.1/MSN, and (d) PdAu0.4/MSN catalysts.

4
0
X. Yang et al. / Journal of Catalysis 291 (2012) 36–43
(
b)
(a)
1.5
2.0
2.5
3.0
3.5
Particle size (nm)
2
0 nm
(c)
(d)
Fig. 4. STEM (a) and HRTEM (b–d) images, with corresponding electron diffraction pattern (inset in d) of the optimized PdAu0.2/MSN.
be beneficial for the homogeneous alloying of metals during ther-
mal treatment. The lattice of the (111) and (200) crystal planes
can be seen in the magnified HRTEM image in Fig. 4d. Due to the
equipment’s detection limitations, we could not identify alloy
formation from changes in the lattice.
The above observations and analysis demonstrate that the pro-
motion of Au, as well as the high surface area of the MSN support,
effectively improved the dispersion of metal components and that
the Au/Pd ratio is an another important factor in controlling the
final particle size. Higher metal dispersion means higher exposed
metal surface areas and consequently higher catalytic activity.
by the addition of Au. The Pd/Si ratio is close to the nominal ratio,
but the Au/Pd ratio is not. It was found that, depending on the
amount of Au added, the XPS-derived Au/Pd ratio varied from
smaller to larger than the nominal value (see Table 3). The surface
composition of an alloy can be related to the inherent microstruc-
ture of a catalyst, such as core–shell structure or homogeneous/
heterogeneous alloy.
The deviation of nanoparticle PdAu catalysts has been investi-
gated by Hutchings et al. [44] with XPS and HRTEM-XEDS. For a
carbon-supported PdAu catalyst they prepared with a sol immobi-
lization technique, it was disclosed that the smaller particles are
Au-rich, whereas the larger ones tend to be Pd-rich. It also
mentioned that these results are diametrically opposite to those
obtained for PdAu nanoparticles prepared with impregnation
methods, which was reported by Kiely and coworkers [45],
implying that the deviation trend may be affected strongly by
the preparation method.
3.1.4. XPS analysis
The surface properties and the chemical states of Pd and Au
were investigated by X-ray photoelectron spectroscopy (XPS).
The Pd3d spectra of the catalysts were fitted by two doublets
II
attributed to the higher Pd3d energy value of Pd O (>336.6 eV)
and the lower energy value of metallic Pd0 (<335.8 eV), as shown
in Fig. 5a. A shift of ca. 0.6–0.8 eV toward high binding energy with
respect to the sample of monometallic Pd (335.1 eV) was observed
for all three PdAu bimetallic catalysts, which is attributable to
charge transfer between Au and Pd [10,39], indicating that the
electronic structure of the surface Pd atoms was modified upon
the addition of Au. Gold binding energies, ca. 84.1 eV, are typical
of the metallic state shown in Fig. 5b.
Although the detection depth of XPS is up to 10 nm, which is
larger than the particle sizes of the PdAu nanoparticles we pre-
pared, we believe the XPS results may strongly reflect the micro-
structure and surface composition information for nanoparticles.
For the sample PdAu0.1/MSN, the XPS-detected ratio of Au/Pd is
only 40% of the nominal ratio, but for the sample PdAu0.4/MSN, this
value is up to 140%, while it is 80% for PdAu0.2/MSN. The more gold
is added, the greater the deviation is. Combined with the results of
XRD, in which the patterns changed gradually from a Pd-like
pattern for PdAu0.1/MSN to an Au-like pattern for PdAu0.4/MSN,
The XPS-derived atomic ratios of Pd/Si and Au/Pd, reported in
Table 3, give evidence of the surface metal composition induced

X. Yang et al. / Journal of Catalysis 291 (2012) 36–43
41
In addition, the fraction of PdII to metallic Pd0 species in the
sample can be reflected roughly by the peak-fitting method. It is
worth noting that increasing the Au/Pd ratio also increased the
(a)
PdAu0.4
PdAu0.2
0
II
metallic Pd content with respect to the Pd species content (PdCl
2
or PdO) from 63% to 88%, seeing Table 3, which is consistent with
the findings of Qian and Huang [46], confirming that adding Au
II
may be helpful to the reduction of Pd to metallic Pd, or the protec-
0
II
II
tion of Pd from to be oxidized into Pd .
0
Pd
Pd
3
.1.5. H
2
TPR analysis
PdAu0.1
Pd
To further understand the interaction between Pd and Au, as
well as between the metals and the support, we investigated the
reducibility of the precursor samples using hydrogen temperature
program reduction (H TPR). In Fig. 6, the mono-Pd catalyst precur-
2
sor presents a single negative peak at 62 °C, caused by dissociation
of palladium hydride [47], indicating the easy reducibility of Pd2+
or PdO species. The mono-Au catalyst precursor shows a broad-
ened hydrogen consumption peak in the temperature range 130–
3
40.3 eV (Pd 3d3/2
)
335.1 eV (Pd 3d5/2
)
3
44 342 340
338 336 334
Binding energy (ev)
332 330
3+
2
10 °C, which could be ascribed to the reduction of Au interacting
(b)
Au 4f7/2
Au 4f5/2
with the SiO support, with different interaction strengths [46]. For
2
PdAu0.1
PdAu0.2
the PdAu bimetallic catalyst precursor, a sharp hydrogen consump-
tion peak was observed, which shifted from 98 to 125 °C depend-
ing on the Au/Pd ratio, revealing that interactions occurred
between Pd and the added Au.
It should be noted that a weak shoulder peak (indicated by the
arrows in Fig. 6) occurred in PdAu0.1/MSN, decreased in PdAu0.2
/
MSN, and finally disappeared for PdAu0.4/MSN. We suggest that
this peak may be related to the reduction of Pd ions existing as oxi-
des, a suggestion supported by the XPS results showing that the
fraction of PdO decreased with the variation in Au/Pd ratio.
It is acknowledged that the dissociative hydrogen adsorption
capacity of the metal catalyst is related to the electronic structure
PdAu0.4
mono-Au
of the metal. Combining this with the results of H
that the addition of Au may modify the electronic properties of Pd.
2
TPR, we suggest
92
90
88
86
84
82
Binding energy (ev)
3.2. Selective hydrogenation of cinnamaldehyde
Fig. 5. XPS spectra of Pd (a) and Au (b) in the different catalysts.
As indicated in Fig. 7, PdAu/MSN showed quite high activity to-
Table 3
ward the selective hydrogenation of CALD.
XPS Pd and Au binding energies (eV) and surface atomic ratios of the catalysts after
Only trace conversion was obtained for mono-Au/MSN under
the given reaction conditions (0.5 MPa, 50 °C), implying low activ-
ity of Au in the hydrogenation process. For the Pd/C and Pd/MSN
catalysts, the conversion of the latter (25%) was almost twice that
of the former (12%), verifying the enhancement arising from the
mesoporous large surface–area support. All three PdAu/MSN cata-
lysts showed greatly enhanced activity compared with the Pd/MSN
reduction at 200 °C.
Catalyst
Binding energy (eV)
Atomic ratio
Pd3d5/2a
Pd/Si (Â10À3
Au/Pdb
Au4f7/2
)
Ã
Pd/MSN
335.1 (0.54)
–
4.3
4.2
4.4
4.2
–
–
3
36.2 (0.46)
335.7 (0.63)
37.2 (0.37)
335.8 (0.72)
37.5 (0.29)
335.6 (0.88)
PdAu0.1/MSN
PdAu0.2/MSN
PdAu0.4/MSN
Au/MSN
84.2
84.3
84.1
84.1
0.04 (0.10)
0.16 (0.21)
0.56 (0.40)
–
3
3
o
177 C
mono-Au
3
–
37.2 (0.12)
o
1
25 C
a
For every sample, line 1 and line 2 present the binding energy of Pd3d5/2 of Pd
PdAu0.4
PdAu0.2
and PdO, respectively, derived from XPS spectrum of the sample with XPS analysis
software, and the datum in parentheses is the relative fraction of Pd or PdO in the
sample.
b
The atomic ratio of Au/Pd measured by XPS, and the datum in parentheses is the
o
9
8 C
ratio value measured by atomic absorption spectroscopy.
PdAu0.1
mono-Pd
we assume that the microstructure of the PdAu alloy catalyst can
be tuned by the Au/Pd ratio, and it seems that as more Au is added,
upon hydrogen reduction at 200 °C, the catalyst is prone to form
Au-rich surface ensembles or clusters. However, further study is
still needed to disclose the mechanism of transformation of the
microstructure of PdAu nanoparticles, as well as to establish a
clearer correlation between the catalyst’s surface composition
and activity.
o
60 C
0
100
200
300
400
500
600
o
Temperature ( C)
Fig. 6. H
2
TPR curves of catalysts with different metal compositions.

4
2
X. Yang et al. / Journal of Catalysis 291 (2012) 36–43
catalyst: the optimal PdAu0.2/MSN catalyst achieved activity four
and eight times higher than that of Pd/MSN and Pd/C, respectively.
It is very interesting that the activities of all five catalysts were
consistent with the particle sizes, hydrogen adsorption capacities,
and XPS results.
As shown in Fig. 7, the selectivity of the alloy catalysts for the
HALD can be improved slightly by the addition of gold. Actually,
we found that the monometallic Au catalyst could reach an excel-
lent selectivity of close to 100% at a reaction temperature of 50 °C,
agreeing well with the results reported by Xu and coworkers
100
1
00
(
a)
8
0
0
8
0
6
60
40
40
react temp., 50 C 20
MCALD:MPd=2500:1
0
o
2
0
[
48,49]; however, monometallic Au catalyst showed only negligi-
ble activity toward the selective hydrogenation of cinnamaldehyde
at this temperature.
0
0
.0
0.6
1.2
1.8
2.4
3.0
Fig. 8 presents the results for the optimal PdAu0.2/MSN catalyst
for the selective hydrogenation of CALD at various hydrogen pres-
sures and temperatures. The hydrogen pressure can influence
activity to some extent, but has little effect on product selectivity.
Clearly, the conversion increased as the hydrogen pressure in-
creased from 0.2 to 1.0 MPa (Fig. 8a); this was because high pres-
sure can enhance hydrogen solubility, making more hydrogen
molecules accessible in the reaction medium. With further in-
creases in the pressure, no obvious conversion increase was ob-
served, implying the absence of a hydrogen transport limitation
at higher pressures.
PH2 (MPa)
1
00
(b)
1
00
80
8
6
4
0
0
0
6
0
0
4
The hydrogenation activity was enhanced significantly with
temperature increase, as shown in Fig. 8b. The conversion of CALD
increased sharply from 13% to 100% when the temperature rose
20
0
pH2=0.5 MPa
MCALD:MPd=5000 :1
20
0
from 25 to 100 °C. The mass activity of Pd and Au is up to
À1 À1
1
66.6 mol h
g
at 100 °C, thereby demonstrating the tempera-
20 30 40 50 60 70 80 90 100 110
PdAu
o
Temperature ( C)
ture-dependent behavior of the hydrogenation reaction and the
high performance and thermal stability of PdAu0.2/MSN.
Fig. 8. Effects of hydrogen pressure (a) and temperature (b) on CALD hydrogenation
performance with the PdAu0.2/MSN catalyst.
Fig. 9 shows the Arrhenius plots of hydrogenation of CALD cat-
alyzed by PdAu0.2/MSN and Pd/MSN catalysts; two high relative
lines has been obtained. The activation energies calculated from
À1
Arrhenius plots are 57 and 62 kJ mol , respectively, further con-
1
.5
1.0
.5
PdAu0.2/MSN
R=0.996
firming the promotion of gold clearly.
The temperature also influenced the selectivity; the best selec-
tivity (92%) of HALD for PdAu0.2/MSN was obtained at 60 °C. Lower
temperatures facilitate the production of completely hydrogenated
hydrocinnamyl alcohol, while higher temperatures may lead to
side products such as aldol condensation products.
The recyclability test for PdAu0.2/MSN was carried out at 50 °C
and 1 MPa hydrogen pressure. It was found that after easy
separation and pretreatment, PdAu0.2/MSN still could achieve a
)
0
mteal
-1
g
0.0
0.5
-1
-
-1.0
nr(molg
-
-
1.5
2.0
Pd/MSN
R=0.995
2.9
000/T
1
8
6
4
2
0
00
1
00
-2.5
Selec.HALD
2.7
2.8
3.0
3.1
1
0
8
6
4
2
0
0
0
0
0
Fig. 9. Arrhenius plots for the hydrogenation of CALD over Pd/MSN and PdAu0.2
/
MSN.
0
conversion of 98% and a selectivity of 85% by the fifth test, demon-
strating its good deactivation resistance and recyclability. In fact,
trace metal leaching (0.12 wt.%) and minimal particle sintering
after the fifth test reaction contributed to its recyclability. Its excel-
lent catalytic performance and good recyclability will make this
catalyst attractive for both fundamental research and practical
applications.
0
0
4
. Conclusions
Fig. 7. Conversion and selectivity for the hydrogenation of CALD with various
catalysts. Reaction conditions: 1.0 g CALD, 50 mg catalyst, 10 ml hexane as solvent,
A high-performance PdAu bimetallic catalyst was prepared by
5
0 °C, 0.5 MPa pH2, reaction for 1 h.
impregnation and hydrogen reduction. It was found that the

X. Yang et al. / Journal of Catalysis 291 (2012) 36–43
43
promotion effect induced by the PdAu alloy synergistic interaction
can strongly influence both the structure and the catalytic behavior
of PdAu/MSN catalysts. The effect of metal composition was
[16] L. Kesavan, R. Tiruvalam, M.H.A. Rahim, M.I. bin Saiman, D.I. Enache, R.L.
Jenkins, N. Dimitratos, J.A. Lopez-Sanchez, S.H. Taylor, D.W. Knight, Science
331 (2011) 195–199.
[
17] B. Pawelec, A. Venezia, V. La Parola, E. Cano-Serrano, J. Campos-Martin, J.
Fierro, Appl. Surf. Sci. 242 (2005) 380–391.
investigated systematically using XRD, H
2
chemisorption, XPS,
[
[
[
18] A. Venezia, R. Murania, G. Pantaleo, G. Deganello, J. Catal. 251 (2007) 94–102.
19] A. Venezia, V.L. Parola, B. Pawelec, J. Fierro, Appl. Catal. A 264 (2004) 43–51.
20] T. Wei, J. Wang, D.W. Goodman, J. Phys. Chem. C 111 (2007) 8781–8788.
and H TPR. The results reveal that alloy formation plays the key
2
role in the structure of PdAu/MSN: on the one hand, alloy forma-
tion can hinder the growth of metal particles; on the other hand,
the electronic properties and compositions of active species can
be tuned. The significantly improved activity can be ascribed to
Pd’s high dispersion and modified electronic properties. However,
the intrinsic mechanism underlying these electronic properties
still requires further study.
[21] W. Huang, J.N. Kuhn, C.K. Tsung, Y. Zhang, S.E. Habas, P. Yang, G.A. Somorjai,
Nano Lett. 8 (2008) 2027–2034.
[
[
22] D. Zhao, J. Sun, Q. Li, G.D. Stucky, Chem. Mater. 12 (2000) 275–279.
23] X. Chen, S. Wang, J. Zhuang, M. Qiao, K. Fan, H. He, J. Catal. 227 (2004) 419–
427.
[
[
[
[
24] G. Liu, M. Hou, T. Wu, T. Jiang, H. Fan, G. Yang, B. Han, Phys. Chem. Chem. Phys.
13 (2011) 2062–2068.
25] J.R.A. Sietsma, J.D. Meeldijk, M. Versluijs-Helder, A. Broersma, A.J. Dillen, P.E.
de Jongh, K.P. de Jong, Chem. Mater. 20 (2008) 2921–2931.
26] S.H. Joo, J.Y. Park, C.K. Tsung, Y. Yamada, P. Yang, G.A. Somorjai, Nature Mater.
Acknowledgments
8
(2008) 126–131.
27] E. Ghedini, F. Menegazzo, M. Signoretto, M. Manzoli, F. Pinna, G. Strukul, J.
Catal. 273 (2010) 266–273.
We thank the National Natural Science Foundation of China
[
[
28] Z. Ma, S. Dai, Nano Res. 4 (2011) 3–32.
29] A.J. Plomp, H. Vuori, A.O.I. Krause, K.P. de Jong, J.H. Bitter, Appl. Catal. A 351
(2008) 9–15.
(
Project s20673040, 20876062, 21076089), the Ministry of Science
and Technology of China (Project 2009AA05Z119), and the Depart-
ment of Science and Technology of Guangdong Province (Project
[
[
[
[
30] H. Ma, L. Wang, L. Chen, C. Dong, W. Yu, T. Huang, Y. Qian, Catal. Commun. 8
(
2007) 452–456.
31] J.P. Tessonnier, L. Pesant, G. Ehret, M.J. Ledoux, C. Pham-Huu, Appl. Catal. A 288
2005) 203–210.
2
004A11004003) for their financial support of this work.
(
32] C. Pham-Huu, N. Keller, G. Ehret, L.C.J. Charbonniere, R. Ziessel, M.J. Ledoux, J.
Mol. Catal. A Chem 170 (2001) 155–163.
33] P. Gallezot, D. Richard, Catal. Rev. 40 (1998) 81–126.
References
[
1] M. Poliakoff, J.M. Fitzpatrick, T.R. Farren, P.T. Anastas, Science 297 (2002) 807–
10.
2] P.T. Anastas, M.M. Kirchhoff, Acc. Chem. Res. 35 (2002) 686–694.
3] H. Wu, Q. Zhang, Y. Wang, Adv. Synth. Catal. 347 (2005) 1356–1360.
4] M. Lamblin, L. Nassar-Hardy, J.C. Hierso, E. Fouquet, F.X. Felpin, Adv. Synth.
Catal. 352 (2010) 33–79.
[34] Q. Fan, Y. Liu, Y. Zheng, W. Yan, Front. Chem. Eng. China 2 (2008) 63–68.
[35] X. Yang, L. Du, S. Liao, Y. Li, H. Song, Catal. Commun. (2011) 29–33.
[36] Y. Han, L. Zhao, J.Y. Ying, Adv. Mater. 19 (2007) 2454–2459.
[37] G. Alvez-Manoli, T.J. Pinnavaia, Z. Zhang, D.K. Lee, K. Marín-Astorga, P.
Rodriguez, F. Imbert, P. Reyes, N. Marín-Astorga, Appl. Catal. A 387 (2010)
26–34.
8
[
[
[
[
[
5] A. Gniewek, J.J. Ziólkowski, A.M. Trzeciak, M. Zawadzki, H. Grabowska, J.
Wrzyszcz, J. Catal. 254 (2008) 121–130.
6] K. Mori, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, J. Am. Chem. Soc. 126 (2004)
[38] A. Sarkany, O. Geszti, G. Safran, Appl. Catal. A 350 (2008) 157–163.
[39] G. Zhang, Y. Wang, X. Wang, Y. Chen, Y. Zhou, Y. Tang, L. Lu, J. Bao, T. Lu, Appl.
Catal. B 102 (2011) 614–619.
1
0657–10666.
[40] A. Lee, C. Ellis, K. Wilson, N. Hondow, Catal. Today (2010) 243–249.
[41] H. Borchert, E.V. Shevchenko, A. Robert, I. Mekis, A. Kornowski, G. Grübel, H.
Weller, Langmuir 21 (2005) 1931–1936.
[42] H. Natter, M. Schmelzer, M.S. Löffler, C. Krill, A. Fitch, R. Hempelmann, J. Phys.
Chem. B 104 (2000) 2467–2476.
[
[
7] C. Della Pina, E. Falletta, M. Rossi, J. Catal. 260 (2008) 384–386.
8] A. Venezia, L. Liotta, G. Pantaleo, V. La Parola, G. Deganello, A. Beck, Z. Koppány,
K. Frey, D. Horváth, L. Guczi, Appl. Catal. A 251 (2003) 359–368.
[
9] D. Wang, A. Villa, F. Porta, L. Prati, D. Su, J. Phys. Chem. C 112 (2008) 8617–
8
622.
[43] B. Lee, Z. Ma, Z. Zhang, C. Park, S. Dai, Micropor. Mesopor. Mater. 122 (2009)
160–167.
[44] 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
(2010) 16568–16569.
[
10] R. Liu, Y. Yu, K. Yoshida, G. Li, H. Jiang, M. Zhang, F. Zhao, S. Fujita, M. Arai, J.
Catal. 269 (2010) 191–200.
11] M. Chen, D. Kumar, C.W. Yi, D.W. Goodman, Science 310 (2005) 291–293.
12] 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–
[
[
[45] A.A. Herzing, M. Watanabe, C.J. Kiely, J. Edwards, M. Conte, Z.R. Tang, G.J.
Hutchings, Faraday Discuss. 138 (2008) 337–351.
3
65.
[
[
13] G.J. Hutchings, Catal. Today 138 (2008) 9–14.
14] J.K. Edwards, E. Ntainjua N, A.F. Carley, A.A. Herzing, C.J. Kiely, G.J. Hutchings,
Angew. Chem. Int. Ed. 48 (2009) 8512–8515.
[46] K. Qian, W. Huang, Catal. Today 164 (2011) 320–324.
[47] C.W. Chou, S.J. Chu, H.J. Chiang, C.Y. Huang, C. Lee, S.R. Sheen, T.P. Perng, C. Yeh,
J. Phys. Chem. B 105 (2001) 9113–9117.
[
15] J.K. Edwards, B. Solsona, A.F. Carley, A.A. Herzing, C.J. Kiely, G.J. Hutchings,
Science 323 (2009) 1037–1041.
[48] H. Shi, N. Xu, D. Zhao, B.Q. Xu, Catal. Commun. 9 (2008) 1949–1954.
[49] K. Sun, Y. Hong, G. Zhang, B.Q. Xu, ACS Catal. 1 (2011) 1336–1346.