Reactivity studies of Au-Pd supported nanoparticles for catalytic applications

Article abstract of DOI:10.1016/j.apcata.2010.05.010

The utilisation of gold-palladium nanoparticles either in the form of colloids or supported nanoparticles has received enormous attention in recent years. These materials are very effective for the transformation of organic compounds to highly useful chem

Full text of DOI:10.1016/j.apcata.2010.05.010

Applied Catalysis A: General 391 (2011) 400–406
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Reactivity studies of Au–Pd supported nanoparticles for catalytic applications
a
a
a
a
a
Jose Antonio Lopez-Sanchez , Nikolaos Dimitratos , Neil Glanville , Lokesh Kesavan , Ceri Hammond ,
a
a
b
a,∗
Jennifer K. Edwards , Albert F. Carley , Christopher J. Kiely , Graham J. Hutchings
School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff CF10 3AT, UK
a
b
Center for Advanced Materials and Nanotechnology, Lehigh University, 5 East Packer Avenue, Bethlehem, PA 18015-3195, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The utilisation of gold–palladium nanoparticles either in the form of colloids or supported nanoparticles
has received enormous attention in recent years. These materials are very effective for the transfor-
mation of organic compounds to highly useful chemical products. The catalytic materials are usually
prepared using deposition–precipitation and impregnation techniques, but recently significant attention
has been focused on the use of colloidal methods. Here we compare and contrast the preparation and cat-
alytic reactivity of Au–Pd supported nanoparticles synthesised by deposition–precipitation and colloidal
methods. The catalyst materials have been evaluated for three different reactions, namely, the oxidation
of benzyl alcohol, the direct synthesis of hydrogen peroxide and the oxidation of carbon monoxide. In
addition, we have focused our attention on the pre-treatment temperature and the improvement of the
deposition–precipitation method by using urea and sodium borohydride for the preparation of highly
active Au–Pd supported nanoparticles.
Received 9 February 2010
Received in revised form 2 May 2010
Accepted 10 May 2010
Available online 19 May 2010
Keywords:
Gold and palladium supported catalysts
Hydrogen peroxide synthesis
Benzyl alcohol oxidation
CO oxidation
Sol-immobilisation
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
hol [36], glycerol [37,38], 1,2-propanediol [39] and the epoxidation
of alkenes [40].
The utilisation of gold and gold–palladium supported nanopar-
In this paper we report for the first time the hydrogen per-
oxide productivity of titania-supported catalysts synthesised by
using the sol-immobilisation methodology and we focus on the
metal loading and present the beneficial role of the addition of
Pd to Au for these catalysts. In addition, we compare and con-
trast the reactivity of supported Au–Pd catalysts prepared using
a deposition–precipitation method using urea as the precipitating
agent in place of sodium hydroxide. We have also explored other
preparation variables such as the effect of an acid pre-treatment on
benzyl alcohol oxidation and the use of zeolites as support materials
for the Au–Pd system for selective oxidation reactions.
ticles in the field of catalysis has grown enormously the recent
years [1–4]. Typical reactions for the use of gold-based catalysts
include the low temperature oxidation of CO [5–12], synthesis of
vinyl chloride by the hydrochlorination of ethyne [13], the selec-
tive oxidation of alcohols [14–27], the selective oxidation of alkenes
to epoxides [28,29] the synthesis of hydrogen peroxide [30–33]
and selective hydrogenation [34]. In our previous papers we have
shown that catalysts based on Au–Pd nanoparticles supported on
a range of supports, e.g. carbon, titania, alumina and iron oxide
can be effective catalysts for the oxidation of alcohols and polyols
[
35], as well as in the synthesis of hydrogen peroxide [33]. More-
over, we have emphasised that the alloying of gold with palladium
leads to a significant enhancement in activity for alcohol oxida-
tion and as well as in the hydrogen peroxide formation [33,35].
The principal method that had been used for the preparation of
the metal-supported catalysts in the earlier publications has been
the impregnation method. Recently, we have shown that the use
of a sol-immobilisation method can also be very effective for the
synthesis of Au and Au–Pd nanoparticles supported on carbon and
titania, and the catalysts typically give very high productivity in the
synthesis of hydrogen peroxide and in the oxidation of benzyl alco-
2. Experimental
2.1. Materials
HAuCl4·3H2O (99.99% purity) and PdCl2 (99.99% purity) were
supplied by Johnson Matthey. Three supports were used namely
titania (Degussa P25), activated carbon (Aldrich G60) and zeolite
H-ZSM-5 (Zeolyst, Si:Al = 30). NaBH4 of purity >96% (Aldrich), urea
(99.0% purity) and polyvinylalcohol (PVA) (Aldrich, MW = 10,000,
8
0% hydrolyzed) were used in the preparation. Fresh stock aque-
ous solutions of PdCl2 (Johnson Matthey) (acidified with HCl),
HAuCl ·3H O of the required concentration, NaBH (0.1 M) and PVA
∗ Corresponding author. Tel.: +44 29 2087 4059; fax: +44 29 2087 4059.
E-mail address: hutch@cardiff.ac.uk (G.J. Hutchings).
4
2
4
(1%, w/w) were prepared.
0
926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.05.010

J.A. Lopez-Sanchez et al. / Applied Catalysis A: General 391 (2011) 400–406
401
2.2. Catalyst preparation
2.3.2. Hydrogen peroxide synthesis
Catalyst testing was performed using a stainless steel autoclave
(Parr Instruments) with a nominal volume of 50 ml and a max-
imum working pressure of 14 MPa. The autoclave was equipped
with an overhead stirrer (0–2000 rpm) and provision for measure-
ment of temperature and pressure. Typically, the autoclave was
charged with the catalyst (0.01 g unless otherwise stated), solvent
A range of monometallic Au and Pd and bimetallic Au–Pd sup-
ported catalysts were prepared by the deposition precipitation
method using urea. To produce a 2.5 wt% Au–2.5 wt% Pd/TiO2 cata-
lyst the detailed procedure was as follows: the support was added
−
3
to an aqueous solution comprising HAuCl and PdCl (4.2 × 10 M)
4
2
and urea (0.42 M). The initial pH was 2. The suspension, whose
temperature was thermostatically held at 80 C, was vigorously
(5.6 g MeOH and 2.9 g H O), purged three times with CO2 (3 MPa)
2
◦
and then filled with 5% H /CO2 and 25% O /CO2 to give a hydro-
2
2
stirred for 4 h (during which time the pH increases and reaches
gen to oxygen ratio of 1:2 at a total pressure of 3.7 MPa. Stirring
(1200 rpm unless otherwise stated) was commenced on reaching
the desired temperature (2 C), and all experiments were carried
7
.2). The material was then recovered by filtration, washed and
◦
dried. The catalyst was chemically reduced at room temperature,
as follows: the required amount of the Au–Pd/TiO catalyst was sus-
pended in distilled water and then a freshly prepared solution of
out for 30 min unless otherwise stated. Gas analysis for H2 and O2
was performed by gas chromatography using a thermal conductiv-
ity detector and a CP – Carboplot P7 column (25 m, 0.53 mm id).
H O yield was determined by titration of aliquots of the final fil-
2
NaBH4 (0.1 M) was added in the required proportion (NaBH /Au
4
(
mol/mol) = 5) with vigorous stirring at room temperature. The
2
2
−
3
slurry was stirred for 30 min and after 2 h the slurry was filtered
and the solid was washed thoroughly with distilled water (2 l of
doubly distilled water). After filtration, a colourless filtrate indi-
cated and verified by UV–vis spectra that all the metal particles
tered solution with acidified Ce(SO4)2 (7 × 10 mol/l). Ce(SO4)2
solutions were standardised against (NH ) Fe(SO ) ·6H O using
4
2
4
2
2
ferroin as indicator.
◦
were embedded onto the titania. The sample was dried at 120 C
2.3.3. CO oxidation
overnight.
The catalytic activity for CO oxidation was determined in a fixed
bed quartz microreactor, operated at atmospheric pressure. The
For the sol-immobilisation method the detailed procedure for
the preparation of Au, Pd and Au–Pd supported catalysts have
been described in detail elsewhere [36,38]. Briefly, for the synthe-
sis of a Au–Pd/C catalyst synthesised by sol-immobilisation method
the following procedure was used. To an aqueous fresh PdCl2 and
feed consisted of CO/O /N2 with a molar ratio of 0.5/19.9/79.6.
2
−
1
The combined flow rate was maintained at 22.5 ml min , and a
constant catalyst loading of 50 mg was employed. The catalyst tem-
perature was maintained at 25 C by immersing the quartz bed in
◦
HAuCl solution of the desired concentration, the required amount
of a PVA solution (1 wt%) was added (PVA/(Au + Pd) (wt/wt) = 1.2);
a thermostatically controlled water bath. Catalysts were tested for
a minimum of 500 min, and analysis of the reaction product was
carried out on-line using gas chromatography. Conversion was cal-
4
a freshly prepared solution of NaBH4 (0.1 M, NaBH /(Au + Pd)
4
(
3
mol/mol) = 5) was then added to form a dark-brown sol. After
0 min of sol generation, the colloid was immobilised by adding
culated on the basis of CO concentration in the effluent, and carbon
balances were 100 ± 2%.
2
activated carbon or titania (acidified at pH 1 by sulfuric acid)
under vigorous stirring conditions. The amount of support mate-
rial required was calculated so as to have a total final metal loading
of 1 wt%. After 2 h the slurry was filtered, the catalyst washed thor-
oughly with distilled water (neutral mother liquors) and dried at
1
2.4. Characterisation of catalysts
The materials prepared by impregnation and immobilisation
have been extensively characterised in our previous studies using
STEM-HAADF imaging and XPS [36,38,41]. We selected them for
this comparative study because they are well characterised and
can act as model systems. For completeness the relevant details
concerning their structures are given here. Catalysts prepared with
the impregnation method produce a broader particle size distribu-
tion and greater median size than the sol-immobilisation method.
Specifically, catalysts synthesised by the impregnation method
show that for both supports they comprise mainly small particles
(ca. 2–10 nm). The AuPd/TiO2 catalyst has particles mainly in the
◦
20 C overnight.
The following notation is used for the catalyst samples used
in this study: I denotes impregnation; acid denotes an acid
pre-treatment of the support with nitric acid; SI denotes sol-
immobilisation; DPU denotes deposition–precipitation using urea;
DPUC denotes deposition precipitation using urea and reduction
with sodium borohydride; w denotes Au and Pd are present in a
1
1
:1 ratio by weight; and m denotes Au and Pd are combined in a
:1 molar ratio.
2
–5 nm range with ca. 8% of larger particles (>20 nm); whereas, the
AuPd/C catalysts comprise particles mainly in the 3–8 nm range
with ca. 20% larger particles (>20 nm). For the impregnated materi-
als the supported Au–Pd alloys have been characterised extensively
by STEM-XEDS [35,36,41] and have been shown to have contrast-
2
2
.3. Catalyst testing
.3.1. Alcohol oxidation
Benzyl alcohol oxidation was carried out in a stirred auto-
ing morphologies. The TiO -supported material has a core–shell
2
clave reactor (100 ml, Autoclave Engineers Inline MagneDrive III).
The vessel was charged with alcohol (40 ml) and catalyst (0.1 g).
The autoclave was then purged five times with oxygen leaving
the vessel at 10 bar gauge. The stirrer was set at 1500 r.p.m.
and the reaction mixture was raised to the required tempera-
ture. The reaction time was measured from the time that the
mixture reached the reaction temperature and samples from the
reactor were taken periodically, via a sampling system. For the
analysis of the products a GC–MS and GC (a Varian star 3400
cx with a 30 m CP-Wax 52 CB column) were employed. The
products were identified by comparison with known standards.
For the quantification of the amounts of reactants consumed
and products generated, an external calibration method was
used.
structure with a palladium-rich shell and a gold-rich core, an
observation that is in common with other oxide-supported Au–Pd
catalysts prepared by wet impregnation. In contrast the carbon-
supported Au–Pd catalyst tends to contain homogeneous alloy
particles. In addition, STEM-XEDS spectrum images showed that
the composition of the metal nanoparticles varied with size, with
the smallest particles being Pd-rich whereas the largest particles
were highly Au-rich. The acid pre-treated Au–Pd/C catalysts (acid
pre-treatment of the support) favoured the formation of a greater
number of the smallest (Pd-rich alloy) particles at the expense of
the intermediate and large particles [41].
We have previously reported the characterisation of cata-
lysts prepared via the sol-immobilisation method [36,38,42]. The
median diameter of the sol-immobilised pure-Au on both supports

4
02
J.A. Lopez-Sanchez et al. / Applied Catalysis A: General 391 (2011) 400–406
Table 1
a
Liquid phase oxidation of benzyl alcohol for catalysts prepared by the sol-immobilisation method. .
Selectivity (%)b
Yield (%)b
Toluene
TOF (h )c
−1
Entry
Catalyst
Conv. (%)
Toluene
Benzaldehyde
Benzoic acid
Benzaldehyde
Benzoic acid
1
2
3
4
5
6
7
8
9
1% Au/CSI
5.2
4.3
2.7
9.4
4.8
5.0
17.7
35.1
26.2
20.6
25.7
26.7
40.9
72.6
86.3
83.3
70.2
55.4
67.3
67.5
63.5
69.2
55.0
2.2
1.6
2.1
4.3
2.6
2.1
2.3
3.6
1.7
1.3
0.5
0.2
0.1
3.8
3.7
2.3
0.1
0.1
0.1
0.2
2.5
1.4
0.4
2.5
1.0
1.1
2500
2900
4200
1% Au/TiO2SI
0.5% Au/TiO2SI
0.5% Au/CSI
4.4
0.8
3.1
8800
1% Pd/CSI
96.4
68.4
15.4
70.0
61.2
81.1
33.8
17.9
3.2
18.0
16.3
33.2
53.4
46.0
10.4
44.5
42.4
44.6
23,300
18,200
7900
38,400
15,400
35,400
1% Pd/TiO2SI
0.5% Pd/TiO2SI
0.5% Pd/CSI
1% (Au–Pd)/TiO2Siw
1% (Au–Pd)/CSIw
1
0
a
b
c
◦
Reaction conditions: benzyl alcohol, 0.1 g of catalyst, T = 120 C, time of reaction = 2 h, pO2 = 150 psi, stirring rate 1500 rpm.
Selectivity to toluene, benzaldehyhde and benzoic acid, benzylbenzoate present as by-product.
Calculation of TOF (h ) after 0.5 h of reaction. TOF numbers were calculated on the basis of total loading of metals.
−1
were about 4.0 nm, while for pure-Pd the median size on C was
increase of TOF by a factor 3.5, and the selectivity towards benzalde-
hyde is not significantly affected (Table 1, entries 1 and 4). The most
important observation concerns the selectivity to toluene which is
formed as a by-product in this reaction. There is a 2-fold increase of
toluene formation on decreasing the Au loading. This trend suggests
that increasing Au loading facilitates the benzaldehyde formation
at the expense of toluene production.
5
.0 nm, which was slightly greater than that for the TiO support at
2
3
.8 nm. The mean diameter and size spread of values for Pd parti-
cles on C were significantly larger than for the corresponding Au on
C sample. The median particle sizes for the sol-immobilised AuPd
alloy were 5.1 and 3.8 nm, respectively, on the C and TiO supports.
2
It is also important to note that the main advantage of the sol-
immobilisation preparation method is that does not produce any
of the large particle sizes that are commonly found in the catalysts
prepared by impregnation. Therefore the particle size distribution
is significantly narrower for the sol-immobilisation method as com-
pared to those made by the impregnation method. STEM-HAADF
images of typical alloy particles in the AuPd/CSI and AuPd/TiO2SI
materials [38] show that the AuPd nanoparticles prepared by sol-
immobilisation are homogeneous alloys.
In the case of the Pd/C catalysts, upon decreasing the Pd load-
ing from 1 to 0.5%, there is an increase of TOF by a factor of 1.7,
while the selectivity to benzaldehyde is increased up to 8% and
the selectivity to toluene decreases by ∼10% (Table 1, entries 5
and 8). To verify this trend the selectivity to toluene is reported
at isoconversion (e.g. 70%) and we observe that for 1 wt% Pd/C and
0.5 wt% Pd/C the selectivity to toluene corresponds to 40 and 26%,
respectively. Increasing the Pd loading also resulted in a significant
increase in conversion with almost all of the initial benzyl alco-
hol being consumed. However, the change in selectivity seems to
indicate that the increase in Pd loading facilitates toluene forma-
tion at the expense of benzaldehyde formation. It should be noted
that Pd as a metal has a significant affinity for H abstraction from
the alcohol to form a surface alkoxide species to give a surface Pd–H
species, and that the oxidation of these Pd–H species is much slower
than for Au–H species [44]. A possible consequence of the presence
of surface hydrogen on the Pd nanoparticles on the reaction pro-
file is that it could catalyse the formation of toluene through the
hydrogenolysis of the C–O bond, as has been observed for Pd cata-
lysts in liquid phase oxidation [45]. Thus, increasing the Pd loading
leads to a higher concentration of Pd–H species (due to the high
affinity of Pd for hydrogen adsorption) on the surface of the catalyst
[46], and this leads to increased toluene formation.
3
. Results and discussion
3.1. Benzyl alcohol oxidation
Monometallic Au, Pd and bimetallic Au–Pd catalysts,
prepared
using
sol-immobilisation,
impregnation
and
deposition–precipitation methods, were evaluated as catalysts for
the oxidation of benzyl alcohol. Table 1 summarises the data for
the oxidation of benzyl alcohol with the monometallic Au and Pd
catalysts and the bimetallic (Au–Pd) catalysts synthesised using
the sol-immobilisation method. In addition, the effect of the metal
loading for the monometallic Au and Pd metals (0.5–1%) was also
◦
studied at the same reaction conditions (pO = 150 psi, T = 120 C,
2
1
500 rpm stirrer speed and 100 mg of catalyst). As reported
previously [42] under these particular experimental conditions
diffusion limitations are avoided. It is clear from the initial turnover
frequencies (TOF) in Table 1 that the bimetallic Au–Pd supported
catalysts exhibit enhanced catalytic activity with respect to the
monometallic Au catalysts. In fact, the 1 wt% (Au–Pd)/C catalysts
give TOF’s that are enhanced by a factor of 10–17 with respect
to the 1 wt% Au/C material, which demonstrates the synergistic
effect of adding Pd to Au [35]. However, with respect to the 1 wt%
Pd/C sample, there is only a slight increase of the TOF by a factor
of 1.4–2.0, which is in agreement with previous observations on
supported catalysts synthesised by the impregnation method [43].
In the case of the 1 wt% (Au–Pd)/TiO2 catalyst, the TOF value is
For the Au/TiO2 catalysts, a decrease of Au loading leads to an
increase of TOF by a factor of 1.4, whereas selectivity to benzalde-
hyde does not vary (Table 1, entries 2 and 3). In addition, toluene
formation is also not affected by the variation of the metal loading,
which is in direct contrast to the behaviour the Au/C catalyst. This
is probably related to different levels of conversion since the Au/C
catalysts are far more active. However, these results emphasise the
importance of the support in controlling the final distribution of
products.
For the Pd/TiO2 catalysts, a decrease in Pd loading decreases the
TOF by a factor of 2.3 (Table 1, entries 6 and 7). This result is different
to the case of Pd/C catalysts, since it retains a similar selectivity to
benzaldehyde at isoconversion, again indicating the important role
played by the support and we anticipate that this may be related
to the dispersion of the Pd.
enhanced by a factor of 7 with respect to the 1 wt% Au/TiO material.
2
For the 1 wt% (AuPd)/TiO catalyst a very similar activity is observed
2
as for the 1 wt% (Pd)/TiO2 catalyst (Table 1), but the selectivity to
benzaldehyde is improved by ∼10% using the bimetallic system.
The effect of metal loading for the monometallic Au and Pd
supported catalysts was also studied (Table 1). In the case of the
Au/C catalysts, decreasing the Au loading from 1 to 0.5% leads to an
Comparison of the catalytic data between monometallic Au and
Pd supported catalysts shows that Pd catalysts are by far more
active than Au catalysts by a factor of 8–10 in terms of TOFs,
whereas selectivity to benzaldehyde is much higher for the Au cat-

J.A. Lopez-Sanchez et al. / Applied Catalysis A: General 391 (2011) 400–406
403
alysts and toluene formation is significantly promoted by the Pd
catalysts.
materials following synthesis is necessary when urea is used, to
enable the removal of the organic precursor residues deposited
onto the support and the ligands on the metal nanoparticles. Hence,
the pre-treatment of the synthesised material is performed at 120
It is evident from the data presented in Table 1 that the most
striking effect between the titania and carbon supports, is an
increase of catalyst activity by a factor of almost 2 (based on TOF
calculated after 30 min) when carbon is the chosen support for
the immobilisation of the (Au–Pd) bimetallic-supported colloids
◦
and 400 C in air to remove residual urea residues. In addition, our
modified method employs in situ chemical reduction with NaBH4
before the filtration of the slurry solution in order to generate
small particles with a narrow distribution, without the need of a
pre-treatment heating step [53]. For the deposition–precipitation
synthesis of Au–Pd supported catalysts we have investigated two
supports, namely zeolite (HZSM-5) and titania.
−
1
−
1
(
24,000–42,000 h for carbon and 15,000–19,000 h for titania).
This increase in catalytic activity can not be explained on the basis
of differences in particle size between the two materials since for
both materials the mean particle size is very similar (4–5 nm). Other
parameters, such as surface area, porosity and support chemical
composition could be playing a role and especially considering the
The catalytic data for the deposition precipitated materials are
presented in Table 3. Comparison at the same heat pre-treatment
temperature shows that H-ZSM5 can, in some instances be a supe-
rior catalyst support to titania. Increasing the heat pre-treatment
2
higher surface area of carbon (∼1000 m /g) as compared to titania
2
(
∼45 m /g) and the very different surface functionality. Another
◦
significant observation is the fact that when a carbon support is
used the toluene formation is increased by a factor between 1.5 and
temperature from 120 to 400 C results in a significant decrease in
catalyst activity for the zeolite-supported catalyst; whereas, for the
titania-supported catalyst there is only a very slight enhancement
in performance. We can attribute this difference in behaviour to
the stronger metal-support interaction in a titania-supported cata-
lyst as compared to the zeolite-supported catalyst, which probably
inhibits nanoparticle sintering. Selectivity to benzaldehyde is in
the range 80-90%, and benzyl benzoate and benzoic acid are the
main by-products. Despite the pore structure and acidity of the H-
ZSM-5, the product distribution is very similar for both catalysts
and, considering its advantage in activity, H-ZSM5 should not be
viewed as an inferior support but rather as an effective support
for Au–Pd for certain selected catalytic applications. In addition,
1 wt% (Au–Pd)/TiO2DPUm synthesised by deposition–precipitation
2
with respect to titania (23–26% selectivity to toluene using carbon
and 38–44% selectivity to toluene using titania at similar conversion
level), indicating the importance of the choice of support.
3.2. Comparison of impregnation versus sol-immobilisation
preparation methods for benzyl alcohol oxidation
5
wt% (Au–Pd)/C catalysts were prepared by the impregnation
method. Acid treatment in carbon-supported catalysts is known to
improve dispersion of the supported metal [54,55], and we have
recently shown that an specific pre-treatment of the carbon sup-
port with dilute nitric acid aids the dispersion of the Au–Pd metals
deposited by impregnation and this reduces the mean particle size
as well as improving selectivity in hydrogen peroxide synthesis
was treated with NaBH , in order to (i) obviate the need for a
4
heat pre-treatment step and (ii) enhance the formation of small
Au–Pd nanoparticles having a narrow particle distribution. In these
preparations the same metal/NaBH4 molar ratio were utilised
as used in the colloidal method. It has been demonstrated that
for monometallic Au supported catalysts this NaBH4 chemical
treatment produces gold particles that are in the same range as
generated by the colloidal method [53]. The result of this synthesis
is very promising since high catalytic activity is obtained with the
1 wt% (Au–Pd)/TiO2DPUCm material which is similar to that found
for the 1 wt% (Au–Pd)/TiO2SIw material (Table 3). In terms of the
selectivity compared at isoconversion (Table 4, entry 5), significant
formation of toluene is observed, as is the case of the bimetallic cat-
alysts prepared by the colloidal method. These results suggest that
similar catalytically active sites are generated with both of these
preparation methods.
[
41]. In this study we have used both non-treated and acid pre-
◦
treated carbons and the final catalysts were calcined at 400 C.
These materials were tested for the liquid phase oxidation of benzyl
alcohol using the same reaction conditions as employed previously.
The treatment of the support with nitric acid adds some insight
into possible correlations between toluene formation and the num-
ber density of acidic sites. A comparison can be made between the
1
wt% (Au–Pd)/C and 1 wt% (Au–Pd)/TiO2 samples synthesised by
the immobilisation method and calcined at the same temperature
◦
(
400 C). The catalytic data for these two materials are presented
in Table 2. In terms of catalyst activity, as given by the TOFs val-
ues, the sol-immobilised 1 wt% (Au–Pd)/C catalysts are more active
by a factor of 8–12 than the 5 wt% (Au–Pd)/C synthesised by the
impregnation method. For the 1 wt% (Au–Pd)/TiO catalyst the TOF
2
is 10–18 times higher than that for the 5 wt% (Au–Pd)/C catalyst
(
(
Table 2, entries 1, 3 and 4). In terms of selectivity, the 1 wt%
Au–Pd)/C catalysts synthesised by the immobilisation method
3.4. H O synthesis and CO oxidation
2
2
gives ca. 8–10% higher selectivity to benzaldehyde, 75–79% with
the 1 wt% (Au–Pd)/C prepared by immobilisation and 62–69% with
the 5 wt% (Au–Pd)/C prepared by the impregnation method. The
acid pre-treated impregnated (Au–Pd)/C catalyst shows a similar
level of activity, but higher selectivity towards toluene with respect
to the untreated acid catalyst (Table 2, entries 4 and 5), suggesting
that surface acidity could play a role in toluene formation.
3.4.1. Monometallic and bimetallic catalysts synthesised by the
immobilisation method
Recently we have demonstrated that bimetallic Au–Pd sup-
ported nanoparticles on carbon prepared by sol-immobilisation are
highly active for the synthesis of hydrogen peroxide [36]. In this
paper we report for the first time the hydrogen peroxide activity
of Au, Pd and bimetallic Au–Pd nanoparticles supported on titania
by the sol-immobilisation method and compare their activity with
the analogous catalysts supported on carbon. The catalytic data
for these materials are presented in Table 5. Comparison between
the monometallic Au, Pd and bimetallic Au–Pd supported cata-
lysts indicates that the addition of Au to Pd markedly enhances
the catalyst activity by factors of 36 and 27 for the carbon and
titania-supported catalysts, respectively. In all the cases the Au–Pd
supported nanoparticles are more active than the corresponding
monometallic Au and Pd supported catalysts.
3.3. Benzyl alcohol oxidation using supported Au–Pd bimetallic
catalysts synthesised by deposition–precipitation
Au and (Au–Pd) catalysts supported on TiO have also been pre-
2
pared using a deposition precipitation method with urea as the
precipitating agent adapted from the original work from Geus and
van Dillen [49]. The choice of urea instead of sodium hydroxide
affords depositing the total nominal loading of the metal onto the
support while retaining small particle sizes (2–5 nm) [50–52]. Nev-
ertheless a pre-treatment of the dried Au/TiO2 and (Au–Pd)/TiO2
Monometallic Pd supported catalysts generate H O2 at a higher
2
rate than monometallic Au supported materials s with the rate

4
04
J.A. Lopez-Sanchez et al. / Applied Catalysis A: General 391 (2011) 400–406
Table 2
Liquid phase oxidation of benzyl alcohol for catalysts prepared by the sol-immobilisation and impregnation methods and calcined at 400 C in air. .
◦
a
Entry
Catalyst
Conv. (%)
Selectivity (%)b
Yield (%)b
Toluene
TOF (h )c
−1
Toluene
Benzaldehyde
Benzoic acid
Benzaldehyde
Benzoic acid
1
2
3
4
1% (Au–Pd)/TiO2SIw
1% (Au–Pd)/CSIw
5% (Au–Pd)/CIw-acid
5% (Au–Pd)/CIw
22.7
6.7
2.4
22.2
2.4
15.7
9.7
69.7
78.7
67.2
68.8
1.9
3.6
0.0
2.6
5.0
0.2
0.4
0.2
15.8
5.3
1.6
0.4
0.2
0.0
0.1
3900
1800
350
2.2
1.5
500
a
◦
Reaction conditions: benzyl alcohol, 0.1 g of catalyst, T = 120 C, time of reaction = 2 h, pO2 = 150 psi, stirring rate 1500 rpm.
b
c
Selectivity to toluene, benzaldehyhde and benzoic acid, benzylbenzoate present as by-product.
Calculation of TOF (h ) after 0.5 h of reaction. TOF numbers were calculated on the basis of total loading of metals.
−1
Table 3
Benzyl alcohol oxidation performance after 6 h reaction with Au–Pd catalysts prepared by deposition precipitation methods: effect of support, treatment (calcination in static
a
air) and chemical treatment (reduction with NaBH4). .
TOF (h 1)b
−
Entry
Catalyst
Catalyst pre-treatment
Conv. (%)
Selectivity (%)
Toluene
Benzaldehyde
Benzoic acid
Benzyl benzoate
1
2
3
4
5
5% (Au–Pd)/TiO2DPUw
5% (Au–Pd)/TiO2DPUw
5% (Au–Pd)/ZSM5DPUw
5% (Au–Pd)/ZSM5DPUw
1% (Au–Pd)/TiO2DPUCm
Dried
Calcined 400 C/air
Dried
Calcined 400 C/air
7.2
8.6
24.9
15.6
61.3
0.6
1.1
1.2
2.0
14.1
83.2
91.9
82.1
81.6
76.5
6.6
2.8
7.6
6.1
6.3
9.6
4.2
9.1
10.3
3.1
1100
1900
3200
1500
39,200
◦
◦
Dried
a
◦
Reaction conditions: benzyl alcohol, 0.05 g of catalyst, T = 120 C, pO2 = 150 psi, stirring rate 1500 rpm.
b
−1
Calculation of TOF (h ) after 0.5 h of reaction. TOF numbers were calculated on the basis of total loading of metals.
Table 4
Benzyl alcohol oxidation results at isoconversion for Au–Pd catalysts prepared by deposition precipitation methods: effect of support, calcination temperature and chemical
a
reduction. .
Entry
Catalyst
Catalyst pre-treatment
Conv. (%)
Selectivity (%)
Toluene
Benzaldehyde
Benzoic acid
Benzyl benzoate
1
2
3
4
5
5% (Au–Pd)/TiO2DPUw
5% (Au–Pd)/TiO2DPUw
5% (Au–Pd)/ZSM5DPUw
5% (Au–Pd)/ZSM5DPUw
1% (Au–Pd)/TiO2DPUCm
Dried
Calcined 400 C/Air
Dried
Calcined 400 C/Air
7
7
7
7
7
0.6
0.9
0.4
0.3
23.6
83.3
90.0
88.0
85.6
72.8
6.5
3.0
3.5
4.0
1.3
9.6
6.1
8.1
10.1
2.3
◦
◦
Dried
a
◦
Reaction conditions: benzyl alcohol, 0.05 g of catalyst, T = 120 C, pO2 = 150 psi.
Table 5
a
Hydrogen peroxide productivity and benzyl alcohol conversion for catalysts prepared by the sol-immobilisation method. .
Entry
Catalyst
Hydrogen peroxide
productivity
Hydrogen peroxide
productivity
(mol H2O2 molmetal
−
1
−1
−1 −1
)
(
mol H2O2 kgcat
h
)
h
1
2
3
4
5
6
1% Au/TiO2SI
1% Pd/TiO2SI
1% (Au–Pd)/TiO2SIm
1% Au/CSI
1% Pd/CSI
1% (Au–Pd)/CSIm
1
13
41
0.2
1.5
54
17
139
585
4
16
820
a
Reaction conditions: reaction time for H2O2 synthesis 30 min; mass of catalyst for H2O2 synthesis: 10 mg.
b
◦
−1
Reaction conditions: benzyl alcohol, catalyst (0.1 g), T = 120 C, pO2 = 10 bar, time of reaction = 6 h, stirring rate 1500 rpm. Calculation of TOF (h ) after 0.5 h of reaction. TOF
numbers were calculated on the basis of total loading of metals.
of hydrogen peroxide formation being from 7 to 15 times higher.
Depending on the support, the rates for hydrogen peroxide for-
mation are increased by a factor of 4–9 when TiO2 is used as the
support instead of the active carbon.
The Au/TiO2 and Au–Pd /TiO2 catalysts prepared using the sol-
immobilisation method were also tested for CO oxidation as shown
in Figs. 1 and 2. For CO oxidation, the dried samples show very
poor CO conversion, i.e. below 1% conversion. As in the case of the
metal-supported catalysts prepared by deposition–precipitation,
a pre-treatment method is necessary for these sol-immobilised
materials. In order to activate the 1 wt% Au/TiO2 it is essential
to perform a heat pre-treatment procedure which removes the
PVA stabilising ligands which can potentially block the active sites
for CO oxidation [54–56]. We selected three different heat pre-
treatments, which have previously been shown to significantly
affect the activity and selectivity in benzyl alcohol oxidation for
monometallic gold and palladium catalysts [25]. These heat pre-
◦
treatments are: (a) heat pre-treatment at 250 C in air for 3 h,
◦
(b) heat pre-treatment at 250 C in nitrogen for 3 h and (c) heat
◦
pre-treatment at 250 C in hydrogen for 3 h. The highest catalyst
activity is observed when the heat pre-treatment is performed
under static air conditions, followed a pre-heat treatment in the
presence of hydrogen and nitrogen. Time-on-line analysis shows
that there is a progressive deactivation of the catalysts over the
first 4 h when treated in the presence of air and nitrogen, after
which the catalytic activity stabilizes at a steady value. It is note-

J.A. Lopez-Sanchez et al. / Applied Catalysis A: General 391 (2011) 400–406
405
Fig. 1. Effect of heat pre-treatment temperature on CO oxidation using the 1 wt%
◦
◦
Au/TiO2SI catalyst: (a) key: ꢀ dried at 120 C, (b) key: ᭹ heat pre-treatment at 250
C
◦
(
static air conditions), (c) key: ꢁ heat pre-treatment at 250 C (flow using H2) and
d) key: ꢂ heat pre-treatment at 250 C (flow using N2).
Fig. 3. Effect of calcination temperature on the catalytic performance of 5 wt%
Au/TiO2DPU (deposition–precipitation/urea): (a) key: ᭹ H2O2 productivity and (b)
key:ꢀ CO conversion.
◦
(
Fig. 2. CO oxidation using 1 wt% Au/TiO2SI, 1 wt% Pd/TiO2SI and 1 wt% (Au–Pd)/TiO2SI
◦
catalysts: (a) 1 wt% Au/TiO2SI key: ꢀ dried at 120 C, (b) 1 wt% Pd/TiO2SI key: ꢁ dried
◦
◦
at 120 C, (c) 1 wt% (Au–Pd)/TiO2SI key: ꢀ dried at 120 C and (d) key: ᭹ 1 wt%
Fig. 4. Effect of calcination temperature 2.5 wt% Au–2.5 wt% Pd/TiO2DPU
◦
(
Au–Pd)/TiO2SI, calcination treatment at 400 C.
(deposition–precipitation/urea): (a) key:
CO conversion.
2 2
ꢂ H O productivity and (b) key: ꢀ
worthy that the monometallic 1 wt% Au/TiO is always more active
2
than the bimetallic 1 wt% Au–Pd)/TiO , independent of the heat
In the case of the 2.5 wt% Au–2.5 wt% Pd/TiO2 catalysts the
variation of heat pre-treatment temperature results in a different
productivity versus calcinations temperature profile in comparison
to the monometallic 5 wt% Au/TiO2 catalysts, as shown in Fig. 4.
Increasing the heat pre-treatment temperature results in a signif-
2
pre-treatment temperature used, indicating that the addition of
palladium to gold by the sol-immobilisation method is not benefi-
cial for improving CO oxidation performance. In addition, the 1 wt%
Pd/TiO2 sol-immobilised catalyst was found to be almost inactive.
◦
icant decrease in the CO conversion, from 99% at 200 C to 10% at
◦
6
00 C. In contrast, the level of productivity for hydrogen peroxide
3.4.2. Mono and bimetallic-supported nanoparticles synthesised
◦
synthesis remains similar over the 200–500 C. However, a signifi-
cant 30% decrease is observed when the pre-treatment temperature
is increased to 600 C.
by DP using urea
The effect of pre-treatment temperature was studied for the
◦
5
wt% Au/TiO2 and 2.5 wt% Au–2.5 wt% Pd/TiO2 samples synthe-
These observations indicate that the CO oxidation reaction
is more sensitive than the hydrogen peroxide synthesis to the
variation of gold particle size, which increases with increasing pre-
treatment temperature. This relative insensitivity of the hydrogen
peroxide reaction can be understood since both the synthesis and
decomposition of hydrogen peroxide can be catalysed by the metal
nanoparticles in the catalyst, and any particle growth sintering that
occurs during heat treatment will affect the activity of the nanopar-
ticles for both steps; i.e. particle growth could decrease the number
sised by the deposition precipitation method using urea instead
of NaOH. Four different pre-treatment temperatures where tested,
namely: 200, 400, 500 and 600 C. The results of the hydrogen per-
◦
oxide synthesis and CO oxidation experiments are presented in
Fig. 3 for the monometallic 5 wt% Au/TiO2 catalysts. The highest
productivity and CO conversion was obtained for a pre-treatment
◦
temperature of 200 C. Increasing the pre-treatment temperature
beyond this leads to a progressive decrease in the CO conversion
◦
◦
(
98% at 200 C to 60% at 600 C) whereas the hydrogen perox-
of H O2 molecules synthesised to some extent over the tempera-
ide productivity is decreased slightly at the heat pre-treatment
2
◦
ture range but simultaneously retain productivity by decreasing the
temperature of 400 C and progressively declines as the heat
◦
number of sites responsible for the undesired H O decomposition.
pre-treatment temperature is increased to 500 C. Even higher
2 2
This contrasts with the simplicity of CO oxidation, where it is well
understood that very small gold nanoclusters can be active and that
activity tends to decrease with increasing particle sizes >2 nm.
heat pre-treatment temperatures do not affect the productivity of
hydrogen peroxide, which is in stark contrast to the situation for
CO conversion.

4
06
J.A. Lopez-Sanchez et al. / Applied Catalysis A: General 391 (2011) 400–406
4
. Conclusions
[15] K. Mori, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, J. Am. Chem. Soc. 126 (2004)
0657–10666.
16] H. Tsunoyama, H. Sakurai, Y. Negishi, T. Tsukuda, Journal of the American Chem-
ical Society 127 (2005) 9374–9375.
[17] A. Abad, C. Almela, A. Corma, H. Garcia, Tetrahedron 62 (2006) 6666–6672.
1
[
The oxidation of benzyl alcohol, hydrogen peroxide synthesis
and oxidation of CO have been investigated using three methods for
the synthesis of monometallic and bimetallic-supported nanopar-
ticles (a) impregnation, (b) a colloidal sol-immobilisation method
and (c) a deposition–precipitation method using urea. A strong syn-
ergistic effect is evident with the addition of Pd to Au, especially
in the case in the synthesis of hydrogen peroxide. For bimetal-
lic nanoparticles, titania is an inferior support to carbon for the
oxidation of benzyl alcohol reaction and hydrogen peroxide syn-
thesis. In addition, the sol-immobilised supported catalysts were
tested for CO oxidation and it was observed that it is possible to
activate the supported colloidal catalysts after a suitable thermal
treatment to partially remove protective PVA ligands. For CO oxida-
tion, the addition of palladium to gold is not beneficial in terms of
catalyst activity. We also report the use a deposition–precipitation
method using urea for the synthesis of gold and gold–palladium
supported nanoparticles and show that by choosing an appropri-
ate heat pre-treatment procedure it is possible to obtain very active
catalysts for CO oxidation and as well as for hydrogen peroxide
synthesis and benzyl alcohol oxidation. For this latter technique
gold–palladium deposited on a zeolite H-ZSM5 support gave a cat-
alyst that had the same product distribution but higher activities
than their titania-supported analogues. Finally, the effect of heat
treatment showed that CO oxidation is more sensitive to the pre-
cise heat treatment temperature than hydrogen peroxide synthesis,
suggesting that the CO oxidation is a reaction that is significantly
more sensitive to the particle size than the other reactions we have
investigated.
[
[
[
18] A. Biffis, S. Cunial, P. Spontoni, L. Prati, J. Catal. 251 (2007) 1–6.
19] P. Haider, A. Baiker, J. Catal. 248 (2007) 175–187.
20] J. Chen, Q. Zhang, Y. Wang, H. Wan, Adv. Synth. Catal. 350 (2008) 453–464.
[21] M. Schrinner, S. Proch, Y. Mei, R. Kempe, N. Miyajima, M. Ballauff, Adv. Mater.
20 (2008) 1928–1933.
[
22] C. Lucchesi, T. Inasaki, H. Miyamura, R. Matsubara, S. Kobayashi, Adv. Synth.
Catal. 350 (2008) 1996–2000.
[23] C.D. Pina, E. Falletta, M. Rossi, J. Catal. 260 (2008) 384–386.
[24] N. Dimitratos, J.A. Lopez-Sanchez, D. Lennon, F. Porta, L. Prati, A. Villa, Catal.
Lett. 108 (2006) 147–153.
[
25] N. Dimitratos, J.A. Lopez-Sanchez, D. Morgan, A.F. Carley, L. Prati, G.J. Hutchings,
Catal. Today 122 (2007) 317–324.
[26] W. Hou, N.A. Dehm, R.W. Scott, J. Catal. 253 (2008) 22–27.
[
[
27] T. Ishida, M. Nagaoka, T. Akita, M. Haruta, Chem. Eur. J. 14 (2008) 8456–8460.
28] M.D. Hughes, Y.-J. Xu, P. Jenkins, P. McMorn, P. Landon, D.I. Enache, A.F. Carley,
G.A. Attard, G.J. Hutchings, F. King, E.H. Stitt, P. Johnston, K. Griffin, C.J. Kiely,
Nature 437 (2005) 1132–1135.
[
[
[
[
29] A. Abad, P. Conception, A. Corma, H. Garcia, Angew. Chem. Int. Ed. 44 (2005)
4066–4069.
30] T. Ishikara, Y. Ohura, S. Yoshida, Y. Hata, H. Nishiguchi, Y. Takita, Appl. Catal. A:
Gen. 291 (2005) 215–221.
31] C. Burato, P. Centomo, M. Rizzoli, A. Biffis, S. Campestrini, B. Corain, Adv. Synth.
Catal. 348 (2006) 255–259.
32] Y. Nomura, T. Ishihara, Y. Hata, K. Kitawaki, K. Kaneko, H. Matsumoto, Chem-
SusChem 1 (2008) 619–621.
[33] J.K. Edwards, G.J. Hutchings, Angew. Chem. Int. Ed. 47 (2008) 9192–9198.
[
[
34] A. Corma, P. Serna, Science 313 (2006) 332–334.
35] D.I. Enache, J.K. Edwards, P. Landon, B. Solsona-Espriu, A.F. Carley, A.A. Herz-
ing, M. Watanabe, C.J. Kiely, D.W. Knight, G.J. Hutchings, Science 311 (2006)
362–365.
[
36] J.A. Lopez-Sanchez, N. Dimitratos, P. Miedziak, E. Ntainjua, J.K. Edwards, D. Mor-
gan, A.F. Carley, R. Tiruvalam, C.J. Kiely, G.J. Hutchings, Phys. Chem. Chem. Phys.
1
0 (2008) 1921–1930.
[37] N. Dimitratos, J.A. Lopez-Sanchez, G.J. Hutchings, Top. Catal. 52 (2009) 258–
68.
2
[
38] 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.
Acknowledgements
[
[
[
[
39] N. Dimitratos, J.A. Lopez-Sanchez, J.M. Anthonykutty, G. Brett, A.F. Carley, S.H.
Taylor, D.W. Knight, G.J. Hutchings, Green Chem. 11 (2009) 1209–1216.
40] S. Bawaked, N.F. Dummer, N. Dimitratos, D. Bethell, Q. He, C.J. Kiely, G.J. Hutch-
ings, Green Chem. 11 (2009) 1037–1044.
41] J.K. Edwards, B. Solsona, E. Ntainjua, A.F. Carley, A.A. Herzing, C.J. Kiely, G.J.
Hutchings, Science 323 (2009) 1037–1041.
We acknowledge the financial support of the Engineering
and Physical Sciences Research Council and the European Union
(
project AURICAT).
References
42] N. Dimitratos, J.A. Lopez-Sanchez, D. Morgan, A.F. Carley, R. Tiruvalam, C.J. Kiely,
D. Bethell, G.J. Hutchings, Phys. Chem. Chem. Phys. 11 (2009) 5142–5153.
[
[
[
[
[
[
[
[
1] A.S.K. Hashmi, G.J. Hutchings, Angew. Chem. Int. Ed. 45 (2006) 7896–7936.
2] G.C. Bond, D.T. Thompson, Catal. Rev. Sci. Eng. 41 (1999) 319–388.
3] M. Haruta, Gold Bull. 37 (2004) 27–36.
[43] D.I. Enache, D. Barker, J.K. Edwards, S.H. Taylor, D.W. Knight, A.F. Carley, G.J.
Hutchings, Catal. Today (2007) 407–411.
[44] A. Abad, C. Almela, A. Corma, H. Garcia, Chem. Commun. (2006) 3178–3180.
[45] T. Mallat, A. Baiker, Chem. Rev. 104 (2004) 3037–3058.
[46] C. Keresszegi, T. Bürgi, T. Mallat, A. Baiker, J. Catal. 211 (2002) 244–251.
[49] J.W. Geus, A.J. van Dillen, in: G. Ert, et al. (Eds.), Preparation of Solid Catalysts,
Wiley-VCH, Weinheim, Germany, 1999, pp. 460–487.
[50] M. Haruta, CATTECH 6 (2002) 102–115.
4] A.S.K. Hashmi, Gold Bull. 37 (2004) 51–65.
5] M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. 16 (1987) 405–408.
6] F. Moreau, G.C. Bond, A.O. Taylor, Chem. Commun. (2004) 1642–1643.
7] F. Moreau, G.C. Bond, A.O. Taylor, J. Catal. 231 (2005) 105–114.
8] W. Yan, B. Chen, S.M. Mahurin, S. Dai, S.H. Overbury, Chem. Commun. (2004)
1
918–1919.
[51] R. Zanella, C. Louis, Catal. Today 107–108 (2005) 768–777.
[52] R. Zanella, S. Giorgio, C.-H. Shin, C.R. Henry, C. Louis, J. Catal. 222 (2004)
357–367.
[
9] S. Carrettin, P. Concepción, A. Corma, J.M. López Nieto, V.F. Puntes, Angew.
Chem. Int. Ed. 43 (2004) 2538–2540.
[
[
10] J. Guzman, B.C. Gates, J. Am. Chem. Soc. 126 (2004) 2672–2673.
11] P. Mohapadra, J. Moma, K.M. Parida, W.A. Jordaan, M.S. Scurrell, Chem. Com-
mun. (2007) 1044–1046.
[53] N. Dimitratos, A. Villa, C.L. Bianchi, L. Prati, M. Makkee, Appl. Catal. A: Gen. 311
(2006) 185–192.
[54] M. Comotti, W.-C. Li, B. Spliethoff, F. Schüth, J. Am. Chem. Soc. 128 (2006)
917–924.
[12] H.-L. Jiang, B. Liu, T. Akita, M. Haruta, H. Sakurai, Q. Xu, J. Am. Chem. Soc. 131
(
2009) 11302–11303.
[55] J.-D. Grunwaldt, C. Kiener, C. Wögerbauer, A. Baiker, J. Catal. 181 (1999)
223–232.
[
[
13] G.J. Hutchings, J. Catal. 96 (1985) 292–295.
14] A.K. Sinha, S. Seelan, S. Tsubota, M. Haruta, Angew. Chem. Int. Ed. 43 (2004)
[56] J.-D. Grunwaldt, M. Maciejewski, O.S. Becker, P. Fabrizioli, A. Baiker, Journal of
Catalysis 186 (1999) 458–469.
1
546–1548.