Au-Pd supported nanocrystals prepared by a sol immobilisation technique as catalysts for selective chemical synthesis

Article abstract of DOI:10.1039/b719345a

Catalysis by gold and gold-palladium nanoparticles has attracted significant research attention in recent years. These nanocrystalline materials have been found to be highly effective for selective and total oxidation, but in most cases the catalysts are prepared using precipitation or impregnation. We report the preparation of Au-Pd nanocrystalline catalysts supported on carbon prepared via a sol-immobilisation technique and these have been compared with Au-Pd catalysts prepared via impregnation. The catalysts have been evaluated for two selective chemical syntheses, namely, oxidation of benzyl alcohol and the direct synthesis of hydrogen peroxide. The catalysts have been structurally characterised using a combination of scanning transmission electron microscopy and X-ray photoelectron spectroscopy. The catalysts prepared using the sol immobilisation technique show higher activity when compared with catalysts prepared by impregnation as they are more active for both hydrogen peroxide synthesis and hydrogenation, and also for benzyl alcohol oxidation. The method facilitates the use of much lower metal concentrations which is a key feature in catalyst design, particularly for the synthesis of hydrogen peroxide. the Owner Societies.

Full text of DOI:10.1039/b719345a

View Article Online / Journal Homepage / Table of Contents for this issue
Physical Chemistry Chemical Physics
This article is part of a PCCP themed issue on:
Analysis, manipulation, and simulation on the nanoscale
Guest Editors: Harald Fuchs, Dominik Marx and Ulrich Simon
Other papers in the issue:
Textile electrodes as substrates for the electrodeposition of porous ZnO
T. Loewenstein, A. Hastall, M. Mingebach, Y. Zimmermann, A. Neudeck and D. Schlettwein,
Phys. Chem. Chem. Phys., 2008, 10, 1844 (DOI: 10.1039/b719691a)
Fluorescence resonance energy transfer in conjugated polymer composites for radiation detection
Y. S. Zhao, H. Zhong and Q. Pei, Phys. Chem. Chem. Phys., 2008, 10, 1848 (DOI: 10.1039/b801375f)
Simulating charge transport in tris(8-hydroxyquinoline) aluminium (Alq3)
J. J. Kwiatkowski, J. Nelson, H. Li, J. L. Bredas, W. Wenzel and C. Lennartz,
Phys. Chem. Chem. Phys., 2008, 10, 1852 (DOI: 10.1039/b719592c)
Functional molecular wires
G. J. Ashwell, P. Wierzchowiec, L. J. Phillips, C. J. Collins, J. Gigon, B. J. Robinson, C. M. Finch, I. R. Grace, C. J. Lambert, P. D.
Buckle, K. Ford, B. J. Wood and I. R. Gentle, Phys. Chem. Chem. Phys., 2008, 10, 1859 (DOI: 10.1039/b719417j)
Forced imbibition—a tool for separate determination of Laplace pressure and drag force in capillary filling experiments
D. I. Dimitrov, A. Milchev and K. Binder, Phys. Chem. Chem. Phys., 2008, 10, 1867 (DOI: 10.1039/b719248g)
Assembly of DNA-functionalized gold nanoparticles studied by UV/Vis-spectroscopy and dynamic light scattering
K. G. Witten, J. C. Bretschneider, T. Eckert, W. Richtering and U. Simon,
Phys. Chem. Chem. Phys., 2008, 10, 1870, (DOI: 10.1039/b719762d)
Metal adsorption on oxide polar ultrathin films
G. Barcaro, A. Fortunelli and G. Granozzi, Phys. Chem. Chem. Phys., 2008, 10, 1876 (DOI: 10.1039/b719346g)
Microstructure analysis of monodisperse ferrofluid monolayers: theory and simulation
S. Kantorovich, J. J. Cerdà and C. Holm, Phys. Chem. Chem. Phys., 2008, 10, 1883 (DOI: 10.1039/b719460a)
Molecular simulation study of temperature effect on ionic hydration in carbon nanotubes
Q. Shao, L. Huang, J. Zhou, L. Lu, L. Zhang, X. Lu, S. Jiang, K. E. Gubbins and W. Shen,
Phys. Chem. Chem. Phys., 2008, 10, 1896 (DOI: 10.1039/b719033f)
Adsorption of small organic molecules on anatase and rutile surfaces: a theoretical study
S. Köppen and W. Langel, Phys. Chem. Chem. Phys., 2008, 10, 1907 (DOI: 10.1039/b719098k)
Electron-induced isomerisation of dichlorobenzene on Cu(111) and Ag(111)
V. Simic-Milosevic, J. Meyer and K. Morgenstern, Phys. Chem. Chem. Phys., 2008, 10, 1916 (DOI: 10.1039/b718847a)
Au–Pd supported nanocrystals prepared by a sol immobilisation technique as catalysts for selective chemical synthesis
J. A. Lopez-Sanchez, N. Dimitratos, P. Miedziak, E. Ntainjua, J. K. Edwards, D. Morgan, A. F. Carley, R. Tiruvalam, C. J. Kiely and G. J.
Hutchings, Phys. Chem. Chem. Phys., 2008, 10, 1921 (DOI: 10.1039/b719345a)
Transport effects in the oxygen reduction reaction on nanostructured, planar glassy carbon supported Pt/GC model electrodes
A. Schneider, L. Colmenares, Y. E. Seidel, Z. Jusys, B. Wickman, B. Kasemo and R. J. Behm,
Phys. Chem. Chem. Phys., 2008, 10, 1931 (DOI: 10.1039/b719775f)
Structure, optical properties and defects in nitride (III–V) nanoscale cage clusters
S. A. Shevlin, Z. X. Guo, H. J. J. van Dam, P. Sherwood, C. R. A. Catlow, A. A. Sokol and S. M. Woodley,
Phys. Chem. Chem. Phys., 2008, 10, 1944 (DOI: 10.1039/b719838h)
Adsorption of oxalate on rutile particles in aqueous solutions: a spectroscopic, electron-microscopic and theoretical study
C. B. Mendive, T. Bredow, A. Feldhoff, M. Blesa and D. Bahnemann,
Phys. Chem. Chem. Phys., 2008, 10, 1960 (DOI: 10.1039/b800140p)
Theoretical evaluation of nano- or microparticulate contact angle at fluid/fluid interfaces: analysis of the excluded area behavior
upon compression
D. O. Grigoriev, H. Möhwald and D. G. Shchukin, Phys. Chem. Chem. Phys., 2008, 10, 1975 (DOI: 10.1039/b719140e)

PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Au–Pd supported nanocrystals prepared by a sol immobilisation
technique as catalysts for selective chemical synthesisw
Jose Antonio Lopez-Sanchez,a Nikolaos Dimitratos,a Peter Miedziak,a
Edwin Ntainjua,a Jennifer K. Edwards,a David Morgan,a Albert F. Carley,a
Ramchandra Tiruvalam,b Christopher J. Kielyb and Graham J. Hutchings*a
Received 14th December 2007, Accepted 23rd January 2008
First published as an Advance Article on the web 18th February 2008
DOI: 10.1039/b719345a
Catalysis by gold and gold–palladium nanoparticles has attracted significant research attention in
recent years. These nanocrystalline materials have been found to be highly effective for selective
and total oxidation, but in most cases the catalysts are prepared using precipitation or
impregnation. We report the preparation of Au–Pd nanocrystalline catalysts supported on carbon
prepared via a sol-immobilisation technique and these have been compared with Au–Pd catalysts
prepared via impregnation. The catalysts have been evaluated for two selective chemical syntheses,
namely, oxidation of benzyl alcohol and the direct synthesis of hydrogen peroxide. The catalysts
have been structurally characterised using a combination of scanning transmission electron
microscopy and X-ray photoelectron spectroscopy. The catalysts prepared using the sol
immobilisation technique show higher activity when compared with catalysts prepared by
impregnation as they are more active for both hydrogen peroxide synthesis and hydrogenation,
and also for benzyl alcohol oxidation. The method facilitates the use of much lower metal
concentrations which is a key feature in catalyst design, particularly for the synthesis of hydrogen
peroxide.
found to be effective for the selective oxidation of alkenes16,17
and alcohols,18 and for the selective hydrogenation of un-
saturated carbonyl compounds and nitro groups.19
Introduction
At present there is a fascination with the design of nanopar-
ticles and nanostructured materials and developing improved
routes for the preparation of these materials is the focus of
much current research, particularly for gold nanoparticles.1
The reason for gold being of particular interest is two-fold.
Firstly, from a preparative point of view, gold is the most
stable noble metal at the nanoscale and has an extensive
surface chemistry. This has provided the impetus to devise
improved preparation routes for gold nanoparticles, and there
is considerable interest in the fabrication of ligand protected
gold nanoparticles.1–5 Secondly, the recent discovery of the
catalytic efficacy of gold nanocrystals for oxidation and
reduction reactions has been a major driving force for research
into novel catalytic applications of gold nanoparticles. This
has enabled a new approach in the search for selective redox
catalysts.6–12 Supported gold catalysts have been shown to be
effective for low temperature oxidation of CO,13 even in the
presence of H2, H2O and CO2 and this reaction is finding some
potential application in fuel cells.14,15 Gold has also been
Although there is a rich catalytic chemistry unfolding for
pure gold nanoparticles, we have recently shown that the
combination of gold and palladium in an alloy nanoparticle
configuration leads to enhanced catalytic activity and selectivity
in redox processes. In particular, we have shown that although
gold alone is a very effective catalyst for the selective oxidation
of an alcohol to an aldehyde under solvent free conditions,18
the alloying of gold with palladium leads to a twenty five-fold
enhancement in activity with simultaneous retention of selec-
tivity.20 While palladium alone is an exceptionally effective
catalyst for the direct hydrogenation of molecular oxygen to
produce hydrogen peroxide,21 the alloying of gold with the
palladium significantly enhances the activity and selectivity for
this challenging reaction.22–27 These two specific reactions
provide the focus for the present study since they are repre-
sentative examples of selective oxidation and hydrogenation
that are of key importance in selective chemical synthesis.
The oxidation of primary alcohols to aldehydes is an
important laboratory and commercial procedure.28–30 Alde-
hydes are valuable both as intermediates and as high value
components for the perfume industry.28 Often oxidations of
this type are carried out using stoichiometric oxygen donors
such as chromate or permanganate, but these reagents are
expensive and have serious toxicity and environmental issues
associated with them.28 Given these limitations, there is sub-
stantial interest in the development of heterogeneous catalysts
that use molecular O2 as the oxidant. Au nanocrystals have
been shown to be highly effective for the oxidation of alcohols
a School of Chemistry, Cardiff University, Main Building, Park Place,
Cardiff, UK CF10 3AT. E-mail: hutch@cardiff.ac.uk; Fax: +44 29
2087 4059; Tel: +44 29 2087 4059
b Center for Advanced Materials and Nanotechnology, Lehigh
University, 5 East Packer Avenue, Bethlehem, PA 18015-3195, USA
w Electronic supplementary information (ESI) available: Bright field
micrograph of the sol-immobilized AuPd/C sample after use as a
catalyst (Fig. S1); bright field micrograph and corresponding histo-
gram of the particle size distribution of the AuPd/C sol-immobilized
sample after calcination at 400 1C (Fig. S2). See DOI: 10.1039/
b719345a
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 1921–1930 | 1921

with O2 in an aqueous base, in particular diols and triols;
although under these conditions, the product is the corre-
sponding mono-acid, not the aldehyde.31–33 However, gold in
the absence of base has recently been shown to be highly
effective for the oxidation of alcohols.12
dried at 110 1C for 16 h and calcined in static air, typically at
400 1C for 3 h.
Sol-immobilisation method. An aqueous solution of PdCl2
(Johnson Matthey) and HAuCl4Á3H2O of the desired concen-
tration was prepared. Polyvinylalcohol (PVA) (1 wt% solu-
tion, Aldrich, MW = 10 000, 80% hydrolyzed) was added
(PVA/Au (by wt) = 1.2); a 0.1 M freshly prepared solution of
NaBH4 (496%, Aldrich, NaBH4/Au (mol/mol) = 5) was
then added to form a dark-brown sol. After 30 min of sol
generation, the colloid was immobilized by adding activated
carbon (acidified at pH 1 by sulfuric acid) under vigorous
stirring conditions. The amount of support material 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
thoroughly with distilled water (neutral mother liquors) and
dried at 120 1C overnight. Two catalysts were prepared using a
1 : 1 molar ratio of Au : Pd (denoted Au–Pd/CSIm) and a 1 : 1
weight ratio of Au : Pd (denoted Au–Pd/CSIw). Monometallic
catalysts containing gold or palladium were also prepared
using similar methodology, and these are denoted Au/CSI
and Pd/CSI.
The direct hydrogenation of O2 to give hydrogen peroxide
remains one of the grand challenges in catalytic chemistry. At
present this important commodity chemical which is produced
on a major scale, ca. 2 M tons per annum, is used mainly as a
bleach and disinfectant. It is manufactured using an indirect
method which is based on the sequential hydrogenation and
oxidation of an alkyl anthraquinone,34 that avoids molecular
hydrogen and oxygen coming into direct contact. However, a
direct reaction could be preferable to the current indirect
process. Indeed, there is a recent report35 that a direct process
using a Pd-based catalyst is being commercialised, which
represents a major step forward in the manufacture of this
important material. The direct reaction between hydrogen and
oxygen has been a research target for many years, with the first
reported study in 1914,36 using a Pd catalyst. Since then, there
have been a number of investigations37–49 emanating mainly
from industrial laboratories and until recently all these catalyst
design studies have been based on supported palladium cata-
lysts. We have recently shown that the addition of gold to the
palladium enhances the formation of hydrogen peroxide in the
direct synthesis process.22–27
Catalyst testing
Benzyl alcohol oxidation. The benzyl alcohol oxidation
reactions were carried out in a stirred 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 5 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 temperature.
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 samples. For the quantifi-
cation of the amounts of reactants consumed and products
generated, the external calibration method was used.
In our previous work, where we demonstrated that the
combination of gold and palladium provides a pronounced
synergistic effect in redox catalysis,22–27 we have prepared the
catalysts via an impregnation procedure in which 5 wt% total
metal loading was utilized. It could, however, be economically
and practically beneficial to utilise lower concentrations of
gold and palladium since this approach may afford improved
dispersions of the nanocrystals. In this paper we investigate the
use of lower concentrations of these precious metals using a
sol-immobilisation method of preparation. We have recently
shown that this methodology is effective for the synthesis of
monometallic gold and palladium catalysts50 and we now
investigate the applicability of this method for generating
gold–palladium alloy nanoparticles supported on carbon.
Hydrogen peroxide synthesis and hydrogenation. Catalyst
testing was performed using a stainless steel autoclave (Parr
Instruments) with a nominal volume of 50 ml and a maximum
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 (5.6 g MeOH and 2.9 g H2O), purged three times with
5% H2/CO2 (3 MPa) and then filled with 5% H2/CO2 and
25% O2/CO2 to give a hydrogen 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
1C), and all experiments were carried out for 30 min unless
otherwise stated. Gas analysis for H2 and O2 was performed
by gas chromatography using a thermal conductivity detector
and a CP À Carboplot P7 column (25 m, 0.53 mm id). H2O2
yield was determined by titration of aliquots of the final
filtered solution with acidified Ce(SO4)2 (7 Â 10À3 mol lÀ1).
Ce(SO4)2 solutions were standardised against (NH4)2-
Fe(SO4)2Á6H2O using ferroin as indicator. Hydrogen peroxide
Experimental
Catalyst preparation
The following notation is used for the catalyst samples we
have prepared: I denotes impregnation, SI denotes sol-immo-
bilisation, w denotes Au and Pd are present as 1 : 1 by weight,
and m denotes Au and Pd are present as 1 : 1 molar ratio.
Impregnation method. 1 wt% Pd-only, 1 wt% Au-only and
Au–Pd bimetallic catalysts were prepared by impregnation of
carbon (Darco G60, Aldrich), via an impregnation method
using aqueous solutions of PdCl2 (Johnson Matthey) and/or
HAuCl4Á3H2O (Johnson Matthey). For the 0.5%Au–0.5%Pd/
carbon catalyst, the detailed preparation procedure employed
is described below. An aqueous solution of HAuCl4Á3H2O
(2 ml, 5 g dissolved in water (250 ml)) and an aqueous solution
of PdCl2 (0.83 ml, 1 g in water (25 ml)) were simultaneously
added to carbon (3.8 g). The paste formed was ground and
ꢀc
This journal is the Owner Societies 2008
1922 | Phys. Chem. Chem. Phys., 2008, 10, 1921–1930

hydrogenation was carried out in an identical manner but in
this case the 25% O2/CO2 was not added so that hydrogen
peroxide could not be synthesised, but part of the water was
replaced by hydrogen peroxide (50 vol%) to give a hydrogen
peroxide concentration of 4 wt%.
Catalyst characterisation
The sols were characterized using a UV spectrometer (V-570,
JASCO) in H2O between 200 and 1200 nm, in a quartz cuvette.
Samples for examination by analytical electron microscopy
(AEM) were prepared by dispersing the catalyst powder in
high purity ethanol, then allowing a drop of the suspension to
evaporate on a holey carbon film supported by a 300 mesh
copper TEM grid. Samples were then subjected to chemical
microanalysis and annular dark-field imaging in a VG Systems
HB603 scanning transmission electron microscope (STEM)
operating at 300 kV equipped with a Nion Cs corrector. This
instrument was also fitted with an Oxford Instruments INCA
TEM 300 system for energy dispersive X-ray (XEDS) analysis.
Bright field diffraction contrast imaging experiments were
carried out on a JEOL 2000FX TEM operating at 200 kV.
X-Ray photoelectron spectra were recorded on a Kratos
Axis Ultra DLD spectrometer employing a monochromatic
AlKa X-ray source (75–150 W) and analyser pass energies of
160 eV (for survey scans) or 40 eV (for detailed scans).
Samples were mounted using a double-sided adhesive tape
and binding energies referenced to the C(1s) binding energy of
adventitious carbon contamination which was taken to be
284.7 eV.
Fig. 1 UV-Vis spectra of sol-immobilised catalysts. Key: 1 wt%Au/
CSI (———), 1 wt% Pd/CSI (Á Á ÁÁ Á ÁÁ Á Á) and 1 wt%Au–Pd/CSIm (TTTT).
+
Pd-NH3 resulted in the disappearance of the gold plasmon
resonance.56
XPS characterization. Fig. 2 shows the Au(4f) and Pd(3d)
spectra for the dried and calcined 1 wt% AuPd (1 : 1 by wt)/C
catalysts, prepared by both sol-immobilisation and impregna-
tion. For all the samples the Au(4f) spectra (Fig. 2a) show no
change in the oxidation state, and the binding energy is
indicative of the presence of metallic gold. In contrast, the
Pd(3d) spectra (Fig. 2b) show significant changes on calcina-
tion. The uncalcined samples exhibit predominantly Pd1 spe-
cies which on calcination transform to mainly Pd2+, but with
a small residual Pd1 component. The dramatic decrease in the
Au(4f) and Pd(3d) spectral intensities for the impregnated
samples as compared with the sol immobilization prepared
samples, as evident in Fig. 2, is reflected in the calculated
molar elemental ratios, as discussed later (Table 3). In order to
calculate the molar ratios, the Pd(3d) intensity is first be
corrected for the presence of an overlapping Au(4d5/2) com-
ponent, by subtracting an intensity calculated from that of the
well-resolved Au(4d3/2) peak.
Results and discussion
Preparation and characterisation of the sols and catalysts
Characterisation of the sols. UV-Vis spectra (200–900 nm) of
the Au and Pd sols were recorded in H2O. For the Au sol the
plasmon resonance band was located at 505 nm (Fig. 1a). This
band is characteristic of the plasmon resonance band of gold
nanoparticles having a size of o10 nm.51,52 In contrast to
gold, Pd metallic sols display no surface plasmon band
(Fig. 1b);53 therefore, we observe the disappearance of the
precursor peak after reduction (lmax = 210 and 238 nm) as has
previously been reported.51 The spectra resulting from the
mixture of the two metal sols indicates the disappearance of
the gold surface plasmon band as previously reported for this
system.51 This is a phenomena commonly found in the for-
mation of sols of bimetallic nanoparticles where one of the
component metals lacks a surface plasmon band. An example
of this phenomena is demonstrated in the formation of Ag–Rh
nanoparticles where the effect was attributed to the sponta-
neous formation of Ag-core/Rh-shell particles.54 In the parti-
cular case of gold and palladium, Deki et al.55 dispersed
nanoparticles of several Au–Pd compositions on a polymer
thin film matrix and concluded that the progressive decrease of
the gold plasmon band in the UV-Vis spectra observed by
increasing the palladium content was the result of changes in
the band structure of the Au particles due to alloying with Pd.
Also the formation of nanocomposites from Au-COOÀ and
Electron microscopy characterization. The size distribution
of the supported metal nanoparticles for both the impregnated
and sol-immobilized AuPd/C samples was assessed from
analysis of bright field TEM micrographs such as those
presented in Fig. 3a and c, respectively. In the sample made
via the impregnation route, the particle size ranged between 2
and 14 nm, with the average size being around 6 nm (Fig. 3b).
Occasional larger particles in excess of 25 nm were also seen in
this material, although these were relatively rare. By compar-
ison, the metal particle size distribution in the catalyst pre-
pared by sol-immobilization was found to be much narrower,
i.e. between 4–7 nm, as shown in Fig. 3d.
The compositions of the metal particles in these two materi-
als were monitored by STEM-XEDS (i.e. X-ray energy dis-
persive analysis) spectrum imaging. A representative annular
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 1921–1930 | 1923

encountered during the experiment due to the degradation of
the PVA ligands bound to particle surface under the high
fluence of electrons in the incident beam. Even so, we can
confirm that the relatively monodisperse particles prepared by
the colloidal route are all bimetallic Au–Pd alloys. A signifi-
cant difference from the impregnated samples, however, can be
seen by comparing the relative intensities of the Au La1 and Pd
La1 signals from particle-to-particle. If one compares the
neighbouring pairs of particles labelled ‘a’ and ‘b’ or ‘c’ and
‘d’ for instance, it is clear that the composition is quite variable
(e.g. particle ‘b’ is considerably more Pd-rich than particle ‘a’).
Hence we conclude that even though the colloidal sol-immo-
bilisation route clearly produces a narrower size distribution
of particles, the degree of control of particle composition
within a particular size interval is not as good as that for the
impregnated sample.
Evaluation of the Au–Pd/C catalysts for selective chemical
synthesis
Benzyl alcohol oxidation. The oxidation of benzyl alcohol is
often used as a model reaction for alcohol oxidation, not only
due to its relatively high reactivity but also because of the
existence of a complex reaction network (Fig. 6). The target
products are benzaldehyde and benzoic acid, but other by-
products e.g. toluene, benzene, benzyl benzoate and acetal can
form depending on the reaction conditions and metal catalyst
employed. These by-products are likely to be the outcome of
side reactions due to hydrogenolysis (toluene formation),
decarbonylation (benzene) and esterification reaction path-
ways (benzyl benzoate) (see Fig. 6).
Fig. 2 Au(4f) spectra (plot a) and Pd(3d) spectra (plot b) for 1 wt%
Au–Pd/C catalysts (Au : Pd = 1 : 1 by wt) prepared by sol immobilisa-
tion [(a) dried and (b) calcined at 400 1C] and impregnation [(c) dried
and (d) calcined at 400 1C].
The catalytic data for Au–Pd/CSIm, Au/CSI and Pd/CSI are
displayed in Fig. 7a and Tables 1 and 2. A comparison
between the monometallic catalysts (Au, Pd) supported on
carbon indicates that these two catalysts gave very similar
activities and selectivities (Table 1). Their selectivities at the
same level of conversion (50%) were also found to be similar
(Table 2). Comparison at iso-conversion permits a direct and
meaningful comparison between the product distributions
obtained with each catalyst. The bimetallic Au–Pd/CSIm cat-
alyst was significantly more active and gave higher yields of
benzaldehyde. In particular, the turnover frequency (mol
benzyl alcohol oxidized per mol of metal) was markedly higher
for the Au–Pd catalyst. However, comparison at iso-conver-
sion (Table 2) showed that the bimetallic catalyst also gave
significantly higher yields of toluene formed by hydrogenoly-
sis. It is interesting to compare the activity of the bimetallic
catalyst prepared by sol-immobilisation and similar catalysts
prepared by impregnation (Fig. 7b and 7c, Tables 1 and 2). It
is apparent that the catalysts prepared by the sol immobiliza-
tion procedure are more active for benzyl alcohol oxidation to
benzaldehyde when compared with catalysts prepared using
the impregnation procedure. Analysis by TEM confirmed that
the particle size distribution for the Au–Pd/CSIm sample did
not change after reaction (see Fig. S1, ESI).w
dark field (ADF) image and the corresponding Au La1 and Pd
La1 elemental maps from the impregnated AuPd/C sample are
shown in Fig. 4. It is clear that there is a direct spatial
correlation between the Au and Pd signals confirming that
all the metal particles are in fact Au–Pd alloys. It is not
possible from our present analysis to discern whether the
Au–Pd particles are random alloys or core–shell in nature. It
is clear however from comparison of the relative Pd and Au
signal intensities that the particle compositions within the 4–8
nm size range shown here are all fairly consistent, with the
particles being slightly Pd-rich in all cases. This is entirely
consistent with previous studies on materials prepared in a
similar manner where the AuPd composition as a function of
particle size was measured by STEM-XEDS.57 In this previous
work we consistently found that large (B50 nm) particles were
Au–rich, intermediate size particles (B15 nm) were closer to
the nominal 1 : 1 at% composition, and the very smallest
(B5 nm) particles were always Pd-rich. For comparison, a
representative ADF image and the corresponding Au La1 and
Pd La1 elemental maps from the area delineated by the box are
presented for the sol immobilized AuPd/C samples shown in
Fig. 5. The quality of the X-ray mapping data is not as good in
this instance because of the significant contamination we
In Table 3 we report elemental atomic ratios and total metal
loadings derived from XPS analysis of these three samples and
also two 1 wt% AuPd/C catalysts prepared by impregnation
and corresponding to molar and weight ratios of Au : Pd =
1 : 1. The Pd : Au molar ratio for the bimetallic catalyst
ꢀc
This journal is the Owner Societies 2008
1924 | Phys. Chem. Chem. Phys., 2008, 10, 1921–1930

Fig. 3 Bright field TEM micrographs and the corresponding histograms of the particle size distributions for the AuPd/C samples prepared by
impregnation ((a) and (b)) and sol-immobilization ((c) and (d)).
Au–Pd/CSIm (0.81) is reasonably close to the expected value of
1.0, indicating that there are few particles present exhibiting
the Au-core Pd-shell morphology observed in several other
systems involving supported Au–Pd nanoparticles,22–26 nota-
bly oxide supported systems, but this observation is in agree-
ment with our recent observations for carbon-supported
Au–Pd nanoparticles prepared by impregnation.27 Since there
is a significant synergistic effect on catalytic activity for the
Au–Pd sample compared to the mono-metallic catalysts
(Table 1), a significant fraction of the particles must be present
as Au–Pd alloys, and this is in agreement with the observa-
tions made by electron microscopy. The impregnated samples
show much higher Pd : Au ratios (by a factor 42) than for the
samples prepared by sol immobilization (Table 3), consistent
with either the development of Pd-rich shell/Au-rich core
particles or a significantly higher Pd dispersion relative to
Au compared with the SI catalysts. Elemental analysis of the
1 wt% catalysts prepared by either impregnation or sol-
Fig. 4 STEM–ADF image and the corresponding Au La1 and Pd La1 elemental maps of the AuPd/C catalyst prepared by the impregnation
method.
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 1921–1930 | 1925

therefore be much higher for the samples prepared by sol-
immobilisation. Differences in particle size are not likely to
make a large contribution to the differences in apparent metal
loading.
AuPd/CSIw was also investigated for benzyl alcohol conver-
sion and the results are presented in Fig. 8a. These data show
significant differences from those for AuPd/CSIm, AuPd/CSIw
being much more active and, when the data are compared at
iso-conversion it is apparent that the selectivities are very
similar (Table 4). We also evaluated Au–Pd/CIm at 120 1C
which was again found to be less active than Au–Pd/CSIw
(Fig. 8b) and we observed a TOF of 35 400 hÀ1 for AuPd/CSIw
and 982 for AuPd/CIm. The XPS derived Pd : Au molar ratio is
1.47 in this instance as compared with the expected value of
1.86. Both AuPd samples show a somewhat lower Pd content
than expected, possibly reflecting a larger average particle size
for Pd.
As noted earlier, the catalysts used in our earlier studies22–27
prepared using impregnation were calcined at 400 1C prior to
use. Hence, we have calcined AuPd/CSIw and have compared
this with catalysts prepared by impregnation, to provide a
comparison for the two methods of catalyst synthesis, and the
data are shown in Table 1. It is very clear that calcination leads
to a decrease in catalyst activity, and the catalyst made by the
sol-immobilisation method is more significantly affected by
heat treatment. Analysis by TEM of the calcined sample
showed that considerable sintering and growth of the Au–Pd
nanoparticles has occurred presumably because of disruption
of the protective ligand shell at elevated temperatures (see Fig.
Fig. 5 STEM–ADF image and the corresponding Au La1 and Pd La1
elemental maps of the AuPd/C catalyst prepared by the sol-immobo-
lization method.
immobilisation confirms that the metal loading of all the
samples is close to the expected nominal value. Although
XPS analysis of the impregnated samples also yields metal
loadings close to the nominal values (Table 3), extremely high
loadings are calculated for the catalysts prepared by sol
immobilisation (Table 3). These anomalous values probably
arise from the coated metal particles preferentially decorating
the exterior surface of the carbon, thereby contributing to the
higher catalytic activity we observe, whereas impregnation
leads to more effective filling of the pores within the support
which in turn means the metal atoms are more homogeneously
distributed throughout the material. Since XPS is very surface
sensitive the metal signal measured by this technique will
Fig. 6 General reaction pathway for benzyl alcohol: (1) benzyl alcohol; (2) benzaldehyde; (3) benzoic acid; (4) benzyl benzoate; (5) benzene;
(6) toluene.
ꢀc
This journal is the Owner Societies 2008
1926 | Phys. Chem. Chem. Phys., 2008, 10, 1921–1930

S2, ESI).w Using TGA we have observed that the PVA is
largely lost by oxidation from the catalyst on heating at
350 1C. Hence, we consider that when we calcine the catalysts
at 400 1C the PVA is largely removed. For the dried materials,
at this stage, we do not know how much of the PVA remains
on the catalyst after drying and clearly this could play a role in
the observed catalysis with the dried materials and we will
attempt to address this issue in subsequent studies.
Hydrogen peroxide synthesis and hydrogenation. A series of
catalysts, prepared either by sol-immobilisation (Au/CSI, Pd/
CSI, AuPd/CSIm, AuPd/CSIm) or by co-impregnation (AuPd/
CIm, AuPd/CIw) all containing 1 wt% total metal, with carbon
as support, were investigated for the direct synthesis of hydrogen
peroxide; the results for reaction at 2 1C are shown in Table 6.
For the so-immobilisation set, the pure Au catalyst did generate
some H2O2 but at a very low rate, whereas the Pd only catalyst
was more active. However, the addition of Au to Pd markedly
enhanced the catalyst activity by a factor of 36. This is the most
dramatic synergistic effect we have observed to date for the
direct synthesis of hydrogen peroxide. Comparison of the
Au–Pd catalysts prepared by sol-immobilisation with the corre-
sponding Au–Pd catalysts prepared by impregnation (Table 5)
shows that the catalysts prepared by sol-immobilisation are
significantly more active. This dramatic enhancement in activity
is a feature that we will investigate in greater detail in subsequent
studies. One of the key problems with respect to the synthesis of
hydrogen peroxide is the stability of the product under reaction
conditions where hydrogen peroxide is present with high pres-
sure H2. It is generally acknowledged that catalysts that are
effective for the synthesis reaction, e.g. supported Pd catalysts,
are all effective hydrogenation catalysts, since their primary role
is the activation of H2 at low temperatures. This creates a major
problem, since catalysts that are effective in the initial hydro-
genation of oxygen to give hydrogen peroxide selectively are also
equally effective at the hydrogenation of hydrogen peroxide to
water in a subsequent hydrogenation reaction (see Fig. 9). We
have found that the decomposition of hydrogen peroxide to give
oxygen and water is not observed during the direct reaction
when hydrogen is present, i.e. k4 is not significant under our
reaction conditions,23,27 and the loss of selectivity we observe is
principally due to the hydrogenation of hydrogen peroxide.27
This is because the selectivity for hydrogen peroxide we observe
at very low conversions approaches 100%, indicating that k3 is
not significant. Hence, under our reaction conditions the even-
tual yield of hydrogen peroxide is governed by the relative ratio
of k1 and k2. In view of this we have investigated the hydro-
genation of hydrogen peroxide (Table 6) and it is evident that
the sol-immobilised catalysts are more active for the sequential
hydrogenation of hydrogen peroxide as well as its synthesis
(Table 6). Clearly, this is a crucial area of catalyst design that
requires further attention if the higher activity of the sol-
immobilised catalysts is to be realised for hydrogen peroxide
synthesis.
Fig. 7 Selective oxidation of benzyl alcohol, (a) 1 wt% Au–Pd/CSIm
;
(b) 1 wt% Au–Pd/CIm; (c) 1 wt% Au–Pd/CIw. Reaction conditions:
benzyl alcohol 40 ml, 0.1 g of catalyst, T = 160 1C, pO2 = 150 psi,
time of reaction = 8 h, stirring rate 1500 rpm. Key:E benzyl alcohol
conversion (%), ’ benzene selectivity (%), m toluene selectivity (%),
Bbenzaldehyde selectivity (%), & benzoic acid selectivity (%) and n
benzyl benzoate selectivity (%).
Conclusions
In our earlier studies based on carbon-supported Au–Pd alloy
nanoparticles prepared by impregnation we have shown these
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 1921–1930 | 1927

Table 1 Liquid phase oxidation of benzyl alcohol for catalysts containing 1 wt% of metal prepared by the sol-immobilization and impregnation
methods
Selectivity/%
Yield/%
Catalyst
Conversion
%
Toluene Benzaldehyde Benzoic acid Toluene Benzaldehyde Benzoic acid TOF/hÀ1 a
b
1%Au–Pd/CSIm
82
48
56
81.1
6.7
44.6
31.0
5.8
8.2
40.9
2.4
7.5
24.0
13.5
19.0
48.0
64.4
61.0
55.0
78.7
65.0
56.6
50.1
73.8
18.0
22.8
23.7
1.3
25.4
2.8
10.9
33.17
0.16
3.2
17.3
10.0
9.1
39.4
30.8
34.2
44.61
5.27
27.6
40.8
37.2
35.3
b
14.8
10.9
13.3
1.05
0.24
8.7
11.3
23.3
1.5
29 900
14 800
16 200
35 404
1770
17 265
12 656
29 914
7154
b
1%Au/CSIb
1%Pd/CSI
1%Au–Pd/CSIw
c
c
1%AuPd/CSIw treated/400 1C
1%Au–Pd/CIm
1%Au–Pd/CIm treated/400 1C 72.1
3.6
b
20.5
15.7
31.3
3.1
b
b
1%Au–Pd/CIw
74.3
b
1%Au–Pd/CIw treated/400 1C 47.8
TOF (hÀ1) at 0.5 h of reaction. TOF numbers were calculated on the basis of total loading of metals. Reaction conditions: benzyl alcohol,
a
c
catalyst (0.1 g) , T = 160 1C, pO2 = 10 bar, time of reaction = 4 h, stirring rate 1500 rpm. Reaction conditions: benzyl alcohol, 0.1 g of catalyst,
T = 120 1C, time of reaction = 2 h, pO2 = 150 psi, stirring rate 1500 rpm.
Table 2 Liquid phase oxidation of benzyl alcohol for catalysts containing 1 wt% total metal loading. Product distribution selectivities were
determined at iso-conversion (50%)
S50%
Catalyst
Benzene
Toluene
Benzaldehyde
Benzoic acid
Benzyl benzoate
Acetal
1%Au–Pd/CSIm
1%Au/CSI
1%Pd/CSI
1%Au–Pd/CIm
1%Au–Pd/CIw
a
0.3
0.7
1.3
0.9
0.2
41.4
5.2
12.6
4.2
53.0
63.5
66.0
61.9
71.1
4.9
24.2
16.2
25.0
6.3
0.7
6.5
3.9
2.7
0.8
0
0
0
0
0
20.9
Reaction conditions: benzyl alcohol, 0.1 g of catalyst, T = 160 1C, pO2 = 10 bar, stirring rate 1500 rpm. S50 indicate the selectivity observed at
50% conversion.
Table 3 XPS derived elemental atomic ratios for carbon supported AuPd, Au and Pd catalysts prepared by sol-immobilisation. Total metal
loading for each sample is 1 wt% and all samples were dried at 120 1C overnight
Elemental ratios and total metal loading derived from XPS peak intensities
Sample
Nominal Pd : Au
Pd/Au
0.81
Pd/C
0.013
Au/C
(Au + Pd)/C
O/C
wt% metal
AuPd/CSIm
Au/CSI
Pd/CSI
AuPd/CSiw
AuPd/CI
AuPd/CI
1/1 molar
0/1
1/0
1/1 Weight (1.85 : 1 molar)
1/1 molar
1/1 Weight (1.85 : 1 molar)
0.017
0.027
0.030
0.027
0.017
0.024
0.001
0.0014
0.144
0.114
0.139
0.136
0.00441
0.0485
39
49
15
29
1.08
1.49
0.017
0.015
0.0007
0.0011
1.47
2.37
3.29
0.010
0.0003
0.0003
Table 4 Liquid phase oxidation of benzyl alcohol for catalysts prepared by sol-immobilization
Selectivity at 50% conversion/%
TOF/hÀ1 c
Benzene
Toluene
Benzaldehyde
Benzoic acid
Benzyl benzoate
Acetal
Catalyst
1%AuPd/CSIma
1%AuPd/CSIwb
a
0.3
0.3
41.4
44.0
52.7
51.3
4.9
1.7
0.7
2.7
0.0
0.1
29 900
35 400
b
Reaction conditions: benzyl alcohol, 0.1 g of catalyst, T = 160 1C, pO2 = 150 psi. Reaction conditions: benzyl alcohol, 0.1 g of catalyst,
T = 120 1C, pO2 = 150 psi. Calculation of TOF (hÀ1) after 0.5 h of reaction. TOF numbers were calculated on the basis of total loading of
metals. S50 indicate the selectivity observed at 50% conversion.
c
to be present in a broad particle size range (typically 3–7 nm
but with a number of particles 410 nm).27,52 We have also
shown these materials to exhibit very high activity for both the
synthesis of hydrogen peroxide27 and the oxidation of alcohols
to aldehydes.20 We used the relatively simple impregnation
method during the catalyst discovery phase of these studies
precisely because it generates a broad range of potential active
species, and hence improves our chances of determining
ꢀc
This journal is the Owner Societies 2008
1928 | Phys. Chem. Chem. Phys., 2008, 10, 1921–1930

Table 6 Hydrogen peroxide hydrogenation results for catalysts pre-
pared by sol-immobilization and impregnation on a carbon supporta
Catalyst
Hydrogen peroxide hydrogenation/%
1%AuPd/CSIm
1%Au–Pd/CSIw
27
18
AuPd/CIm
Au–Pd/CIw
a
4
7
Reaction conditions, all catalysts uncalcined. MeOH (5.6 g), H2O
(2.22 g), H2O2 (0.68 g) and 0.1 g catalyst, 5%H2/CO2, P = 420 psi.
T = 2 1C, stirring speed = 1200 rpm and reaction time = 30 min.
Fig. 9 Reaction scheme for the synthesis of hydrogen peroxide from
the reaction of O2 and H2.
particles. However, both methods lead to the formation of
alloy nanoparticles and X-ray photoelectron spectroscopy
together with STEM-XDS show that in both cases the parti-
cles are homogeneous alloys. The sol-immobilised catalysts
display different activities when compared with the materials
prepared by impregnation, being more active for hydrogen
peroxide synthesis and also for alcohol oxidation. The sol-
immobilisation method certainly permits a greater control of
the particle size distribution design parameter, but controlling
the compositional homogeneity as a function of particle size
still needs some further development. Studies have shown that
Au–Pd alloy nanoparticles are of interest for the formation of
vinyl acetate58 and the cyclisation of acetylene59 and hence the
results of this study may have general applicability.
Fig. 8 Selective oxidation of benzyl alcohol, (a) 1 wt% Au–Pd/CSIw
;
(b) 1 wt% Au–Pd/CIw (reaction conditions: benzyl alcohol 40 ml, 0.1 g
of catalyst, T = 120 1C, pO2 = 150 psi, time of reaction = 2 h, stirring
rate 1500 rpm. Key: E benzyl alcohol conversion (%), ’ benzene
selectivity (%), m toluene selectivity (%), B benzaldehyde selectivity
(%), & benzoic acid selectivity (%), n benzyl benzoate selectivity (%)
and J acetal selectivity (%).
whether or not catalyst activity is enhanced with this metho-
dology. In contrast, the sol-immobilisation method leads to a
distinctly narrower particle size range of Au–Pd alloy nano-
Acknowledgements
Table 5 Hydrogen peroxide productivity for catalysts prepared by
sol-immobilization and impregnation on a carbon supporta
This work formed part of the EU AURICAT project (Con-
tract HPRN-CT-2002-00174) and the EPSRC/Johnson Mat-
they funded ATHENA project and we thank them for funding
this research. We also thank the World Gold Council (through
the GROW scheme), and Cardiff University (AA Reed stu-
dentship) for providing support for J.K.E. C.J.K. and R.T.
would also like to acknowledge the financial support of the
NSF Nanomanufacturing program under contract #CMMI-
0457602).
Hydrogen peroxide productivity
Catalyst
Mol H2O2 kgÀ1
(cat) hÀ1
Mol H2O2 molÀ1
(metal) hÀ1
1%Au/CSI
1%Pd/CSI
1%AuPd/CSIm
1%Au–Pd/CSIw
0.21
1.51
54.1
128.0
4.1
16.1
820
1775
AuPd/CIm
Au–Pd/CIw
a
12.0
60.0
181
832
Reaction conditions: reaction time for H2O2 synthesis 30 min; Mass
of catalyst for H2O2 synthesis: 10 mg, all catalysts uncalcined.
References
1 J. Shan and H. Tenhu, Chem. Commun., 2007, 4580.
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 1921–1930 | 1929

2 M.-C. Daniel and D. Astrue, Chem. Rev., 2004, 104, 293.
3 L. M. Liz-Mnarzan, Mater. Today, 2004, 7, 26.
4 J. Turkevich, P. C. Stevenson and J. Hillier, Discuss. Faraday Soc.,
1951, 11, 55.
28 R. A. Sheldon and J. K. Kochi, Metal-Catalyzed Oxidations of
Organic Compounds, Academic Press , New York, 1981.
29 G. ten Brink, I. W. C. E. Arends and R. A. Sheldon, Science, 2000,
287, 1636.
´
5 M. Brust, M. Walker, D. Bethell, D. J. Schriffrin and C. J. Kiely, J.
Chem. Soc., Chem. Commun., 1994, 801.
30 M. Pagliaro, S. Campestrini and R. Ciriminna, Chem. Soc. Rev.,
2005, 34, 837.
6 G. C. Bond and D. T. Thompson, Catal. Rev. Sci. Eng., 1999, 41,
319.
7 G. C. Bond and D. T. Thompson, Gold Bull., 2000, 33, 41.
8 M. Haruta, Gold Bull., 2004, 37, 27.
9 A. S. K. Hashmi, Gold Bull., 2004, 37, 51.
10 R. Meyer, C. Lemaire, Sh. K. Shaikutdinov and H.-J. Freund,
Gold Bull., 2004, 37, 72.
11 G. J. Hutchings, Gold Bull., 2004, 37, 37.
31 L. Prati and M. Rossi, J. Catal., 1998, 176, 552.
32 F. Porta, L. Prati, M. Rossi and G. Scari, J. Catal., 2002, 211, 464.
33 F. Porta and L. Prati, J. Catal., 2004, 224, 397.
34 H. T. Hess, in Kirk-Othmer Encyclopaedia of Chemical Engineer-
ing, ed. I. Kroschwitz and M. Howe-Grant, Wiley, New York,
1995, vol. 13, p. 961.
35 ‘Degussa Headwaters builds peroxide demonstrator’, The
Chemical Engineer, 2005, 766, p. 16.
12 A. S. K. Hashmi and G. J. Hutchings, Angew. Chem., Int. Ed.,
2006, 45, 7896.
36 H. Henkel and W. Weber, US Pat., 1108752, 1914.
37 G. A. Cook, US Pat., 2368640, 1945.
13 M. Haruta, T. Kobayashi, H. Sano and N. Yamada, Chem. Lett.,
1987, 16, 405.
38 Y. Izumi, H. Miyazaki and S. Kawahara, US Pat., 4009252,
1977.
14 P. Landon, J. Ferguson, B. E. Solsona, T. Garcia, A. F. Carley, A.
A. Herzing, C. J. Kiely, S. E. Golunski and G. J. Hutchings, Chem.
Commun., 2005, 3385.
15 P. Landon, J. Ferguson, B. E. Solsona, T. Garcia, S. Al-Sayari, A.
F. Carley, A. Herzing, C. J. Kiely, M. Makkee, J. A. Moulijn, A.
Overweg, S. E. Golunski and G. J. Hutchings, J. Mater. Chem.,
2006, 16, 199.
16 A. K. Sinha, S. Seelan, S. Tsubota and M. Haruta, Angew. Chem.,
Int. Ed., 2004, 43, 1546.
17 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 and C. J. Kiely, Nature, 2005, 437,
1132.
39 Y. Izumi, H. Miyazaki and S. Kawahara, US Pat., 4279883, 1981.
40 H. Sun, J. J. Leonard and H. Shalit, US Pat., 4393038, 1981.
41 L. W. Gosser and J.-A. T. Schwartz, US Pat., 4772458, 1988.
42 L. W. Gosser, US Pat., 4889705, 1989.
43 C. Pralins and J.-P. Schirmann, US Pat., 4996039, 1991.
44 T. Kanada, K. Nagai and T. Nawata, US Pat., 5104635, 1992.
45 J. Van Weynbergh, J.-P. Schoebrechts and J.-C. Colery, US Pat.,
5447706, 1995.
46 S.-E. Park, J. W. Yoo, W. J. Lee, J.-S. Chang, U. K. Park and
C. W. Lee, US Pat., 5972305, 1999.
47 G. Paparatto, R. d’Aloisio, G. De Alberti, P. Furlan, V. Arca,
R. Buzzoni and L. Meda, EP Pat., 0978316A1, 1999.
48 B. Zhou and L.-K. Lee, US Pat., 6168775, 2001.
49 M. Nystrom, J. Wangard and W. Herrmann, US Pat., 6210651,
2001.
18 A. Abad, P. Conception, A. Corma and H. Garcia, Angew. Chem.,
Int. Ed., 2005, 44, 4066.
19 A. Corma and P. Serna, Science, 2006, 313, 332.
20 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
and G. J. Hutchings, Science, 2006, 311, 362.
50 N. Dimitratos, J. A. Lopez-Sanchez, D. Morgan, A. Carley, L.
Prati and G. J. Hutchings, Catal. Today, 2007, 122, 317.
51 C. L. Bianchi, P. Canton, N. Dimitratos, F. Porta and L. Prati,
Catal. Today, 2005, 102, 203.
21 V. R. Choudhary and C. Samanta, J. Catal., 2006, 238, 28.
22 P. Landon, P. J. Collier, A. J. Papworth, C. J. Kiely and G. J.
Hutchings, Chem. Commun., 2002, 2058.
23 P. Landon, P. J. Collier, A. F. Carley, D. Chadwick, A. J.
Papworth, A. Burrows, C. J. Kiely and G. J. Hutchings, Phys.
Chem. Chem. Phys., 2003, 5, 1917.
24 B. E. Solsona, J. K. Edwards, P. Landon, A. F. Carley, A. Herzing,
C. J. Kiely and G. J. Hutchings, Chem. Mater., 2006, 18, 2689.
25 J. K. Edwards, B. Solsona, P. Landon, A. F. Carley, A. Herzing,
M. Watanabe, C. J. Kiely and G. J. Hutchings, J. Mater. Chem.,
2005, 15, 4595.
26 J. K. Edwards, B. E. Solsona, P. Landon, A. F. Carley, A. Herzing,
C. J. Kiely and G. J. Hutchings, J. Catal., 2005, 236, 69.
27 J. K. Edwards, A. F. Carley, A. A. Herzing, C. J. Kiely and G. J.
Hutchings, Faraday Discuss., 2008, in press.
52 S. Link and M. A. El-Sayed, J. Phys. Chem. B, 1999, 103
4212.
53 J. A. Creighton and D. G. Eadon, J. Chem. Soc., Faraday Trans.,
1991, 87, 3881.
54 K. Hirakawa and N. Toshima, Chem. Lett., 2003, 32, 78.
55 S. Deki, K. Akamatsu, Y. Hatakenaka, M. Mizuhata and A.
Kajinami, Nanostructured Mater., 1999, 11, 59.
56 X. Wang, K. Naka, M. Zhu, H. Itoh and Y. Chujo, Langmuir,
2005, 21, 12395.
57 A. A. Herzing, M. Watanabe, C. J. Kiely, J. Edwards, M. Conte,
Z. R. Tang and G. J. Hutchings, Faraday Discuss., 2008, in press.
58 M. Chen, D. Kumar, C.-W. Yi and D. W. Goodman, Science,
2005, 310, 291.
59 A. F. Lee, C. J. Baddeley, C. Hardacre, R. M. Ormerod, R. M.
Lambert, G. Schmid and H. West, J. Phys. Chem., 1995, 99, 6096.
ꢀc
This journal is the Owner Societies 2008
1930 | Phys. Chem. Chem. Phys., 2008, 10, 1921–1930