Physical mixing of metal acetates: Optimisation of catalyst parameters to produce highly active bimetallic catalysts

Article abstract of DOI:10.1039/c3cy00263b

We have investigated the optimisation of the catalytic parameters for the preparation of catalysts by the simple mixing and thermal treatment of a support and metal acetate precursors. We have studied the effect of metal ratio and metal loading to produce a catalyst which has the optimum activity for a variety of reactions including benzyl alcohol oxidation, glycerol oxidation and the direct synthesis of hydrogen peroxide. We have demonstrated the high activity of these catalyst's for a variety of substrates and performed XPS and XRD studies on the catalysts to help elucidate the origin of the catalysts improved activity.

Full text of DOI:10.1039/c3cy00263b

Catalysis
Science & Technology
View Article Online
View Journal | View Issue
PAPER
Physical mixing of metal acetates: optimisation of
catalyst parameters to produce highly active
bimetallic catalysts
Cite this: Catal. Sci. Technol., 2013, 3,
2910
Peter J. Miedziak,a Simon A. Kondrat,a Noreen Sajjad,ab Gavin M. King,a
Mark Douthwaite,a Greg Shaw,a Gemma L. Brett,a Jennifer K. Edwards,a
a
b
a
*
David J. Morgan, Ghulam Hussain and Graham J. Hutchings
We have investigated the optimisation of the catalytic parameters for the preparation of catalysts by the simple
mixing and thermal treatment of a support and metal acetate precursors. We have studied the effect of metal
ratio and metal loading to produce a catalyst which has the optimum activity for a variety of reactions including
benzyl alcohol oxidation, glycerol oxidation and the direct synthesis of hydrogen peroxide. We have demonstrated
the high activity of these catalyst's for a variety of substrates and performed XPS and XRD studies on the catalysts
to help elucidate the origin of the catalysts improved activity.
Received 17th April 2013,
Accepted 7th May 2013
DOI: 10.1039/c3cy00263b
www.rsc.org/catalysis
Introduction
associated solely with the Al2O3 support. The blocking of sites
on the Al2O3 support with phosphate showed that a Cl : Au
Heterogeneous catalysis is a key process in the manufacture
of a range of bulk and fine chemicals and is at the heart of
green chemical processes. The use of catalysts for oxidation
reactions can lead to a greener process when compared with
alternative routes. Gold nanoparticles have been shown to be
effective for the oxidation of alkenes1–3 alcohols,4,5 the oxida-
tion of CO6,7 and for the direct synthesis of hydrogen perox-
ide.8 It has also been demonstrated that alloying gold with
palladium leads to a substantial enhancement in activity for
alcohol oxidation9,10 and also significantly improves the yield
of hydrogen peroxide in the direct synthesis reaction.10,11
The catalysts reported for these oxidations are usually pre-
pared by three main methods; namely, wet impregnation,
deposition precipitation and sol-immobilisation. However,
all these preparation techniques use HAuCl4 as the gold pre-
cursor, making it difficult to completely remove chloride
from the final catalyst. It is known that the presence of chlo-
ride in a catalyst can lead to particle agglomeration, through
Au–Cl–Au bridges12 and also leads to the blocking of active
sites.13 The result of these effects is a loss of catalyst activity,
which is well documented for CO oxidation with gold mono-
metallic catalysts.14,15 It has been observed that a Cl/Au atom
ratio of 0.1 is sufficient to decrease the activity by half for a
CO oxidation catalyst, despite most of the Cl− being
atom ratio even as low as 0.0006 could affect activity. A
method of limiting chloride levels in catalysts involves tuning
the pH of the deposition, to ensure significant hydrolysis of
the HAuCl4 and also to minimize Cl− deposition onto the
support.15 Unfortunately, a pH that facilitates total hydrolysis
(∼pH 10) also results in a decrease in the Au loading
levels.15,16 An alternative method of reducing chloride levels
is by heat treatment of the catalyst, but unfortunately this
causes agglomeration through sintering.15 Modification of the
impregnation and deposition techniques by the addition of
ammonia has also been shown to minimize Cl− content in
catalysts to less than 200 ppm.17–19 However, the addition of
ammonia is undesirable in an applied scaled up process,
due to its corrosive and toxic nature.
The simple physical mixing of metal acetates and subse-
quent heat treatment under an inert atmosphere has been
shown to be effective for the formation of supported metal
nanoparticles.20 Recently we clearly demonstrated the poten-
tial for the chloride-free formation of supported Au and
Au–Pd alloy catalysts by the physical mixing of metal acetates
with a support.21 The removal of aqueous chloride ions from
the preparation medium significantly decreases the waste
material and ensures that the catalyst preparation is com-
pleted in a greener manner. These Au–Pd alloy catalysts were
shown to have exceptionally high activity, compared to wet
impregnation and deposition precipitation techniques, for
both the direct synthesis of hydrogen peroxide and benzyl
alcohol oxidation. Although the preparation technique pro-
duced significant amounts of highly dispersed sub 10 nm
a Cardiff Catalysis Institute, School of Chemistry, Cardiff University, Main Building,
Park Place, Cardiff, CF10 3AT UK. E-mail: hutch@cardiff.ac.uk;
Fax: +44 (0)29 2087 4059
b Department of Chemistry, University of Sargodha, Sargodha, 40100 Pakistan
2910
|
Catal. Sci. Technol., 2013, 3, 2910–2917
This journal is © The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
particles, thought to be responsible for the high activity,
large μm particles of Au and Pd were also observed.
The charged autoclave was then purged three times with 5%
H2/CO2 (7 bar) before filling with 5% H2/CO2 to a pressure of
29 bar at 20 °C. The pressure was allowed to drop to 26 bar as
the gas dissolved in the solvent at 20 °C. This was followed by
the addition of 25% O2/CO2 (11 bar). The temperature was
then allowed to decrease to 2 °C followed by stirring (at
1200 rpm) of the reaction mixture for 30 min. The above
reaction parameters represent the optimum conditions we
have previously used for the synthesis of H2O2. H2O2 pro-
ductivity was determined by titrating aliquots of the final
solution after reaction with acidified Ce(SO4)2 (0.01 M) in
the presence of two drops of ferroin indicator.
In our previous work, we showed that catalysts that were
analogous in terms of metal ratio and weight loading to pre-
viously reported optimised catalysts prepared by ‘wet’ prepa-
ration methods, such as impregnation or deposition9 displayed
superior activities for both alcohol oxidation and hydrogen per-
oxide synthesis. It was noted that the high activity for hydrogen
peroxide synthesis was practically limited by the high hydroge-
nation rates. In this paper we focus on the optimisation of the
physically ground catalysts with the aim of improving metal
dispersion, we show that the optimisation of the Au: Pd ratio
and metal loading can lead to significantly different activities
for alcohol oxidations. We demonstrate that significant enhance-
ments of the TOF (turn over frequency) are achieved when the
optimised catalysts are used.
H2O2 hydrogenation. Hydrogen peroxide hydrogenation was
evaluated using a Parr Instruments stainless steel autoclave
with a nominal volume of 100 mL and a maximum working
pressure of 140 bar. To test each catalyst for H2O2 hydrogenation,
the autoclave was charged with catalyst (0.01 g) and a solution
containing 4 wt% H2O2 (5.6 g CH3OH, 2.22 H2O and 0.68 g
H2O2 50% w/w). The charged autoclave was then purged three
times with 5% H2/CO2 (7 bar) before filling with 5% H2/CO2
to a pressure of 29 bar at 20 °C. The temperature was then
allowed to decrease to 2 °C followed by stirring (at 1200 rpm)
the reaction mixture for 30 min. The amount of hydrogenated
H2O2 was determined by titrating aliquots with acidified
Ce(SO4)2 (0.0288 M) in the presence of a ferroin indicator.
Glycerol oxidation. Catalytic reactions were carried out
using a 50 mL glass reactor. The glycerol solution (0.3 mol L−1)
was admitted into the reactor with a solution of base (NaOH,
substrate/base = 2) and the desired amount of catalyst
(glycerol/metal mole fraction = 500) was suspended in the
solution. The glass reactor was purged with oxygen three times
and adjusted to the desired pressure of 3 bar. This pressure
was maintained at a constant level throughout the experiment;
hence as the oxygen was consumed in the reaction it was
continuously replenished. The reaction mixture was heated to
the desired temperature (60 °C) and stirred for the requisite
amount of time (between 0.5 h and 4 h). The reactor vessel was
then cooled to room temperature and the reaction mixture
analysed. This analysis was carried out using high-pressure liq-
uid chromatography (HPLC) equipped with ultraviolet and
refractive index detectors. Reactants and products were sepa-
rated using a Metacarb 67H column eluted with 0.01 mol L−1
aqueous H3PO4 with a flow rate of 0.3 mL min−1. Samples of
the reaction mixture (0.5 mL) were diluted (to 5 mL) using the
eluent. Products were identified by comparison with authentic
samples. For the quantification of the reactants consumed
and products generated, an external calibration method was used.
Experimental
Catalyst preparation
Physical grinding. Palladium(II) acetate (Sigma Aldrich
99.9%) and gold(III) acetate (Alfa Aesar 99.9%) were added to
the TiO2 (Degussa, P25) and the mixture was ground manually
in a pestle and mortar for 10 min. The mass of acetate salts
used was varied to give 1–10 wt% total metal loading. Au : Pd
metal ratio was varied, with 0 : 1, 1 : 4, 1 : 1, 4 : 1 and 1 : 0 being
investigated. The resultant mixture was placed in a 4 inch
ceramic boat and heat treated at 350 °C for 2 h in a tubular
furnace under flowing helium (ramp rate 5 °C min−1).
Impregnation. Aqueous solutions of PdCl2 (Johnson
Matthey, 6 mg in 1 mL) and HAuCl4·3H2O (Johnson Matthey,
12.25 g in 1000 mL) were added to titania. The mixture was
stirred for 5 min, until a paste was formed, dried at 110 °C
for 16 h and calcined at 400 °C for 3 h in static air.
Catalyst testing
Benzyl alcohol oxidation. Catalyst testing was performed
using a stainless steel autoclave (Autoclave Engineers In-line
MagneDrive III) with a nominal volume of 100 mL and a maxi-
mum working pressure of 140 bar. The vessel was charged with
benzyl alcohol (40 mL) and catalyst (25 mg). The autoclave was
then purged three times with oxygen leaving the vessel at 10 bar.
The pressure was maintained at a constant level throughout
the experiment; as the oxygen was consumed in the reaction,
it was replenished. The stirrer speed was set at 1500 r.p.m.
and the reaction mixture was raised to the required tempera-
ture of 140 °C. Samples from the reactor were taken periodi-
cally via a sampling pipe, ensuring that the volume purged
before sampling was higher than the tube volume, and analysed
by GC (Varian 3800) using a CP-wax column.
Catalyst characterisation
Direct H2O2 synthesis. Hydrogen peroxide synthesis and
Powder X-ray diffraction. Characterisation was performed
using powder X-ray diffraction (XRD) on a (θ–θ) PANalytical
X'pert Pro powder diffractometer using a Cu Kα radiation
source operating at 40 KeV and 40 mA. Standard analysis was
performed using a 40 min scan with a back filled sample. Dif-
fraction patterns of phases were identified using the ICDD data
hydrogenation was evaluated using
a Parr Instruments
stainless steel autoclave with a nominal volume of 100 mL
and a maximum working pressure of 140 bar. To test each
catalyst for H2O2 synthesis, the autoclave was charged with
catalyst (0.01 g) and 8.5 g solvent (5.6 g MeOH and 2.9 g H2O).
This journal is © The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2910–2917 | 2911

View Article Online
Paper
Catalysis Science & Technology
base. Diffraction patterns were analyzed using full pattern
refinement, with background functions, zero shift and pseudo-
Voigt profile functions. Reference crystal structure profiles were
selected from the inorganic crystal structure database.
X-ray photoelectron spectroscopy. XPS measurements were
made on a Kratos Axis Ultra DLD spectrometer using
monochromatic Al Kα radiation (120 W source power). An
analyser pass energy of 160 eV was used for survey scans,
while 40 eV was employed for detailed regional scans.
Samples were mounted using double-sided adhesive tape, and
binding energies were referenced as discussed in the text.
Results and discussion
Au and Pd catalysts were all prepared supported on titania as
we have previously demonstrated it to be an extremely effec-
tive support for oxidation and reduction reactions.10,11 Previ-
Fig. 1 The conversion of benzyl alcohol by catalysts (5 wt% metal) prepared
by the physical grinding methodology with varying gold : palladium (weight) ratios.
◆ 5%Pd, ● 4%Pd–1%Au, ▲ 2.5%Pd–2.5%Au, 1%Pd–4%Au, ■ 5%Au.
ously 2.5wt%Au
+ 2.5wt%Pd/TiO2 catalysts prepared by
physical grinding (PG) were found to be more active than
analogous catalysts prepared by either deposition precipita-
tion (DP) or wet impregnation (IMP) for benzyl alcohol oxida-
tion.21 A study into the effect of metal ratio for impregnation
prepared catalysts has shown that a gold : palladium ratio of
1 : 1 weight (∼1 : 1.85 molar) afforded the catalysts with the
highest initial activity for both alcohol oxidation (benzyl alco-
hol)22 and the direct synthesis of hydrogen peroxide,23 how-
ever when catalysts are prepared by the sol-immobilisation
method and tested for glycerol oxidation a gold : palladium
ratio of ∼1 : 1.6 weight (1 : 3 molar) was found to be the opti-
mum.24 It is clear from these results that for different prepa-
ration methods and different substrates the optimum metal
ratio varies. To investigate the effect of Au : Pd ratio on the
activity of physically ground catalysts for benzyl alcohol oxi-
dation, several titania supported 5 wt% metal catalysts with
varying Au : Pd ratios were prepared and tested. The results of
these tests are shown in the time-on-line data presented in
Fig. 1, with the corresponding selectivities to benzyaldehyde
and toluene being shown in Fig. 2. A catalyst comprising of
Pd only and a catalyst with a 1 : 4 Au : Pd composition had
very similar benzyl alcohol conversion profiles and, as has
been shown previously for impregnation catalysts,22 the cata-
lyst with a 1 : 1 Au : Pd ratio displayed the highest conversion.
The catalysts comprised predominantly (4 : 1 Au : Pd) of Au
had significantly lower conversion than the aforementioned
catalysts. However, there was a clear synergistic effect demon-
strated with the addition of only a small amount of palla-
dium to the gold catalyst, after 3 h the 4%Au–1%Pd catalyst
had converted over twice as much benzyl alcohol as the 5%
Au catalyst. As all the catalysts tested comprised 5 wt% total
metal and the Au : Pd ratio was varied the catalysts with the
highest gold content will have the least number of mols of
metal due to gold having almost twice the molar mass of pal-
ladium, therefore it is important to standardise the initial
activity. The simplest way of standardising activity is by turn
over frequency (TOF), Table 1 shows TOFs (mol substrate
converted per mol of metal per hour) calculated at 0.5 h,
Fig. 2 The selectivity towards the major products during the oxidation of benzyl
alcohol by catalysts (5 wt% metal) prepared by the physical grinding
methodology with varying gold : palladium (weight) ratios. ▲ 5%Pd, ◆ 4%Pd–1%Au,
● 2.5%Pd–2.5%Au,
1%Pd–4%Au, ■ 5%Au. Filled symbols = toluene, open
symbols = benzoic acid.
Table 1 The TOFs during the conversion of benzyl alcohol by catalysts (5 wt%
metal) prepared by the physical grinding methodology with varying gold :
palladium (weight) ratios
Au : Pd ratio (wt)
TOF (calculated at 0.5 h) (h−1
)
0 : 1
1 : 4
1 : 1
4 : 1
1 : 0
34 164
30 892
42 156
27 958
18 752
2912
|
Catal. Sci. Technol., 2013, 3, 2910–2917
This journal is © The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
Fig. 3 The selectivity at iso-conversion (30%) towards benzaldehyde and toluene
by catalysts with various Au : Pd ratios. Black 5%Pd, dark grey 4%Pd–1%Au, light
grey 2.5%Au–2.5%Pd, white 1%Pd–4%Au. The 5%Au catalysts did not reach 30%
conversion in our experiments.
Fig. 4 The conversion of benzyl alcohol by catalysts (1 : 1 weight Au : Pd) prepared
by the physical grinding methodology with varying metal loading. ◆ 10 wt%, ● 5 wt%,
▲ 3 wt%, 2.5 wt%, ■ 1 wt%.
which emphasise the fact that a synergistic effect for gold
palladium catalysts is observed when using the physical
grinding preparation methodology. As previously observed
for catalysts prepared by physical grinding and other tech-
niques a monometallic Pd catalyst was significantly more
active (TOF at 0.5 h of 34 164 h−1) than monometallic Au
(TOF at 0.5 h of 18 752 h−1).21 However, the high activity of
monometallic Pd catalysts can be partially attributed to the
disproportionation reaction, as demonstrated by the poor
selectivity towards benzaldehyde with significant selectivity
towards toluene.25 The 1 : 4 Au : Pd catalyst had only a slightly
lower TOF, at 0.5 h (30 892 h−1) than that observed for the
monometallic Pd catalyst. This observation along with the
comparably high selectivity towards toluene suggests that the
dispersed active species is predominantly monometallic Pd in
this catalyst. The rationale for this is that Au–Pd alloying is
known to produce a synergistic effect on activity and partially
inhibit disproportionation, neither of which were observed
in this catalyst. The highest activity observed was that of the
1 : 1 Au : Pd catalyst (TOF at 0.5 h of 42 156 h−1), which also
had the lowest toluene selectivity of all Pd containing cata-
lysts. This clear synergistic effect is indicative of alloy forma-
tion, which has been confirmed from STEM analysis
performed on a PG 1 : 1 Au : Pd catalyst used in the previous
publication.21 Interestingly the lower activity of the gold rich
4 : 1 Au : Pd catalyst relative to the 1 : 1 Au : Pd catalyst was also
accompanied by an increase in toluene selectivity; this is
again implicit of minimal alloying, with the Pd being in a
monometallic state suggesting that extensive alloying between
gold and palladium will only occur if the metals are present
in a certain ratio range. To provide a fair comparison of the
selectivites to benzaldeyde and toluene Fig. 3 shows the selec-
tivities at 30% iso-conversion; the 5%Au catalyst is excluded
from this comparison as the maximum activity reached was
27% in this study, however the remaining data emphasises
the clear synergistic effect present in the 2.5%Au–2.5%Pd
catalyst as discussed above.
In our previous work we extensively characterised the 5 wt%
metal catalyst (1 : 1 Au : Pd) and we demonstrated that
although very small bimetallic particles were formed, the vast
majority of the metal was present as large (>50 nm) parti-
cles.21 To try and obtain more effective metal distribution a
series of catalysts with varying total metal loading were pre-
pared, having the previously determined optimal 1 : 1 Au : Pd
metal ratio. A comparison of 1 : 1 Au : Pd catalysts prepared
with total loadings between 1 wt% and 10 wt% for the oxida-
tion of benzyl alcohol is shown in Fig. 4. The highest conver-
sion of benzyl alcohol was obtained with the 5 and 10 wt%
catalysts, which have almost identical time-on-line profiles.
As expected conversions then drop with lowering of metal
loading. It is immediately apparent from this data is that the
additional metal in the 10 wt% catalyst relative to the 5 wt%
catalysts is providing no additional activity. To obtain greater
clarity between the trends of metal loading and activity the
TOFs, at 0.5 h reaction time, were calculated (Table 2). The
previous observation made about the 10 wt% catalyst from
the time-on-line data is confirmed by significant difference in
TOFs calculated at 0.5 h, between the 10 wt% catalysts TOF of
18 395 h−1 and the 5 wt% catalysts TOF of 42 156 h−1. The TOF
of catalysts continued to increase as loading decreased up to
the total 2.5 wt% metal catalyst demonstrating that even in the
Table 2 The TOFs during the conversion of benzyl alcohol by catalysts prepared
by the physical grinding methodology with varying total metal loading and
1 : 1 Au : Pd (weight) ratio
Metal loading (Au : Pd 1 : 1 wt ratio)
TOF (calculated at 0.5 h) (h−1
)
10
5
3
2.5
1
18 395
42 156
55 031
57 861
36 087
This journal is © The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2910–2917 | 2913

View Article Online
Paper
Catalysis Science & Technology
Table 3 XPS derived metal concentrations for the physically ground catalysts
palladium was observed in all catalysts as well, with the
exception of the lowest loaded 1 wt% catalyst. It was observed
that, as total metal loadings increased, gold unit cell size
decreased by a total of 0.2 Å3 over the entire weight loading
range. Conversely the palladium unit cell size increased
slightly with increasing weight loading. These changes in
unit cell size suggest some limited alloying, which increased
with the metal weight loading. It should be noted that a unit
cell size associated with bulk 1 : 1 Au : Pd was not observed in
any samples indicating a system in which small incorpora-
tion of one element into another was prevalent. Average crys-
tallite size of the gold rich phases was found to decrease
from 165 nm in the 0.5%Au–0.5wt%Pd TiO2 to 51 nm in the
5%Au–5wt%Pd TiO2, suggesting increased gold dispersion
with increased metal loading. Alternatively palladium rich
phases were found to have no variation in crystallite size, at
ca. 12 nm, with changing weight loading. These observations
corroborate those seen in XPS; expressly that gold disperses
poorly with high concentration being required to achieve any
significant dispersion and that palladium disperses signifi-
cantly better, with concentration having a less obvious effect
on dispersion. It is apparent that increasing palladium load-
ing does little to affect its relative dispersion, therefore low
loadings of this metal limit the effective wasting of metal.
However, a minimum amount of gold appears necessary to
facilitate its dispersion. Therefore samples with loadings
around 1.25%Au–1.25%Pd TiO2 appear to provide the best
compromise with enough gold present to facilitate some dis-
persion and alloy formation and a low enough palladium
concentration to minimise wasted metal.
Catalyst
% at conc Pd % at conc Au Pd/Au ratio
5%Au–5%Pd TiO2
2.5%Au–2.5%Pd TiO2
1.5%Au–1.5%Pd TiO2
1.25%Au–1.25%Pd TiO2 0.93
0.5%Au–0.5%Pd TiO2 0.29
2.94
1.93
1.34
0.28
0.21
0.11
0.11
n/a
10
8.7
11.7
8.0
n/a
5 wt% catalyst a significant amount of the metals are not effec-
tive for catalysis. There is a significant drop in the TOF for the
1 wt% catalyst however suggesting that this catalyst may con-
tain metals in quantities below the minimum required for
alloying to occur.
XPS analysis of the 1–10 wt% metal loaded catalysts was
carried out and the results are shown in Table 3. In all cases
the atomic surface concentration of gold was considerably
lower than bulk concentration, while the surface concentra-
tion of palladium was relatively higher than the gold. This
suggests, as we have previously reported, the distribution of
palladium in these catalysts is significantly better than the
distribution of gold. It was noted that no observable signal
associated with gold was seen in the 1 wt% catalyst,
explaining the drop in TOF as no dispersion of gold and
hence no alloying was achieved. The calculated Pd : Au ratios
for the catalysts with gold and palladium dispersion did not
follow any discernible trend with ratios varying between 8 and
11.7. A high Pd : Au ratio can be indicative of a core shell
structure to bimetallic nano particles however the Pd : Au
ratios are considerably higher than those previously reported
for catalysts prepared by the impregnation or deposition pre-
cipitation techniques9 suggesting it is a function of metal dis-
tribution rather than core shell formation as noted in our
previous publication.9 Interestingly, the most active 2.5 wt%
loaded catalyst clearly had the lowest Pd : Au ratio, showing a
greater relative gold contribution and potentially indicating
the greatest degree of alloy formation amongst the catalysts.
XRD patterns and subsequent refinement are shown in
Fig. 5 and Table 4. A cubic phase with unit cell size that
strongly matches that of gold was observed in all catalysts
prepared with variable total weight loadings. The cubic
We have previously shown that gold palladium catalysts
are active for the oxidation of glycerol, therefore we tested
the physically ground catalysts along with an analogous
impregnation catalyst, the results of this testing are shown in
Fig. 6. The amount of catalyst used in these experiments was
varied in order to maintain a constant substrate : metal ratio
of 500. It is interesting to note in the case of glycerol there
seems to be two distinct trends, between the catalysts pre-
pared by the impregnation method and the catalysts made by
the physical grinding method. As the reactions were carried
Fig. 5 X-ray diffraction of Au–Pd TiO2 (Au : Pd 1 : 1) catalysts prepared by physical grinding with total metal loadings between 1–10 wt%. (a) 10 wt% Au–Pd TiO2,
(b) 5 wt% Au–Pd TiO2, (c) 3 wt% Au–Pd TiO2, (d) 2.5 wt% Au–Pd TiO2, (e) 1 wt% Au–Pd TiO2
2914
|
Catal. Sci. Technol., 2013, 3, 2910–2917
This journal is © The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
a
Table 4 Calculated Au and Pd particle properties from XRD
Catalyst
Metal phase
Unit cell volume (Å3)
Crystallite size (nm)
Weight fraction (%)
R value
5%Au–5%Pd TiO2
Au
Pd
Au
Pd
Au
Pd
Au
Pd
Au
Pd
67.70
59.03
67.79
59.08
67.85
58.88
67.86
58.82
67.94
n/a
51
10
56
12
98
11
144
14
165
n/a
3.2
5.3
1.5
1.6
0.7
0.4
0.4
0.6
0.1
n/a
2.5 (1.8)
2.5%Au–2.5%Pd TiO2
1.5%Au–1.5%Pd TiO2
1.25%Au–1.25%Pd TiO2
0.5%Au–0.5%Pd TiO2
2.7 (1.9)
2.9 (2.0)
2.7 (2.0)
2.9 (2.0)
Errors: unit cell volume 0.1 Å3, crystallite size 2 nm, weight fraction 10%.
a
Fig. 7 H2O2 productivity and hydrogenation data for the 1 : 1 Au : Pd catalysts with
various total metal loadings expressed in terms of mass of catalyst. ■ Productivity,
◇ hydrogenation.
Fig. 6 The oxidation of glycerol by catalyst prepared by the physical grinding
methodology (open symbols) and impregnation methodology (filled symbols) total
metal loading: 10% ◇, 5% □, 3% △, 2.5% ○, 2% .
out at constant substrate : base ratio the TOFs in this case are
directly proportional to the conversion. It is immediately
clear that the catalysts prepared by the physical grinding
method are considerably more active than those prepared by
the impregnation method. Interestingly in contrast to the case
of benzyl alcohol the metal loading does not seem to make a
difference for glycerol oxidation, irrespective of the preparation
method used. The selectivity towards C3 products was lower for
the physically ground catalysts, however this is most likely due
to the activity of these catalysts and the reaction conditions
used leading to over-oxidation of the desired products.
The trend found in the H2O2 synthesis, for the catalysts
prepared by physical grinding with 1–10 wt% metal loading
(Fig. 7), was that of increasing H2O2 productivity and hydro-
genation. The notable exception being the lower productivity
of the 10 wt% catalyst which indicates the additional metal is
not beneficial for the direct synthesis of H2O2. To clarify
these catalytic properties the results were calculated with
respect to mass of metal as opposed to the conventional
mass of catalyst (Fig. 8). The resulting trend is in
Fig. 8 H2O2 productivity and hydrogenation data for the 1 : 1 Au : Pd catalysts
with various total metal loadings expressed in terms of mass of metal. ■ Productivity,
◇ hydrogenation.
This journal is © The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2910–2917 | 2915

View Article Online
Paper
Catalysis Science & Technology
Table 5 The application of the physically ground catalysts to various substrates
TOF that are comparable and occasionally better than those
previously reported by other preparation methods.
Substrate
TOF (calculated at 0.5 h) (h−1
)
Benzyl alcohol
Benzyl alcohola
1-Phenylethanol
1,4-Butanediol
Cinnamyl alcoholb
25 079
93 875
26 448
52 833
1149
Note added after first publication
This article replaces the version published on 20th May 2013,
in which Fig. 4 did not appear correctly. Additionally, Fig. 3
and Fig. 4 have been exchanged with one another, with
respect to the previous version.
Reaction conditions: 40 mL of substrate, 160 °C, 14 mg 1.25%Au–1.25%
Pd/TiO2 catalyst, 10 bar O2.a 7 mg 2.5%Au–2.5%Pd/TiO2 used as catalyst.
b
40 mL 0.2 M aqueous solution, 90 °C.
References
concurrence with TOFs of the catalysts for benzyl alcohol oxi-
dation, with the optimal loading for activity being 2.5 wt%.
Evidence from catalytic testing is suggestive of increased
metal dispersion, relative to total loading, as total metal load-
ing is decreased. However, with both the oxidation of benzyl
alcohol and H2O2 synthesis activity dropped dramatically for
the 1 wt% loaded catalyst, indicating a more complex change
in catalyst properties with varying metal loading. The low pro-
ductivity and decreased hydrogenation activity of the 10 wt%
metal loaded catalyst indicates that the majority of the addi-
tional metal in this sample is not providing any activity.
Finally we carried out a series of tests to demonstrate the
general applicability of the physically ground catalysts. It is
clear from the previous data that the optimised loading for the
highest activity with these catalysts is 1.25wt%Au–1.25wt%Pd.
A series of catalytic oxidations were carried out on a range of
substrates and the TOFs were calculated, these results are
shown in Table 5. The TOF data presented demonstrates that
these catalysts display high initial activities for a range of dif-
ferent oxidation substrates, the TOFs recorded for benzyl alco-
hol oxidation and cinnamyl alcohol (93875 and 1149) are higher
than those previously reported in the literature for either
impregnation or deposition preparation methods demonstrat-
ing that by tuning the preparation method to obtain
optimised catalysts significant enhancements in the activity
of gold–palladium catalysed oxidations can be achieved.9,10
1 A. K. Sinha, S. Seelan, S. Tsubota and M. Haruta, Angew.
Chem., Int. Ed., 2004, 43, 1546–1548.
2 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, 1132–1135.
3 L. Kesavan, R. Tiruvalam, R. M. H. Ab, S. M. I. bin,
D. I. Enache, R. L. Jenkins, N. Dimitratos, J. A. Lopez-
Sanchez, S. H. Taylor, D. W. Knight, C. J. Kiely and
G. J. Hutchings, Science, 2011, 331, 195–199.
4 S. Biella, L. Prati and M. Rossi, J. Catal., 2002, 206, 242–247.
5 F. Porta and L. Prati, J. Catal., 2004, 224, 397–403.
6 M. Haruta, T. Kobayashi, H. Sano and N. Yamada, Chem.
Lett., 1987, 405–408.
7 M. Haruta, N. Yamada, T. Kobayashi and S. Iijima, J. Catal.,
1989, 115, 301–309.
8 P. Landon, P. J. Collier, A. J. Papworth, C. J. Kiely and
G. J. Hutchings, Chem. Commun., 2002, 2058–2059.
9 P. J. Miedziak, Q. He, J. K. Edwards, S. H. Taylor,
D. W. Knight, B. Tarbit, C. J. Kiely and G. J. Hutchings,
Catal. Today, 2011, 163, 47–54.
10 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, 362–365.
11 J. K. Edwards, B. E. Solsona, P. Landon, A. F. Carley,
A. Herzing, C. J. Kiely and G. J. Hutchings, J. Catal.,
2005, 236, 69–79.
12 A. Schulz and M. Hargittai, Chem.–Eur. J., 2001, 7,
3657–3670.
Conclusions
We have shown that further development of the physical
grinding method of preparing chloride-free catalysts can lead
to significant enhancements in the activity of the catalysts. By
reducing the overall metal content of the catalyst higher TOFs
can be achieved, indicating that in the higher loaded catalysts
significant amounts of the metals deposited are ‘spectators’
to the active metal particles. It is also noted that if the total
metal content is too low the synergistic effect of combining
the two metals is lost, indicating that at low metal loadings
the metals do not form the alloy particles that we believe to
be responsible for the high activity of the catalyst. These
observations from the catalytic data were confirmed by XRD
and XPS characterisation. We have further shown the general
applicability of this simple, scalable preparation method by
demonstrating its ability to oxidise a variety of substrates with
13 H. S. Oh, J. H. Yang, C. K. Costello, Y. M. Wang, S. R. Bare,
H. H. Kung and M. C. Kung, J. Catal., 2002, 210, 375–386.
14 I. Dobrosz-Gomez, I. Kocemba and J. M. Rynkowski, Appl.
Catal., B, 2009, 88, 83–97.
15 H. H. Kung, M. C. Kung and C. K. Costello, J. Catal.,
2003, 216, 425–432.
16 F. Moreau, G. C. Bond and A. O. Taylor, J. Catal., 2005, 231,
105–114.
17 L. Delannoy, H. N. El, A. Musi, N. N. L. To, J.-M. Krafft and
C. Louis, J. Phys. Chem. B, 2006, 110, 22471–22478.
18 F. Somodi, I. Borbath, M. Hegedus, A. Tompos, I. E. Sajo,
A. Szegedi, S. Rojas, F. J. L. Garcia and J. L. Margitfalvi, Appl.
Catal., A, 2008, 347, 216–222.
19 W.-C. Li, M. Comotti and F. Schueth, J. Catal., 2006, 237,
190–196.
2916
|
Catal. Sci. Technol., 2013, 3, 2910–2917
This journal is © The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
20 Y. Lin, K. A. Watson, M. J. Fallbach, S. Ghose, J. G. Smith,
D. M. Delozier, W. Cao, R. E. Crooks and J. W. Connell, ACS
Nano, 2009, 3, 871–884.
21 S. A. Kondrat, G. Shaw, S. J. Freakley, Q. He, J. Hampton,
J. K. Edwards, P. J. Miedziak, T. E. Davies, A. F. Carley,
S. H. Taylor, C. J. Kiely and G. J. Hutchings, Chem. Sci.,
2012, 3, 2965–2971.
22 D. I. Enache, D. Barker, J. K. Edwards, S. H. Taylor,
D. W. Knight, A. F. Carley and G. J. Hutchings, Catal. Today,
2007, 122, 407–411.
23 D. Enache, D. Knight and G. Hutchings, Catal. Lett.,
2005, 103, 43–52.
24 G. L. Brett, P. J. Miedziak, N. Dimitratos, J. A. Lopez-
Sanchez, N. F. Dummer, R. Tiruvalam, C. J. Kiely,
D. W. Knight, S. H. Taylor, D. J. Morgan, A. F. Carley and
G. J. Hutchings, Catal. Sci. Technol., 2012, 2, 97–104.
25 S. Meenakshisundaram, E. Nowicka, P. J. Miedziak,
G. L. Brett, R. L. Jenkins, N. Dimitratos, S. H. Taylor,
D. W. Knight, D. Bethell and G. J. Hutchings, Faraday
Discuss., 2010, 145, 341–356.
This journal is © The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2910–2917 | 2917