Cinnamaldehyde hydrogenation using Au-Pd catalysts prepared by sol immobilisation

Article abstract of DOI:10.1039/c7cy02556d

We report the catalytic performance of Au-Pd nanoparticles prepared via a sol immobilisation technique for the catalytic hydrogenation of cinnamaldehyde under mild reaction conditions. We synthesised a series of bimetallic Au-Pd colloidal supported nanoparticles with different Au: Pd molar ratios and optimized the experimental parameters to achieve the best catalyst performance. The optimum catalytic activity for the hydrogenation of cinnamaldehyde was observed for Au₅₀Pd₅₀/TiO₂ (with a Au:Pd molar ratio of 1:1), while the monometallic Pd/TiO₂ was the most selective towards hydrocinnamaldehyde. The catalysts have been structurally characterised and FTIR analysis showed that the presence of adsorbed carbonyl surface species in used catalyst materials is coupled with Pd leaching, which is the main reason for catalyst deactivation. The effect of calcination on the most active Au-Pd/TiO₂ was studied in the range 110-400 °C and a direct correlation between the rise in calcination temperature and catalyst stability and selectivity was observed. These results emphasise the importance of tuning the Au-Pd molar ratio and understanding the metal-support interaction of catalysts synthesised for hydrogenation reactions, such as cinnamaldehyde hydrogenation.

Full text of DOI:10.1039/c7cy02556d

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: S. Cattaneo, S.
Freakley, D. J. Morgan, M. Sankar, N. Dimitratos and G. Hutchings, Catal. Sci. Technol., 2018, DOI:
10.1039/C7CY02556D.
This is an Accepted Manuscript, which has been through the
Volume 6 Number 1 7 January 2016 Pages 1–308
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
Accepted Manuscripts are published online shortly after
www.rsc.org/catalysis
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
ISSN 2044-4753
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/catalysis

Page 1 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Cinnamaldehyde hydrogenation using Au-Pd catalysts prepared by
sol immobilisation
Stefano Cattaneo, Simon J. Freakley, David J. Morgan, Meenakshisundaram Sankar, Nikolaos
Dimitratos, and Graham J. Hutchings*
Cardiff Catalysis Institute, School of Chemistry, Cardiff University, Cardiff, CF10 3AT,
United Kingdom.
* Hutch@cardiff.ac.uk
1

Catalysis Science & Technology
Page 2 of 27
View Article Online
DOI: 10.1039/C7CY02556D
Abstract
We report the catalytic performance of AuꢀPd nanoparticles prepared via a sol immobilisation
technique for the catalytic hydrogenation of cinnamaldehyde under mild reaction conditions. We
synthesised a series of bimetallic AuꢀPd colloidal supported nanoparticles with different Au:Pd
molar ratios and optimized the experimental parameters to achieve best catalyst performance. The
optimum catalytic activity for the hydrogenation of cinnamaldehyde was observed for the
Au50Pd50/TiO2 (Au:Pd 1:1 molar ratio), while the monometallic Pd/TiO2 was the most selective
towards hydrocinnamaldehyde. The catalysts have been structurally characterised and FTIR
analysis showed that the presence of adsorbed carbonyl surface species in used catalyst materials
coupled with Pd leaching, which is the main reason for catalyst deactivation. The effect of
calcination on the most active AuꢀPd/TiO2 was studied in the range 110ꢀ400 °C and we observed a
direct correlation between the rise in calcination temperature and catalyst stability and selectivity.
These results emphasise the importance of tuning the AuꢀPd molar ratio and understanding the
metalꢀsupport interaction of catalysts synthesised for hydrogenation reactions, such as
cinnamaldehyde hydrogenation.
2

Page 3 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Introduction
Supported palladium nanoparticles based catalysts are widely reported as heterogeneous
catalysts for many organic transformations such as selective alcohol oxidation,1 unsaturated
hydrocarbon hydrogenation,2 and CꢀC coupling reactions.3 It has been shown that alloying
palladium with gold enhances its catalytic activity for the oxidation of benzyl alcohol4
hydrogenation of levulinic acid to gammaꢀvalerolactone5 and the direct synthesis of
hydrogen peroxide,6 demonstrating a synergistic in both oxidation and hydrogenation
reactions. α,βꢀunsaturated aldehydes are important molecules in the pharmaceutical,
fragrance and fine chemical industry;7,8 cinnamaldehyde (CAL) is used extensively in the
fragrance industry and also as a polymerisation and corrosion inhibitor in metal coatings.9,10
The selective hydrogenation of the vinyl (C=C) or carbonyl (C=O) groups of CAL lead to
hydrocinnamaldehyde (HCAL) and cinnamyl alcohol (COH) respectively, that are useful
intermediates in pharmaceuticals and perfumery.9,11 Further hydrogenation of the products,
results in hydrocinnamyl alcohol (HCOH) and phenyl propane (PPR) by hydrogenolysis
(Figure 1). Several transition and noble metal based catalysts are reported in literature to be
active in the hydrogenation of cinnamaldehyde.7 Gold nanoparticles typically require harsh
reaction conditions to obtain high conversion and selectively hydrogenate the carbonyl
group leading to the formation of COH. Bus et al., reported the conversion of CAL with
Au/γꢀAl2O3 to COH with a selectivity up to 90 % at 85 bar of H2 and 100 °C. In their
studies, high selectivity to COH were produced from the initial stages of the reaction, and
the selectivity remained unaltered up to 85 % of conversion, after which all the COH was
further hydrogenated into HCOH.12 Similar results were obtained by other research groups
13
for Au nanoparticles supported on different supports such as CeO2 and αꢀFe2O3.14 Au
supported on CNT, SiO2 and ZrO2, however, has been reported to selectively hydrogenated
the C=C bond.15,16 On the other hand, Pd nanoparticles are more selective towards the
hydrogenation of the C=C bond both when supported17 or stabilised in ionic liquids.18
Numerous studies on the effect of support materials, solvents and hydrogen sources have
been reported in literature.12–16,19,20 However, very few articles have been focused on
supported AuPd catalysts. Dash et al. reported equimolar production of HCAL and HCOH
using bimetallic AuPd nanoparticles stabilised in imidazolium ionic liquids when the Au/Pd
molar ratio was 3:1 and 1:3. However, the selectivity of HCOH increased substantially for
the 1:1 AuPd catalyst.21 Similar results were reported for the AuPd/C catalysts prepared by
reverse waterꢀinꢀoil microemulsion method.22 On the other hand, Parvulescu et al. observed
3

Catalysis Science & Technology
Page 4 of 27
View Article Online
DOI: 10.1039/C7CY02556D
23
an increase in COH production using AuPd colloids embedded in SiO2 . Very high
selectivity towards HCAL was obtained by Yang et al. with bimetallic nanoparticles
supported on mesoporous SiO2.24
In the present work, preꢀformed colloidal AuPd nanoparticles with narrow particle size
distribution were prepared via a solꢀimmobilisation method. These preꢀformed nanoparticles
were then immobilised onto titanium dioxide and tested for the selective hydrogenation of
cinnamaldehyde under relatively mild reaction conditions. In this extensive study we report
the effect of reaction temperature, hydrogen partial pressure, solvent and Au/Pd molar ratio
has been on the catalytic activity, selectivity and the stability of the solꢀimmobilised catalyst.
Finally, an efficient 2ꢀstep heat treatment protocol is reported to increase the metalꢀsupport
interaction, which is crucial for increased activity, selectivity and stability for this selective
hydrogenation reaction.
Experimental
Materials
HAuCl4 and PdCl2 (99.999 % purity, SigmaꢀAldrich) were used as metal precursors and,
commercial TiO2 (Degussa, P25) was used as the support. NaBH4 (granular, 99.99 % purity,
SigmaꢀAldrich) and polyvinyl alcohol (PVA, Mw = 9,000ꢀ10,000, 80 % hydrolysed, Sigmaꢀ
Aldrich) were used for the catalyst preparation. Cinnamaldehyde (SigmaꢀAldrich, 99 %),
cinnamyl alcohol (Alfa Aesar, 95 %), hydrocinnamaldehyde (SigmaꢀAldrich, 98 %),
hydrocinnamyl alcohol (SigmaꢀAldrich, 98%) and phenylpropane (SigmaꢀAldrich, 98 %)
were used as substrates and/or as standards. Toluene (SigmaꢀAldrich, > 99 %), ethanol
(VWR, 99.8 %) and 2ꢀpropanol (SigmaꢀAldrich, 99.5 %) were used as solvents.
Catalyst preparation
Solꢀimmobilisation catalysts were prepared as follows: To an aqueous solution of HAuCl4
and/or PdCl2 of the desired concentration (total metal concentration 0.127, 0.130, 0.143,
0.165, 0.194, and 0.235 mmol/L for Au/Pd molar ratios of 100:0, 95:5, 75:25, 50:50, 25:75
and 0:100 respectively), PVA solution (1 wt %) was added (PVA/(Au + Pd) (w/w) = 1.2 for
bimetallic catalysts, 0.65 for monometallic catalysts) under constant stirring. A freshly
prepared aqueous solution of NaBH4 (0.1 M, NaBH4/(Au + Pd) (mol/mol) = 5) was then
added to form a rubyꢀred sol when monometallic gold was prepared or a darkꢀbrown sol
4

Page 5 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
when AuPd or Pd sols were prepared. After 30 minutes of sol generation, the colloid was
immobilised by adding the TiO2 and acidified at pH 1 by sulfuric acid under continuous
vigorous stirring. The amount of support material required was calculated so as to have a
total final metal loading of 1 wt. %, and the molar ratio of Au/Pd was varied by adjusting the
relative concentrations of the two metals in aqueous solution. After 1 h of stirring, the slurry
was filtered and the catalyst was washed thoroughly with 2 L of distilled water and dried at
120 °C for 16 h. Unless specified this dried catalysts were used without any further
treatment. In some cases, the dried catalyst was first calcined in static air for 4 hours at
either 200, 300 and 400 °C with a heating rate of 10 °C/min and finally reduced at the same
heat treatment temperature under a 5 % H2/Ar flow for another 4 hours.
Catalyst testing
The hydrogenation reactions were carried out in a Radleys carousel reactor using 5 parallel
50 mL glass reactors stirred using magnetic bars. In a typical reaction, catalyst (50 mg,
typical substrate/metal molar ratio 1200:1), substrate (500 µL, 4 mmol) and solvent (5 mL of
toluene, ethanol or 2ꢀpropanol) were charged into the reactors at room temperature, which
were then purged with H2 (3 times) before the reactors were sealed. The reactors were
connected to the gasꢀline to replenish the consumed H2 gas. The reactors were loaded into a
preheated heating block, which was maintained at the desired reaction temperature and
pressure (typically 100 °C and 1 H2 bar). The reaction started by commencing stirring inside
the reactors at 1000 rpm. After the desired reaction time, the reactors were taken out the
carousel and cooled down in an ice bath for a period of 5 minutes to quench the reaction.
The contents were then centrifuged and an aliquot of the clear supernatant reaction mixture
(0.5 mL) was diluted with an external standard (0.5 mL of a 0.7 M solution of oꢀxylene in
the appropriate solvent) for GC measurement. For the analysis of the products, a GCꢀMS
(Shimadzu, GCMSꢀQP2010SE) was employed and the resulting fragmentation peaks
compared with standards present in the software database, while for the quantification of the
amounts of reactants consumed and products generated, a GCꢀFID (Agilent 7820A equipped
with an Agilent HPꢀ5, 30 m x 320 µm x 0.25 µm column) was employed and the external
calibration method was used. Turnover frequency (TOF) were calculated according to the
following equation:
ꢃꢄꢅꢆꢇꢈꢉ
ꢀꢁꢂ =
ꢃꢄꢅꢊꢋ ∗ ꢌ
where molconv is the moles of CAL converted, molMe is the moles of the active sites and t is
the reaction time in hours. The total metal content was obtained from MPꢀAES analysis.
5

Catalysis Science & Technology
Page 6 of 27
View Article Online
DOI: 10.1039/C7CY02556D
The catalyst reusability was carried out using the following experimental procedure. After
90 minutes of reaction, the flask was cooled to room temperature in an ice bath and the
catalyst allowed to settle at the bottom of the flask. The supernatant liquid reaction mixture
was removed carefully and the solid catalyst was washed with acetone 2 times and allowed
to dry in air. A second reaction was then started with the washed catalyst and this procedure
was repeated 4 times.
Catalyst characterisation
The metal colloids were analysed by UVꢀVis (Agilent Cary 60) spectroscopy to study the
surface plasmon resonance peak of gold and the completion of metal reduction. The UVꢀvis
spectra were recorded in quartz cuvettes after 30 minutes of sol generation.
TEM experiments were carried out on a JEOL JEMꢀ2100 electron microscope with a 200
kV accelerating voltage. The samples were first dispersed in methanol and sonicated for 5
minutes and then a drop was placed onto a 300 mesh carbonꢀcoated copper grid.
Nitrogen physisorption analysis was performed at – 196 °C with a Quantachrome
Quadrasorb SI after evacuation of the samples at 120 °C for 2 hours. Surface area and pore
volume were calculated using BrunauerꢀEmmetꢀTeller (BET) theory.
Powder XRD measurements were carried out at room temperature in a PANalytical
X’PertPRO instrument using a Cu Kα radiation source, (Kα, λ= 1.5418 Ao). The
diffractograms were collected over the 10ꢀ80 ° 2θ range at a rate of 1 °/min. The metal
loading of the catalysts was verified using an Agilent 4100 MPꢀAES. The samples were
firstly dissolved in aqua regia, diluted and filtered to remove the undissolved support. Xꢀray
photoelectron spectroscopy (XPS) measurements were performed on a Thermo Scientific Kꢀ
alpha+ spectrometer. Samples were analysed using a monochromatic Al Xꢀray source
operating at 72 W (6 mA x 12 kV), with the signal averaged over an ovalꢀshaped area of
approximately 600 x 400 microns. Data was recorded at pass energies of 150 eV for survey
scans and 40 eV for high resolution scan with a 1eV and 0.1 eV step size respectively.
Charge neutralisation of the sample was achieved using a combination of both low energy
electrons and argon ions (less than 1 eV) which gave a C(1s) binding energy of 284.8 eV.
All data were analysed using CasaXPS (v2.3.17 PR1.1) using Scofield sensitivity factors
and an energy exponent of ꢀ0.6.
Diffuse reflection infrared spectroscopy (DRIFTS) was carried out with a Bruker Tensor 27
spectrometer fitted with a HgCdTe (MCT) detector and operated with OPUS software. The
6

Page 7 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Harrick Praying Mantis HVCꢀDRPꢀ4 cell was equipped with ZnSe windows and included
gas inlet and outlet and vacuum ports as well as capabilities for heating and cooling. The
required gas flow was introduced using a 2 % CO/N2 gas mixture at 40 cm3 minꢀ1 of, over a
period of 60 minutes. The mixture composition and flow rate was controlled by massꢀflow
controllers. Each absorbance spectrum represents an average of 64 scans with a spectral
resolution of 2 cmꢀ1. Prior to analysis, the gasꢀphase CO signal was removed by subtracting
the spectra recorded under CO containing atmosphere, followed by background and
normalisation of the spectra. No high temperature pretreatments were performed on the
catalysts in order to avoid sintering of metal nanoparticles.
Results and discussion
Catalyst characterisation
A series of AuPd catalysts with varying metal ratios were prepared. The AuPd metal loading
and ratios were confirmed by XPS & MPꢀAES (total metal content = 1 wt. %) (Supporting
information, Table S1). The average metal particle size and particle size distribution of the
catalysts were determined by TEM (supporting information, Figure S1 and Table S2).
Representative bright field TEM micrographs of monometallic and bimetallic catalysts are
presented in Figure 2. The mean particle sizes of all the supported bimetallic AuxPdy
nanoparticles were between 2.1 (±0.4) ꢀ 2.7 (± 1.0) nm without any appreciable size
variation for different AuPd ratios. The presence of small metal nanoparticles was evidenced
by the low intensity of Au (200) and Pd (200) reflexions in the powder XRD of the catalysts
(supporting information, Figure S2). BET analysis on the bare support TiO2 and on the final
Au50Pd50/TiO2 shows no substantial difference in surface area and porosity due to metal
deposition (Table S3).
XPS data of the supported monometallic and bimetallic catalysts are presented in Figure 3.
For the whole series of samples, the Au4f7/2 peak were between 83.3ꢀ83.7 eV confirming
that Au is in the metallic state.25 The Pd(3d) spectra showed the presence of metallic Pd
(334.9 eV) as the major species and some Pd2+, attributed to PdO (337.0 eV) since no
chloride is detected, (Figure 3). For the monometallic Au/TiO2 sample the B.E. of Au was
83.7 eV, lower than that of bulk gold (ca. 84.0ꢀ84.2 eV). This slight decrease in B.E. is be
attributed to (i) particle size effects and (ii) charging of Au particles (presence of Au with
partially negative charge Auδꢀ). The addition of Pd to the Au samples caused a downward
7

Catalysis Science & Technology
Page 8 of 27
View Article Online
DOI: 10.1039/C7CY02556D
shift in the BE of Au from 83.7 to 83.3 eV indicating an electronic interaction between the
26
Au and Pd as noted by Lee et al
.
In common with that study the Pd B.E. shifts were no
greater than 0.2 eV, which we consider to be the experimental confidence limit for absolute
B.E. (supporting information, Figure S3). With the small size of the nanoparticles, the
photoelectron attenuation lengths are of the same order as the particle sizes, therefore
absolute information on the formation of coreꢀshell particles from XPS alone is difficult,
however analysis of the bimetallic catalysts revealed the Pd/Au atomic ratios were close to
the expected nominal values suggesting the presence of random alloys rather than coreꢀshell
structures and in agreement with our previous reports (supporting information, Table S1)27.
COꢀDRIFTS analysis was performed on the bimetallic AuꢀPd and monometallic Au, Pd
nanoparticles supported on TiO2, Figure 4. The monometallic Au catalyst shows a primary
peak at around 2120 cmꢀ1 comprised of two peaks at 2124 and 2112 cmꢀ1 assigned to CO
linearly bonded on Au0 step sites on top of nanoparticles and on peripheral Au0 step sites
respectively (deconvolution shown in supporting information, Figure S4).28 The presence of
the latter peak, in particular, suggests the presence of small nanoparticles on the catalyst
surface. The peak at 2072 cmꢀ1, on the other hand, has been assigned to linear CO adsorbed
on Auδꢀ species.29 These partially negative species form during the gradual reduction of Au0
to Auδꢀ by the adsorbed CO at high CO partial pressure,30 as shown by comparison of the
spectra at low CO partial pressure (supporting information, Figure S4). The broad peak at
2050ꢀ1950 cmꢀ1 has been assigned to bridge bonded CO on Auδꢀ species. The presence of
this broad band is a direct effect of very well dispersed small Au nanoparticles on the
surface of support, as reported by Tabakova et al 31
.
The COꢀDRIFTS spectra of monometallic Pd catalyst (deconvolution shown in supporting
information, Figure S5), consists of two broad groups of peaks. The first group, is comprised
of 3 peaks at 2130, 2096 and 2083 cmꢀ1; the latter two peaks have been assigned to linear
CO coordinated on Pd0,32–34 while the peak at 2130 cmꢀ1 suggests the presence of cationic Pd
species,33,34 most likely PdO, in agreement with XPS analysis. The second group at lower
wavenumbers, is comprised by multiple and broad peaks assigned to µ2 bridge bonded CO
(ca. 1980ꢀ1930 cmꢀ1) and µ3 bridge bonded CO (ca. 1930ꢀ1820 cmꢀ1) in accordance with
previous literature.32,34
8

Page 9 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
In the case of bimetallic AuPd catalysts, it has been previously reported that Pd carbonyls
are much more stable than Au carbonyls due to the higher πꢀbackꢀdonation of the PdꢀCO
bond.35 For this reason, the typical PdꢀCO bands dominate the DRIFTS spectra in AuPd
alloys when Pd is present in significant amount and it is therefore difficult to observe bands
related to the presence of Au sites. Indeed, for the Au35Pd65, Au50Pd50 and Au65Pd35
catalysts, all the COꢀPd features observed for the monometallic Pd catalyst are present,36
with varying intensities corresponding to the amount of Pd (supporting information, Figure
S6). The same explanation applies to the Au65Pd35 catalyst, where the nanoparticle mean
size is very similar to the Pd/TiO2. Interestingly, when the Au/Pd ratio is 95:5, the CO
stretching modes are dominated by the COꢀAu interactions (supporting information, Figure
S7). The appearance of a strong peak at 2123 cmꢀ1 with a small shoulder at 2103 cmꢀ1 is
evident and these peaks are characteristic of linear CO bonded on Au0 step sites on top of
nanoparticles and of peripheral Au0 step sites.
Cinnamaldehyde hydrogenation over Au50Pd50/TiO2
Cinnamaldehyde (CAL) hydrogenation reactions were initially performed at 100 °C and
under 1 bar of H2 using a bimetallic goldꢀpalladium catalyst (Au50Pd50/TiO2, 1 wt.%, 1:1
Au/Pd molar ratio) as our chosen model catalyst, and the main products observed were
cinnamyl alcohol (COH), hydrocinnamaldehyde (HCAL), hydrocinnamyl alcohol (HCOH)
and phenylpropane (PPR) (Figure 5). The conditions to avoid diffusion limitations were
identified to ensure reactions were not under mass transfer limitations (supporting
informations, Figure S8).
The influence of hydrogen pressure on both activity and selectivity was also investigated. Several
experiments were carried out varying the hydrogen pressure from 1 to 3 bar and using the optimised
reaction conditions reported above. The increase of H2 pressure enhances the concentration of the
gas in the reaction medium, increasing the accessibility of the H2 molecule. Figure 6a shows that the
TOF for cinnamaldehyde hydrogenation increased linearly from 600 to 1080 hꢀ1 as the partial
pressure increased in the studied range, confirming that the reaction is a first order in respect of the
hydrogen partial pressure.24 The reaction temperature effect was subsequently evaluated by varying
the reaction temperature from 40 to 100 °C and maintaining the hydrogen partial pressure constant
at 1 bar. It is well known that hydrogenation rates usually increase with increasing temperatures,7
and here too there is a direct correlation between TOF and temperature (Figure 6b). It should be
9

Catalysis Science & Technology
Page 10 of 27
View Article Online
DOI: 10.1039/C7CY02556D
noted, however, that the hydrogenation selectivity to HCAL at isoꢀconversion level over the
Au50Pd50/TiO2 catalyst remained unchanged with variation of both H2 pressure and reaction
temperature (supporting information, Figure S9 c and d, 40 – 45 % selectivity at 30 % of isoꢀ
conversion). This behaviour is in agreement with previous studies on bimetallic catalysts.24,37,38 The
apparent activation energy, from Arrhenius plot, was 24 kJ molꢀ1 (supporting information, Figure
S10). The obtained value is in good agreement with the data reported in the literature for various
metal based catalysts (supporting information, Table S4).38–42
3.4 Solvent effects
Toluene was initially employed as solvent in order to avoid potential transfer
hydrogenations of CAL with hydrogen donor solvents.It has been shown previously that the
solvent plays a very important role in this reaction in terms of activity and
selectivity.12,14,19,22,23. Besides toluene, polar solvents (ethanol and isopropanol) were also
tested for the hydrogenation of cinnamaldehyde hydrogenation. It has been reported for the
CAL hydrogenation reaction over Pd/C, Pt/C and Co/Al2O3, that polar solvents promote the
hydrogenation of carbonyl group, while nonꢀpolar solvents favour the hydrogenation of the
vinyl group.43 Our results (Figure 7) show a small variation of catalytic activity (± 7 %
conversion after 2 hours), between the three solvents studied, however significant difference
in selectivity is observed between the solvents. It is interesting to notice the strong decrease
in selectivity to HCAL, with ethanol (only 3 % is produced at 85 % conversion), and the
increase of PPR selectivity (from 3 % with toluene up to 18 % with ethanol). In presence of
Lewis acid sites and small chain alcohols, it has been previously reported for this reaction
the formation of acetals.14,23 Indeed, when both ethanol and isopropanol were used,
considerable amounts of acetals were produced (33 % and 25 % respectively).
3.5 Effect of Au/Pd molar ratio
The effect of Au/Pd molar ratio on (i) cinnamaldehyde hydrogenation activity (adsorption of
cinnamaldehyde on metal active sites) and (ii) the selectivity towards to the C=O or C=C
hydrogenation was studied. A series of catalysts with different Au/Pd metal molar ratios
were prepared (Au, Au95Pd5, Au65Pd35, Au50Pd50, Au35Pd65, and Pd, supporting information,
Table S1). Monometallic Au/TiO2 and Au95Pd5/TiO2 catalysts were inactive (Figure 8); an
increase of catalytic activity based on TOF was observed only with a further addition of Pd,
meaning that a minimum amount of this noble metal is required to initiate the reaction. This
is in agreement with the COꢀDRIFT results (Figure 4), where no Pd features were detected
in the Au95Pd5 catalyst. When the molar ratio between Au and Pd is approximately 50:50,
10

Page 11 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
the activity reached a maximum with a TOF of 836 hꢀ1, and a further addition of Pd
decreased the catalytic performance. Activity enhancement due to alloying effect were also
reported by other groups, but no effect on selectivity was reported.21,24 Only Parvulescu et
al., reported an increase in COH selectivity using AuPd colloid embedded in SiO2 as
catalyst,23 while Szumelda and coꢀworkers observed a progressive decrease in both activity
and selectivity towards HCAL starting from a Pd/C catalyst and increasing the amount of
Au.22 In both cases, changes in selectivity are explained as geometric and electronic effects.
It is clear from these results that Au, although totally inactive when is present alone under
our experimental conditions, plays an important role when alloyed with Pd, increasing the
catalytic activity of more than 50 % with respect to the monometallic Pd (from 541 to 836 hꢀ
1). In terms of selectivity, contrary of what previously reported by Parvulescu and coꢀ
workers for AuPd on SiO2, an increase in the content of Pd leads to higher production of
HCAL up to 82 % in selectivity, with a consequential decrease in HCOH selectivity at isoꢀ
conversion levels (Figure 9). It is possible that this selectivity improvement is simply caused
by the increase in Pd availability on the surface of the nanoparticles with the increase of total
Pd content.
3.6 Catalyst stability
Au50Pd50/TiO2 catalyst was tested in multiple consecutive reactions in order to evaluate its
reusability. The results are reported in Figure 10. After the first run, the conversion
decreased from 73 % to 50 % and it stabilised at a value of 45 % for the following runs. The
selectivity, however, changed accordingly to the conversion level (HCOH decreased from
56 % to 45 %), indicating that the deactivation process affects mainly the catalytic activity
rather than the selectivity. MPꢀAES analysis showed a significant loss (about 22 %) of the
Pd content after 4 runs that partially explain the loss in activity. Pd leaching was also
detected when the reaction was performed with the monometallic Pd/TiO2 catalyst, with a
28 % loss after 4 consecutive reactions. A control experiment using the hot filtrate solution
without catalyst confirmed that eventual leached species do not take active part to the
reaction. Sintering of metal nanoparticles could be another reason for catalyst deactivation.
TEM analysis of the spent catalyst (Table 1) showed an increase in the mean particle size as
well as a broadening in size distribution. The FTIR spectra of the fresh and spent catalysts
(supporting information, Figure S11), show a weak band at 1670 cmꢀ1 accompanied by a
broad band in the range of 3700ꢀ3000 cmꢀ1 that can be attributed to the OH stretching and
bending typical for the stabilising agent used (supporting information, Figure S12).
11

Catalysis Science & Technology
Page 12 of 27
View Article Online
DOI: 10.1039/C7CY02556D
However, the enhanced peak at 1670 cmꢀ1 in the spent catalyst can be also attributed to the
alcoholic functional group of the CAL hydrogenation products and a series of small peaks at
around 1500ꢀ1400 cmꢀ1 typical account for the presence of aromatic C=C stretches that
indicate the presence of either COH or HCOH adsorbed on the surface of the catalyst
(supporting information, Figure S13). The stronger band at 3700ꢀ3000 cmꢀ1 in the fresh
catalyst is only due to remaining adsorbed water on the catalyst surface.
In order to improve the catalyst reusability, the Au50Pd50/TiO2 catalyst was heated at 200,
300 and 400 °C. High temperature calcination/reduction steps improve the metalꢀsupport
interaction via a mechanism known as strong metalꢀsupport interaction (SMSI);44 SMSI has
been accounted for stability and selectivity improvement in several reactions by a thin layer
of TiOx species that moves over and partially covers supported metal nanoparticles.45 The
results on the effect of heat treatment are reported in Table 1. Increasing the reduction
temperature decreases the activity from 73 for the dried only sample to 50 % for the catalyst
heated at 400 °C. However, the stability improves considerably, with the Au50Pd50/TiO2–300
°C and Au50Pd50/TiO2–400 °C catalysts maintain the catalytic activity for 4 consecutive
reactions. A small decrease in activity was observed for the Au50Pd50/TiO2–200 catalyst
(from 70 to 58 %). MPꢀAES tests on the final reaction solutions, revealed the absence of Pd,
confirming that high temperature treatments help to increase the stability of the catalyst by
increasing the metalꢀsupport interaction. XPS data (supporting information, Table S5) of the
heatꢀtreated catalysts do not suggest significant changes in metal composition with the heat
treatment, even though, as mentioned before, these data cannot be considered as a true
representation of the real surface composition. The overall loss in activity is probably due to
both the suppressed hydrogen chemisorption due to the SMSI and increase in mean particle
size with the increased reduction temperature (supporting information, Table S6), with a
measured average mean diameter that varies from 2.1 ± 0.6 nm to 5.3 ± 1.2 nm for the
Au50Pd50/TiO2 and Au50Pd50/TiO2–400 °C catalysts respectively. On the other hand, the
significant improvement in stability may be due to the protective TiOx surface layer prevent
metal sintering phenomena during reaction (Figure 11), and this is confirmed again by TEM
analysis (supporting information, Table S6): for high temperature treated catalysts, the
nanoparticles dimension changes only slightly (from 5.3 ± 1.2 nm to 5.9 ± 1.3 nm for the
Au50Pd50/TiO2–400 catalyst). The high temperature treatments not only influence the
activity of the catalysts, but also the selectivity: increasing the reduction temperature, the
selectivity towards CAL increases drastically from 40 to 83 % at isoꢀcoversion, with
Au50Pd50/TiO2 and Au50Pd50/TiO2–400 respectively. It is well known that high temperature
12

Page 13 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
treatments in AuPd alloy systems may cause Pd segregation on the surface of the
nanoparticles to form coreꢀshell structures.46–48 This hypothesis is also consistent with our
observations at different Au:Pd ratios, where a Pd rich surface led to higher selectivity
towards CAL. XPS analysis cannot clear out all doubts, since the penetration depth of the Xꢀ
ray is in the same order of magnitude of the nanoparticle dimensions, and the quantification
analysis are good only to evaluate the overall content of the two metals.
The Au50Pd50/TiO2ꢀ300 catalyst represents a good compromise between activity, selectivity
towards HCAL and stability to reuse, and it is among the most active and selective
bimetallic AuPd catalyst reported in literature.21–24
4. Conclusions
In this study, the catalytic hydrogenation of cinnamaldehyde has been investigated using
supported AuꢀPd nanoparticles prepared via sol immobilisation method. Systematic studies
have been carried out to optimise catalyst properties and reaction conditions. We have
shown that the most active bimetallic AuꢀPd catalyst is with 1:1 molar ratio. In addition,
nature of solvent significantly affects the selectivity for this hydrogenation reaction. When
nonꢀpolar toluene was used as the solvent higher selectivity of HCAL and HCOH was
observed, whereas when polar solvents (ethanol and isopropanol) were used a significant
increase in the formation of acetal and PPR was observed. Finally, the effect of heat
treatments was studied and by increasing the heat treatment temperature under air and
hydrogen we switched off the leaching of Pd observed at lower heat treatment temperatures
without significantly affecting the catalytic activity, therefore improving the longꢀterm
stability of bimetallic catalysts. The selectivity towards HCAL improved significantly as
well, probably due to Pd segregation on the surface of the nanoparticles.
References
1
Z. Guo, B. Liu, Q. Zhang, W. Deng, Y. Wang and Y. Yang, Chem. Soc. Rev., 2014, 43,
3480–524.
2
3
4
G. C. Bond, MetalꢀCatalysed Reactions of Hydrocarbons, 2005.
L. Yin and J. Liebscher, Chem. Rev., 2007, 107, 133–173.
D. I. Enache, J. K. Edwards, P. Landon, B. SolsonaꢀEspriu, A. F. Carley, A. A. Herzing, M.
Watanabe, C. J. Kiely, D. W. Knight and G. J. Hutchings, Science (80ꢀ. )., 2006, 311, 362–
365.
W. Luo, M. Sankar, A. M. Beale, Q. He, C. J. Kiely, P. C. A. Bruijnincx and B. M.
Weckhuysen, Nat. Commun., 2015, 6, 1–10.
5
6
7
8
J. K. Edwards and G. J. Hutchings, Angew. Chemie Int. Ed., 2008, 47, 9192–9198.
P. MäkiꢀArvela, J. Hájek, T. Salmi and D. Y. Murzin, Appl. Catal. A Gen., 2005, 292, 1–49.
P. Gallezot and D. Richard, Selective Hydrogenation of α,βꢀUnsaturated Aldehydes, 1998,
vol. 40.
13

Catalysis Science & Technology
Page 14 of 27
View Article Online
DOI: 10.1039/C7CY02556D
9
Ullmann’s Encyclopedia of Industrial Chemistry, 7th edn., 2014.
US 4209643, 1978.
10
11
12
13
US 5632980 A, 1997.
E. Bus, R. Prins and J. a. van Bokhoven, Catal. Commun., 2007, 8, 1397–1402.
L. He, J. Ni, L. C. Wang, F. J. Yu, Y. Cao, H. Y. He and K. N. Fan, Chem. ꢀ A Eur. J., 2009,
15, 11833–11836.
14
C. Milone, C. Crisafulli, R. Ingoglia, L. Schipilliti and S. Galvagno, Catal. Today, 2007, 122,
341–351.
15
16
H. Shi, N. Xu, D. Zhao and B.ꢀQ. Xu, Catal. Commun., 2008, 9, 1949–1954.
X. Zhang, Y. C. Guo, Z. Cheng Zhang, J. Sen Gao and C. M. Xu, J. Catal., 2012, 292, 213–
226.
17
18
A. M. R. Galletti, C. Antonetti, A. M. Venezia and G. Giambastiani, Appl. Catal. A Gen.,
2010, 386, 124–131.
K. Anderson, S. Cortiñas Fernández, C. Hardacre and P. C. Marr, Inorg. Chem. Commun.,
2004, 7, 73–76.
19
20
21
22
Z. Tian, X. Xiang, L. Xie and F. Li, Ind. Eng. Chem. Res. 2013, 2013, 52, 288−296.
S. Fujiwara, N. Takanashi, R. Nishiyabu and Y. Kubo, Green Chem., 2014, 3230–3236.
P. Dash, N. A. Dehm and R. W. J. Scott, J. Mol. Catal. A Chem., 2008, 286, 114–119.
T. Szumełda, A. Drelinkiewicz, R. Kosydar and J. Gurgul, Appl. Catal. A Gen., 2014, 487,
1–15.
23
V. I. Pârvulescu, V. Pârvulescu, U. Endruschat, G. Filoti, F. E. Wagner, C. Kübel and R.
Richards, Chem. ꢀ A Eur. J., 2006, 12, 2343–2357.
24
25
X. Yang, D. Chen, S. Liao, H. Song, Y. Li, Z. Fu and Y. Su, J. Catal., 2012, 291, 36–43.
N. Dimitratos, J. A. LopezꢀSanchez, D. Morgan, A. F. Carley, R. Tiruvalam, C. J. Kiely, D.
Bethell and G. J. Hutchings, Phys. Chem. Chem. Phys., 2009, 11, 5142–5153.
Y. Lee and Y. Jeon, J. korean Phys. Soc., 2000, 37, 451–455.
N. Dimitratos, J. A. LopezꢀSanchez, J. M. Anthonykutty, G. Brett, A. F. Carley, R. C.
Tiruvalam, A. A. Herzing, C. J. Kiely, D. W. Knight and G. J. Hutchings, Phys. Chem.
Chem. Phys., 2009, 11, 4952.
M. Manzoli, A. Chiorino and F. Boccuzzi, Surf. Sci., 2003, 532–535, 377–382.
M. Compagnoni, S. A. Kondrat, C. E. ChanꢀThaw, D. J. Morgan, D. Wang, L. Prati, A. Villa,
N. Dimitratos and I. Rossetti, ChemCatChem, 2016, 8, 2136–2145.
K. Chakarova, M. Mihaylov, S. Ivanova, M. A. Centeno and K. Hadjiivanov, J. Phys. Chem.
C, 2011, 115, 21273–21282.
T. Tabakova, F. Boccuzzi, M. Manzoli and D. Andreeva, Appl. Catal. A Gen., 2003, 252,
385–397.
T. Lear, R. Marshall, J. A. LopezꢀSanchez, S. D. Jackson, T. M. Klapötke, M. Bäumer, G.
Rupprechter, H. J. Freund and D. Lennon, J. Chem. Phys., 2005, 123, 174706.
H. Zhu, Z. Qin, W. Shan, W. Shen and J. Wang, J. Catal., 2004, 225, 267–277.
S. M. Rogers, C. R. A. Catlow, C. E. ChanꢀThaw, A. Chutia, N. Jian, R. E. Palmer, M.
Perdjon, A. Thetford, N. Dimitratos, A. Villa and P. P. Wells, ACS Catal., 2017, 7, 2266–
2274.
26
27
28
29
30
31
32
33
34
35
36
A. Villa, N. Dimitratos, C. E. ChanꢀThaw, C. Hammond, G. M. Veith, D. Wang, M.
Manzoli, L. Prati and G. J. Hutchings, Chem. Soc. Rev., 2016, 45, 4953–4994.
J. H. Carter, S. Althahban, E. Nowicka, S. J. Freakley, D. J. Morgan, P. M. Shah, S.
Golunski, C. J. Kiely and G. J. Hutchings, ACS Catal., 2016, 6, 6623–6633.
K. Q. Sun, Y. C. Hong, G. R. Zhang and B. Q. Xu, ACS Catal., 2011, 1, 1336–1346.
W. O. Oduro, N. Cailuo, K. M. K. Yu, H. Yang and S. C. Tsang, Phys. Chem. Chem. Phys.,
2011, 13, 2590–602.
37
38
39
40
S. Mahmoud, A. Hammoudeh, S. Gharaibeh and J. Melsheimer, J. Mol. Catal. A Chem.,
2002, 178, 161–167.
L. J. Durndell, C. M. a. Parlett, N. S. Hondow, M. a. Isaacs, K. Wilson and A. F. Lee, Sci.
Rep., 2015, 5, 9425.
H. Li, X. Chen, M. Wang and Y. Xu, Appl. Catal. A Gen., 2002, 225, 117–130.
J. P. Breen, R. Burch, J. GomezꢀLopez, K. Griffin and M. Hayes, Appl. Catal. A Gen., 2004,
268, 267–274.
41
42
43
44
45
H. Yamada and S. Goto, J. Chem. Eng. Japan, 2003, 36, 586–589.
D. D. Eley, H. Pines and P. B. Weisz, Advances in Catalysis, Academic Press Inc., 1989.
J. C. Matsubu, S. Zhang, L. DeRita, N. S. Marinkovic, J. G. Chen, G. W. Graham, X. Pan
14

Page 15 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
and P. Christopher, Nat. Chem., 2016, 9, 120–127.
R. Sharpe, J. Counsell and M. Bowker, Surf. Sci., 2017, 656, 60–65.
46
47
A. A. Herzing, A. F. Carley, J. K. Edwards, G. J. Hutchings and C. J. Kiely, Chem. Mater.,
2008, 20, 1492–1501.
48
L. Hilaire, P. Légaré, Y. Holl and G. Maire, Surf. Sci., 1981, 103, 125–140.
15

Catalysis Science & Technology
Page 16 of 27
View Article Online
DOI: 10.1039/C7CY02556D
Table 1. Catalysts reusability in the CAL hydrogenation. Reaction conditions 90 minutes,
100 °C, 1 H2 bar, catalyst amount 50 mg, CAL 4 mmol, substrate/metal molar ratio 1200:1.
Catalyst/Catalyst
preꢀtreatment
Run Mean particle Conv. (%)
size (nm)
Selectivity at 50 % conversion
HCAL
COH
HCOH
PPR
Au50Pd50/TiO2 ꢀ dried
Au50Pd50/TiO2 ꢀ dried
Au50Pd50/TiO2 ꢀ 200 °C
Au50Pd50/TiO2 ꢀ 200 °C
Au50Pd50/TiO2 ꢀ 300 °C
Au50Pd50/TiO2 ꢀ 300 °C
Au50Pd50/TiO2 ꢀ 400 °C
Au50Pd50/TiO2 ꢀ 400 °C
1
4
1
4
1
4
1
4
2.1 ± 0.6
3.3 ± 0.9
2.5 ± 0.8
3.4 ± 0.8
3.7 ± 1.0
3.8 ± 0.8
5.3 ± 1.2
5.9 ± 1.3
73
46
70
58
56
55
50
49
40
49
67
62
81
78
83
79
6
52
45
31
34
18
21
16
20
2
1
1
3
2
2
1
1
5
1
0
0
0
0
0
16

Page 17 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Figure 1. Cinnamaldehyde hydrogenation pathways.
CAL
COH
PPE
OH
OH
O
O
H2
H2
ꢀH2O
H2
H2
H2
H2
H2
ꢀH2O
HCAL
HCOH
PPR
17

Catalysis Science & Technology
Page 18 of 27
View Article Online
DOI: 10.1039/C7CY02556D
Figure 2. Representative TEM images of the fresh catalysts (a) Au/TiO2 (b) Au95Pd5/TiO2
(c) Au65Pd35/TiO2 (d) Au50Pd50/TiO2 (e) Au35Pd65/TiO2 (f) Pd/TiO2.
18

Page 19 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Figure 3. Pd (3d) XPS spectra for the AuxPdy/TiO2 catalysts: (a) Pd/TiO2 (b) Au35Pd65/TiO2
(c) Au50Pd50/TiO2 (d) Au65Pd35/TiO2 (e) Au95Pd5/TiO2 (f) Au/TiO2.
Pd(3d) / Au(4d)
(a)
(b)
(c)
(d)
(e)
(f)
370 365 360 355 350 345 340 335 330 325 320
Binding Energy / eV
19

Catalysis Science & Technology
Page 20 of 27
View Article Online
DOI: 10.1039/C7CY02556D
Figure 4. COꢀDRIFT of (a) Pd/TiO2, (b) Au35Pd65/TiO2, (c) Au50Pd50/TiO2, (d)
Au65Pd35/TiO2, (e) Au95Pd5/TiO2 and (f) Au/TiO2.
(a)
(b)
(c)
(d)
(e)
(f)
2300
2200
2100
2000
1900
1800
Wavenumber (cm-1)
20

Page 21 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Figure 5. Cinnamaldehyde hydrogenation profile. Reaction conditions 100 °C, 1 H2 bar,
catalyst amount 50 mg, CAL 4 mmol, substrate/metal molar ratio 1200:1.
0.8
CAL
HCAL
COH
HCOH
PPR
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0
30
60
90
120
150
Time (min)
21

Catalysis Science & Technology
Page 22 of 27
View Article Online
DOI: 10.1039/C7CY02556D
Figure 6. Effect of (a) H2 pressure and (b) temperature on the TOF of the reaction. Reaction
conditions (a) 100 °C, catalyst amount 50 mg, CAL 4 mmol, (b) 1 H2 bar, catalyst amount
50 mg, CAL 4 mmol.
1000
1400
800
600
400
200
0
1200
1000
800
600
400
200
0
313
333
353
373
1
2
3
T (K)
H2 Pressure (bar)
22

Page 23 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Figure 7. Solvent effect in the CAL hydrogenation. Reaction conditions 100 °C, 1 H2 bar,
catalyst amount 50 mg, CAL 4 mmol, substrate/metal molar ratio 1200:1.
100
80
60
40
20
0
100
80
60
40
20
0
HCAL
COH
HCOH
PPR
Acetal
Conv
Toluene
EtOH
i-PrOH
Solvent
23

Catalysis Science & Technology
Page 24 of 27
View Article Online
DOI: 10.1039/C7CY02556D
Figure 8. Au/Pd ratio effect in the CAL hydrogenation. Reaction conditions 100 °C, 1 H2
bar, catalyst amount 50 mg, CAL 4 mmol, substrate/metal molar ratio 1200:1.
1000
800
600
400
200
0
100
80
60
40
20
0
Au content (%)
24

Page 25 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Figure 9. Au/Pd molar ratio effect in the CAL hydrogenation. Reaction conditions 100 °C, 1
H2 bar, catalyst amount 50 mg, CAL 4 mmol, substrate/metal molar ratio 1200:1.
100
80
HCAL
COH
60
HCOH
PPR
40
20
0
65
50
35
0
Au content
25

Catalysis Science & Technology
Page 26 of 27
View Article Online
DOI: 10.1039/C7CY02556D
Figure 10. Reusability of the Au50Pd50/TiO2 catalyst in the CAL hydrogenation. Reaction
conditions 100 °C, 1 H2 bar, catalyst amount 50 mg, CAL 4 mmol, substrate/metal molar
ratio 1200:1.
100
80
60
40
20
0
100
80
60
40
20
0
HCAL
COH
HCOH
PPR
Conv
1
2
3
4
Run (#)
26

Page 27 of 27
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02556D
Figure 11. TEM evidence of the presence of a TiOx layer covering the metal NPs.
27