Alumina-grafted SBA-15 as a high performance support for Pd-catalysed cinnamyl alcohol selective oxidation

Article abstract of DOI:10.1016/j.cattod.2013.11.056

Ultrathin alumina monolayers grafted onto an ordered mesoporous SBA-15 silica framework afford a composite catalyst support with unique structural properties and surface chemistry. Palladium nanoparticles deposited onto Al-SBA-15 via wet impregnation exhibit the high dispersion and surface oxidation characteristic of pure aluminas, in conjunction with the high active site densities characteristic of thermally stable, high-area mesoporous silicas. This combination confers significant rate enhancements in the aerobic selective oxidation (selox) of cinnamyl alcohol over Pd/Al-SBA-15 compared to mesoporous alumina or silica supports. Operando, liquid-phase XAS highlights the interplay between dissolved oxygen and the oxidation state of palladium nanoparticles dispersed over Al-SBA-15 towards on-stream reduction: ambient pressures of flowing oxygen are sufficient to hinder palladium oxide reduction to metal, enabling a high selox activity to be maintained, whereas rapid PdO reduction and concomitant catalyst deactivation occurs under static oxygen. Selectivity to the desired cinnamaldehyde product mirrors these trends in activity, with flowing oxygen minimising CO cleavage of the cinnamyl alcohol reactant to trans-β-methylstyrene, and of cinnamaldehyde decarbonylation to styrene.

Full text of DOI:10.1016/j.cattod.2013.11.056

Catalysis Today 229 (2014) 46–55
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Alumina-grafted SBA-15 as a high performance support for
Pd-catalysed cinnamyl alcohol selective oxidation
Christopher M.A. Parletta,∗, Lee J. Durndella, Andreia Machadoa, Giannantonio Cibinc,
Duncan W. Bruceb, Nicole S. Hondowd, Karen Wilsone, Adam F. Leea,f,∗
a Department of Chemistry, University of Warwick, Coventry CV4 7AL, UK
b Department of Chemistry, University of York, Heslington, York, YO10 5DD, UK
c Diamond Light Source, Harwell Science and Innovation Campus, Didcot OX11 0DE, UK
d Institute for Materials Research, School of Process, Environmental and Materials Engineering, University of Leeds, Leeds LS2 9JT, UK
e European Bioenergy Research Institute, Aston University, Aston Triangle B4 7ET, UK
f
School of Chemistry, Monash University, Victoria 3800, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 July 2013
Received in revised form 16 October 2013
Accepted 13 November 2013
Available online 7 January 2014
Ultrathin alumina monolayers grafted onto an ordered mesoporous SBA-15 silica framework afford a
composite catalyst support with unique structural properties and surface chemistry. Palladium nanopar-
ticles deposited onto Al-SBA-15 via wet impregnation exhibit the high dispersion and surface oxidation
characteristic of pure aluminas, in conjunction with the high active site densities characteristic of ther-
mally stable, high-area mesoporous silicas. This combination confers significant rate enhancements in
the aerobic selective oxidation (selox) of cinnamyl alcohol over Pd/Al-SBA-15 compared to mesoporous
alumina or silica supports. Operando, liquid-phase XAS highlights the interplay between dissolved oxy-
gen and the oxidation state of palladium nanoparticles dispersed over Al-SBA-15 towards on-stream
reduction: ambient pressures of flowing oxygen are sufficient to hinder palladium oxide reduction to
metal, enabling a high selox activity to be maintained, whereas rapid PdO reduction and concomitant
catalyst deactivation occurs under static oxygen. Selectivity to the desired cinnamaldehyde product mir-
rors these trends in activity, with flowing oxygen minimising C O cleavage of the cinnamyl alcohol
reactant to trans-␤-methylstyrene, and of cinnamaldehyde decarbonylation to styrene.
Keywords:
Palladium
Selective oxidation
Operando
XAS
Alumina
SBA-15.
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
acid. Aldehyde selectivity is also influenced by competitive hydro-
genation of the C C bond or C O cleavage in the parent alcohol,
Cinnamaldehyde is an allylic aldehyde widely employed as a
chemical intermediate in organic synthesis and as a desirable prod-
uct in its own right for application as an insecticide and food
and perfume additive [1]. Related aromatic and aliphatic allylic
aldehydes are also commercially important fine chemical and
agrochemical products [2–4], hence their green synthesis through
low-energy and atom-economical routes has been the focus of
extensive academic and commercial research. The direct aerobic
selective oxidation (selox) of allylic alcohols is a potential route to
such unsaturated aldehydes that avoids recourse to stoichiometric
oxidants and can be undertaken in either the liquid or vapour phase
under mild conditions [5,6]. Such chemistry is believed to proceed
via oxidative dehydrogenation of the alcohol to the corresponding
allylic aldehyde which may undergo subsequent oxidation to the
and decarbonylation of the desired aldehyde product.
A number of heterogeneous catalysts capable of facilitating
alcohol selox using air or dioxygen as the oxidant have been
developed, with nanoparticulate palladium and/or gold the active
components of choice [7–11]. In recent years, an array of ex
situ and in situ analytical techniques and parallel kinetic pro-
filing has identified electron-deficient Pd2+, present as PdO at
the surface of nanoparticles, as the active species participating
in palladium-catalysed allylic alcohol selox. Pd2+species are pro-
posed to play a critical role in aerobic alcohol oxidations over AuPd
core-shell nanoparticles [12] and intercalated Pd/NiZn layered
double hydroxides [13], while Pd2+ complexes are ubiquitous in
homogeneously catalysed selox chemistry [14,15]. Time-resolved,
liquid and vapour phase XAS measurements [16–19] on carbon
and oxide supported nanoparticles respectively, and complemen-
tary in situ XPS measurements over O/Pd(1 1 1) single crystals
[20,21] have highlighted PdO reduction to metallic palladium as
a potential deactivation pathway, and hence the importance of
generating highly dispersed nanoparticles which favour an oxide
phase [20,22–24]. Temperature-programmed desorption and XPS
∗
Corresponding author at: University of Warwick, Department of Chemistry,
Gibbet Hill Road, Coventry CV47AL, UK. Tel.: +44 0 2476 151839/29208 74778.
E-mail addresses: c.m.a.parlett@warwick.ac.uk (C.M.A. Parlett),
a.f.lee@warwick.ac.uk (A.F. Lee).
0920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.11.056

C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
47
in water (75.5 cm3) and HCl (2 M, 291.5 cm3) and stirred at 35 ◦C.
Tetramethoxysilane (15.5 cm3) was then added, and the mixture
left for 20 h under agitation. The resulting gel was aged for 24 h at
100 ◦C without agitation. The remaining solid was filtered, washed
with water (1000 cm3) and dried at room temperature before cal-
cining at 500 ◦C for 6 h under flowing air (ramp rate 1 ◦C min−1). The
protocol of Landau and co-workers was used to deposit a uniform
alumina coating onto SBA-15 (Al-SBA-15(n) where n corresponds
to the number of grafting cycles) [36]. Aluminium-tri-sec-butoxide
(14.5 g) was dissolved in anhydrous toluene (100 cm3) at 85 ◦C with
stirring. Triethylamine (2.1 cm3) was added to the solution fol-
lowed by dried SBA-15 (1 g). After 6 h stirring at 85 ◦C the solution
was filtered under vacuum (∼0.1 bar) to recover the solid, which
was washed three times with toluene (or 3 × 35 cm3 total). The alu-
mina precursor was hydrolysed in ethanol (318 cm3) containing
water (1.6 cm3) for 24 h at 25 ◦C. The solid product was recov-
ered by filtration under vacuum (∼0.1 bar), washed with ethanol
(300 cm3), and dried under vacuum (0.25 bar) at 50 ◦C on a rotary
evaporator. The solid was further dried at 120 ◦C in air, before a
three-step calcination sequence. The powdered material was ini-
tially heated to 250 ◦C for 1 h, then 400 ◦C for 1 h and finally 500 ◦C
for 4 h (common ramp rate 1 ◦C min−1). Consecutive alumina graft-
ing cycles were carried out using an identical protocol each time,
with adjustments to the quantities to maintain the initial Al:Si
ratios.
A nominal 1 wt.% Pd/Al-SBA-15(4) catalyst was prepared using
tetraamine palladium(II) nitrate solution (10 wt.% Pd salt) by
incipient wetness impregnation. A slurry of the palladium pre-
cursor and Al-SBA-15(4) support was stirred for 18 h at room
temperature before heating to 50 ◦C. After ∼5 h, agitation was
ceased and the solids aged at 50 ◦C for a further 24 h to yield
a dry powder. Dried samples were calcined at 500 ◦C (ramp
rate 1 ◦C min−1) in static air for 2 h, then reduced under flow-
ing H2 (10 cm3 min−1) at 400 ◦C (ramp rate 10 ◦C min−1) for 2 h.
The resulting materials were stored in air, and not re-activated
prior to use. Full synthesis, characterisation and reactivity of
Pd/SBA-15 and Pd/meso-Al2O3 controls are described in references
[22–24].
of crotyl alcohol (but-2-en-1-ol) and crotonaldehyde (but-2-en-1-
al) over Pd(1 1 1) and palladium-rich Au/Pd surface alloys suggest
deactivation arises from crotonaldehyde decarbonylation and sub-
sequent site-blocking by strongly adsorbed propylidyne and CO
[20,21]. Computational modelling predicts Pd(1 1 1) can also drive
propanoic acid decarbonylation [25]. Vapour phase studies of crotyl
alcohol oxidation show that such in situ palladium reduction is
reversible under oxidising environments [17,19], enabling selox to
be maintained, however the ability of dissolved oxygen to preserve
palladium as PdO during liquid phase selox is unknown.
Improved palladium selox catalysts require enhanced surface
PdO concentration. A number of approaches have been explored
to disperse palladium nanoparticles throughout nanoporous sup-
ports, e.g. Pd coordination to pyridyls [26] or amines [27,28]
covalently tethered to silicas. However, the grafting densities
achievable via such organic surface functionalisation are low (typ-
ically 0.1 ligands nm−2), and palladium is only weakly bound
and prone to leaching in situ. Direct impregnation of inorganic
oxides with palladium precursors (and subsequent thermochem-
ical processing) or pre-formed colloidal nanoparticles remain
the simplest, scalable routes to achieving high Pd dispersion,
although the presence of surfactant/polymer stabilisers can hin-
der good adhesion of pre-prepared colloids and/or modify their
catalytic reactivity [29]. Tailoring the support architecture offers
an alternative means to influencing palladium dispersion. Ordered
mesoporous silicas, prepared via surfactant templating routes
[30,31], possess high surface areas and interconnected mesopores
that permit the synthesis of highly oxidised 1–2 nm Pd particles
[22,23]. We recently discovered that mesoporous alumina is supe-
rior to silica for stabilising sub-nanometre palladium clusters and
even atomically dispersed Pd2+ centres [24,32], presumably reflect-
ing a stronger metal-support interaction (a function of the acid/base
character of metal oxides [33–35]). Unfortunately, pure alumina
supports are plagued by low hydrothermal stability, and typically
significantly lower surface areas and mesopore ordering than their
silica counterparts. The growth of conformal alumina coatings over
a mesostructured silica framework opens the way to new catalyst
supports combining the structural stability (and diverse topolo-
gies) accessible through silica materials chemistry, with the surface
chemistry of alumina. Alumina coated MCM-41 [36] and SBA-15
[37] have been reported via hydrolysis of alkoxyaluminium species.
Such dual-oxide supports have shown enhanced immobilisation of
heteropoly acids [38] and attendant activity in o-xylene alkylation
[39], and strong metal support interactions in bimetallic PtSn cat-
alysts promoting propane dehydrogenation [40]. Alumina grafting
also affords a direct means to introduce surface acidity into silica
[41] and thus control activity and selectivity in cobalt-catalysed
Fischer–Tropsch synthesis [42].
2.2. Catalyst characterisation
Accurate Pd loadings were determined by MEDAC Analytical and
Chemical Consultancy Service Ltd. Samples were digested in HF
prior to ICP analysis on a Varian Vista MPX inductively coupled
plasma-optical emission spectrometer (ICP-OES).
Nitrogen porosimetry was undertaken on a Quantachrome Nova
2000e porosimeter, with analysis carried out using NovaWin v11
software. Samples were degassed at 120 ◦C for 2 h prior to recording
N2 adsorption/desorption isotherms. BET surface areas were calcu-
lated over the relative pressure range 0.02–0.2, with microporosity
assessed by the t-plot method over the relative pressure range of
0.2–0.5; a linear relationship was observed for both. Pore diame-
ters and volumes were calculated applying the BJH method to the
desorption isotherm for relative pressures >0.35.
Here we report the first use of alumina-functionalised,
mesoporous silica as a support for catalytic oxidation. Palladium-
impregnated Al-SBA-15 exhibits significantly higher nanoparticle
dispersions and surface PdO concentrations than either pure meso-
porous alumina or SBA-15 supports, and correspondingly higher
activities in the liquid phase aerobic selox of cinnamyl alcohol
to cinnamaldehyde. Complementary operando XAS measurements
reveal a strong correlation between activity and in situ palladium
oxidation state, with dissolved oxygen maintaining an active PdO
species and thereby retarding on-stream catalyst deactivation.
Powder X-ray diffraction (XRD) patterns were recorded on
a Panalytical X’pertPro diffractometer fitted with an X’celerator
´
˚
detector using Cu K␣ (1.54 A) sources calibrated against a crystalline
Si standard. Low-angle patterns were recorded for 2ꢀ = 0.3–8◦
with a step size of 0.01◦. Wide-angle patterns were recorded for
2ꢀ = 25–75◦ with a step size of 0.02.
2. Experimental
Pd dispersions were estimated by CO chemisorption on
a Quantachrome ChemBET 3000 assuming a 1:2 CO:Pd stoi-
chiometry. Samples were outgassed at 150 ◦C under flowing He
(20 cm3 min−1) for 1 h, then reduced at 100 ◦C under flowing
hydrogen (20 cm3 min−1) for 1 h before room temperature anal-
ysis. This reduction protocol is far milder than that employed
2.1. Catalyst synthesis
The parent SBA-15 silica framework was synthesised following
the method of Zhao et al. [30]. Pluronic P123 (10 g) was dissolved

48
C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
during the catalyst synthesis, and hence does not induce particle
sintering.
2.4. Operando selox XAS measurements
High-resolution (scanning) transmission electron microscopy
(S)TEM images were recorded on a FEI Tecnai F20 FEG TEM operat-
ing at 200 kV equipped with a Gatan Orius SC600A CCD camera.
Samples were prepared by dispersion in methanol and drop-
casting onto a copper grid coated with a holey carbon support
film (Agar Scientific Ltd). Images were analysed using ImageJ 1.41
software. High-angle annular dark-field (HAADF) imaging, which
is particularly sensitive to heavy atoms [43], was used to image
Pd-impregnated supports. Scanning electron microscopy (SEM)
images were recorded on a Carl Zeiss Evo-40 SEM operating at
25 kV. Samples were supported on aluminium stubs each backed
with carbon tape.
X-ray photoelectron spectra were acquired on a Kratos AXIS HSi
spectrometer equipped with a charge neutraliser and monochro-
mated Al K␣ excitation source (1486.7 eV). Binding energy (BE)
referencing was employed using the adventitious carbon peak at
284.6 eV. Survey scans were recorded for surface elemental analy-
sis (pass energy 160 eV), with high resolution spectra recorded at
40 eV pass energy. Spectral fitting was performed using CasaXPS
Version 2.3.5. A common Gaussian/Lorentzian (90:10) mix with
asymmetry based on a Donaich-Sunjic mix of 0.005 was determined
from fitting a PdO standard, and used for all Pd chemical environ-
ments, with a common full-width half maximum (FWHM) adopted
for all components. Two Pd 3d5/2 species were observed at 335.4
and 336.8 eV, assigned as Pd metal and PdO respectively, both with
spin-orbit doublet separations of 5.3 eV. It is important to recall
that the inelastic mean free path of Pd 3d5/2 photoelectrons excited
by Al K␣ radiation is ∼1.3 nm, i.e. comparable to the diameter of
our nanoparticles, hence significant contributions may be expected
from palladium metal residing within nanoparticle cores. Spectral
fitting of Si, Al and O species was undertaken adopting a common
Gaussian/Lorentzian (70:30) lineshape and FWHM. A maximum
of two Si and two Al components were observed, assigned to the
pure oxide and interfacial alumina-silicate species, at 103.4 and
102.3 eV for Si, and 73.8 and 74.7 eV for Al, with spin-orbit dou-
blet separations of 0.61 eV (Si) and 0.41 eV (Al). O 1s XP spectra
were fit to three components associated with the silica framework,
alumina coating and alumina-silicate interface at 532.9, 530.9 and
531.8 eV respectively. All binding energies and doublet separations
are consistent with those reported in the NIST surface database
[44].
Fluorescence Pd K-edge (24.35 KeV) X-ray absorption spectra
were collected at Diamond Light Source on beamline B18 employ-
ing a Si [3 1 1] monochromator, Pt coated mirrors and a Vortex
multichannel fluorescence detector. Operando XAS spectra were
continuously acquired over a 4 h time period, spectra acquisi-
tion length of ∼0.15 h, using a bespoke PTFE operando cell fitted
with 25 ␮m Kapton windows. A reaction slurry of 250 mg catalyst,
84 mmol cinnamyl alcohol (11.23 g), 50 cm3 solvent and 0.5 cm3
mesitylene was continuously circulated though the cell via PTFE
transfer tubing from a stirred glass reaction reservoir held at 90 ◦C
under either static or flowing O2 though the reaction solution
(15 cm3 min−1 at 1 bar) with magnetic stirrer at 1000 rpm. These
reaction conditions represent a doubling of the alcohol:catalyst
ratio with respect to the laboratory selox conditions above in order
to observe the impact of oxygen upon deactivation processes over a
reasonable timescale. Spectra were processed using the IFEFFIT ver-
sion 1.2.11d open source software suite, employing Athena version
0.9.1 for normalisation, background subtraction and linear combi-
nation fitting of XANES, and Artemis version 0.9.1 for EXAFS fitting.
Reference spectra of PdO and a Pd foil standard were also recorded.
3. Results and discussion
3.1. Material characterisation
Successful synthesis of the parent SBA-15 mesoporous silica
support, which exists in a characteristic p6mm space group [30],
and its retention following alumina grafting was first confirmed
by low-angle XRD and nitrogen adsorption isotherms shown in
Fig. 1. XRD patterns were essentially unchanged after the grafting
procedure (Fig. 1A), with common reflections and peak intensities
evidencing ordered mesostructures that were stable against pore
expansion/contraction and collapse during the additional chemical
and thermal processing associated with alumina coating. Porosime-
try showed type IV isotherms with H1 hysteresis for SBA-15 and all
Al-SBA-15 samples (Fig. 1B), highlighting the presence of meso-
pores. The corresponding surface areas decreased with alumina
grafting cycle (Fig. 1 C), due to initial micropore filling and sub-
sequent loss of mesopore area coincident with shrinking mesopore
diameter (WBJH). The latter is clear from the downshift in the
unimodal BJH pore size distributions (Fig. 1D), with mesopores nar-
2.3. Cinnamyl alcohol selox
rowing by 0.6 nm between the parent SBA-15 and Al-SBA-15(4).
√
The constant unit cell parameter a (calculated from a = 2d(100)
/
3
Catalyst screening was performed using a Radleys Starfish
carousel batch reactor on a 10 cm3 scale at 90 ◦C under either
static oxygen (1 bar), or with oxygen bubbled through the reac-
tion solution (3 cm3 min−1 at 1 bar) via 0.5 mm i.d. PTFE tubing
while magnetic stirring was maintained at 1000 rpm. 50 mg of
catalyst was added to a reaction mixture of 8.4 mmol cinnamyl alco-
hol (1.123 g, Aldrich), 0.1 cm3 mesitylene (Aldrich) as an internal
standard, and 10 cm3 HPLC grade toluene solvent. Control reactions
in the absence of any solid phase, or presence of bare supports,
were conducted in parallel, and all gave negligible conversions.
Reactions were periodically sampled, with 0.25 cm3 aliquots with-
drawn, filtered, and diluted with 1.75 cm3 toluene for triplicate
analysis on a Varian 3900GC with CP-8400 autosampler (CP-Sil5
CB column, 15 m × 0.25 mm × 0.25 ␮m). Initial rates were calcu-
lated from the early linear region of the alcohol conversion profiles
(typically 20–30 min reaction), with selectivity and mass balances
calculated using calibrated response factors for reactants and prod-
ucts. Conversion and selectivity values are reported within 3%
error, with mass balances in all cases ≥95% during the first hour
and ≥90% after 24 h.
for hexagonal close-packed mesopores, with the plane spacing d
obtained from Bragg’s Law) determined by XRD and narrowing
mesopore diameters upon grafting, provide strong evidence for
uniform alumina deposition within the mesopore matrix, apparent
from the increase in wall thickness (a - WBJH) with grafting cycle in
Fig. 1E.
SEM images of the corresponding SBA-15 template and Al-
SBA-15(4) material show a single, common straw-like crystallite
morphology with matching dimensions (Fig. 2), precisely as antici-
pated if alumina had grafted into the parent SBA-15 framework,
and not a second, discrete morphology that would indicate the
formation of a separate, pure alumina phase.
The high surface sensitivity and chemical specificity of XPS ren-
ders it the ideal technique with which to follow the formation of
alumina adlayers as a function of grafting number. As noted in
the Experimental, the Al 2p, Si 2p and O 1s spectra for the Al-
SBA-15 materials exhibit multiple chemical states associated with
the underlying silica template, alumina-silica interface, and alu-
mina multilayers, as highlighted in Fig. 3. Aluminium and silicon
atoms at the interface can be readily differentiated, since they

C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
49
1800
A
B
Adsorption
Desorption
SBA-15
1600
SBA-15
1400
1200
1000
800
600
400
200
0
Al-SBA-15(4)
Al-SBA-15(3)
Al-SBA-15(4)
Al-SBA-15(3)
Al-SBA-15(2)
Al-SBA-15(1)
Al-SBA-15(2)
Al-SBA-15(1)
0.0 0.2 0.4 0.6 0.8 1.0
0
1
2
3
4
5
6
Relative pressure / P/P0
2θ / °
SBA-15
9
8
7
6
5
4
3
2
1
0
Total
Micropore
Mesopore
C
D
E 12
1000
900
800
700
600
500
400
300
200
100
0
Al-SBA-15(1)
Al-SBA-15(2)
Al-SBA-15(3)
Al-SBA-15(4)
11
10
Unit cell
Pore Spacing
Pore Diameter
Wall thickness
9
8
7
6
5
4
0
1
2
3
4
0
1
2
3
4
0
2
4
6
8
Alumina grafting cycle
Pore diameter / nm
Alumina grafting cycle
Fig. 1. Structural properties of SBA-15 and Al-SBA-15 composites: (A) low angle XRD patterns; (B) N2 adsorption isotherms; (C) surface areas (total from BET, micropore
from t-plot, and mesopore from BET-micropore- external surface area); (D) BJH mesopore distribution; and (E) mesopore dimensions.
exhibit unique BEs reflecting their different Pauling electronega-
tivities and associated induced dipoles mediated via the bridging
oxygen. Considering the Al 2p XP spectra (Fig. 3B), aluminium
atoms within Al-SBA-15(1), i.e. deposited during the first grafting
cycle, exhibit BE characteristic of Al3+ but occur at 74.7 eV, approxi-
mately 1.1 eV higher energy than observed for the pure mesoporous
alumina (meso-Al2O3) reference. The absence of any bulk alumina
features for Al-SBA-15(1) discounts the possibility that aluminium
introduced during the grafting procedure is hydrolysed to an alu-
mina phase spatially-isolated from the SBA-15 framework. The
intensity of this interfacial alumina feature grows with each suc-
cessive grafting cycle, and is accompanied by the emergence of bulk
alumina features, consistent with the formation of alumina multi-
layers not in direct contact with the silica support. The converse
behaviour is observed in the Si 2p XP spectra (Fig. 3A), wherein the
pure parent SBA-15 component at 103.4 eV is progressively atten-
uated with each grafting cycle, concomitant with the growth of a
new silicon environment at 102.3 eV attributed to electronically
perturbed silicon atoms at the interface with the evolving alumina
overlayer. The stoichiometry of the alumina coating was estimated
from the ratio of the ‘bulk-like’ Al 2p3/2 state at 73.3 eV to the associ-
ated O 1s state at 530.9 eV, corrected for their relative instrumental
response factors, and was between Al2O2.9 and Al2O3.2 demonstrat-
ing the successful deposition of a stoichiometric alumina adlayer.
The thickness of the alumina overlayer can be estimated from the
variation in the overall Si 2p XP intensity with excitation source
energy. The inelastic mean free path for Si 2p3/2 photoelectrons
excited by Mg K␣ radiation is 2 nm versus 2.33 nm for excitation
by Al K␣ [45]. Applying these values to the Al-SBA-15(4) sample
according to Eq. S1 (derived from [46]) yields an alumina thickness
of 0.62 nm, which equates to ∼2.7 alumina monolayers assuming a
thickness of 0.23 nm [47], slightly greater than the pore shrinkage
determined by porosimetry.
Having established that the Al-SBA-15(4) material comprised
a 2–3 monolayer, stoichiometric alumina coating permeating
throughout the SBA-15 template, palladium nanoparticles were
subsequently introduced via incipient wetness impregnation to
achieve a loading of 0.77 wt.%. Low-angle XRD, porosimetry and
XPS (Fig. S1) showed that the ordered mesostructure and alumina
surface composition of the alumina-coated SBA-15 support were
preserved in the resulting Pd/Al-SBA-15(4) sample. This contrasts
with palladium incorporation by an identical protocol into a meso-
porous alumina support, wherein alumina crystallisation and a
consequent loss of mesopore ordering and pore dimensions was
observed (Fig. S2), highlighting the greater stability of the Al-SBA-
15 support.
Palladium nanoparticles and their size distribution and disper-
sion were evaluated by HAADF-STEM and CO chemisorption (Fig. 4
and Table 1). HAADF-STEM shows uniformly distributed nanopar-
ticles with a narrow size range and mean particle diameter of 1.1
(
0.5 nm), while bright-field TEM shows that the ordered meso-
pore architecture of the Al-SBA-15(4) support is preserved, with
elemental analysis (in STEM configuration) confirming the pres-
ence of both Pd and Al throughout the pore network. Palladium
metal dispersions were used to estimate average particle diam-
eters, assuming a spherical morphology, for comparison against
similar Pd loadings on mesoporous alumina and SBA-15 supports
[22,24]. Table 1 shows that far higher dispersions and smaller
nanoparticles were achieved over the Al-SBA-15(4) support than
pure SBA-15 or meso-Al2O3 materials, as reported for Pt on Al-SBA-
15 [40]. The palladium oxidation state within Pd/Al-SBA-15(4) was
probed by XPS, with both metal and oxide observed (Fig. S3). The

50
C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
Fig. 2. Representative SEM images of (A) Al-SBA-15(4); and (B) the parent SBA-15.
Table 1
Physicochemical properties of Pd/Al-SBA-15(4) versus Pd/meso-Al2O3 and Pd/SBA-15.
Support
Pd loading (wt.%)a
Pd dispersion (%)b
Pd particle size (nm)b
Surface PdO content (%)c
Al-SBA-15(4)
0.77
0.74
0.89
88 ( 1)
74 ( 1)
52 ( 1)
1.0 ( 0.1)
1.5 ( 0.1)
2.3 ( 0.1)
44.3 ( 4.2)
21.9 ( 2.2)
6.0 ( 0.1)
d
meso-Al2O3
SBA-15e
a
ICP analysis.
b
Pulse chemisorption assuming stoichiometry of 1:2 CO:Pd.
PdO% from fitted Pd 3d XP spectra.
From reference [24].
c
d
e
From reference [22].
surface PdO content quantified in Table 1 was significantly higher
than that observed for comparable metal loadings over mesoporous
silica or alumina, consistent with the higher palladium dispersion.
We have previously noted that these properties are closely corre-
lated over alumina supports [24], and Fig. 5 confirms that in this
respect, palladium nanoparticles on Al-SBA-15(4) behave akin to
those on pure meso-Al2O3.
transport limitations. The mass-normalised initial rate of cinnamyl
alcohol selox for Pd/Al-SBA-15(4) was exceptional, being twice that
observed over the pure meso-Al2O3 support, and almost 20 times
that obtained for Pd/SBA-15 (Table 2, Fig. S4). Although Pd/Al-SBA-
15(4) affords a higher rate of cinnamyl selox than obtained for the
Pd/meso-Al2O3 catalyst, their corresponding turnover frequencies
(TOFs) per surface PdO sites are almost identical (∼13,500 h−1),
suggesting a common active species and palladium-support inter-
action; the principal impact of the more highly ordered Al-SBA-15
support is simply to increase the density of such PdO sites. A
similar explanation has been advanced for the superior electrocat-
alytic performance of Pt supported on ordered mesoporous CMK-3
versus a disordered wormhole-like mesoporous carbon for ethanol
oxidation and oxygen reduction [48]. If cinnamyl selox activity
was solely a function of the concentration of PdO present at the
surface of oxidised palladium nanoparticles, then one might expect
a common TOF for all three catalysts, independent of support type.
3.2. Cinnamyl alcohol selox
The catalytic activity of Pd/Al-SBA-15 towards the aerobic selox
of cinnamyl alcohol was determined in order to benchmark its
performance against Pd/SBA-15 or Pd/meso-Al2O3 [22–24], and
thus assess support effects. Identical reaction conditions, includ-
ing stirring speed, reaction vessel dimensions, O2 flow rate and
temperature, were selected to match our previous studies and
facilitate direct comparison of catalysts in the absence of bulk mass

C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
51
A
B
Al-SBA-15(4
Al-SBA-15(3
)
)
Al-SBA-15(4)
Al-SBA-15(3)
Al-SBA-15(2
)
Al-SBA-15(2)
Al-SBA-15(1)
Al-SBA-15(1)
Mesoporous
Al2O3(x0.5)
SBA-15(x0.5
100 98
)
108
106
104
102
80
78
76
74
72
70
Binding energy / eV
Binding energy / eV
Fig. 3. Background-subtracted, fitted (A) Si 2p; and (B) Al 2p XP spectra of Al-SBA-15 as a function of alumina grafting cycles. Comparative spectra for SBA-15 and mesoporous
alumina references are also shown.
In fact, the TOF for the Pd/SBA-15 catalyst was only 6,108 h−1
.
to disperse and preserve smaller, and hence more heavily oxidised,
PdO nanoparticles during cinnamyl selox. A gradation of ‘PdO’
active centres may thus exist over oxide supports, which though all
formally Pd2+, may differ in their dimensionality and thus reducibil-
ity [18]. The unsaturated allylic aldehyde, cinnamaldehyde, was the
dominant product for all three catalysts (formed with ∼70% selec-
tivity), with 3-phenylpropan-1-ol and trans-␤-methylstyrene the
Since the morphology of Al-SBA-15 is identical to that of its SBA-15
parent, such different TOFs are not easily explained by differing
mass-transport properties, but rather suggest a palladium-silica
interaction distinct from that obtained with the alumina-grated
surface. We propose that the enhanced activity and TOFs for alu-
mina versus silica supported palladium reflects a superior ability
Fig. 4. (A) Representative HAADF-STEM image of Pd/Al-SBA-15(4) with Pd particle size distribution derived from 100 particles shown inset; and (B) Bright field TEM image
with corresponding EDX spectrum shown inset (unlabelled peaks correspond to C and Cu from the sample grid).

52
C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
Table 2
Cinnamyl alcohol selox over Pd/Al-SBA-15(4) versus Pd/meso-Al2O3 and Pd/SBA-15.
Initial activity (mmol gPd−1 h−1
)
TOF (h−1
)
b
c
Catalyst
Conversion (%)a
Cinnamaldehyde (%)b
Pd/Al-SBA-15
Pd/meso-Al2O3
Pd/SBA-15e
95 (100)
56 (98)
21 (41)
64 (66)
39 (66)
16 (31)
53260
25595
2950
13600
13391
6108
d
a
Conversion and cinnamaldehyde selectivity after 30 min (and 3 h in parentheses) flowing O2.
Calculated over the initial 20–30 min for conversions <50%.
Normalised to PdO active site density (XPS).
From reference [20].
b
c
d
e
From reference [18].
major by-products (both ∼14% after 3 h reaction), with styrene and
preserve palladium in the +2 oxidation state under full selox reac-
tion conditions.
ethylbenzene minor by-products (<5%) via cinnamaldehyde decar-
bonylation and subsequent hydrogenation (Fig. S5).
The time-dependent evolution of palladium oxidation state and
concomitant impact upon instantaneous selox activity during the
first 2 h of reaction are closely correlated (Fig. 7A). Under static
conditions, the concentration of PdO, determined in solution dur-
ing cinnamyl alcohol oxidation, falls rapidly during the first 20 min
reaction before reaching a plateau, mirrored by a sharp drop in cat-
alyst reactivity which reaches a low, stable selox rate. When this
reaction is repeated with an oxygen flow introduced before alcohol
addition, the initial high [PdO] is maintained for the first 20 min
of the reaction, coincident with a constant, high selox rate. Longer
reaction times result in a monotonic decrease in both palladium
oxide concentration and activity, although neither parameter falls
to the low values observed under static oxygen.
3.3. Operando XAS studies of cinnamyl alcohol selox
Additional laboratory studies of cinnamyl alcohol selox over
Pd/Al-SBA-15 revealed rapid on-stream deactivation in under
10 min when a static oxygen pressure was maintained above the
reaction solution, instead of oxygen flowing through the solution
(Fig. S6 and 7). We therefore undertook time-resolved, operando
Pd K-edge XAS measurements on the 0.77 wt.% Pd/Al-SBA-15 cat-
alyst to assess the interplay between static versus flowing oxygen
and the corresponding oxidation state of palladium nanopar-
ticles. Previous work on the vapour phase oxidation of crotyl
alcohol, an analogue of cinnamyl alcohol, has shown the impor-
tance of high oxygen concentrations in maintaining steady state
catalytic selox over palladium nanoparticles and single crystals
[19,21,23]. Normalised Pd K-edge XAS spectra of as-prepared
Pd/Al-SBA-15 in toluene, and after 1 h cinnamyl alcohol selox
under static or flowing oxygen, are presented in Fig. 6. XANES
of the as-prepared catalyst exhibits a significant white line con-
sistent with the presence of substantial PdO (∼74% PdO from
fitting to standards). After 1 h selox under flowing oxygen (i.e.
a higher concentration of dissolved oxygen) there is little vis-
ible change in the operando XANES spectra, and indeed fitting
suggests around 66% of palladium remains as PdO. In contrast,
after 1 h reaction under static oxygen, significant changes are
apparent in the XANES consistent with reduction to palladium
metal, with only 41% PdO retained. Flowing oxygen thus helps
If catalyst deactivation quantitatively tracked palladium reduc-
tion, then the corresponding instantaneous TOFs (per PdO site)
should remain constant throughout reaction under static and flow-
ing oxygen. Fig. 7B reveals that a constant TOF is only observed for
the first 30 min under flowing oxygen, before a steep decrease to the
same low steady state value obtained under static conditions, indi-
cating the genesis of a new, common, poorly active species, which
vida supra, is assigned to Pd0. Palladium metal is known to favour
aldehyde decarbonylation [49], resulting in catalyst self-poisoning
by strong-bound hydrocarbons and CO [20,21]. Carbonaceous
residues may irreversibly block subsequent alcohol adsorption,
while CO can serve as both a reversibly adsorbed site-blocker and
reductant [50,51]. Our operando measurements confirm that on-
stream deactivation occurs via a transition from highly active and
selective cinnamyl alcohol oxidation over PdO, to the slow and
unselective conversion of cinnamyl alcohol over metallic Pd with
attendant self-poisoning by undesired by-products of decarbony-
lation. Although the low flow of oxygen (at atmospheric pressure)
employed in this study was insufficient to fully suppress in situ
catalyst reduction, it is clear that the local concentration of (dis-
solved) oxygen is critical in regulating the rate of Pd/Al-SBA-15
deactivation. This is consistent with the oxidative dehydrogena-
tion reaction mechanism generally accepted for allylic alcohol selox
[52]. Fig. 7 shows that maintaining palladium catalysts in a high
oxidation state during operando liquid phase selox is essential for
optimising their activity, in agreement with our observations of gas
phase crotyl alcohol selox [17,19]. Throughout our research on the
selective oxidation of allylic alcohols, whether in the liquid or gas
phase, and under oxygen-rich or deficient conditions, a direct cor-
relation between selox activity and palladium oxide concentration
determined in situ has been observed [16,18,21], hence there is no
convincing evidence for catalyst deactivation during allylic alcohol
selox due to palladium ‘over-oxidation’.
60
50
40
30
20
y = 1.3258x - 75.586
R² = 0.9469
10
0
60
70
80
90
100
The preceding XANES analysis is supported by EXAFS fitting
of the as-prepared 0.77 wt.% Pd/Al-SBA-15 catalyst in toluene,
and after 2 h cinnamyl alcohol selox under static or flowing
oxygen (Fig. 8). An excellent fit to the as-prepared Pd/Al-SBA-15
catalyst was obtained for a mix of PdO and Pd (Table S1), with
Pd dispersion / %
Fig. 5. Linear correlation between palladium dispersion (from CO titration) and
surface PdO content (from XPS fitting) for Pd/Al-SBA-15(4) and mesoporous
Al2O3 (taken from reference [21]). Red and blues data points represent 0.77 wt.%
Pd/Al-SBA-15 and 0.74 wt.% Pd/meso-Al2O3 respectively. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version
of this article.)
˚
Pd-O nearest neighbours at 2.02 A the dominant scatterers. We
propose the as-prepared catalyst comprised a metallic palladium

C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
53
A
1.5
1.2
0.9
0.6
0.3
0.0
Fresh
Flowing O2
Static O2
Raw
data
Fit
PdO
Pd
B
24200 24400 24600 24800 25000
Energy / eV
24300 24350 24400 24450
Energy / eV
Fig. 6. (A) Operando XANES of the as-prepared Pd/Al-SBA-15 in toluene (superior data quality reflects the average of 10 spectra), and after 1 h cinnamyl alcohol selox reaction
under static or flowing O2 (single spectrum); (B) linear combination fitted XANES spectrum of Pd/Al-SBA-15 under flowing O2.
PdO (O2)
0.015
0.014
0.013
0.012
0.011
0.01
45000
40000
35000
30000
25000
20000
15000
10000
5000
8000
7000
6000
5000
4000
3000
2000
1000
0
PdO (static)
Activity (O2)
Activity (static)
A
B
O2
Static
0.009
0.008
0.007
0.006
0
0
0.5
1
1.5
2
0
0.5
1
1.5
2
Reaction time/ h
Reaction time/ h
Fig. 7. (A) Impact of flowing versus static oxygen on palladium oxidation state and catalytic activity of Pd/Al-SBA-15 during liquid phase cinnamyl alcohol selox. (B) Evolution
of corresponding TOFs (normalised per PdO surface sites) as a function of reaction time (250 mg catalyst and 84 mmol alcohol substrate). Note, lower TOFs relative to laboratory
batch conditions reflect heating losses from circulating solution during transfer between cell and reservoir.
Raw data
fit
Raw data
fit
14
12
10
8
14
12
10
8
Fresh
Fresh
Flowing O2
6
4
6
Flowing O2
Static O2
2
4
0
2
-2
-4
Static O2
5
0
3
4
5
6
7
8
9
10
0
1
2
3
4
6
Wavenumber / Å-1
Radial distance / Å
Fig. 8. k3-weighted EXAFS data and corresponding Fourier transforms and fits for as-prepared 0.77 wt.% Pd/Al-SBA-15 in toluene and after 1 h cinnamyl alcohol selox reaction
under static O2 or flowing O2.

54
C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
core encapsulated by a multilayer palladium oxide shell, akin to
Pd/SBA-15 [22]. The extremely low Pd–Pd 1st shell coordination
properties. The resulting composite Al-SBA-15 support facilitates
enhanced dispersion of palladium nanoparticles, and thus a greater
proportion of surface PdO than either mesoporous silica or alumina
for comparable palladium loadings. Consequently, Pd/Al-SBA-15
exhibits superior activity in the liquid phase aerobic selective oxi-
dation of cinnamyl alcohol to cinnamaldehyde than Pd/SBA-15 or
Pd/meso-Al2O3. Turnover frequency analysis indicates that ultra-
thin and bulk alumina surfaces generate higher levels of a common
(PdO) active species than can be achieved over silica, indicating sig-
nificant metal-support interactions. Operando XAS measurements
reveal strong correlations between the palladium oxidation state
and selox activity which identify rapid, on-stream PdO reduction
as the major deactivation pathway, which can be ameliorated by an
increase in the local oxygen partial pressure. PdO reduction opens
up undesired C O cleavage and decarbonylation pathways (with
attendant loss of cinnamaldehyde selectivity), with the resulting
hydrocarbon and CO by-products liberated driving surface pal-
ladium carbide formation. Knowledge of the redox chemistry of
palladium nanoparticles under reaction conditions, and advantages
of employing a nanostructured Al-SBA-15 support in generating
and maintaining high surface PdO concentrations is critical to the
design of next-generation palladium-derived catalysts for energy
efficient alcohol selox and process optimisation.
˚
number at 2.7 A (and absence of longer range shells expected for
˚
˚
Pd metal), and presence of Pd–Pd scatterers at 2.68 A and 3.04 A
consistent with ‘bulk-like’ PdO, indicate a relatively thick oxide
shell and small metal core compared to the silica analogue, in
keeping with the higher selox activity of Pd/Al-SBA-15.
EXAFS spectra of Pd/Al-SBA-15 in solution following 2 h reaction
under either static or flowing oxygen could only be successfully fit-
ted by including a third phase, palladium carbide (Fig. 8 and Table
S1); fits to a two-phase mixture of oxide, metal or carbide were
poor (Fig. S8 and Table S2). Palladium carbide is known to form via
surface-mediated reactions with gaseous hydrocarbons at ambient
temperature [53,54] or CO decomposition, resulting in a character-
˚
istic Pd–C scattering feature ∼2.16 A and concomitant palladium
lattice expansion [55]. Good fits were obtained only though includ-
˚
ing additional shells, Pd–C at 2.16 A and an expanded Pd–Pd shell at
˚
∼2.84 A [54]. Carbide formation from metallic palladium can usu-
ally be discerned through subtle shifts in the first and second XANES
features due to the interaction between the excited 1s photoelec-
tron with empty 5p and 4f orbitals [55,56], however the complex
mixture of oxide/metal/carbide present under reaction conditions
prohibited such observations. The first shell Pd–Pd coordination
number of the metallic phase observed operando under either oxy-
gen environment of ∼1.8 is essentially identical to that of the fresh
catalyst in toluene, suggesting a common metallic core. This leads
us to hypothesise that the carbide represents an amorphous sur-
face phase, formed as a result of in situ PdO reduction (apparent
from the fall in first shell Pd–O coordination number from 3.2 under
toluene, to 1.8 and 1.4 under flowing and static oxygen respec-
tively) and subsequent carbon laydown over metallic surface sites.
The concomitant rise in selectivity to trans-␤-methylstyrene from
∼5% after 10 min reaction to ∼15% after 2 h (and to a lesser degree
styrene), associated with a partial switchover from oxidative dehy-
drogenation to C O cleavage pathways, could be the trigger for
carbide formation [49].
Acknowledgement
We thank the EPSRC (EP/E046754/1; EP/G007594/2; and
EP/F009488/1) for financial support, a Leadership Fellowship
(A.F.L.) and studentships (L.J.D. and A.M.) and Diamond Light Source
for beamtime (SP7668-1). K.W. acknowledges the Royal Society for
the award of an Industry Fellowship. Electron microscopy access
was provided through the Leeds EPSRC Nanoscience and Nanotech-
nology Research Equipment Facility (LENNF) (EP/F056311/1).
Appendix A. Supplementary data
It is interesting to consider whether dissolved oxygen could
re-oxidise palladium nanoparticles reduced in situ, and thus regen-
erate catalytic selox activity. We have only ever observed palladium
re-oxidation and associated reactivation during vapour phase
selox experiments [17,18] at (i) significantly higher temperatures
(140 ◦C) than those used in the present liquid phase study (90 ◦C),
and (ii) local oxygen concentrations far exceeding those employed
during our solution phase experiments (mol fractionvapour = 0.025
versus mol fractiontoluene = 0.0009 [57]), or following a deliberate
post-reaction 500 ◦C calcination treatment [22] wherein the cal-
cined catalyst exhibits a comparable PdO content to that of fresh
catalysts. The forcing conditions required to re-oxidise palladium
nanoparticles suggest that the deactivated metal/carbide surfaces
may be poisoned by strongly adsorbed hydrocarbons, and hence are
resistant to mild re-oxidation treatments – this hypothesis is sup-
ported by comparing C 1s XP spectra of fresh and spent 0.42 wt.%
Pd/SBA-16 catalysts (Fig. S9). Pd(0) reoxidation back to PdO thus
likely requires thermal activation; this poses a problem for solution
phase regeneration, with Henry’s Law dictating a decrease in oxy-
gen solubility with increasing solution temperature. We therefore
predict that palladium re-oxidation at 90 ◦C would only be feasible
under pO2 > 20 bar for which the mol fraction of dissolved oxygen
rises to 0.025 [57].
Supplementary material related to this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.cattod.
2013.11.056.
References
[1] S.S. Cheng, J.Y. Liu, K.H. Tsai, W.J. Chen, S.T. Chang, Chemical composition and
mosquito larvicidal activity of essential oils from leaves of different Cinnamo-
mum osmophloeum provenances, J. Agr. Food Chem. 52 (2004) 4395–4400.
[2] G.O. Onawunmi, Evaluation of the antimicrobial activity of citral, Lett. Appl.
Microbiol. 9 (1989) 105–108.
[3] T. Hattori, Y. Suzuki, O. Uesugi, S. Oi, S. Miyano, Cationic palladium(II)
complex-catalyzed [2 + 2] cycloaddition and tandem cycloaddition-allylic
rearrangement of ketene with aldehydes: an improved synthesis of sorbic acid,
Chem. Commun. (2000) 73–74.
[4] C. Mercier, P. Chabardes, Organometallic chemistry in industrial vitamin A and
vitamin E synthesis, Catal. Org. React. 62 (1994) 213.
[5] J.H. Clark, Green chemistry: challenges and opportunities, Green Chem. 1
(1999) 1–8.
[6] R.A. Sheldon, Fundamentals of green chemistry: efficiency in reaction design,
Chem. Soc. Rev. 41 (2012) 1437–1451.
[7] A. Corma, H. Garcia, Supported gold nanoparticles as catalysts for organic reac-
tions, Chem. Soc. Rev. 37 (2008) 2096–2126.
[8] X.Y. Liu, R.J. Madix, C.M. Friend, Unraveling molecular transformations on sur-
faces: a critical comparison of oxidation reactions on coinage metals, Chem.
Soc. Rev. 37 (2008) 2243–2261.
[9] T. Takei, T. Akita, I. Nakamura, T. Fujitani, M. Okumura, K. Okazaki, J. Huang, T.
Ishida, M. Haruta, Chapter one-heterogeneous catalysis by gold, in: C.G. Bruce,
C.J. Friederike (Eds.), Advances in Catalysis, Academic Press, 2012, pp. 1–126.
[10] C.P. Vinod, K. Wilson, A.F. Lee, Recent advances in the heterogeneously catal-
ysed aerobic selective oxidation of alcohols, J Chem. Technol. Biotechnol. 86
(2011) 161–171.
4. Conclusions
[11] S.E. Davis, M.S. Ide, R.J. Davis, Selective oxidation of alcohols and aldehydes
over supported metal nanoparticles, Green Chem. 15 (2013) 17–45.
[12] T. Balcha, J.R. Strobl, C. Fowler, P. Dash, R.W.J. Scott, Selective aerobic oxida-
tion of crotyl alcohol using AuPd core-shell nanoparticles, ACS Catal. 1 (2011)
425–436.
Deposition of ultrathin alumina films on mesoporous SBA-15
silica provides an effective route to the synthesis of a highly
ordered, high-area catalyst support exhibiting alumina-like surface

C.M.A. Parlett et al. / Catalysis Today 229 (2014) 46–55
55
[13] T. Hara, M. Ishikawa, J. Sawada, N. Ichikuni, S. Shimazu, Creation of highly sta-
ble monomeric Pd(II) species in an anion-exchangeable hydroxy double salt
interlayer: application to aerobic alcohol oxidation under an air atmosphere,
Green Chem. 11 (2009) 2034–2040.
[14] R.A. Sheldon, I. Arends, A. Dijksman, New developments in catalytic alcohol
oxidations for fine chemicals synthesis, Catal. Today 57 (2000) 157–166.
[15] S.S. Stahl, Palladium oxidase catalysis: selective oxidation of organic chemi-
cals by direct dioxygen-coupled turnover, Angew. Chem. Int. Edit. 43 (2004)
3400–3420.
[16] A.F. Lee, K. Wilson, Structure-reactivity correlations in the selective aerobic
oxidation of cinnamyl alcohol: in situ XAFS, Green Chem. 6 (2004) 37–42.
[17] A.F. Lee, C.V. Ellis, J.N. Naughton, M.A. Newton, C.M.A. Parlett, K. Wilson,
Reaction-driven surface restructuring and selectivity control in allylic alcohol
catalytic aerobic oxidation over Pd, J. Am. Chem. Soc. 133 (2011) 5724–5727.
[18] C.V. Gaskell, C.M.A. Parlett, M.A. Newton, K. Wilson, A.F. Lee, Redox-controlled
crotyl alcohol selective oxidation: in situ oxidation and reduction dynamics
of catalytic Pd nanoparticles via synchronous XANES/MS, ACS Catal. 2 (2012)
2242–2246.
[19] C.M.A. Parlett, C.V. Gaskell, J.N. Naughton, M.A. Newton, K. Wilson, A.F. Lee,
Operando synchronous DRIFTS/MS/XAS as a powerful tool for guiding the
design of Pd catalysts for the selective oxidation of alcohols, Catal. Today 205
(2013) 76–85.
[36] M.V. Landau, E. Dafa, M.L. Kaliya, T. Sen, M. Herskowitz, Mesoporous alumina
catalytic material prepared by grafting wide-pore MCM-41 with an alumina
multilayer, Micropor. Mesopor. Mat. 49 (2001) 65–81.
[37] M. Baca, E. de la Rochefoucauld, E. Ambroise, J.M. Krafft, R. Hajjar, P.P. Man,
X. Carrier, J. Blanchard, Characterization of mesoporous alumina prepared by
surface alumination of SBA-15, Micropor. Mesopor. Mat. 110 (2008) 232–241.
[38] P. Madhusudhan Rao, A. Wolfson, S. Kababya, S. Vega, M.V. Landau, Immobi-
lization of molecular H3PW12O40 heteropolyacid catalyst in alumina-grafted
silica-gel and mesostructured SBA-15 silica matrices, J. Catal. 232 (2005)
210–225.
[39] X. Sheng, Y. Zhou, Y. Zhang, Y. Duan, Z. Zhang, Y. Yang, Immobilization of
12-Tungstophosphoric acid in alumina-grafted mesoporous LaSBA-15 and its
catalytic activity for alkylation of o-xylene with styrene, Micropor. Mesopor.
Mat. 161 (2012) 25–32.
[40] Y. Duan, Y. Zhou, Y. Zhang, X. Sheng, S. Zhou, Z. Zhang, Effect of aluminum
modification on catalytic properties of PtSn-based catalysts supported on SBA-
15 for propane dehydrogenation, J. Nat. Gas Chem. 21 (2012) 207–214.
[41] E.J.M. Hensen, D.G. Poduval, P. Magusin, A.E. Coumans, J.A.R. Veen, Formation
of acid sites in amorphous silica–alumina, J. Catal. 269 (2010) 201–218.
[42] Q.-Q. Hao, Y.-H. Zhao, H.-H. Yang, Z.-T. Liu, Z.-W. Liu, Alumina grafted to SBA-15
in supercritical CO2 as a support of cobalt for Fischer–Tropsch synthesis, Energy
Fuels 26 (2012) 6567–6575.
[20] A.F. Lee, Z. Chang, P. Ellis, S.F.J. Hackett, K. Wilson, Selective oxidation of crotyl
alcohol over Pd(1 1 1), J. Phys. Chem. C 111 (2007) 18844–18847.
[21] A.F. Lee, J.N. Naughton, Z. Liu, K. Wilson, High-pressure XPS of crotyl alcohol
selective oxidation over metallic and oxidized Pd(1 1 1), ACS Catal. 2 (2012)
2235–2241.
[43] S.J. Pennycook, L.A. Boatner, Chemically sensitive structure-imaging with a
scanning-transmission electron-microscope, Nature 336 (1988) 565–567.
[44] NIST X-ray Photoelectron spectroscopy database, NIST standard reference
database 20, Version 3.5 (2003), http://srdata.nist.gov/xps/. Accessed January
2012.
[22] C.M.A. Parlett, D.W. Bruce, N.S. Hondow, M.A. Newton, A.F. Lee, K. Wilson, Meso-
porous silicas as versatile supports to tune the palladium-catalyzed selective
aerobic oxidation of allylic alcohols, Chem. Cat. Chem. 5 (2013) 939–950.
[23] C.M.A. Parlett, D.W. Bruce, N.S. Hondow, A.F. Lee, K. Wilson, Support-enhanced
selective aerobic alcohol oxidation over Pd/mesoporous silicas, ACS Catal. 1
(2011) 636–640.
[45] P.J. Cumpson, M.P. Seah, Elastic scattering corrections in AES and XPS. II. Esti-
mating attenuation lengths and conditions required for their valid use in
overlayer/substrate experiments, Surf. Interface Anal. 25 (1997) 430–446.
[46] A.F. Lee, C.V. Ellis, K. Wilson, N.S. Hondow, In situ studies of titania-supported Au
shell-Pd core nanoparticles for the selective aerobic oxidation of crotyl alcohol,
Catal. Today 157 (2010) 243–249.
[24] C.M.A. Parlett, K. Wilson, D.W. Bruce, N.S. Hondow, A.F. Lee, Selective oxida-
tion of allylic alcohols over highly ordered Pd/meso-Al2O3 catalysts, Catal.
Commun. (2013).
[47] V. Maurice, I. Bennour, S. Zanna, L.H. Klein, P. Marcus, Modifications and growth
mechanisms of ultrathin aluminum oxide films on NiAl in water, J. Phys. Chem.
C 114 (2010) 7132–7140.
[25] J. Lu, S. Behtash, M. Faheem, A. Heyden, Microkinetic modeling of the decar-
boxylation and decarbonylation of propanoic acid over Pd(1 1 1) model surfaces
based on parameters obtained from first principles, J. Catal. 305 (2013) 56–66.
[26] B. Karimi, A. Zamani, S. Abedia, J.H. Clark, Aerobic oxidation of alcohols using
various types of immobilized palladium catalyst: the synergistic role of func-
tionalized ligands, morphology of support, and solvent in generating and
stabilizing nanoparticles, Green Chem. 11 (2009) 109–119.
[27] Y.T. Chen, Z. Guo, T. Chen, Y.H. Yang, Surface-functionalized TUD-1 mesoporous
molecular sieve supported palladium for solvent-free aerobic oxidation of ben-
zyl alcohol, J. Catal. 275 (2010) 11–24.
[48] S.Q. Song, Y.R. Liang, Z.H. Li, Y. Wang, R.W. Fu, D.C. Wu, P. Tsiakaras, Effect of
pore morphology of mesoporous carbons on the electrocatalytic activity of Pt
nanoparticles for fuel cell reactions, Appl. Catal. B-Environ. 98 (2010) 132–137.
[49] M. Bowker, L. Cookson, J. Bhantoo, A. Carley, E. Hayden, L. Gilbert, C. Morgan, J.
Counsell, P. Yaseneva, The decarbonylation of acetaldehyde on Pd crystals and
on supported catalysts, Appl. Catal. A-Gen. 391 (2011) 394–399.
[50] T. Maillet, C. Solleau, J. Barbier Jr., D. Duprez, Oxidation of carbon monoxide,
propene, propane and methane over a Pd/Al2O3 catalyst efect of the chemical
state of Pd, Appl. Catal. B-Environ. 14 (1997) 85–95.
[51] T. Lear, N.G. Hamilton, D. Lennon, The application of temperature-programmed
desorption, adsorption isotherms and temperature-programmed oxidation to
investigate the interaction of CO with alumina-supported palladium catalysts,
Catal. Today 126 (2007) 219–227.
[28] Z.C. Ma, H.Q. Yang, Y. Qin, Y.J. Hao, G.A. Li, Palladium nanoparticles confined in
the nanocages of SBA-16: enhanced recyclability for the aerobic oxidation of
alcohols in water, J. Mol. Catal. A-Chem. 331 (2010) 78–85.
[29] C.-J. Jia, F. Schuth, Colloidal metal nanoparticles as a component of designed
catalyst, Phys. Chem. Chem. Phys. 13 (2011) 2457–2487.
[52] M. Besson, P. Gallezot, Selective oxidation of alcohols and aldehydes on metal
catalysts, Catal. Today 57 (2000) 127–141.
[30] D.Y. Zhao, J.L. Feng, Q.S. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D.
Stucky, Triblock copolymer syntheses of mesoporous silica with periodic 50 to
300 angstrom pores, Science 279 (1998) 548–552.
[53] M.W. Tew, M. Janousch, T. Huthwelker, J.A. van Bokhoven, The roles of carbide
and hydride in oxide-supported palladium nanoparticles for alkyne hydrogena-
tion, J. Catal. 283 (2011) 45–54.
[31] F. Kleitz, S.H. Choi, R. Ryoo, Cubic Ia3d large mesoporous silica: synthesis and
replication to platinum nanowires, carbon nanorods and carbon nanotubes,
Chem. Commun. (2003) 2136–2137.
[32] S.F.J. Hackett, R.M. Brydson, M.H. Gass, I. Harvey, A.D. Newman, K. Wilson,
A.F. Lee, High-activity, single-site mesoporous Pd/Al2O3 catalysts for selec-
tive aerobic oxidation of allylic alcohols, Angew. Chem. Int. Edit. 46 (2007)
8593–8596.
[33] A.Y. Stakheev, L.M. Kustov, Effects of the support on the morphology and elec-
tronic properties of supported metal clusters: modern concepts and progress
in 1990, Appl. Catal. A-Gen. 188 (1999) 3–35.
[34] S.J. Tauster, Strong metal-support interactions, Accounts Chem. Res. 20 (1987)
389–394.
[54] M. Bauer, R. Schoch, L. Shao, B. Zhang, A. Knop-Gericke, M. Willinger, R. Schlögl,
D. Teschner, Structure–activity studies on highly active palladium hydrogena-
tion catalysts by X-ray absorption spectroscopy, J. Phys. Chem. C 116 (2012)
22375–22385.
[55] M.A. Newton, M.D. Michiel, A. Kubacka, M. Fernández-García, Combining time-
resolved hard X-ray diffraction and diffuse reflectance infrared spectroscopy
to illuminate CO dissociation and transient carbon storage by supported Pd
nanoparticles during CO/NO cycling, J. Am. Chem. Soc. 132 (2010) 4540–4541.
[56] J.A. McCaulley, In situ X-ray absorption spectroscopy studies of hydride and
carbide formation in supported palladium catalysts, J. Phys. Chem. 97 (1993)
10372–10379.
[57] K. Fischer, M. Wilken, Experimental determination of oxygen and nitrogen sol-
ubility in organic solvents up to 10 MPa at temperatures between 298 K and
398 K, J. Chem. Thermodyn. 33 (2001) 1285–1308.
[35] K. Hayek, R. Kramer, Z. Paal, Metal-support boundary sites in catalysis, Appl.
Catal. A-Gen. 162 (1997) 1–15.