Enhancing the performance of SBA-15-supported copper catalysts by chromium addition for the chemoselective hydrogenation of trans-cinnamaldehyde

Article abstract of DOI:10.1039/c3cy00198a

SBA-15-supported copper-chromium mixed oxide nanoparticles (CuCr/SBA-15) were prepared by incipient wetness impregnation followed by mild drying at 25°C and calcination. The Cu:Cr weight ratios were 1:1, 5:1, and 10:1, at a constant total loading of 5 wt%. Monocomponent SBA-15-supported Cu-oxide (Cu/SBA-15) and Cr-oxide (Cr/SBA-15) were prepared as reference samples. The materials were systematically characterized by XRD at low and high angles, N₂ physisorption, DR UV-vis, FT-IR, and XPS spectroscopies, and TPR. XRD at low angles and N₂ physisorption confirmed the preservation of the mesoporous structure of the SBA-15 support after impregnation and calcination. In the case of monocomponent samples, CuO appeared poorly dispersed, while the Cr species (Cr₂O₃, mono- and polychromates) appeared highly dispersed on the surface of the SBA-15 support. The progressive addition of chromium to copper had positive effects on the average crystallite size of CuO, which decreased from ～28 nm (Cu/SBA-15) to ～3 nm (CuCr/SBA-15; Cu:Cr = 1:1), and on reducibility, as well. Metallic active phases were obtained by reducing of the oxide phases under a hydrogen flow at 350°C. By comparison to the monometallic catalysts, the reduced CuCr/SBA-15 materials were active in the hydrogenation of cinnamaldehyde and chemoselective towards cinnammyl alcohol (>50 mol%). The activity can be correlated with the particle size of copper, whereas the high selectivity to unsaturated alcohol can be associated with the presence of dual Cu ⁰-Crⁿ⁺ sites. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3cy00198a

Catalysis
Science & Technology
View Article Online
View Journal
PAPER
Enhancing the performance of SBA-15-supported
copper catalysts by chromium addition for the
chemoselective hydrogenation of trans-
cinnamaldehyde
Cite this: DOI: 10.1039/c3cy00198a
B. Dragoi,*a A. Ungureanu,a A. Chirieac,a V. Hulea,b S. Royerc and E. Dumitriua
SBA-15-supported copper–chromium mixed oxide nanoparticles (CuCr/SBA-15) were prepared by incipi-
ent wetness impregnation followed by mild drying at 25 1C and calcination. The Cu : Cr weight ratios
were 1 : 1, 5 : 1, and 10 : 1, at a constant total loading of 5 wt%. Monocomponent SBA-15-supported
Cu–oxide (Cu/SBA-15) and Cr–oxide (Cr/SBA-15) were prepared as reference samples. The materials
were systematically characterized by XRD at low and high angles, N2 physisorption, DR UV-vis, FT-IR, and
XPS spectroscopies, and TPR. XRD at low angles and N2 physisorption confirmed the preservation of the
mesoporous structure of the SBA-15 support after impregnation and calcination. In the case of mono-
component samples, CuO appeared poorly dispersed, while the Cr species (Cr2O3, mono- and polychro-
mates) appeared highly dispersed on the surface of the SBA-15 support. The progressive addition of
chromium to copper had positive effects on the average crystallite size of CuO, which decreased from
B28 nm (Cu/SBA-15) to B3 nm (CuCr/SBA-15; Cu : Cr = 1 : 1), and on reducibility, as well. Metallic active
phases were obtained by reducing of the oxide phases under a hydrogen flow at 350 1C. By comparison
to the monometallic catalysts, the reduced CuCr/SBA-15 materials were active in the hydrogenation of
cinnamaldehyde and chemoselective towards cinnammyl alcohol (450 mol%). The activity can be corre-
lated with the particle size of copper, whereas the high selectivity to unsaturated alcohol can be asso-
ciated with the presence of dual Cu0–Crn+ sites.
Received 24th March 2013,
Accepted 28th May 2013
DOI: 10.1039/c3cy00198a
www.rsc.org/catalysis
toxicity of the precursors and/or the solvents used, or the low
activity displayed by the final materials due to the encapsulation
Introduction
During the last decades, much attention has been paid to bulk
or supported copper catalysts due to their intrinsic activity
in many reactions such as toluene oxidation,1 hydrogenolysis
of esters,2 steam reforming of methanol,3 water gas shift,4
dehydrogenation of alcohols,5,6 hydroamination of alkynes,7
hydrogenation of cinnamaldehyde8–12 and so on. Usually, the
copper based materials have been prepared by precipitation,
chemical vapor deposition and sol–gel synthesis because these
methods allow the achievement of high loadings of copper com-
bined with high dispersions and satisfying thermal stability.13,14
However, some disadvantages of these methods, such as the high
of the active phase in the sol–gel catalysts, make them unattractive
for the preparation of the copper supported nanoparticles.
In this context, the preparation of the copper supported
nanoparticles via incipient wetness impregnation seems to be
an appropriate alternative due to its simplicity, low cost and
lack of waste. However, the main drawback of the method
consists of the formation of large particles during the thermal
treatments, including the drying step at 120 1C and the calcination
at elevated temperatures (generally ranging from 350 to 700 1C).
Moreover, the high mobility of the copper species15 makes them
even more difficult to be stabilized on a support. In this context,
many efforts are made nowadays to improve the dispersion and
thermal stability of the supported copper (oxide) nanoparticles.
A first approach has been reported by de Jong et al. and
consists of the control of the calcination atmosphere,
for instance, the replacement of air with a mixture of 2%
NO/N2.16 Also, the direct reduction of the supported copper
hydroxinitrate proved to be an effective method to get highly
a Gheorghe Asachi Technical University of Iasi, Faculty of Chemical Engineering and
Environmental Protection, 71A D. Mangeron Bdv., 700050 Iasi, Romania.
E-mail: bdragoi@tuiasi.ro; Fax: +40-232 271311; Tel: +40-232 278683
b Institut Charles Gerhardt, UMR 5253, CNRS-UM2-ENSCM-UM1, MACS,
8 rue de l’Ecole Normale, 34 296 Montpellier Cedex 5, France
c
´
Universite de Poitiers, CNRS UMR 7285, IC2MP, 4 Rue Michel Brunet,
86022 Poitiers Cedex, France
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
dispersed and stable copper nanoparticles but the ammonia of the bicomponent particles as well as the formation of the
generated during the nitrate decomposition still limits its wide copper–chromium spinel structures. However, the mechanism
application.16 Toupance et al. have reported that the drying of for the improvement of the copper dispersion in binary Cu–Cr
the impregnated hydrated copper nitrate–SiO2 composite at materials still remains unknown.
ambient temperature prevents the agglomeration of the copper
hydroxinitrate crystals into larger particles, but the calcination from the great importance of these materials as effective non-
step still generates
broad particle size distribution.17 precious metal catalysts for CO23 and VOCs elimination.24
The interest for the copper–chromium oxides also arises
a
Recently, we reported a study on the dispersion and stabili- Moreover, some authors reported that the addition of chro-
zation of copper in CuNi/SBA-15 materials by incipient wetness mium to copper that was supported on zirconia by impregna-
impregnation followed by gentle drying at 25 1C and calcination tion or copper that was deposited on carbon L2S appreciably
under stagnant air. Under these conditions, we demonstrated improved the catalytic performances of the resulting materials
that both the oxide and the metal nanoparticles display in the hydrogenation of a,b-unsaturated aldehydes.25
improved dispersion, besides high thermostability during the
The hydrogenation of cinnamaldehyde (CNA) is a challenge in
calcination and reduction steps.18 It is worth mentioning that fine chemicals synthesis due to the presence of conjugated double
the presence of a second metal (i.e., nickel) was an important bonds in the side chain of the phenyl ring, i.e., olefin (CQC) and
factor that prevented the sintering and migration of the copper aldehyde (CQO) bonds. From the thermodynamics point of view,
oxide outside the pores. Indeed, extended studies on the role of the hydrogenation of the CQC bond with the formation of the
the chemical composition, i.e., Cu : Ni ratio, on the morphology hydrocinnamaldehyde (HCNA) is easier to perform while the hydro-
and structure of the oxide nanoparticles highlighted the positive genation of the CQO bond requires a chemoselective catalyst to
role of the nickel on the dispersion, stability and reducibility of produce the unsaturated alcohol, cinnamyl alcohol (CNOL). Cur-
copper and vice versa.19 These results indicate a synergistic effect rently, the industrial production of CNOL is centered on three
between the two metal components that reached a maximum for methods: (i) Meerwein–Ponndorf–Verley reduction using aluminium
an optimum Cu: Ni weight ratio of about 1 : 4.
alcoholates as catalysts, (ii) selective hydrogenation of cinnamalde-
The feasibility of the copper stabilization via formation of hyde using an alkali borohydride and (iii) to a much lesser extent, the
a solid solution was previously reported by Marchi et al.20 selective hydrogenation using osmium on carbon as a catalyst.26–28
They included copper in ternary and quaternary oxides adopting In this context, researchers are making efforts to find an appropriate
spinel structures. Accordingly, they observed improved reducibility heterogeneous catalyst for the selective hydrogenation of cinnamal-
of the copper cations highly dispersed in the super-stoichiometric dehyde with potential applications in industry. So far, a significant
zinc aluminate spinel in comparison with copper supported on number of publications have been generated by academic media,
silica or even a Cu–Al mixed oxide prepared by co-precipitation. whose importance is still limited to the fundamental research. In the
Moreover, the authors observed an effect of the second transition majority of cases noble metals are reported as the most efficient
metal on the dispersion and reducibility of copper. Therefore, the catalysts.29 Nevertheless, reports on the non-noble metals such as
reduction of copper in the calcined Cu–Ni–Zn–Al took place at copper or nickel as catalysts for the hydrogenation of CNA have
Tmax = 277 1C, while the copper in Cu–Co–Zn–Al sample was become more frequent in the last few years.11,12,20,30–32
reduced at Tmax = 217 1C. It was also reported that in the binary
Herein, we show that using the incipient wetness impreg-
Cr2O3–CuO systems, Cr2O3 prevents the agglomeration of the CuO nation followed by mild drying, a high dispersion of copper–
into larger particles.21,22 However, this promoting effect is still not chromium mixed oxide nanoparticles on SBA-15 can be
well understood. It was suggested that this stabilization occurs via achieved. Accordingly, improved reducibility and catalytic prop-
(a) formation of separated CuO and CuCr2O4 or (b) Cr2O3 acts as erties are obtained, as compared with the monocomponent
an obstacle between CuO crystallites hindering the growing of the counterparts. Mono- and bicomponent CuCr/SBA-15 samples
larger particles. Accordingly, it was also pointed out that Cr2O3 has with various Cu : Cr weight ratios were prepared and the oxide
not only the role of promoter of the copper structure, but also forms were characterized to study their chemical composition,
facilitates electron transfer between copper and chromium species, texture, particle size and structure, and reducibility. After
and thus modifies the final properties of the system.
reduction under hydrogen flow, the catalysts were evaluated
The nature of the solid used as a support for the deposition in the hydrogenation of cinnamaldehyde, a reaction that is
of metal (oxide) nanoparticles is also a key factor to control the highly sensitive to the properties of the supported bimetallic
dispersion and stability of the nanoparticles. In this context, nanoparticles (e.g., the nature of metals and promoters and
the ordered mesoporous silicas such as SBA-15 are of particular their bulk and surface chemical composition, nature of sup-
interest as host materials for nanoparticles due to their two- port, and particle morpho-structural properties, in relation to
dimensional hexagonally arranged channel system, typically of the catalyst preparation method and pretreatment conditions).
5–9 nm in mean pore diameter, and large specific surface areas.
Recently, Tsoncheva et al.22 reported the deposition of copper–
chromium particles on SBA-15 by incipient wetness impregna-
Experimental
Materials
tion technique and acetylacetonate precursors in toluene. They
showed that the chemical composition of the samples, as well All the chemicals required to prepare the support and the catalysts
as the texture of the support, strongly influence the dispersion were used without any additional purification: tetraethylorthosilicate
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
(Si(OC2H5)4, TEOS, 98%, Aldrich), non-ionic triblock co-polymer The mesopore size distribution was determined by NLDFT
Pluronic P123 (poly(ethylene oxide)-block-poly(propylene oxide)- equilibrium algorithm for cylindrical pores.
block-poly(ethylene oxide), EO20PO70EO20, molecular weight =
Temperature programmed reduction was carried out on an
5800, BASF Corp.), copper nitrate (Cu(NO3)2ꢀ3H2O, 98%, Autochem 2910 analyzer from Micromeritics. 100 mg of calcined
Aldrich), chromium nitrate (Cr(NO3)2ꢀ9H2O, 99%, Sigma- catalyst was placed in a U-shaped microreactor and activated at
Aldrich), distilled water and hydrochloric acid. The chemicals 500 1C for 1 h (heating ramp of 15 1C minꢁ1, air flow of
used for the hydrogenation reaction were also used as received: 30 mL minꢁ1). After cooling to 50 1C, 3% Ar/H2 mixture was
trans-cinnamaldehyde (C6H5–CHQCH–CHO, 98%, Merck) as stabilized at 20 mL minꢁ1 flow rate into the sample tube. During
reagent and propylene carbonate (C4H6O3, 99%, Sigma-Aldrich) the analysis, the temperature was increased up to 900 1C at a rate of
as solvent.
10 1C minꢁ1. The effluent gas was monitored with a TCD.
Diffuse reflectance UV-visible spectra were recorded on a
Shimadzu UV-2450 spectrometer equipped with an integrating
Support and catalysts preparation
SBA-15 silica was synthesized following the classical method.33 sphere unit (ISR-2200). The spectra were collected in the range
The solid was obtained by calcination in air at 500 1C for 6 h of 200–800 nm and using BaSO4 as a reference.
with a heating rate of 2.5 1C minꢁ1
.
Infrared spectra were collected on a Vector 22 FT-IR spectro-
Mono- and bicomponent catalysts were prepared by incipi- meter from Bruker over the wave number range of 400–4000 cmꢁ1
,
ent wetness impregnation method as follows: the freshly cal- using pellets formed from a mixture of 99 mg KBr and 1 mg of
cined SBA-15 support was impregnated with an aqueous sample. Each spectrum was obtained by averaging 40 scans with a
solution of the corresponding hydrated nitrates to obtain, resolution of 2 cmꢁ1
.
assuming that both cations are completely reduced to the
XPS spectra were recorded on an ESCALAB 250 (Thermo
metallic phase, a total metal loading of 5 wt%. For the bicom- Electron, Thermo Fisher Scientific, WI) photoelectron spectro-
ponent samples, three Cu : Cr weight ratios were taken into meter equipped with a monochromatic Al-Ka source (1486.6 eV,
account, i.e., 1 : 1, 5 : 1, and 10 : 1. All the impregnated materials powered at 20 mA and 10 kV). The binding energies were
were submitted to a gentle drying step at 25 1C for 5 days. The determined relative to the Si 2p XPS peak set at 103.2 eV.
oxides of copper and chromium were obtained after calcination Quantification of the elements and their valences was based on
under stagnant air at 500 1C for 6 h (heating ramp of 1.5 1C minꢁ1). a non-linear least squares fitting program using a sum of
The monocomponent samples were labeled as M/S, where M = Cu Lorentzian and Gaussian component curves after background
or Cr and S = SBA-15 while the bicomponent samples were labeled subtraction and using the Scofield coefficients as correction
as CuxCr/S, where x = 1, 5 and 10.
factors.
Physico-chemical characterization
Hydrogenation of trans-cinnamaldehyde
All the solid materials were systematically characterized by XRD The hydrogenation of trans-cinnamaldehyde was performed in
at low and high angles, nitrogen physisorption, TPR and FT-IR, liquid phase at atmospheric pressure, in a thermostated three-
DR UV-vis, and XPS spectroscopies.
neck glass reactor equipped with a reflux condenser, hydrogen
XRD measurements were performed on a Bruker AXS D8 bubbler and magnetic stirrer. The following reaction condi-
apparatus with monochromatic CuKa radiation (l = 0.15406 nm tions were applied: 265 mg of sample that had been reduced in
wavelength) at room temperature. For small angles analysis, all H2 flow (1 L hꢁ1) at 350 1C for 10 h (heating ramp of 6 C minꢁ1),
patterns were recorded over the 2y range from 0.5 to 61 with a step reaction temperature of 150 1C, 1 mL of CNA, 25 mL of
of 0.011. For high angle interval, the patterns were recorded from propylene carbonate, hydrogen flow of 1 L hꢁ1 and constant
15 to 801 2y with a step of 0.021. The crystallite sizes of the oxides stirring speed of 900 rpm to achieve the kinetic regime.
were calculated by using the Scherrer equation: dhkl = l/bcosy, Samples were periodically taken off and analyzed by GC
where l = incident ray wavelength (0.15406 nm); b = peak width (HP5890 equipped with DB-5 capillary column (25 m ꢂ
at half height (rad) after Warren’s correction for instrumental 0.20 mm ꢂ 0.33 mm) and FID) with the following temperature
broadening, y = Bragg angle.
program: initial temperature of 150 1C (1 min) followed by
Nitrogen physisorption was carried out at 77 K on an Autosorb heating (15 1C minꢁ1) up to 270 1C. The identification of the
1-MP automated gas sorption system (Quantachrome Instru- products was achieved from the retention times of the pure
ments). Before the analysis, the samples were pretreated as compounds and occasionally by GC-MS (Agilent 6890N system
follows: 50 mg of calcined sample was out-gassed under high equipped with Agilent 5973 MSD detector and 30 m DB-5 ms
vacuum at 350 1C for 3 h. After recording the adsorption/ column). Quantitative analyses were performed by taking into
desorption isotherms, textural properties were calculated using account the FID response factors for each compound.
Autosorb 1 software. The BET surface area, SBET, was assessed
using the multipoint BET algorithm for the relative pressure,
p/po, ranging between 0.10 and 0.30. The total pore volume, Vp,
was evaluated at a relative pressure of about 0.975. The micro-
Results and discussion
Structural and textural characterization
pore volume and the surface of the micropores were evaluated
XRD at high and low 2h angles. Fig. 1 displays the diffraction
by the t-plot method (de Boer statistical thickness = 3.6–8.3 Å). patterns collected at high 2y angles for the chromium and/or
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
very close to the average pore size (8.5 nm) suggesting that
the most part of crystallites are located inside the pores of the
SBA-15 silica.
When copper is substituted by chromium, the XRD patterns
change as a function of the Cu : Cr ratio. Therefore, the gradual
replacing of copper by chromium predominantly leads to a
decrease in the intensity of CuO diffraction peaks, while the
FWHM value slightly increases in comparison to the Cu/S
monocomponent sample, suggesting the formation of smaller
CuO particles in the bicomponent samples. The diffraction
peaks corresponding to the Cr2O3 phase, and easily observed
in the monocomponent material, strongly decreased for the
bicomponent samples. Additionally, the formation of CuCr2O4
spinel structure was observed in the bicomponent samples.
The intensity of those peaks increases with the amount of
chromium in the sample and reached a maximum for the
Cu1Cr/S sample. According to these results, the stabilization
of copper seems to occur via two routes: (i) the formation of the
copper–chromium spinel and (ii) the interposition of Cr2O3
between CuO crystallites preventing, thus, the agglomeration of
Fig. 1 HA-XRD patterns for the calcined samples.
copper oxides supported on SBA-15. It can be observed that CuO into larger particles.21,22
these patterns contain different reflections depending on the
The periodicity of the pore structure in the SBA-15 support
sample composition. Therefore, at the same total content in and in Cu–Cr containing samples was followed by XRD at low
metal (5 wt%), the copper containing sample exhibits reflec- 2y angles and the collected diffractograms are illustrated in
tions characteristic to CuO (tenorite34) while the chromium Fig. 2. The pattern of the SBA-15 presents three diffraction
containing sample exhibits typical reflections of Cr2O3. Also, it peaks at 2y below 21, indexed as the (100), (110), and (200) hkl
was noted that the peaks of Cr2O3 are much less intense and reflections typical for a p6mm symmetry. The corresponding
broader than those of CuO, indicating differences in the crystal- d-spacing of the (100) plane is 9.59 nm and the ao parameter is
lite sizes for the two oxides. Moreover, the crystallite size as well 11.08 nm (Table 1). Therefore, the support used to prepare the
as the shapes of the peaks give information on the localization catalysts is well ordered at long range and contains mesopores
of the oxide particles. The average crystallite size evaluated by in a hexagonal array, as previously described.33 After the
Scherrer equation shows values of 28 nm for CuO (Table 1) that impregnation of the support with nitrate precursors (chromium
are much larger than the average pore size of the parent SBA-15 and copper) and calcination, the typical reflections of SBA-15
while the reflection peaks at B35 and B381 are very intense were identified in all diffractograms, indicating the preserva-
and thin. These observations suggest a localization of the oxide tion of the long-range order of the mesopores upon oxide
particles outside of the pores. For Cr2O3, the diffraction peaks formation. Nevertheless, it is known that the differences of
are not very intense and their lorentzian shape indicates a wide the electron density distribution are sensitive to the distribu-
distribution of crystallite size with a maximum at 8.7 nm that is tion of the matter in the solid.35,36 Therefore, the electronic
Table 1 Structural and textural properties of the samples
Cu/(Cu + Cr)
(wt ratio)
SBET (m2 gꢁ1
)
Smicro (m2 gꢁ1
)
Vp (cm3 gꢁ1
)
Vmicro (cm3 gꢁ1
)
APSg (nm)
8.5
a
b
c
d
e
f
Sample
SBA-15
d100 (nm)
ao (nm)
—
9.59
9.59
11.08
11.08
10.96
11.59
11.72
11.45
695
113
109
88
1.11
1.09
1.04
0.98
0.98
0.99
0.052
0.050
0.042
0.038
0.045
0.042
(100%)
690
(99%)
633
(91%)
600
(86%)
606
Cu/S
1
8.5
8.5
8.5
8.5
8.5
Cu10Cr/S
Cu5Cr/S
Cu1Cr/S
Cr/S
0.91
0.84
0.50
0
9.49
10.04
10.15
9.92
79
93
(87%)
616
93
(89%)
pffiffiffi
a
b
c
d100 = The lattice spacing obtained by low angle XRD. a0 = The hexagonal unit cell parameter calculated using the equation: a0 ¼ 2d100 3. SBET
d
is the total specific surface area obtained using the BET equation. Smicro is the micropore surface area obtained by the t-method applied on the
linear region of the first part of the isotherm. Vp is the total pore volume measured at p/p0 = 0.97. Vmicro is the micropore volume obtained by the
t-method. APS is the average pore size evaluated by NL-DFT for cylindrical pores/equilibrium model.
e
f
g
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
Fig. 3 N2 physisorption isotherms (left) and pore sizes distribution curves (right)
for the calcined samples.
thus, the electronic density of the pores is higher than air. In
this case, the intensity of the (100) peaks is diminished. If the
results of XRD at high angles are considered, which show very
small Cr2O3 crystallites probably grown inside the pores, the
formation of a very thin film of chromium oxide species on the
internal surface of the mesopores might be admitted, due to
the absence of a significant (100) peak intensity decrease.
N2 physisorption. The porosity of the samples was further
characterized by nitrogen physisorption at 77 K. The N2 physi-
Fig. 2 LA-XRD patterns for the calcined samples.
density of the empty pores is equal to the density of air while sorption isotherms and pore size distribution curves of the
the electronic density of the walls is higher than that of the air. samples are shown in Fig. 3. Details of the textural properties
Accordingly, changes in the electronic density of the pores and/ derived from these isotherms are summarized in Table 1. The
or walls could be observed in the collected diffractograms, isotherm of the parent SBA-15 is of type IV according to IUPAC
more precisely in the intensity of the diffraction peaks as well classification and it reflects the mesoporous structure also
as their corresponding 2y values. Indeed, the patterns included being in good agreement with the XRD pattern recorded at
in Fig. 2 show small changes in the intensities of the (100) peak. low 2y angles.33 The single step capillary condensation at
In addition, shifts in 2y peak position were noticed depending relative pressure, p/po = 0.73, reflects the presence of cylindrical
on the composition. The Cu/S sample has a similar XRD pattern mesopores with a narrow pore size distribution (PSD). The
and identical ao parameter (11.8 nm) as the parent SBA-15, maximum of the PSD curve is centered at 8.5 nm and repre-
suggesting that the most significant part of CuO crystallites are sents the average pore size of the sample. The surface area of
located on the external surface of the silica grains, as already the Cu/S sample is close to that of the parent SBA-15, indicating
suggested by XRD at high angle. For the Cr/S sample, the no significant changes on the internal surface of this sample
intensity of the first reflection does not significantly change and sustaining the XRD results, which stand for the external
in comparison with the parent SBA-15 but a minor shift towards localization of the CuO particles. After impregnation with
lower 2y is observed that corresponds to a slight increase in nitrates of copper and/or chromium and calcination, the shape
interplanar distance (9.92 nm vs. 9.59 nm). Based on these of the nitrogen physisorption isotherms did not change, the
results and the physisorption measurements, the wall thick- capillary condensation taking place at the same relative pres-
ness was calculated and found to be of 2.95 nm for Cr/S, instead sure as for the SBA-15 support. Hence, the pore size distribu-
of 2.58 nm for SBA-15.
tion curves exhibit the same maximum at 8.5 nm.
It seems that the metal precursors introduced in the pores
These general observations indicate the absence of any
changed the electron density distribution with direct conse- structural collapse of the ordered mesophase during the
quences on the reflection intensities as previously proposed by impregnation of the SBA-15 and subsequent thermal treat-
Wang et al.37 Accordingly, the formation of a thin layer of oxide ments (i.e., drying and calcination). Also, no delay on the
precursors on the internal walls does not significantly change desorption branch was identified in any of the isotherms,
the intensity of the first reflection in comparison with the which points out that the principal mesopores remain open,
parent SBA-15. Moreover, a spatial order could be responsible and that no significant constriction forms in the pores after
for an increased intensity in comparison with that of the parent calcination. However, the decreases in specific surface areas
support. By contrast, when impregnated species are chaotically and pore volumes suggest the localization of the oxides pre-
oriented inside the pores, they practically fill the pores and, cursors inside the mesopores of the SBA-15.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Spectroscopic investigation
of the Cr/S sample at 270 and 350 nm are attributed to O2ꢁ
-
Cr6+ LMCT transitions in highly dispersed Cr6+ in tetrahedral
coordination. The bands at 460 and 600 nm are attributed to
d–d transitions of Cr3+ in octahedral symmetry. Therefore, the
DR UV-vis spectroscopy revealed the simultaneous formation of
mono- and polychromate species of Cr6+ and Cr2O3 (Cr3+), the
last one already being identified by XRD.
DR UV-vis spectroscopy. First information on the metal
oxides species generated by calcination of the copper and
chromium nitrates impregnated on SBA-15 was provided by
DR UV-vis spectroscopy. The spectra are illustrated in Fig. 4.
The spectrum of the Cu/S sample presents two distinct
absorption bands: (i) a LMCT (ligand-to-metal-charge-transfer)
band with a maximum at 250 nm and (ii) a very broad absorp-
tion band at 360–800 nm. The first low intense band, due to
O2ꢁ - Cu2+ LMCT in octahedral coordination of Cu2+, is
associated with a minor population of Cu2+ in highly dispersed
CuO.38 The second band represents overlapped absorptions
coming from Cu2+ in a variety of CuO species usually found in
bulk CuO particles for example, Cu–O–Cu oligomers, low-
symmetry Cu species such as square-planar CuOx, and Cu2+
ions in a distorted octahedral environment caused by the Jahn–
Teller effect.38 The spectrum of the Cr/S sample contains
several well-resolved absorption bands with maxima at 270,
350, 460 and 600 nm.
These bands characterize several chromium species,
although only Cr2O3 was detected by XRD. The other chromium
species are either present in a low amount or remain amor-
phous, avoiding their detection by XRD. According to previous
works on chromium supported on different supports (i.e.,
silica, SBA-15, MCM-41, Al2O3), different chromium species
(i.e., Cr6+ and Cr3+) were generated depending on the chro-
mium amount and the nature of the support.22,39–46 Cr6+ is
usually found tetrahedrally coordinated by oxygen in CrO3,
[CrO4]2ꢁ and [Cr2O7]2ꢁ, while Cr3+ is found in distorted octa-
hedrally coordinated environments such as in Cr2O3. Based on
these studies, the absorption bands identified in the spectrum
The spectra collected for the bicomponent samples contain
similar absorption bands as for the monocomponent counter-
parts. However, the intensities of the bands as well as the blue
or red shifts depend on the amount of each element as well as
their dispersion on the SBA-15 surface. The first absorption
band was observed at 250 nm (Cu10Cr/S) and 260 nm (Cu5Cr/S
and Cu1Cr/S), respectively while a second band was observed at
360 nm for all bicomponent samples. These bands are assigned
to highly dispersed copper and chromium (mainly Cr6+) spe-
cies, probably interacting together. The intensities of the
absorption bands between 400 and 600 nm are diminished in
comparison with the monocomponent samples and they are
attributed to Cu2+ and Cr3+ in CuO, Cr2O3, and CuCr2O4 spinel.
The absorption band with the maximum at B760 nm slowly
increases with the amount of copper and is assigned to d–d
transitions of Cu2+ ions in octahedral coordination geometries
in CuO particles.
FT-IR spectroscopy. The further investigation of the oxide
species were performed by analyzing the vibrational spectra
collected for the chromium containing samples and illustrated
in Fig. 5.
Due to the presence of the silica support, these spectra show
contributions of both the silica and the metal oxides supported
on the silica. Therefore, the infrared bands specific to the silica
framework (i.e., 1200, 1100, 812 and 470 cmꢁ1) are not dis-
cussed herein.39 Supplementary bands were identified at 560,
625 and 920 cmꢁ1. Interestingly, the intensities of all these
Fig. 4 DR UV-vis spectra for the calcined samples.
Fig. 5 FT-IR spectra for the calcined samples.
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
bands decrease as the amount of chromium decreases, [CrO4]2ꢁ and [Cr2O7]2ꢁ, or even [CrnO3n+1
]
2ꢁ). Moreover, surface
and they are therefore considered as ‘‘fingerprints’’ of the quantification shows that the majority of chromium exists as
chromium species. The vibration bands observed at 560, 625 Cr2O3 (68.5% – Table 2) while the Cr/Si ratio (0.020 from XPS vs.
and the shoulder at 720 cmꢁ1 in the Cr/S sample originate from 0.140 from ICP) indicates that only a small part of the chro-
Cr–O vibration in crystalline Cr2O3.47,48 For the bicomponent mium is found in the near surface region.
samples, the band at 560 cmꢁ1 is shifted to 597 cmꢁ1 and is
In the bicomponent samples, a change in the BE of Cr6+
attributed to the chromite anion.49 This is in good agreement was observed in comparison with the Cr/S monocomponent
with XRD data showing the formation of CuCr2O4, especially for sample. Such a BE shift indicates a change in the chemical
the samples with a high amount of chromium (Cu5Cr/S environment of the Cr6+ cations, which can be attributed to the
and Cu1Cr/S). As concerns the absorption at 920 cmꢁ1, it presence of copper in close interaction. However, this shift of
might correspond to CrQO stretching vibrations of mono- the binding energy remains very limited, irrespective of
and polynuclear surface chromates.50
the copper content. In addition, the relative part of Cr6+ is
constantly increasing from 31.5% (Cr/S) to 100% (Cu10Cr/S),
All the spectra contain a vibration band at around 970 cmꢁ1
.
It can be observed that the absorption is blue shifted as the while the contribution of Cr3+ (as Cr2O3 and CuCr2O4 according
amount of chromium decreases in the sample, so that for the to XRD) decreases from 68.5% (Cr/S sample) to 0% (Cu10Cr/S).
Cu10Cr/S the band is observed at 977 cmꢁ1. Due to the dual Additionally, the chromium surface concentration (see Cr/Si
composition of the samples, i.e., silica–metal oxides, the attri- ratio in the Table 2) is progressively decreasing with the total
bution of this band is more difficult. Usually, this band is amount of chromium in the material.
observed at 950–970 cmꢁ1, more often at 960 cmꢁ1, and is
For copper, BE from 934.25 to 933.66 eV, with a shake-up
attributed to the stretching vibration of free silanols on the peak at 942.20 eV, were observed. According to the literature,
silica surface. Frequently, the same band was attributed to Si–O these energy values depend on the sample composition, crystal-
stretching vibration in the Si–OR group, where R = Ti,51–55 Cr,56 linity, and particularly on the copper micro-environment.1,64–67
or Cu.57 In accord with these results, the band observed at For instance, the BE for bulk CuO is 932.4 eV.68 Accordingly, the
B970 cmꢁ1 is attributed to Si–O–Cr vibrations. Additionally, it decrease in the binding energy from 934.25 eV to 933.6 eV can
was observed that the shift to higher wave numbers of this arise from the decrease in CuO dispersion with the increase in
band is proportional to the increase in copper amount. There- copper content, which is in good agreement with the XRD
fore, this band could be a contribution of the two metals (i.e., results. Unlike chromium, the quantitative data listed in
copper and chromium) interacting with the silica surface.
Table 2 show that the amount of copper (Cu/Si) in B10 nm
XPS analysis. The surface composition in the copper depth from the surface decreases when the amount of copper
and chromium containing materials was investigated by XPS increases. However, the measured values indicate that a signifi-
analysis. The binding energies (BE) besides the valence state cant part of the copper is located deep inside the samples, and is
assignment and atomic surface composition are listed in Table 2. not visible by XPS.
Two oxidation states were observed for chromium in the
The XPS results thus confirmed the formation of different
Cr/S sample, with BE of 576.40 eV and 578.83 eV. According to copper and chromium oxides species, those distributions at the
the literature, these energies are related to Cr3+ and Cr6+
,
surface are depending on the composition of the sample.
respectively.58–63 This result is in agreement with XRD, FT-IR Therefore, a Cu : Cr weight ratio of 1 : 1 is favorable for the
and DR UV-vis data, all indicating the presence of crystalline formation of copper–chromite spinel while a ratio of 10 : 1
Cr2O3 as well as species of amorphous Cr6+ (such as CrO3, generates only hexavalent chromium species. Moreover, it was
Table 2 XPS data from fitting of Cr 2p and Cu 2p and the surface composition of samples
Cr 2p3/2
B.E. (eV)
579.07
Cu 2p3/2
B.E. (eV)
At%
Cr
Cr/Si
ICP
Cu/Si
ICP
VSAa
Cr6+
VSAa
Cu2+
Cu
XPS
XPS
Sample
Cu10Cr/S
933.66
942.20
934.05
942.20
0.08
(100%)
0.03
(21%)
0.11
(79%)
0.12
(40%)
0.17
(60%)
0.39
(68.5%)
0.18
0.14
0.013
0.003
0.139
0.005
Cu5Cr/S
Cu1Cr/S
Cr/S
576.36
579.34
Cr3+
Cr6+
Cu2+
Cu2+
—
0.17
0.17
—
0.024
0.069
0.140
0.007
0.010
0.02
0.124
0.071
—
0.009
0.012
—
576.41
579.21
Cr3+
Cr6+
934.25
942.20
576.40
578.73
Cr3+
Cr6+
—
(31.5%)
a
Valence state assignment.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
shown that a good dispersion of CuO can be achieved regard- by gentle drying of the samples after wetness impregnation of
less of the Cu : Cr ratio. Accordingly, a positive effect of the Cr6+ the nitrate precursors.
species on the CuO dispersion could be also assumed.
The TPR curve for the Cr/S sample shows a minor reduction
peak at 270 1C and a major one at 450 1C, which may be
attributed to the reduction of Cr6+ to Cr3+ 43,73–80
mentioning that the complete reduction of chromium to zero
.
It is worth
Reducible properties of the samples
The degree of interaction between the support and the metallic valent chromium is not possible in the TPR experimental
species, as well as metal–metal interactions, was further inves- conditions.81,82
tigated by TPR. The recorded TPR profiles for the mono-
For the bicomponent samples, the TPR profiles contain two
component and bicomponent samples are displayed in Fig. 6. main reduction peaks attributed to the reduction of Cu2+ to Cu0
The TPR curve for the Cu/S sample contains a main reduction and Cr6+ to Cr3+, respectively. It can be noticed that the
peak centered at 345 1C and a shoulder at 500 1C. Normally, reduction temperatures of the copper cations shifted to much
these two maxima should correspond to the reduction of Cu2+ lower values in comparison with the monocomponent sample
to Cu0. Because the XRD results have already shown the (Cu/S). It is known that in the case of CuO, the temperature of
presence of large crystalline CuO in the calcined Cu/S sample, reduction is strongly influenced by the particle size and dis-
for this sample the peak centered at 345 1C can be attributed to persion, i.e., the reducibility is improved when the dispersion is
the reduction of poorly dispersed CuO. The shoulder identified high.83–85 As revealed by XRD, the bicomponent samples also
at 500 1C is not usual for CuO reduction. This peak was contain CuCr2O4 spinels. Therefore, the reduction maxima at
observed during the reduction of Cu/MCM-48, and copper in 222, 258, and 268 1C for Cu1Cr/S, Cu5Cr/S, and Cu10Cr/S,
mesoporous silica–alumina, and was attributed to difficult to respectively have contributions from two precursors, i.e. the
reduce copper species such as copper silicate.69–71 Indeed, Van reduction of Cu2+ in highly dispersed CuO and Cu2+ in spinel
Der Grift et al.72 performed TPR analysis on different copper structure, respectively. Thus, the temperatures of copper oxide
hydrosilicates and they identified varying reducible behavior reduction, which are lower than for the monocomponent
for copper species depending on the composition and structure material, indicate an interaction between copper species and
of the silicate. For example, they observed a reduction tempera- chromium species in close proximity to each other. For the Cr6+
ture of 527 1C for dioptase (CuSiO2(OH)2).
cations, no significant change of the reducibility was observed
Our results thus evidence the possibility to stabilize a part of showing that the oxides interaction only affects Cu2+
the copper in the silica matrix as copper silicate, obtained only reducibility.
The physico-chemical characterization discussed above
clearly shows that the addition of chromium to copper depos-
ited on SBA-15 has a positive effect on the dispersion and
reducibility of copper. Moreover, according to these results, the
observed effects also depend on the amount of chromium
introduced in the sample. Therefore, a smaller amount of
chromium (i.e., Cu : Cr = 10 : 1) favors the stabilization of copper
probably via a Cr2O3 barrier between CuO particles that does
not allow the sintering of the CuO into larger particles. Equal
amounts of copper and chromium in the sample (i.e., Cu : Cr =
1 : 1) allow the stabilization of copper via formation of CuCr2O4
spinel, as well. Accordingly, the reducibility of copper is much
improved as shown above. Taking into account the sensitivity
of the hydrogenation of cinnamaldehyde to the properties of
the catalyst, these observed structural and reducible effects
were evaluated in this reaction and the obtained results are
discussed further on.
Calcined Cu–Cr/SBA-15 materials as catalytic precursors for the
hydrogenation of trans-cinnamaldehyde
After reduction at 350 1C under H2 flow, the catalytic perfor-
mances of the CuxCr/S materials were evaluated in the hydro-
genation of cinnamaldehdye (CNA). According to Scheme 1, the
hydrogenation of this aldehyde involves parallel and consecutive
reactions. The parallel reactions (1 and 2) lead to the partial
hydrogenated products, cinnamyl alcohol (CNOL) and hydrocinn-
Fig. 6 TPR profiles for the calcined samples.
amaldehyde (HCNA), respectively, whose further hydrogenation
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
Scheme 1 Reaction pathways for the hydrogenation of CNA.
generates the completely hydrogenated product, hydrocinnamyl
alcohol (HCNOL).
It is already accepted that the catalytic activity and selectivity
in the hydrogenation of cinnamaldehyde is governed by the
competition between the two double bonds during the mole-
cule chemisorption on the catalyst surface. Therefore, the
nature and number of the catalytic active sites as well as
their microenvironment are the key factors that control the
adsorption mode of the CNA molecule, and consequently the
selectivities in reaction. First, when the hydrogenation of CNA
was performed over the monocomponent samples, i.e., Cu/S
and Cr/S, an extremely low activity (below 1.5 mol% after 360
min of reaction, Cu/S) or even the lack of catalytic activity (Cr/S)
were noted. Generally, copper is not active in hydrogenation
reactions due to its fully occupied d10 orbitals. However, when
the particles are sufficiently small and highly dispersed, the
electronic density increases making copper an active
catalyst.19,29,31 As expected, the Cu/S sample, containing large
particles of the Cu species, exhibited very low activity in the
CNA hydrogenation. For the Cr/S sample, as the TPR results
indicated, it is difficult to reduce chromium oxides to Cr0.
Therefore, the dissociative chemisorption of hydrogen, a pre-
requisite to the CNA hydrogenation, cannot be accomplished.
When both elements, copper and chromium precursors, are
supported on the SBA-15 silica, the reduced materials become
active in the hydrogenation of CNA. Fig. 7 displays the typical
evolution of the CNA conversion for the mono- and bimetallic
CuxCr/S catalysts while Fig. 8 illustrates the conversion of CNA
Fig. 7 Typical evolution of CNA conversion vs. reaction time for the mono- and
bimetallic CuxCr/S catalysts (test conditions: Treaction = 150 1C, 0.264 g catalyst,
1 mL of CNA, solvent = propylene carbonate (25 mL), H2 flow = 1 L hꢁ1, stirring
rate = 900 rpm. The full lines represent the reaction products while the dotted
lines correspond to CNA conversion).
Fig. 8 CNA conversion vs. reaction time and the evolution of the reaction
products over time with the Cu1Cr/S catalyst reduced at 350 1C.
vs. reaction time as well as the evolution of the products for the the reaction is mainly directed to CNOL (B50 mol%) as
Cu1Cr/S catalyst. illustrated for the Cu1Cr/S sample (Fig. 8). Note that this value
The introduction of a small amount of chromium in the is higher than that previously reported for catalysts containing
sample (Cu10Cr/S) was favorable to the catalytic activity (Fig. 7). Cu0 or Cu+ species (30 mol%).10 For all three catalysts, the
Thus, CNA conversion of about 5 mol% was obtained and this selectivity to HCNA reached a maximum of B38 mol%, which
is in line with the TPR results, which show the higher reduci- thereafter slowly decreased while the HCNOL increased. These
bility for Cu2+ cations in this material. A further increase in trends indicate that similar active sites exist in all samples, on
chromium content (Cu5Cr/S), generated a catalyst with a higher which the adsorption of the two double bonds (CQC and CQO)
activity (B8 mol% of CNA conversion after 360 min of reac- takes place in the same manner. The higher selectivity to CNOL
tion), despite the lower copper content. When copper and can be considered as an evidence of the preferred CQO
chromium are in equal amounts (Cu1Cr/S) the activity is further adsorption on the catalyst surface. An explanation for this
improved (B11 mol% of CNA conversion after 360 min of preferred CQO adsorption on active sites can be given based
reaction). This behavior is probably due to the initial copper on the theoretical study of Delbecq and Sautet.86 According to
dispersion and reducibility for these samples (according to their theory, it is possible to improve the selectivity to the
XRD and TPR results). As concerns the products distribution, unsaturated alcohol either by activation of the CQO group or
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
7 G. V. Shanbhag, T. Joseph and S. B. Halligudi, J. Catal.,
2007, 250, 274.
8 J. Barrault, A. Derouault, G. Courtois, J. M. Maissant,
J. C. Dupin, C. Guimon, H. Martinez and E. Dumitriu, Appl.
Catal., A, 2004, 262, 43.
9 M. Chatterjee, T. Iwasaki, Y. Onodera, H. Hayashi,
Y. Ikushima, T. Nagase and T. Ebina, Appl. Clay Sci., 2004,
25, 195.
10 S. Valange, A. Derouault, J. Barrault and Z. Gabelica, J. Mol.
Catal. A: Chem., 2005, 228, 255.
by hindering the adsorption via the CQC bond. The activation
of the carbonyl bond is based on the theory of the two-electron
stabilizing interactions and involves the addition of an electro-
positive metallic cation in the close vicinity of the catalytically
active metal. This second metallic cation, acting as an acid
Lewis site, is an electrophilic center that will attract the free
electrons of the oxygen allowing the hydrogenation of the CQO
bond. For the Cu–Cr catalysts synthesized in this study, the
partially reduced chromium cations (Cr3+ and/or Cr2+) are
acting as electrophilic sites, which favors the formation of the
unsaturated alcohol. Accordingly, the CNA molecules are
adsorbed linearly on the cationic Crn+ sites and, thus, the
hydrogenation of the CQO bond is achieved by the activated
hydrogen on the neighboring Cu0 (the Cu0–Crn+ dual sites).
´
11 V. S. Gutierrez, A. S. Diez, M. Dennehy and M. A. Volpe,
Microporous Mesoporous Mater., 2011, 141, 207.
12 V. Gutierrez, M. Alvarez and M. A. Volpe, Appl. Catal., A,
2012, 413, 358.
13 W. V. Knowles, M. O. Nutt and M. S. Wong, in Catalyst
Preparation, Science and Engineering, ed. J. Regalbuto, CRC
Press, Boca Raton, 2007, p. 254.
Conclusions
14 J. R. Creighton and P. Ho, in Chemical vapor Deposition, ed.
J.-H. Park and T. S. Sudarshan, Surface Engineering Series,
ASM International, 2001, vol. 2, p. 13.
15 A. Carrero, J. A. Calles and A. J. Vizcaino, Appl. Catal., A,
2007, 327, 82.
16 P. Munnik, M. Wolters, A. Gabrielsson, S. D. Pollington,
G. Headdock, J. H. Bitter, P. E. de Jongh and K. P. de Jong,
J. Phys. Chem. C, 2011, 115, 14698.
17 T. Toupance, M. Kermarec and C. Louis, J. Phys. Chem. B,
2000, 104, 965.
18 A. Ungureanu, B. Dragoi, A. Chirieac, S. Royer, D. Duprez
and E. Dumitriu, J. Mater. Chem., 2011, 21, 12529.
19 A. Ungureanu, B. Dragoi, A. Chirieac, C. Ciotonea, S. Royer,
D. Duprez, A. Sophie Mamede and E. Dumitriu, ACS Appl.
Mater. Interfaces, 2013, 5, 3010.
The use of chromium as a second metal positively influences the
dispersion of CuO particles and their reducibility, when sup-
ported on SBA-15. Two mechanisms were taken into account to
explain these results: (i) the formation of copper–chromite spinel
and (ii) the interposition of Cr2O3 or even Cr6+ species between
CuO crystallites, preventing thus their aggregation and the
formation of bulky particles. The stabilizing effect of chromium
was further confirmed by the evolution of catalytic activity on
copper–chromium samples in the hydrogenation of cinnamal-
dehye. Therefore, CNA conversions up to 12 mol% and selectiv-
ities to CNOL of B50 mol% were noted for CuCr/SBA-15,
whereas the monocomponent samples showed negligible activ-
ity. This high level of chemoselectivity was explained by the
existence of dual Cu0–Cr3+/Cu0–Cr2+ active-adsorption sites,
which favor the adsorption of CNA molecules via the CQO bond.
20 A. J. Marchi, D. A. Gordo, A. F. Trasarti and C. R. Apesteguia,
Appl. Catal., A, 2003, 249, 53.
21 M. Wojciechowska, J. Haber, S. Łomnicki and J. Stoch,
J. Mol. Catal. A: Chem., 1999, 141, 155.
Acknowledgements
B. Dragoi acknowledges the project PERFORM-ERA ‘‘Postdoc-
toral Performance for Integration in the European Research
Area’’ (ID-57649), financed by the European Social Fund-
Romanian Government. The Romanian authors acknowledge
the project PN-II-ID-PCE-2011-3-0868, nr. 264/2011 financed by
UEFISCDI – Romania.
¨
22 T. Tsoncheva, M. Jarn, D. Paneva, M. Dimitrov and I. Mitov,
Microporous Mesoporous Mater., 2011, 137, 56.
23 R. Prasad and P. Singh, Catal. Rev. Sci. Eng., 2012, 54, 224.
24 A. Z. Abdullah, M. Z. A. Bakar and S. Bhatia, J. Hazard.
Mater., 2006, 129, 39.
25 Z. Ferhat, A. Derouault, J. Barrault and M. Bettahar, React.
Kinet. Catal. Lett., 2002, 76, 249.
26 S. H. Liu, S. Jaenicke and G. K. Chuah, J. Catal., 2002, 206, 321.
27 H. Surburg and J. Panten, Common Fragrance and Flavor
Materials, Preparation, Properties and Uses, Wiley-VCH
VerlagGmbH & Co. KGaA, Weinheim, 2006, p. 111.
28 A. Y. Hammoudeh, S. M. Saada and S. S. Mahmoud, Jordan
J. Chem., 2007, 2, 53.
29 P. Maki-Arvela, J. Hajek, T. Salmi and D. Y. Murzin, Appl.
Catal., A, 2005, 292, 1.
30 H. Wang, Y. Shu, M. Zheng and T. Zhang, Catal. Lett., 2008,
124, 219.
References
1 M.-J. Suh and S.-K. Ihm, Top. Catal., 2010, 53, 447.
2 L. Chen, P. Guo, L. Zhu, M. Qiao, W. Shen, H. Xu and K. Fan,
Appl. Catal., A, 2009, 356, 129.
3 S. Velu, K. Suzuki, M. Okazaki, M. P. Kapoor, T. Osaki and
F. Ohashi, J. Catal., 2000, 194, 373.
4 T. P. Maniecki, P. Mierczyski and W. K. Jowiak, Kinet. Catal.,
2010, 51, 843.
5 J. N. Keuler, L. Lorenzen and S. Miachon, Appl. Catal., A,
2001, 218, 171.
31 B. Dragoi, A. Ungureanu, D. Meloni, M. Casula, A. Chirieac,
A. Sasu, V. Solinas and E. Dumitriu, Environ. Eng. Manage. J.,
2010, 9, 1203.
6 F.-W. Chang, W.-Y. Kuo and K.-C. Lee, Appl. Catal., A, 2003,
246, 253.
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
32 A. Chirieac, B. Dragoi, A. Ungureanu, A. Moscu Corcodel, 60 A. B. Gaspar, C. A. C. Perez and L. C. Dieguez, Appl. Surf. Sci.,
C. Rudolf, A. Sasu and E. Dumitriu, Environ. Eng. Manage. J.,
2012, 11, 47.
33 D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson,
B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548.
34 http://database.iem.ac.ru/mincryst/s_carta.php?TENORITE.
35 W. Schmidt, Microporous Mesoporous Mater., 2009, 117, 372.
2005, 252, 939.
61 L. Zhang, J. Deng, H. Dai and C. T. Au, Appl. Catal., A, 2009,
354, 72.
62 M. Gomez-Cazalilla, A. Infantes-Molina, J. Merida-Robles,
E. Rodrıguez-Castellon and A. Jimenez-Lopez, Energy Fuels,
2009, 23, 101.
´
36 M. Imperor-Clerc, D. Bazin, M.-D. Appay, P. Beaunier and 63 V. R. Elıas, E. V. Sabre, E. L. Winkler, M. L. Satuf,
´
A. Davidson, Chem. Mater., 2004, 16, 1813.
37 Y. M. Wang, Z. Y. Wu and J. H. Zhu, J. Solid State Chem.,
2004, 177, 3815.
E. Rodriguez-Castellon, S. G. Casuscelli and G. A. Eimer,
Microporous Mesoporous Mater., 2012, 163, 85.
64 L. Xiaowei, S. Mingmin, H. Xi, Z. Haiyang, G. Fei, K. Yan,
D. Lin and C. Yi, J. Phys. Chem. B, 2005, 109, 3949.
65 H. Zhu, Y. Wu, X. Zhao, H. Wan, L. Yang, J. Hong, Q. Yu,
L. Dong, Y. Chen, C. Jian, J. Wei and P. Xu, J. Mol. Catal. A:
Chem., 2006, 243, 24.
´
38 J. J. Bravo-Suarez, B. Subramaniam and R. V. Chaudhari,
J. Phys. Chem. C, 2012, 116, 18207.
39 M. I. Zaki, N. E. Fouad, J. Leyrer and H. Knozinger, Appl.
Catal., 1986, 21, 359.
40 M. Casalboni, V. Ciafardone, G. Giuli, B. Izziy, E. Paris and 66 A. Kaddouri, N. Dupont, P. Gelin and P. Delichere, Catal.
P. Prosposito, J. Phys.: Condens. Matter, 1996, 8, 9059. Lett., 2011, 141, 1581.
41 E. Rombi, M. G. Cutrufello, V. Solinas, S. De Rossi, 67 K. Yoshida, C. Gonzalez-Arellano, R. Luque and P. L. Gai,
G. Ferraris and A. Pistone, Appl. Catal., A, 2003, 251, 255. Appl. Catal., A, 2010, 379, 38.
42 P. Michorczyk, J. Ogonowski, P. Kustrowski and 68 X. Xu, J.-W. He and D. W. Goodman, Surf. Sci., 1993, 284, 103.
L. Chmielarz, Appl. Catal., A, 2008, 349, 62.
43 L. Zhang, Y. Zhao, H. Dai, H. He and C. T. Au, Catal. Today,
2008, 131, 42.
44 T. Tsoncheva and E. Sarkadi-Priboczki, Microporous Meso-
porous Mater., 2012, 148, 1.
69 O. P. Tkachenko, K. V. Klementiev, M. W. E. van den Berg,
H. Gies and W. Gru¨nert, Phys. Chem. Chem. Phys., 2006, 8, 1539.
70 O. P. Tkachenko, K. V. Klementiev, N. Koc, X. Yu,
M. Bandyopadhyay, S. Grabowski, H. Gies and W. Gru¨nert,
Stud. Surf. Sci. Catal., 2004, 154, 1670.
´
45 M. Popova, A. Szegedi, Z. Cherkezova-Zheleva, A. Dimitrova 71 G. Moretti, C. Dossi, A. Fusi, S. Recchia and R. Psaro, Appl.
and I. Mitov, Appl. Catal., A, 2010, 381, 26. Catal., B, 1999, 20, 67.
46 V. R. Elias, E. V. Sabre, E. L. Winkler, M. L. Satuf, 72 C. J. G. Van Der Grift, A. Mulder and J. W. Geus, Appl. Catal.,
E. Rodriguez-catellon, S. G. Casuscelli and G. A. Eimer,
Microporous Mesoporous Mater., 2012, 163, 85.
47 S. M. El-Sheikh, R. M. Mohamed and O. A. Fouad, J. Alloys
Compd., 2009, 482, 302.
48 M. Aghaie-Khafri and M. H. Kakaei Lafdani, Powder Tech-
nol., 2012, 222, 152.
1990, 60, 181.
73 A. B. Gaspar, J. L. F. Brito and L. C. Dieguez, J. Mol. Catal. A:
Chem., 2003, 203, 251.
74 G. Calleja, J. Aguado, A. Carrero and J. Moreno, Appl. Catal.,
A, 2007, 316, 22.
75 X. Zhao and X. Wang, J. Mol. Catal. A: Chem., 2007, 261, 225.
´
49 E. A. Campos, R. C. L. Dutra, L. C. Rezende, M. F. Diniz, W. M. D. 76 P. Michorczyk, J. Ogonowski, P. Kustrowski and
Nawa and K. Iha, J. Aerosp. Technol. Manage., 2010, 2, 323. L. Chmielarz, Appl. Catal., A, 2008, 349, 62.
50 D. L. Hoang, A. Dittmar, M. Schneider, A. Trunschke, H. Lieske, 77 M. Trejda, A. Wojtaszek, M. Ziolek and J. Kujawa, Appl.
K.-W. Brzezinka and K. Witke, Thermochim. Acta, 2003, 400, 153. Catal., A, 2009, 365, 135.
51 B. S. Uphade, Y. Yamada, T. Akita, T. Nakamura and 78 G. Wang, L. Zhang, J. Deng, H. Dai, H. He and C. T. Au, Appl.
M. Haruta, Appl. Catal., A, 2001, 215, 137. Catal., A, 2009, 355, 192.
52 W. Y. Jung, S. H. Baek, J. S. Yang, K.-T. Lim, M. S. Lee, G.-D. Lee, 79 R. Zhang, D. Shi, Y. Zhao, B. Chen, J. Xue, X. Liang and
S. S. Park and S.-S. Hong, Catal. Today, 2008, 131, 437. Z. Lei, Catal. Today, 2011, 175, 26.
53 N. Phonthammachai, M. Krissanasaeranee, E. Gulari, 80 L. Mahoney, C.-M. Wu, H. S. Kibombo, E. Thiruppathi,
A. M. Jamieson and S. Wongkasemjit, Mater. Chem. Phys.,
2006, 97, 458.
J. Baltrusaitis, S. Rasalingam and R. T. Koodali, Microporous
Mesoporous Mater., 2012, 170, 211.
54 J. C. van der Waal and H. van Bekkum, J. Mol. Catal. A: 81 K. Pelzer, J.-P. Candy, G. Godard and J.-M. Bassetpag, in
Chem., 1997, 124, 137.
55 A. Ungureanu, D. Trong On, E. Dumitriu and S. Kaliaguine,
Appl. Catal., A, 2003, 254, 203.
Nanoparticles and catalysis, ed. D. Astruc, Wiley-VCH Verlag
GmbH & Co KGaA, Weinheim, 2008, vol. 1, p. 613.
82 http://www.webelements.com/chromium/compounds.html.
56 K.-T. Li and W.-T. Weng, J. Taiwan Inst. Chem. Eng., 2009, 40, 48. 83 C. H. Tu, A. Q. Wang, M. Y. Zheng, X. D. Wang and T. Zhang,
57 V. S. Garcia-Cuello, L. Giraldo and J. C. Moreno-Pirajan,
J. Chem. Eng. Data, 2011, 56, 1167.
58 G. Pantaleo, L. F. Liotta, A. M. Venezia, G. Deganello,b,
Appl. Catal., A, 2006, 297, 40.
84 X. C. Zheng, S. H. Wu, S. P. Wang, S. R. Wang, S. M. Zhang
and W. P. Huang, Appl. Catal., A, 2005, 283, 217.
E. M. Ezzo, M. A. El Kherbawi and H. Atia, Mater. Chem. 85 Z. Liu, M. D. Amiridis and Y. J. Chen, J. Phys. Chem. B, 2005,
Phys., 2009, 114, 604. 109, 1251.
59 B. Liu and M. Terano, J. Mol. Catal. A: Chem., 2001, 172, 227. 86 F. Delbecq and P. Sautet, J. Catal., 1995, 152, 217.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.