Facile preparation of 3D ordered mesoporous CuOx-CeO2 with notably enhanced efficiency for the low temperature oxidation of heteroatom-containing volatile organic compounds

Article abstract of DOI:10.1039/c3ra42566e

SBA-16 silica with intact surface hydroxyl groups was quickly obtained (5 min) via a fast-microwave-assisted method, and further adopted as an efficient template for the synthesis of three-dimensional (3D) ordered mesoporous CuO _x-CeO₂ catalysts (htpCCx) through a simple and reproducible host-guest interaction. XRD, XPS, H₂-TPR, and Raman results reveal that many of the Cu²⁺ ions in htpCCx are incorporated into the CeO₂ lattice, leading to the formation of a Cu _xCe_1-xO_2-δ solid solution, which produces a large number of oxygen vacancies and enhances the reducibility of the metal. The interaction of Cu and Ce is essential to the reaction as it maintains the Cu²⁺/Cu¹⁺ and Ce⁴⁺/Ce³⁺ redox couples. The catalyst has a 3D mesostructure and possesses remarkably enhanced low-temperature activity for the combustion of epichlorohydrin. HtpCC20 has been identified as the most powerful catalyst for this reaction, with the reaction rate at 165 °C being about 6.3 and 33.3 times higher than those of catalysts synthesized using conventional incipient impregnation and thermal combustion methods, respectively. Furthermore, htpCC20 shows superior CO₂ selectivity (>99%) and stability (no deactivation occurs after 50 h reaction). It is believed that the dispersion of the active phase, density of surface active oxygen, and low-temperature reducibility are the dominant factors governing the catalytic performance.

Full text of DOI:10.1039/c3ra42566e

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Facile preparation of 3D ordered mesoporous CuOx–
CeO2 with notably enhanced efficiency for the low
temperature oxidation of heteroatom-containing
volatile organic compounds3
Cite this: RSC Advances, 2013, 3,
19639
Chi He,ab Yanke Yu,a Changwei Chen,a Lin Yue,c Nanli Qiao,c Qun Shen,d
Jinsheng Chen*a and Zhengping Hao*c
SBA-16 silica with intact surface hydroxyl groups was quickly obtained (5 min) via a fast-microwave-
assisted method, and further adopted as an efficient template for the synthesis of three-dimensional (3D)
ordered mesoporous CuOx–CeO2 catalysts (htpCCx) through a simple and reproducible host–guest
interaction. XRD, XPS, H2-TPR, and Raman results reveal that many of the Cu2+ ions in htpCCx are
incorporated into the CeO2 lattice, leading to the formation of a CuxCe12xO22d solid solution, which
produces a large number of oxygen vacancies and enhances the reducibility of the metal. The interaction
of Cu and Ce is essential to the reaction as it maintains the Cu2+/Cu1+ and Ce4+/Ce3+ redox couples. The
catalyst has a 3D mesostructure and possesses remarkably enhanced low-temperature activity for the
combustion of epichlorohydrin. HtpCC20 has been identified as the most powerful catalyst for this
reaction, with the reaction rate at 165 uC being about 6.3 and 33.3 times higher than those of catalysts
synthesized using conventional incipient impregnation and thermal combustion methods, respectively.
Furthermore, htpCC20 shows superior CO2 selectivity (.99%) and stability (no deactivation occurs after 50
h reaction). It is believed that the dispersion of the active phase, density of surface active oxygen, and low-
temperature reducibility are the dominant factors governing the catalytic performance.
Received 25th May 2013,
Accepted 7th August 2013
DOI: 10.1039/c3ra42566e
www.rsc.org/advances
NOx and other more harmful by-products.2 Moreover, catalytic
oxidation techniques are feasible at any fuel-to-oxidant ratio,
1. Introduction
The development of effective methods for the environmentally
friendly removal of VOCs emitted from industrial processes
and municipal incineration is of great importance considering
their high toxicity and contribution to major environmental
problems such as ozone formation/depletion, aerosols, and
global warming.1 The catalytic oxidation of VOCs offers
interesting advantages over conventional flame combustion
or other treatment approaches. It can be carried out at low
temperature, thereby reducing or even avoiding the use of an
auxiliary fuel, as well as inhibiting the formation of thermal
thus allowing for the treatment of streams with very low
hydrocarbon concentrations.3,4 The key challenges are still
how to design and prepare low-cost and highly efficient
catalysts for the low-temperature catalytic combustion of
VOCs, especially for heteroatom (Cl, N, S, etc.) containing
pollutants.5 Supported or unsupported noble metal catalysts
(mainly Pd and Pt) and transition metal oxides (Co, Cu, Mn, Cr
and Ni) that can exist as single or mixed oxides are usually
employed for the complete oxidation of VOCs. Noble metals
are catalytically more active at low temperature in comparison
to other metal oxides, however they are expensive and rapid
deactivation by chlorine or sulfur poisoning greatly limits their
large-scale application for the elimination of S–VOCs and Cl–
VOCs.3,6,7 CeO2 has recently attracted considerable attention
as an efficient catalyst due to its abundant oxygen vacancy
defects, high oxygen storage capacity and the relatively easy
shuttle between its Ce3+ and Ce4+ oxidation states.8–10
However, CeO2 is deactivated quickly when used for the
decomposition of Cl–VOCs due to the strong adsorption of
HCl or Cl2 on to its surface, which blocks the active sites.11
Researchers have found that the stability of CeO2 could be
aKey Lab of Urban Environment and Health, Institute of Urban Environment, Chinese
Academy of Sciences, Xiamen 361021, P. R. China. E-mail: jschen@iue.ac.cn;
Fax: +86 592 6190765; Tel: +86 592 6190765
bDepartment of Environmental Science and Engineering, School of Energy and Power
Engineering, Xi’an Jiaotong University, Xi’an 710049, P. R. China
cDepartment of Environmental Nano-materials, Research Center for Eco-
Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, P. R. China.
E-mail: zpinghao@rcees.ac.cn; Fax: +86 10 62923564; Tel: +86 10 62923564
dResearch Center for Greenhouse Gases and Environmental Engineering, Shanghai
Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, P. R.
China
Electronic supplementary information (ESI) available. See DOI: 10.1039/
3
c3ra42566e
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19639

View Article Online
Paper
RSC Advances
improved if the adsorbed Cl2 or Cl2 was removed or
transferred rapidly enough.1,9 Among transition metal oxides,
chromium, manganese and copper oxides are estimated to be
the most active catalysts for the oxidation of heteroatom-
containing VOCs (HCVOCs). Copper and its compounds have
been widely applied in the study of dechlorination reactions
and have achieved good results.12–14 For instance, Huang et al.
have reported that a CuO modified USY zeolite possessed good
activity and high selectivity for Cl–VOC decomposition.15
Adding ceria as a promoter to copper oxide has many
advantages for catalytic performance: 1) it stabilizes the
dispersion of the active Cu phase;16 2) it acts as an oxygen
storing phase (Ce3+/Ce4+) and strongly affects the copper redox
properties;17–20 and 3) it creates structural defects and
improves the oxygen storage capacity and diffusivity.19,21,22
However, studies concerning the low-temperature catalytic
combustion of HCVOCs using CuO–CeO2 have seldom been
reported, despite the material being proven to have excellent
catalytic activities for the oxidation of many VOC pollu-
tants.23,24
to a large loss of Si–OH groups during the organic template
removal process that involved a thermal calcination,40,41 and
this phenomenon has also been reported by other
groups.35,42,43 Other approaches such as using solvents and
supercritical fluid extraction require long periods of operation
(.24 h) and it is difficult to remove the organic template
completely.44 Tian et al. proposed microwave irradiation as an
effective technique for making organic templates such as
Pluronic P123 quickly and with total removal of the one-
dimensional (1D) SBA-15 material with the greatest preserva-
tion of the original mesoporous structure and surface hydroxyl
groups.45 In this work, we report a fast-microwave-assisted
(FMA) method, which can completely remove triblock copoly-
mers P123 and F127 within 5 min. We demonstrate that the
obtained material is a powerful host for the production of
crystalline 3D ordered CuOx–CeO2 with a high BET surface
area (.100 m2 21) and large pore volume (.0.17 cm3 21).
g g
Moreover, the application of this FMA technique for the
convenient and fast preparation of other mesostructured
metal oxides could be expected.
Numerous active materials such as porous clay/silica-
supported catalysts and hydrotalcite-derived metal oxides have
been successfully developed for the elimination of VOCs (e.g.,
alkane and arene) in our previous work.46–50 Herein, for the
At present, the primary issues for the oxidation of HCVOCs
using transition metal oxides are how to improve the catalytic
activity, reaction durability, and thermal stability. Different
protocols have been reported for the synthesis of highly
efficient transition metal oxide-loaded catalysts in previous
studies, such as sol–gel, incipient impregnation, urea-nitrate
combustion, coprecipitation, and one-pot polymeric precur-
sors, however, the surface areas and total pore volumes of the
obtained catalysts were rather low.9,25–27 In order to overcome
the aforementioned problems, porous supports (e.g., c-Al2O3
and USY zeolite) are usually adopted, however, the metal oxide
content has to always be controlled in order to avoid
aggregation of the active phase.15,28–30 Mesoporous transition
metal oxides with d-shell electrons confined to nanosized
walls, redox active internal surfaces, and connected pore
networks demonstrate extremely fascinating prospects for use
in catalysis, energy conversion and storage, as well as gas
sensing, adsorption, and separation.31–33 The use of soft
templates (e.g., chelating agents and surfactants) and hard
templates (e.g., mesoporous materials and carbon nanotubes)
has led to some wonderful contributions.34–36 Compared with
soft templates, hard templates have some advantages for the
preparation of ordered mesoporous arrays. Supramolecular
mesoporous silica templates (MCM41S, SBA-15, KIT-6, SBA-16,
etc.) have uniform mesocavities, high surface areas, and large
pore volumes, which makes them the perfect candidates to
serve as ‘‘hard templates’’.35,37,38 Among these, SBA-16 has
three-dimensional (3D) channels and a cubic Im3m (body-
centered cubic arrangement of cages with 8 apertures to the
nearest neighbors) structure, which could be more favorable
for mass and thermal transfer than a unidirectional pore
system.39 In order to introduce the maximum quantity of
metal precursors, as many silanol groups as possible must be
retained during the preparation of the silica template. In our
previous work, we found that the mesochannels of SBA-15 and
KIT-6 were far from completely filled by metal precursors due
first time, we have prepared
a series of 3D ordered
mesoporous CuOx–CeO2 catalysts through reverse replication
of the structure of FMA-SBA-16 for use in the low temperature
oxidation of epichlorohydrin (ECH: a representative and highly
toxic HCVOC widely used in industry) with the aim of
developing a more efficient catalyst for HCVOC removal. The
catalytic performance of these materials was extensively
compared to that of the corresponding single oxide, and the
CuOx–CeO2 binary oxides synthesized via traditional incipient
impregnation, coprecipitation and thermal combustion meth-
ods. The central issues addressed in this study include: 1) the
proposition of a simple and reproducible approach for the
synthesis of ordered mesoporous metal oxides; 2) the devel-
opment of practical and highly effective transition metal-
loaded catalysts for the low-temperature removal of HCVOCs;
3) the illumination of the role and contribution of physico-
chemical properties (e.g., surface area, pore structure, surface
state, reducibility, and active oxygen mobility) to the catalytic
performance and durability; and 4) the investigation of the
nature of the active sites and the role of copper and ceria in the
binary metal oxides in order to elucidate a possible reaction
mechanism.
2. Experimental
2.1. Synthesis of the SBA-16 silica template
Stucky et al.51 found that the EO/PO ratio of a Pluronic
copolymer could be used to tailor the pore diameter, pore
entrance size, and structural order of SBA-16. Ryoo et al.38
proposed that the obtained SBA-16 possessed the largest
possible pore diameter and pore window size as well as the
19640 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
highest possibility of forming an ordered structure when a
amount of nitrogen adsorbed at a relative pressure (P/P0) of ca.
0.99. The specific surface area was calculated using the
Brunauer–Emmett–Teller (BET) method, and the pore size
distribution was derived from the desorption branch of the N2
isotherm using the Barrett–Joyner–Halenda (BJH) method. The
actual Cu content was determined using inductively coupled
mixture of P123 (EO20PO70EO20) and F127 (EO106PO70EO106
)
with an EO/PO ratio of 16/7 was adopted. Therefore, in the
present work, SBA-16 was prepared under acidic conditions
using two poly(alkylene oxide)-type triblock copolymers,
Pluronic 123 and F127, as structure-directing agents and
tetraethyl orthosilicate (TEOS) as a silica source. Briefly, 0.48 g
of P123 and 2.41 g of F127 were dissolved in 142 mL of 2 M
HCl solution at 35 uC. Afterward, 11.6 ml of TEOS (98%) was
added to the solution, and magnetic stirring was continued for
20 min at 35 uC. The mixture was kept under static conditions
in an oven for 24 h at the same temperature, and then aged at
100 uC for 24 h. The recovered precipitate was filtered and
dried at 100 uC without washing to obtain SBA-16. The P123
and F127 organic templates in the as-made SBA-16 were totally
removed within 5 min with a FMA technique using an XT-9912
microwave preparation system (Shanghai Xintuo Co., China).
Briefly, 1.0 g of as-made SBA-16, 6.67 ml of concentrated
HNO3, and 3.25 ml of H2O2 (30 wt%) were digested in a Teflon
vessel for 5 min using an operating power, reaction tempera-
ture and pressure of 1300 W, 250 uC and 2.0 MPa, respectively.
The obtained sample was washed with deionized water and
ethanol, and then dried at 40 uC in order to obtain FMA-SBA-
16.
plasma optical emission spectroscopy (ICP-OES) with
a
PerkinElmer OPTIMA 7000 DV spectrometer.
Thermogravimetric analysis (TGA) was performed using a
Netzsch 209 F3 Tarsus thermo-microbalance from 50 to 900 uC
at a heating rate of 10 uC min21 under a stream of O2/N2.
Fourier transform infrared spectra (FT-IR) were obtained using
a Nicolet IS10 IR spectrometer by measuring the absorbance
on KBr pellets (2 wt% sample and 98 wt% KBr). Transmission
electron microscopy (TEM) images were obtained using a
Hitachi H-7650 microscope operating at an acceleration
voltage of 80 kV. Field emission scanning electron microscopy
(FE-SEM) images were obtained using a Hitachi S-4800
microscope equipped with an EDX detector for composition
analysis of the synthesized catalysts. Temperature pro-
grammed reduction (H2-TPR) experiments were performed
using a Quantachrome ChemBET 3000 TPR-TPD analyzer.
Prior to reduction, 50 mg of the catalyst was heated in a 5% O2/
He flow (50 mL min21) from room temperature to 500 uC and
held at this temperature for 30 min, before treatment with a
stream of He at room temperature for 30 min in order to
remove any residual oxygen. Finally, the sample was reduced
under a 5% H2/He flow (50 mL min21) from 50 to 700 uC at a
heating rate of 10 uC min21. The temperature programmed
desorption of O2 (O2-TPD) was carried out using the same
apparatus as that for H2-TPR. 50 mg of the catalyst was pre-
heated in a 5% O2/He flow (50 mL min21) at 500 uC for 30 min,
then the sample was heated from 50 to 800 uC at a heating rate
of 10 uC min21 under a stream of pure He. X-ray photoelectron
spectroscopy (XPS) experiments were carried out with a
Quantum 2000 instrument under high vacuum (,5 6 10210
torr) using Al Ka as the exciting radiation at a constant pass
energy of 1486.6 eV. The C 1s peak was used to calibrate the
binding energy. Raman spectra were obtained using a Horiba
Jobin Yvon S.A.S. LabRAM Aramis spectrometer at ambient
temperature and under moisture-free conditions. The emis-
sion line at 532 nm from an Ar+ ion laser was focused in order
to analyze a spot of about 1 mm. The time for acquisition was
varied according to the intensity of the Raman scattering, and
the wave numbers obtained from the spectra were accurate to
within 1 cm21. Diffuse reflectance UV-visible (DRUV-Vis)
spectra were obtained using a Hitachi UV-3000 spectro-
photometer under ambient conditions with spectral grade
BaSO4 as a reference material. X-ray absorption fine structure
(XAFS) data for the htpCCx materials were measured using a
4W1B beam line at the Beijing Synchrotron Radiation Facility
(BSRF). The spectral range from 8.820 to 9.640 keV for the
K-absorption edge at 8.990 keV was collected under ambient
conditions. A Si (111) monochromator crystal was used and
the XAFS spectra of the adsorption samples were collected with
a Lytle ionization detector in fluorescence mode. An average of
three scans was performed in order to achieve an adequate
signal/noise ratio, and the energy of the Cu K-absorption edge
was calibrated with Cu foil. The XAFS data were processed
2.2. Catalyst preparation
CuOx–CeO2 catalysts with 3D ordered mesoporous structures
were synthesized using
a simple thermolysis process.
Typically, different amounts of Cu(NO3)2?3H2O (0.13–0.41 g)
and Ce(NO3)3?6H2O (1.75–2.25 g) were dissolved in 18 mL of
ethanol. Dry FMA-SBA-16 samples (0.43 g per group) were then
added to the above solutions. After stirring for 6 h, the
solutions were transferred to clean Petri dishes and the solvent
was evaporated. Next, the hybrid samples were thermally
calcined at 550 uC for 4 h at a heating rate of 2 uC min21. The
metal oxides were obtained in very high yields by solution
etching the silica framework using 0.1 M NaOH at 80 uC for 12
h. The CuOx–CeO2 materials were washed with distilled water
and ethanol and then recovered using centrifugation. The final
catalysts with different Cu content are denoted as htpCCx,
while the used catalysts are denoted as htpCCxu. For
comparison, CuOx–CeO2 binary oxides with Cu loading of 30
wt% were prepared using conventional incipient impregna-
tion, coprecipitation, and thermal combustion methods
following literature procedures,27,28,52 and the related catalysts
are denoted as impCC30, copCC30 and tcbCC30, respectively.
Moreover, bulk-CuO and bulk-CeO2 single oxides were also
synthesized by thermal calcination of the corresponding
hydrated metal nitrates at 550 uC for 4 h.
2.3. Catalyst characterization
X-ray diffraction (XRD) patterns were recorded using a Philips
X’Pert powder diffraction system with Cu-Ka radiation (l =
0.15418 nm) in the 2h ranges of 0.6–5u (scanning rate of 0.10u
min21) and 10–80u (scanning rate of 4u min21). N2 adsorption/
desorption isotherms of the catalysts were measured at 77 K
using a Quantachrome NOVA 1200e gas sorption analyzer. All
samples were degassed under vacuum at 300 uC for 3 h before
measurement. The total pore volume was estimated from the
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19641

View Article Online
Paper
RSC Advances
following the standard procedure of background absorption
removal: normalization, conversion to k-space and Fourier
transformation. Extended X-ray absorption fine structure
(EXAFS) geometric information was extracted using WINXAS
3.1 software. A linear function fit for the pre-edge region and a
second-order polynomial fit in the post-edge region were used
for the normalization in order to remove the background
absorption. The normalized EXAFS spectra were converted to
the EXAFS function x(k) (where k is the wavenumber of the
photoelectron) using a cubic spline-fit and k3-weighting. The
k3x(k) spectra were then converted into R space by Fourier
transforming the region between 2.6 to 9.9 Å21 without using a
window function.
2.4. Catalytic activity measurement
All evaluation experiments were performed at atmospheric
pressure in a continuous-flow fixed-bed reactor, containing a
stainless steel tube (6 mm internal diameter) that was filled
with the catalyst. VOC-containing gas was generated by
bubbling air through a VOC saturator, which was then further
diluted with another air stream, before reaching the reaction
bed. The temperatures of the catalyst bed and tubular electric
furnace were monitored automatically using E-type thermo-
couples. In each test, 300 mg of the catalyst (40–60 mesh) was
placed at the middle of the tube reactor and the total flow rate
was maintained at 200 mL min21, i.e., a gas hourly space
velocity (GHSV) of 18 400 h21, and the O2 feed concentration
was kept at about 21 v/v%. In each test, the catalyst bed
temperature was first raised to 120 uC with the feed stream
passing and stabilized for 30 min. The temperature was then
increased to the next desired value at a heating rate of 5 uC
min21 and stabilized for 20 min prior to online analysis of the
Fig. 1 FT-IR spectra of (a) as-made SBA-16, (b) FMA-SBA-16, (c) calcined SBA-16,
and (d) htpCC20.
relative intensity of the Si–OH bending vibrations centered at
957 cm21 is almost the same for the as-made SBA-16 and FMA-
SBA-16 samples, but much weaker for the calcined sample,
indicating that a high level of silanol groups are preserved on
the interior and exterior pore wall surfaces during the FMA
process while the thermal calcination inevitably decreases
their number. The thermal stability of the aforementioned
samples was investigated using TGA (Fig. S1, ESI ). TGA curves
3
reveal weight losses of approximately 9.0% for the FMA-SBA-16
and 7.9% for the calcined SBA-16, which are much lower than
that of the as-made SBA-16 (about 34.8%). The results from the
FT-IR spectra and TGA clearly indicate that the FMA technique
leads to fast and thorough removal of the template, while
simultaneously preserving the integrity of the surface hydroxyl
groups
effluent gas composition using
a gas chromatograph
(GC9890A, Shanghai Linghua Co., China) equipped with an
electron capture detector (ECD) and flame ionization detector
(FID). The concentrations of organic chlorinated compounds
in the feed and effluent streams were determined and
quantified by the ECD, while other by-products were identified
by the FID and a Pfeiffer Omnistar online mass spectrometer.
Briefly, the formation of the final nanostructured mesopor-
ous CuOx–CeO2 catalysts required the following four steps:
firstly, the 3D mesoporous SBA-16 siliceous templates with
abundant surface silanol groups were synthesized using the
FMA approach. Secondly, the related hydrated metal nitrate
precursors were introduced via highly enhanced host–guest
interactions such as hydrogen bonding, Coulombic interac-
tions, coordinating interactions, and van der Waals interac-
tions.53 Thirdly, the mesostructured pattern was formed and
the guest materials crystallized using the thermal calcination
method. Fourthly, the silica host was etched with a dilute
NaOH aqueous solution in order to obtain the final mesos-
tructured binary metal oxides. Fig. 2A shows the small-angle
XRD (SA-XRD) patterns of the synthesized materials in the 2h
range of 0.7–5u. Both the calcined and FMA treated samples
have one intense diffraction peak at around 0.8u, alongside
two other weak peaks between 1.1–2.0u, which can be assigned
to the (110), (200) and (211) reflections of the cubic Im3m
mesostructure.38 The htpCC10 and htpCC20 oxide catalysts
possess (110) and (200) reflections between 0.7–1.1u, which
indicate that the uniform pore structure of FMA-SBA-16 is well
retained. Furthermore, the diffraction peaks appear to become
3. Results
3.1. Template removal and preparation of 3D mesoporous
CuOx–CeO2 oxides
FT-IR transmission spectra of the as-made SBA-16, FMA-SBA-
16, and calcined SBA-16 samples are shown in Fig. 1. A broad
band at around 3400 cm21 that appears in all of the samples is
partially caused by the O–H stretching vibration of adsorbed
water molecules, whose bending vibration mode is also
responsible for the band recorded at around 1630 cm21. The
IR absorption bands centered at 2936 and 1466 cm21, shown
in the spectrum of as-made SBA-16 can be assigned to the C–H
bond stretching and bending vibrations of the P123 tem-
plate,45 however these absorptions are nearly indiscernible in
the spectra of the FMA-SBA-16 and calcined SBA-16 samples.
This suggests that both the FMA and calcination methods can
efficiently remove the organic P123 template. Remarkably, the
19642 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Fig. 2 (A) SA-XRD patterns of (a) calcined SBA-16, (b) FMA-SBA-16, (c) as-made SBA-16, (d) htpCC10, (e) htpCC20, and (f) htpCC30; (B) WA-XRD profiles of (a) bulk-
CeO2, (b) htpCC10, (c) htpCC20, (d) htpCC30, (e) copCC, (f) impCC, (g) tcbCC, and (h) bulk-CuO.
weaker or even disappear with increasing Cu content, which
indicates that the ordered mesoporous structure decreases to
some extent.
small clusters on the ceria surface, which cannot be detected
using XRD;54,55 2) the incorporation of CuO into the CeO2
lattice, or a combination of these two phenomena.56,57 The
lattice parameters of the CeO2 unit cell were obtained from the
position of the (111) diffraction peak and are listed in Table 1.
The slight decrease in the bulk-CeO2 (5.4276 Å) lattice
parameter after the addition of the copper species is evidence
that Cu ions are partly incorporated into the CeO2 crystallites,
probably forming Cu–O–Ce species, especially considering
that the radius of a Cu2+ ion (0.073 nm) is smaller than that of
a Ce3+ (0.110 nm) or Ce4+ ion (0.097 nm).57 Additionally, the
lattice microstrain was calculated, and the value related to
bulk-CeO2 (8.7 6 1023) was found to be somewhat lower than
those of the CuOx–CeO2 binary oxides (8.8 6 1023–14.3 6
3.2. Crystal phase and textural properties of the prepared
samples
Fig. 2B shows the wide-angle XRD (WA-XRD) patterns of the
samples synthesised using different preparation methods and
with differing Cu/(Cu + Ce) loading. The distinct fluorite-type
oxide structure of CeO2 was present in each sample (JCPDS No.
81-0792). No diffraction peaks corresponding to CuO or Cu2O
could be identified, except in the cases of impCC and tcbCC
(JCPDS No. 80-1268). The absence of CuO peaks can probably
be attributed to: 1) the fine dispersion of CuO in the form of
Table 1 Physicochemical characteristics of the synthesized catalysts
Nom.a (wt%) ICPb (wt%) XPSc (wt%) d110 (nm) a111 (Å) Cs (nm) Ls (61023
)
SBET (m2 g21
)
Dv (cm3 g21
h
)
Dp (nm)
d
e
f
g
i
Sample
htpCC10
htpCC20
htpCC30
copCC
impCC
tcbCC
10
20
30
30
30
30
—
—
9.1
11.7
22.5
35.3
32.2
37.1
42.6
—
6.4
6.1
5.7
—
—
—
5.4217
5.4179
5.4173
5.4152
5.4148
5.4125
5.4276
—
15.6
12.2
10
8.9
128.8
117.9
108.2
86.5
59.1
3.9
0.28
0.24
0.17
0.14
0.16
0.01
0.18
—
5.2
4.9
4.0
3.1
5.5
5.3
6.1
5.4
20.6
28.9
31.1
29.3
30.5
—
11.6
14.3
14.1
8.8
2.2
8.7
10
15.7
64.1
16
bulk-CeO2
bulk-CuO
—
—
60.8
1.1
—
—
—
—
a
b
c
Nominal Cu content in the synthesized samples. The Cu value estimated using XPS analysis. Actual Cu content obtained from the ICP-
d
e
f
OES analysis. Interplanar spacing of the (110) plane, calculated using Bragg’s law. Crystallite size. Lattice microstrain value calculated
from the (111) reflection of CeO2 using the Scherrer equation. BET specific surface area calculated at P/P0 = 0.05–0.25. Total pore volume
estimated at P/P0 = 0.99. BJH pore diameter calculated from the desorption branch.
g
h
i
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19643

View Article Online
Paper
RSC Advances
1023) due to the incorporation of Cu. The ICP-OES results
listed in Table 1 reveal that the actual Cu content in the final
catalyst is approximately equal to the theoretical Cu loading.
Moreover, the surface composition of the synthesized samples
was further estimated using XPS (Table 1). It can be concluded
that the Cu species are uniformly distributed in htpCCx and
copCC from the surface through to the interior. In contrast,
there is a surface enrichment of Cu oxide for the impCC
(37.1%) and tcbCC (42.6%) samples, which is consistent with
the XRD results (Fig. 2B (f and g)).
than those of the samples synthesized via other approaches
(0.01–0.16 cm3 21). However, the Dp values of the htpCCx
g
samples (ca. 4.0–5.2 nm) are a little lower than those of impCC
(ca. 5.5 nm) and tcbCC (ca. 5.3 nm). The values for the
interplanar spacing (d110) of the mesoporous CuCeOx samples
are listed in Table 1, and the results reveal that the
mesoporous mixed metal oxides possess larger interplanar
spacings (ca. 5.7–6.4 nm) than that of the Si-based SBA-16
template (ca. 5.5 nm). Noticeably, the SBET values (108.2–128.8
m2
those of the other catalysts and single metal oxides (0.1–86.5
m2 21). All of the SBET, Dv, and Dp values are gradually
g
21) of all of the htpCCx samples are much higher than
The N2 adsorption/desorption isotherms and pore size
distribution of all of the synthesized catalysts are shown in
g
reduced with the increasing copper loading, as shown in
Table 1.
Fig. S2 (ESI ), while the specific surface area (SBET), mean pore
3
diameter (Dp), and total pore volume (Dv) are listed in Table 1.
It is clear that htpCC10 and htpCC20 possess type IV
isotherms with a typical H2 type hysteresis loop, characteristic
of an ordered mesoporous cage-like (Im3m) structure accord-
ing to IUPAC classification.58 This result demonstrates good
reverse replication of the mesostructure of the FMA-SBA-16
template by the metal nitrate precursors via a simple host–
guest interaction. However, the pore order is decreased when
the Cu content is further increased, which is in good
agreement with the XRD results (Fig. 2A (f)). In general, the
Dv values (0.17–0.28 cm3 g21) of the htpCCx catalysts are larger
Further information about the microstructure of the as-
prepared CuOx–CeO2 mixed oxides, as well as the FMA-SBA-16,
is provided using TEM (Fig. 3). Long-range ordered pores
arranged in a 3D mesoporous structure can be observed for
FMA-SBA-16, htpCC10, and htpCC20, with average pore
diameters of ca. 4.7, 5.5, and 5.2 nm, respectively (Fig. 3a–d).
TEM also shows that htpCC30 has a smaller pore diameter (ca.
5.0 nm) and a lack of long-range order, which is consistent
with the XRD and N2 sorption results (Fig. 2, Fig. 3e, and Fig.
S2, ESI ). The nanoparticles of the copCC sample, synthesized
3
Fig. 3 Representative TEM images of the various catalysts: (a) FMA-SBA-16, (b) htpCC10, (c) htpCC20 [100], (d) htpCC20 [111], (e) htpCC30, (f) copCC, (g) impCC, (h)
tcbCC, and (i) htpCC20u.
19644 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Fig. 4 FE-SEM images of (a) FMA-SBA-16, (b) bulk-CeO2, (c) htpCC20 (insets: magnified SEM patterns and EDS spectra), (d) copCC, (e) impCC, and (f) tcbCC.
using a coprecipitation protocol, exhibit a well dispersed
morphology with an average crystallite size of ca. 17 nm
(Fig. 3f), while the metal precursors are prone to agglomerate
into clusters with calculated diameters greater than 40 and 200
nm for the impCC and tcbCC catalysts, respectively (Fig. 3g
and h). Fig. 4 shows typical FE-SEM images of the prepared
samples. The SEM image of FMA-SBA-16 (Fig. 4a) indicates
that the most commonly observed particle morphology is that
of rhombidodecahedron-shaped single crystals, which is in
accordance with the crystal model of the cage-like Im3m
structure.59 The htpCC20 sample possesses a porous and
coarse surface morphology, which is different to that of the
pure siliceous FMA-SBA-16 template probably due to breakage
of the long-range order of the original mesopores (Fig. 4c).
EDX results reveal that the Cu content of htpCC20 (22.13 wt%)
is close to the nominal value (Fig. 4c, inset). The copCC,
impCC, and tcbCC samples are composed of nanocrystallites
with diameters ranging from dozens to hundreds of nan-
ometers, moreover these nanosized particles are badly
aggregated in the tcbCC sample (Fig. 4d–f).
3.3. Surface composition and reducibility of the synthesized
catalysts
The surface composition and chemical state of the elements
that exist in the CuOx–CeO2 catalysts were further investigated
using XPS. Fig. 5 shows a comparison of the Cu 2p, Ce 3d, and
O 1s XPS binding energies (BE), as well as the Cu LMM Auger
kinetic energy, of the catalysts with different Cu loadings and
synthesis methods. The presence of a shake-up peak and a
high Cu 2p3/2 binding energy (BE . 933.1 eV) are two major
XPS features that are characteristic of Cu2+ species, while a
lower Cu 2p3/2 binding energy (BE = 932.2–933.1 eV) and the
absence of a shake-up peak are characteristic of reduced
copper (Cu1+ or Cu0).60 The Cu 2p spectra of the CuOx–CeO2
(Fig. 5) catalysts possessed shake-up peaks of weak intensity
(especially for the htpCCx samples) between 940–947 eV and
Cu 2p3/2 binding energy peaks at 934.5 (the principal peak)
and 932.4 eV. The lower Cu 2p3/2 binding energy (932.4 eV) and
weak intensity of the shake-up satellite structure suggest the
presence of reduced copper species. The degree of reduction of
the Cu species can be studied by determining the Isat/Ipp ratios
(Table 2).61 The Isat/Ipp values in this work are in the range of
0.11–0.37, which are between the values of 0.57 for pure Cu2+
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19645

View Article Online
Paper
RSC Advances
Fig. 5 Cu 2p, Auger LMM kinetic energy, Ce 3d, and O1s XPS spectra of (a) htpCC10, (b) htpCC20, (c) htpCC30, (d) copCC, (e) impCC, and (f) tcbCC.
and zero for pure Cu1+,62 indicating the presence of a reduced
Cu phase in all of the samples. The Auger Cu LMM lines were
further investigated for all of the CuOx–CeO2 samples in order
to clarify the composition of the reduced copper species as the
Cu 2p3/2 binding energy cannot be used to distinguish between
Cu2O and Cu0. The broad features of the kinetic energy spectra
Table 2 Characteristic data and a summary of the XPS and H2–TPR analysis results
f
Osd (%)
Ce3+/Cetot
Tfr (uC)
LH2 (mmol g21
)
HH2 (mmol g21
)
Cu (%)
Ce (%)
O (%)
Olatt
Osur
Oads
a
b
c
d
e
Sample
Isat/Ipp
htpCC10
htpCC20
htpCC30
copCC
impCC
tcbCC
0.17
0.14
0.11
0.22
0.28
0.37
0.37
0.39
0.38
0.33
0.31
0.23
159
103
95
99
152
251
194.3
868
979.5
693.7
425.9
96.4
679.9
1416.6
1547.3
1172.1
988.2
2.2
5.2
7.7
8.0
6.4
15.1
13.6
14.1
16.8
18.1
21.0
82.7
81.2
78.2
75.2
75.6
63.4
61.1
42
51.4
58.3
64.3
67.8
36.5
45.9
42.1
35.4
31.5
25.3
2.4
12.1
6.5
6.3
4.2
1014.3
15.6
6.9
a
The ratio of the Cu 2p satellite peaks to the principal peaks of the CuOx–CeO2 catalysts.
Ce3z u0zu0zv0zv0
b
c
Ce3zratio~
~
000 , and the related peak areas obtained from Fig. 6. Temperature corresponding
Ce3zzCe4z u0zuzu0zu00zu000zv0zvzv0zv00zv
d
to the first reduction peak in the H2–TPR analysis. H2 consumption of the reduction peaks below 200 uC. e Total H2 consumption of the reduction
f
peaks from room temperature to 600 uC. Oxygen species distribution in the various catalysts.
19646 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
for the Auger Cu LMM electrons, shown in Fig. 5, are
composed of the contribution from two Cu species. The peaks
in the Auger kinetic spectra at 917.6 and 914.6 eV correspond
to Cu2+ and Cu1+ species, respectively.63 The coexistence of
Cu1+ and Cu2+ ions in the CuO–CeO2 catalysts prepared using
various methods have been previously reported.27,60,62
Avgouropoulos et al.27 and Flytzani-Stephanopoulos et al.60
have proposed that Cu1+ species result from the strong
interactions of the Cu clusters with Ce oxide. The formation
of Cu1+ might be induced by substitution at the interface of the
two oxide phases due to the similarity between the ionic radii
of Cu1+ and Ce4+.
The experimental and fitted Ce 3d spectra of the synthesized
catalysts are shown in Fig. 5, and the calculated Ce3+ ratios are
listed in Table 2. It should be noted that there is no significant
difference in the Ce3+/Cetot ratios among the htpCCx samples
with different Cu/Ce ratios, while the quantity of Ce3+ in the
htpCCx catalysts (0.37–0.38) is slightly higher than in copCC
(0.33), impCC (0.31), and tcbCC (0.23). Previous work has
reported that the existence of Ce3+ in CeO2 implies the
formation of an oxygen vacancy, and thus, it can be inferred
that the htpCCx samples possess more oxygen vacancies than
the other catalysts. Furthermore, the ratios of Ce3+ for all of
the prepared samples are much higher than that of pure CeO2
reported in the literature (Ce3+ = 0.067), which leads to the
larger lattice parameters of these samples compared to that of
the reported one (Table 1). The corresponding O 1s XPS
spectra of the synthesized catalysts can be resolved into three
peaks by deconvolution (Fig. 5): 1) the lattice oxygen (Olatt, BE
= 529.3 eV) and surface oxygen (Osur, BE = 531.5 eV) assigned to
the defect oxide or surface oxygen ions in low coordination
situations and weakly bonded oxygen species; 2) adsorbed
oxygen species (Oads, BE = 532.8 eV) from hydroxyl species and
adsorbed water as contaminants on the surface.9,64 Table 2
shows the oxygen species distribution in the various samples,
and the results reveal that the Osur values of htpCCx (.36.5%)
are much higher than those of the other samples (25.3–35.4%).
The htpCC20 sample has the highest percentage of Osur
(45.9%), followed by htpCC30 (Osur = 42.1%). In sharp
contrast, tcbCC possesses the lowest amount of surface oxygen
(Osur = 25.3%).
The H2-TPR profiles of the CuOx–CeO2 catalysts are shown
in Fig. 6, and the reducibility of bulk-CeO2 and bulk-CuO is
also investigated for comparison purposes. The reduction of
pure CuO commences at about 250 uC, and its reduction
profile is characterized by a single peak centered at 373 uC
(Fig. 6g). Pure CeO2 shows two weak reduction peaks with
apexes at 415 and 517 uC (Fig. 6h), which can be ascribed to
the reduction of the surface capping and bulk oxygen of CeO2,
respectively.65 The reduction peaks of the CuOx–CeO2 mixed
oxides are systematically shifted towards the lower tempera-
ture region compared with bulk-CeO2 and bulk-CuO, implying
that the reduction of copper and cerium oxide is promoted due
to the interaction between the Cu and Ce oxide. The htpCCx
catalysts have two to four reduction peaks below 500 uC, with
the reductive peaks at around 100 and 170 uC attributed to the
reduction of the well dispersed Cu species and Cu incorpo-
rated into the CeO2 lattice, respectively.66 The reduction peaks
higher than 200 uC correspond to the contribution of crystal-
Fig. 6 H2-TPR profiles of (a) htpCC10, (b) htpCC20, (c) htpCC30, (d) copCC, (e)
impCC, (f) tcbCC, (g) bulk-CuO, and (h) bulk-CeO2.
line CuO,23 and signal peaks centered at reduction tempera-
tures exceeding 300 uC are probably associated with the
reduction of CeO2 surface oxygen. The first reduction peak
decreases from 159 to 95 uC with the increasing Cu loading of
the htpCCx samples (Fig. 6a–c). This is probably due to an
increase in Cu–O–Ce species, which facilitate the reduction of
copper species due to the reducibility of Cu–O–Ce.22 In
addition, the copCC sample is more easily reduced than the
impCC and tcbCC samples due to the better dispersion of the
copper species,67 which is in good agreement with the XRD
and TEM results (Fig. 2 and Fig. 3). The H2 consumption of the
reduction peaks (the bulk-CuO reduction peak was used for
calibration) below 200 uC and 600 uC, as well as the
temperatures that correspond to the first reduction peaks are
listed in Table 2. Taking into account both the H2 consump-
tion and reduction temperature, htpCC20 and htpCC30
possess superior reducibility compared to the other samples.
3.4. Structure characteristics of the CuOx–CeO2 catalysts
Fig. 7 shows the UV-Vis diffuse reflectance spectra of the
prepared catalysts. Bulk-CeO2 shows three characteristic peaks
centered at 226, 281, and 345 nm, which can be assigned to the
f A d transitions of Ce3+ species, low coordination Ce4+–O
charge transfer, and interband transitions in CeO2, respec-
tively.68 The absorption peak at around 461 nm can be
attributed to the emission and absorption of one photon by a
Cu1+ complex,25 indicating the presence of Cu1+ in the
synthesised catalysts. The absorption in the 636–675 nm
range corresponds to the 2d transitions of Cu2+ in an
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19647

View Article Online
Paper
RSC Advances
Fig. 8 Copper K-edge XANES spectra of htpCC20, CuO, Cu2O, and Cu foil (inset:
the Fourier transformed EXAFS spectra of htpCC20 and htpCC20u). The peak
positions are uncorrected for phase shifts.
weighted Fourier transform magnitudes of the EXAFS spectra
for the htpCC20 and htpCC20u catalysts. The tiny decrease of
the Cu–O peak intensity after ECH oxidation at 1.5 Å suggests
the reduction of a small fraction of Cu2+ to Cu1+.73 The tiny
decrease of the Cu–O peak intensity during the oxidation
reaction, indicating the loss of Cu–O bonds, was probably due
to a small amount of the active oxygen (*O) being used to
oxidize ECH without being replaced by gas phase oxygen.
Moreover, no higher Cu–Cu shell peak was observed at 2.5 Å
for the htpCC20u sample, indicating that the Cu ions could
not be reduced to metallic Cu during the ECH oxidation, and
that two redox couples, i.e., Ce4+/Ce3+ and Cu2+/Cu+, coexisted
within the htpCCx catalysts during the catalysis.
Fig. 7 DRUV-Vis profiles of (a) htpCC10, (b) htpCC20, (c) htpCC30, (d) copCC, (e)
impCC, (f) tcbCC, and (g) bulk-CeO2.
octahedral Oh configuration, which is more or less tetragonally
distorted and the wavelength decreases as the distortion
increases. For square pyramidal or tetrahedral complexes, the
2d transitions are shifted towards higher wavelengths.69 The
above results indicate that the htpCC10 sample features
square pyramidal or tetrahedral copper complexes, while
htpCC20 and htpCC30 have copper species in octahedral Oh
configurations.
3.5. Catalytic activity and reaction by-product distribution
The Cu K-edge XANES spectrum of the fresh htpCC20
catalyst is compared with that of the CuO, Cu2O, and Cu foil
standards in Fig. 8. The different edge energies and edge
shapes of these standard materials provide important infor-
mation, which can be used to identify the oxidation state and
local structure of the Cu species. The weak pre-edge peak of
the CuO standard at 8.978 keV can be ascribed to the dipole-
forbidden electronic transition from 1s A 3d in Cu2+, and the
more intense peak between 8.982 and 8.987 keV can be
attributed to dipole-allowed 1s A 4p electron transitions in
Cu1+ and Cu2+, respectively.70,71 It can be observed that a small
edge transition peak centered at about 8.981 keV appears in
the spectrum of htpCC20, indicating the existence of Cu1+
species. Additionally, the absence of the edge transition peaks
(8.978 keV and 8.982–8.987 keV), the higher white line
intensity (8.992–9.007 keV), and the different shape of
htpCC20 XANES spectrum stems from the interaction between
the copper species and CeO2 because both the incorporation of
Cu into the CeO2 lattice and the strong interaction of isolated
Cu and CeO2 can lead to different electronic environments and
structural properties.72 The inset of Fig. 8 shows the k3-
The catalytic behavior of the synthesized catalysts for ECH
oxidation was evaluated, as shown in Fig. 9. In the temperature
range of our experiments no thermal or homogeneous ECH
destruction could be observed during the blank test. The
CuOx–CeO2 catalysts have better catalytic activity for the ECH
oxidation than bulk-CeO2 and bulk-CuO. All of the htpCCx
samples possess much higher low-temperature activities than
those synthesized via other traditional protocols in this work.
In particular, htpCC20 has the highest activity with the
complete combustion of ECH achieved at ca. 204 uC, which
is a much lower temperature than the other catalysts (Fig. 9A
and Fig. S3, ESI ). Fig. S4 (ESI ) shows T10, T50, and T90 (the
3
3
temperatures needed for the 10%, 50%, and 90% conversion
of ECH) as a function of the catalyst. The temperature decrease
at 50% ECH conversion for htpCC20 (T50 = 143 uC) is about 135
uC when compared with bulk-CeO2 (T50 = 278 uC). The catalytic
activity sequence of all of the prepared samples is as follows:
htpCC20 . htpCC30 . htpCC10 . copCC . impCC & tcbCC
. bulk-CuO . bulk-CeO2. Fig. 9B shows Arrhenius plots for
the ECH oxidation under different conditions (ECH conversion
, 10%). The reaction rates (r) and activation energies (Ea) were
19648 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
that of htpCC10 (r = 93.9 mmol g21 h), tcbCC (r = 12.4 mmol g21
h), and bulk-CuO (r = 7.2 mmol g21 h), respectively. Moreover,
htpCC20 has the lowest Ea value (Ea = 113 kJ mol21) when
compared to the other samples (Ea = 141.2–246.5 kJ mol21),
which also indicates its superior low-temperature activity. The
distribution of the reaction by-products and CO2 selectivity
using various catalysts were further investigated, as shown in
Table 3 and Fig. 10. It was found that the primary reaction
products were CO2, H2O, Cl2, HCl, and a trace of CO. The
organic chlorinated compounds could not be detected during
the ECH oxidation under the reaction conditions, while a
small amount of propanal and propanoic acid were observed
as reaction intermediates with some of the prepared catalysts.
It should be noted that there are no reaction by-products when
htpCC20 and htpCC30 are used, and that the by-product
concentration is directly proportion to the catalytic activity. In
addition, the CO2 selectivities listed in Table 3 prove that
htpCC20 is a powerful catalyst for the oxidation of ECH at low
temperature (S50 . 99%). The reaction determining factor and
related ECH oxidation processes for the CuOx–CeO2 catalysts
will be further discussed in the following section.
Some researchers have proposed that hydrocarbons could
be used to remove the Cl species from the surface of Pt loaded
catalysts and thus enhance the catalytic performance.74 The
effect of an additional hydrocarbon (hexane (HE) is selected as
a typical VOC) on the activity of the most active catalyst was
investigated in order to reveal the applicability of the related
promotion effect in our work (Fig. 11 and Fig. S5, ESI ). For
3
comparison purposes, the mutual effect of using ECH and HE
over pure bulk-CeO2 was also studied. The addition of HE has
an obvious positive influence on ECH oxidation with the T50
temperature decreasing from 279 to 242 uC accordingly (Fig.
S5, ESI ). This promotion effect may be explained as follows:
3
Fig. 9 (A) The r–T curves of ECH oxidation using the synthesized catalysts; (B)
Arrhenius plot of ECH oxidation reaction (300 mg catalyst; 300 ppm ECH; 21%
O2 balanced with N2; GHSV = 18 400 h21; reaction temperature: 120–195 uC).
an H-atom is abstracted from an HE molecule, which can then
react with the strongly adsorbed Cl species on the surface of
bulk-CeO2 to form HCl, resulting in the recovery of some of the
active sites for ECH oxidation. For htpCC20, the presence of
HE has negligible influence on ECH conversion at low reaction
temperatures (,150 uC), while the catalytic activity decreases
at higher reaction temperatures with the T90 temperature
increasing by about 17 uC. This decrease in the catalytic
activity suggests that the consumption of surface oxygen
during HE oxidation decreases the oxygen species needed to
estimated according to the slopes of the Arrhenius plots and
are listed in Table 3. As can be seen, the htpCC20 sample
possesses much higher ECH catalytic activity at low tempera-
ture. For instance, the reaction rate (r) of htpCC20 at 165 uC (r
= 413.3 mmol g21 h) is 4.04, 33.3, and 57.4 times greater than
Table 3 The catalytic performance and representative results of the kinetic measurements
rc (mmol g21 h)
Ea (kJ mol21
d
)
R2
a
a
a
b
Sample
T10 (uC)
T50 (uC)
T90 (uC)
S50 (%)
htpCC10
htpCC20
htpCC30
copCC
impCC
tcbCC
146
126
135
141
152
183
213
198
191
143
144
198
208
232
278
237
215
174
195
225
240
—
98.1
99.4
99.2
97.4
92.3
89.9
81.7
86.4
93.9
413.3
351.7
120.7
66.1
12.4
9.4
170.1
113
141.2
159.3
201
205.5
189.3
246.5
0.975
0.986
0.927
0.968
0.982
0.998
0.988
0.976
bulk-CeO2
bulk-CuO
—
—
7.2
a
b
Temperatures at which 10%, 50% and 90% conversion of ECH occurred. CO2 selectivity at which 50% conversion of ECH occurred:
mole number of CO2 produced
S50
~
|100%. c Reaction rate achieved at 165 uC. d Apparent activation energy obtained from the Arrhenius plot.
3|(mole number of ECH fed)
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19649

View Article Online
Paper
RSC Advances
Fig. 11 The activity of htpCC20 and bulk-CeO2 (300 mg catalyst; 300 ppm ECH;
1000 ppm n-hexane; 21% O2 balanced with N2; GHSV = 18 400 h21).
enhanced catalytic performance.75 Herein, the role of each of
these features is further elucidated on the basis of the
experimental results.
4.1. Oxygen vacancies and the mobility of active oxygen in the
composite oxides
The formation of oxygen vacancies has an important influence
on catalytic activity. These defective sites can act as active
centers in the catalytic reaction because the oxygen molecules
of the defective oxides tend to be readily released and
transferred, which can enhance the reactivity of the catalyst
for HVOC oxidation.76 The Ce 3d XPS spectra demonstrate that
the htpCCx samples have larger numbers of Ce3+ ions,
compared to copCC, impCC, and tcbCC (Table 2 and Fig. 5),
indicating that there are more oxygen vacancies in the htpCCx
catalysts. The oxygen vacancies in htpCCx as well as bulk-CeO2
were further investigated using Raman spectroscopy, as shown
Fig. 10 Primary reaction product distribution: (A) propanal, and (B) propanoic
acid.
remove the adsorbed Cl species, which blocks the active sites
and probably deactivates the catalyst.
4. Discussion
It is generally accepted that there are many factors that
influence the catalytic performance of a transition-metal oxide,
including the specific surface area, pore structure, dispersion
of the active phase, number of oxygen vacancies, and mobility
of the active oxygen. Table 1 lists the related textural data of all
of the synthesized catalysts. No relationship between SBET, Dv,
Dp and ECH oxidation is observed, indicating that both the
surface area and pore structure are not key factors governing
the catalytic performance, although the 3D mesoporous
structure of the htpCCx samples could be favorable for mass
and thermal transfer. It is known that the presence of oxygen
vacancies favors the activation of oxygen molecules to give
active oxygen adspecies, and that a good reducibility or oxygen
mobility can provide a facile redox process that leads to
Fig. 12 Raman spectra of (a) htpCC10, (b) htpCC20, (c) htpCC30, and (d) bulk-
CeO2 (inset table: integrated areas for the peaks at 601 cm21).
19650 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
O2(ad) A O22(ad) A O2(ad) A O22(lattice). The physically
2
adsorbed oxygen O2(ad) and chemically adsorbed oxygen O2
/
O2(ad) species are much easier to desorb than the lattice O22
species.50 The desorption of physically adsorbed oxygen is not
observed for all of the catalysts. Bulk-CeO2 shows two
desorption peaks centered at 301 and 576 uC, assigned to the
chemically adsorbed oxygen and lattice oxygen species,
respectively, while there is no obvious oxygen species
desorption from the bulk-CuO sample (Fig. 13g and h). The
CuOx–CeO2 catalysts possess two to four oxygen desorption
peaks. It is worth noting that the CuOx–CeO2 samples (except
impCC) have oxygen species desorption peaks due to
chemically adsorbed molecular oxygen, O22(ad), at very low
temperatures (,220 uC), while no such desorption peak
around this temperature is found for the pure bulk-CuO and
bulk-CeO2. Moreover, the htpCCx samples possess large
quantities of chemically adsorbed oxygen, and the total
amount of chemically adsorbed oxygen (O22/O2(ad)) species
in the present work is in the order htpCC20 . htpCC30 .
htpCC10 . copCC . tcbCC # impCC . bulk-CeO2 & bulk-
CuO according to the deconvolution peak areas under the O2–
TPD curves below ca. 500 uC. It can be inferred that the
desorbed oxygen species below 220 uC should be related to the
interaction between the CuO and CeO2 oxides. Ce 3d XPS and
Raman spectra (Fig. 5 and Fig. 12) revealed that surface oxygen
vacancies are always present in the CuOx–CeO2 oxides, which
can be centers for oxygen activation. Oxygen molecules (O2)
from the gas phase can adsorb into the oxygen vacancies
under the reaction conditions and be activated to form O22 or
O2 species, which are easily desorbed below 500 uC81 and are
important for the ECH oxidation reaction.
Fig. 13 O2–TPD patterns of (a) htpCC10, (b) htpCC20, (c) htpCC30, (d) copCC, (e)
impCC, (f) tcbCC, (g) bulk-CeO2, and (h) bulk-CuO.
in Fig. 12. Pure bulk-CeO2 shows one sharp peak centered at
460 cm21, characteristic of the Raman shift of a fluorite-type
structure.77 The Raman spectra of bulk-CuO can be found in
4.2. Redox properties and durability of the synthesized CuOx–
CeO2 catalysts
the literature, with three peaks at 295, 340, and 626 cm .
21 78 All
of the htpCCx samples exhibit the peak corresponding to bulk-
CeO2, which is shifted to lower frequencies as a result of
changes to the cell parameters and the formation of oxygen
vacancies in the lattice.79 The oxygen vacancy concentration in
pure CeO2 and other CuO–CeO2 catalysts has been indirectly
The redox characteristics of the metal oxides (especially the
active phase reducibility) are critical for the catalytic perfor-
mance, according to the Mars–van Krevelen mechanism,
which has been proven to be suitable for describing the VOC
oxidation over CuO–CeO2 type catalysts.82,83 According to the
results of H2–TPR and in situ XANES, Menon and coworkers28
proposed that both Cu2+ and Ce4+ can be reduced during the
reaction. A similar result was also found by Martinez-Arias
et al. for CO/O2 redox over a CuO–CeO2 catalyst.84 It is well-
known that additional elements incorporated into a metal
oxide can weaken the bonds between the metal and oxygen
atoms, making the latter more reactive and mobile.20 This
effect will be even stronger if the added element has a lower
valence than the cation it replaces (such as Cu2+ and Ce4+). The
results of XRD, XPS and Raman spectroscopy have demon-
strated that a portion of the Cu2+ ions (particularly in htpCCx)
are incorporated into the CeO2 lattice probably in the form of a
CuxCe12xO22d solid solution (Fig. 2, Fig. 5, and Fig. 12). The
resulting CuxCe12xO22d solid solution exhibits a lower Cu
redox potential, i.e., an enhanced reducibility, due to the Cu2+–
O22–Ce4+ connection, and the vacancy can be considered as a
bridge for the transfer of O from Ce to Cu. Likewise, the
reduction process of CeO2 is also shifted towards lower
temperatures when Ce4+ ions are substituted with Cu2+ ions.
The H2–TPR results in this work (Fig. 6) revealed that all of the
measured in the literature to be in the region of 585 cm21 52,80
.
The relevant peaks of the htpCCx catalysts were deconvoluted
using a Gaussian method, and the corresponding peak areas
are shown in Fig. 12 (see the inset table). It can be seen that
the oxygen vacancy concentration follows the order htpCC20 .
htpCC30 . htpCC10 & bulk-CeO2, which correlates well with
the catalytic performance (Table 3, Fig. 9, and Fig. 10). It is
known that the higher the number of oxygen vacancies, the
higher number of Cu species incorporated into the ceria
lattice, which causes distortion and defective sites in the
structure, according to the following equation:
xCuO + (1 2 x)CeO2 « x*% +
2+
4+
(2 2 x)O22 + xCu2Ce + (1 2 x)Ce2Ce
(1)
where *% is an oxygen vacancy.
O2–TPD is an effective method to determine the mobility of
oxygen species, and the results for all of the synthesized
catalysts are shown in Fig. 13. Generally, the adsorbed (ad)
oxygen species changes according to the following sequence:
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19651

View Article Online
Paper
RSC Advances
CuOx–CeO2 mixed oxides can be reduced at lower tempera-
tures compared to bulk-CeO2 and bulk-CuO. In addition, the
htpCCx samples possess higher reducibilities than those of
the other catalysts, which is in good agreement with the O2–
TPD results (Fig. 13). Furthermore, the reduction temperature
of copCC is much lower than that of impCC and tcbCC, which
is probably due to the aggregation of CuO particles (mono-
crystalline CuO) on the surfaces of impCC and tcbCC (proven
using WA-XRD, Fig. 2B). Huang and coworkers have proposed
that the well dispersed Cu could be connected to CeO2 via a so-
called nested oxygen ion, which has a weakened CuLO bond,
thus allowing for easy reduction.85 EXAFS spectroscopy of
CuxCe12xO2 catalysts and the density functional theory (DFT)
results of a Cu0.25Ce0.75O2 catalyst from Hegde and cow-
orkers22 have revealed the structure and chemical environ-
ment of Cu species in CeO2. They found that there are six O
atoms around each incorporated Cu2+, the nearest in the first
shell at a distance of 1.92–1.95 Å. The rearrangement of the
CeO2 lattice around incorporated Cu2+ induces distortion and
vacancies, leading to a weak Fourier transformed signal from
the medium range shells.22 Likewise, the Cu2+ of
CuxCe12xO22d in the present samples would weaken the
surface oxygen bonds of the solid solution, making them
more reactive and reducible (Table 2 and Fig. 6). The
monocrystalline CuO particles on the impCC and tcbCC
catalysts are reduced at relatively high temperatures (see
Fig. 6 for the H2–TPR results), indicating that the monocrystal-
line CuO phase will only be active at higher reaction
temperatures. This result suggests that the inferior catalytic
performance of impCC and tcbCC can be partly ascribed to the
inertia properties of crystalline CuO at low temperatures and
the suppressive effect of monocrystalline CuO towards the
surface active species.61
was carried out by feeding a stream of gas containing 300 ppm
of ECH and 21% O2 at reaction temperatures of 225 and 300 uC
(Fig. 14). As shown in Fig. 14, during the first 10 h, the ECH
conversion over tcbCC and pure bulk-CeO2 falls from 41% and
27% to about 32% and 11%, respectively, due to the strong
adsorption of HCl or Cl2 produced from the decomposition of
ECH on the CeO2 surface.9 However, the original activities of
tcbCC and bulk-CeO2 are rather low at 225 uC. An obvious
activity drop during the first 15 h was noticed for the copCC
and impCC catalysts, decreasing from 95% and 80% to 69%
and 52%, respectively. ECH conversion over htpCC20 also
decreases from 100% to 92% during the first 3 h, but this
remains at about 91% for the rest of the reaction time. The
lower activity loss (,10%) and faster ECH conversion
equilibrium period (,3 h) of htpCC20 indicate that the 3D
ordered mesostructures of the htpCCx samples evidently
enhance the catalysts’ durability. Wang et al. found that
higher reaction temperatures have a significant positive effect
on the removal of surface Cl species over MnOx–CeO2 catalysts
in the combustion of chlorobenzene.1,9 Therefore, the dur-
ability of the catalysts with higher catalytic performance (i.e.,
htpCC20 and copCC) was further evaluated at a reaction
temperature of 300 uC. For copCC, stable activity can be
achieved within 6 h, and less activity reduction (about 17%) is
also observed. Noticeably, ECH conversion over htpCC20, is
maintained at almost 98% or higher for 50 h. In addition, the
ordered mesoporous structure of the used htpCC20 catalyst,
shown in Fig. 3i, further confirms the high stability of the
mesostructured copper and ceria composite metal oxide.
Analysis with GC and online MS shows that the htpCC20
catalyst under study demonstrates more than 99.5% selectivity
for CO2 with no other C-containing by-products detected. Zhou
et al. demonstrated that the interaction of CeO2 and CuO can
improve the stability of a CuO–CeO2–USY catalyst used in the
oxidation of 1,2-chloroethane, and that the abundant oxygen
species inhibit the production of coke during long time
reaction.15 In the oxidation of ECH, ECH was first adsorbed on
to the surface of the catalysts, then the C–Cl bond dissociation
occurs, followed by the reaction of the dissociated fragments
and active oxygen to produce COx and H2O. Chlorine species,
which can block the active sites and deactivate the catalysts,
need to be removed from the catalyst surface in the form of Cl2
or HCl. Highly dispersed CuO nanoparticles over the surface of
CeO2 can offer plenty of active sites for ECH oxidation.
Moreover, the incorporation of Cu2+ into the lattice of CeO2
results in a large number of oxygen vacancies in the CuOx–
CeO2 catalysts, especially the htpCCx samples, and hence
creates large quantities of active oxygen species (O22/O2) that
can efficiently react with the adsorbed chlorine species. As
shown in Fig. 14, the htpCC20 catalyst possesses good
durability for ECH combustion, suggesting that the removal
of Cl species is rapid enough to maintain the active sites that
are accessible to reactant molecules at 300 uC.
In general, catalyst deactivation is
a hurdle in the
commercial application of the catalytic elimination of
HVOCs. The catalytic stability during the decomposition of
ECH, as a function of the conversion versus time-on-stream,
4.3. Proposed reaction process and mechanism for ECH
oxidation
The results derived from XRD, H2–TPR, XPS and Raman
spectroscopy (Fig. 2, Fig. 5, Fig. 6, and Fig. 12) reveal that a
large amount of Cu2+ species migrated into the CeO2 lattice of
Fig. 14 The conversion of ECH over htpCC20, copCC, impCC, tcbCC, and bulk-
CeO2 as a function of time-on-stream at various reaction temperatures (300 mg
catalyst; 300 ppm ECH; 21% O2, balanced with N2; GHSV = 18 400 h21).
19652 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Scheme 1 The formation of oxygen vacancies and proposed mechanism for ECH oxidation over the htpCCx catalysts.
the htpCCx catalysts in the form of a CuxCe12xO22d solid
solution. A schematic of the crystal structure of Cu substituted
CeO2 can be found in Scheme 1 (left image). The Cu2+–O22Ce4+
connection in the CuxCe12xO22d solid solution reduced the
redox potential of Cu species as well as acting as a bridge for O
transfer between Cu and Ce. Thus, the interaction between Cu
and Ce in the solid solution ensures that the reducibility is
adsorbed Cl species (eqn (3)). Simultaneously, a portion of the
HCl produced returns to the catalyst surface and is adsorbed
into the oxygen vacancies or some of the active sites for the
oxygen transformation (eqn (4)). After this, the reduced Cu
sites can be recovered (Cu2+) via migration of the adsorbed O2/
O22 species or lattice oxygen from CeO2 grains (eqn (5) and
(6)). The role of Ce4+ in the ECH oxidation is similar to that of
Cu2+ in the well-known Wacker reaction for the direct
oxidation of alkenes to aldehydes or ketones, which is
catalyzed by CuCl2 and PdCl2. In the Wacker reaction, the
reduced Pd0 is re-oxidized by Cu2+ regenerating Pd2+, i.e.,
contact between Cu2+ and Pd0 is essential in order to close the
enhanced for both compounds, and this allows an effective
2e{
ꢀ
ꢀ
redox cycle (i.e., Cu2z Cuz
Ce4z Ce3z) to occur
O2{ O{
=
during oxidation reactions. The proposed mechanism for low
temperature ECH oxidation with the htpCCx catalysts is shown as
a schematic in order to obtain deeper insight into this reaction
process (Scheme 1), and the primary elementary reaction routes
can be listed as follows:
catalytic cycle of Cu1+/Cu2+ P Pd0/Pd2+ 86 Reduced Ce3+ can be
.
re-oxidized by the reduction of oxygen at the Ce sites or by
CO2/H2O in the gas phase (eqn (7) and (8)).28 The reoxidized
Ce4+ sites can then act as oxygen suppliers at the ECH
oxidation sites for the re-oxidation of nearby reduced Cu1+. As
a consequence, the coexistence of Cu and Ce species is
essential to the reaction, maintaining the Cu2+/Cu1+ P Ce4+/
Ce3+ redox cycle according to the Mars–van Krevelen mechan-
ism.83 The consumed active oxygen species can be replenished
from gas phase oxygen on the surface of the CeO2 crystals,
probably around the oxygen vacancies (eqn (9)), and the
desorption of the adsorbed chlorine species (in the form of
Cl2) can be promoted by the active oxygen species (eqn (10)).
Some partial oxidation products (i.e., propanal and propanoic
acid) were formed during ECH combustion, especially over
tcbCC, bulk-CuO, and bulk-CeO2 (Fig. 10). It is known that C–
Cl bonds are weaker than C–H bonds, and hence, they are
more prone to attack from nucleophiles.87 Consequently, the
first step in the oxidation of ECH molecules is probably their
dissociative adsorption on to active Cu sites via Cl abstraction.
The adsorbed Cl (*Cl) in the tcbCC, bulk-CuO, and bulk-CeO2
samples may be difficult to remove until an equilibrium of
adsorbed and desorbed Cl species is achieved due to the
absence of oxygen vacancies and active oxygen species (Table 2
and Fig. 12). Therefore, the elementary reaction steps in eqn
(7), (9), and (10) were slower for these catalysts than for the
htpCCx catalysts, thus, resulting in the formation of small
ECH(g) + * A *ECH(ad)
(2)
*ECH(ad) + Cu2+ A Cu1+ + COx +
H2O(g) + HCl(g) + *Cl(ad)
(3)
(4)
HCl(g) + * A *Cl2(ad)
Ce4+ + Cu1+ A Cu2+ + Ce3+
Cu1+ + *O A Cu2+
(5)
(6)
Ce3+ + *O A Ce4+
(7)
Ce3+ + H2O/CO2 A Ce4+ + H2/CO
(8)
O2(g) + *% A *O
(9)
*O + *Cl2(ad) A Cl2(g) + *%
(10)
Firstly, ECH molecules in the gas phase must come into
contact and be adsorbed (ECH(ad)) into the active Cu2+ sites
(eqn (2)). ECH(ad) can then react with active oxygen species
from the Cu sites to form COx, H2O and HCl as well as
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19653

View Article Online
Paper
RSC Advances
amounts of organic reaction by-products. Obviously, the active
oxygen species, reducibility, and oxygen vacancies are impor-
tant for efficient ECH low-temperature combustion as well as
for *Cl species removal.
References
1 Y. Dai, X. Y. Wang, Q. G. Dai and D. Li, Appl. Catal., B, 2012,
111–112, 141–149.
2 S. L. Alderman, G. R. Farquar, E. D. Poliakoff and
B. Dellinger, Environ. Sci. Technol., 2005, 39, 7396–7401.
3 C. He, J. J. Li, P. Li, J. Cheng, Z. P. Hao and Z.-P. Xu, Appl.
Catal., B, 2010, 96, 466–475.
5. Conclusions
´
˜
4 G. Arzamendi, V. A. de la Pena O’Shea, M. C. Alvarez-
3D ordered mesoporous CuOx–CeO2 catalysts (htpCCx) were
prepared via a simple metal nitrate precursor host–guest
interaction with SBA-16 silica as a hard template. We propose
that the fast-microwave-assisted (FMA) method is a universal
approach for the quick removal (about 5 min) of the organic
template in the siliceous materials, and the employment of
this FMA technique in the synthesis of other mesostructured
metal oxides with widespread applications in energy and
environmental fields can be anticipated. The experimental
results show that Cu species are uniformly distributed over the
mesoporous CuOx–CeO2 samples, and a large number of Cu2+
ions are incorporated into the CeO2 lattice in the form of a
CuxCe12xO22d solid solution, leading to structural defects and
improved oxygen mobility. The interaction of Cu and Ce
increases the reducibility of the CuOx–CeO2 catalysts, and the
coexistence of Cu and Ce is required in the reaction to preserve
the two redox couples, i.e., Cu2+/Cu1+ and Ce4+/Ce3+. The
surface area and pore structure are not the key factors
governing the catalytic performance, although the 3D mesos-
tructure could be favorable for mass and thermal transfer. The
catalysts with 3D mesoporous structures have much higher
low-temperature activity for epichlorohydrin oxidation, e.g.,
the T50 temperature of htpCC20 decreases more than 130 uC
compared to bulk-CeO2. HtpCC20 has been shown to be the
most powerful catalyst for epichlorohydrin removal due to its
superior activity, selectivity, and stability. The better activity of
the htpCCx catalysts can be ascribed to the high dispersion of
the active phase, improved mobility of oxygen species, and
large amount of surface active oxygen. We can rationally expect
these novel materials to be effective catalysts for the elimina-
tion of HVOCs, while simultaneously having other potential
applications in catalysis, as well as separation and adsorption
processes.
´
´
Galvan, J. L. G. Fierro, P. L. Arias and L. M. Gandıa, J.
Catal., 2009, 261, 50–59.
5 H. F. Li, G. Z. Lu, Q. G. Dai, Y. Q. Wang, Y. Guo and Y.
L. Guo, Appl. Catal., B, 2011, 102, 475–483.
6 C. H. Cho and S. K. Ihm, Environ. Sci. Technol., 2002, 36,
1600–1606.
7 V. H. Vu, J. Belkouch, A. Ould-dris and B. Taouk, AIChE J.,
2008, 54, 1585–1591.
8 P. Zimmer, A. Tschope and R. Birringer, J. Catal., 2002, 205,
339–345.
9 X. Y. Wang, Q. Kang and D. Li, Appl. Catal., B, 2009, 86,
166–175.
10 J. Qi, J. Chen, G. D. Li, S. X. Li, Y. Gao and Z. Y. Tang,
Energy Environ. Sci., 2012, 5, 8937–8941.
11 H. F. Li, G. Z. Lu, Q. G. Dai, Y. Q. Wang, Y. Guo and Y.
L. Guo, ACS Appl. Mater. Interfaces, 2010, 2, 838–846.
12 Y. Fan, X. B. Lu, Y. W. Ni, H. J. Zhang, M. W. Zhu, Y. Li and
J. P. Chen, Appl. Catal., B, 2011, 101, 606–612.
13 G. J. Song, S. H. Kim, Y. C. Seo and S. C. Kim, Chemosphere,
2008, 71, 248–257.
14 T. Fujimori and M. Takaoka, Environ. Sci. Technol., 2009,
43, 2241–2246.
15 Q. Q. Huang, X. M. Xue and R. X. Zhou, J. Mol. Catal. A:
Chem., 2011, 344, 74–82.
16 B. Skarman, T. Nakayama, D. Grandjean, R. E. Benfield,
E. Olsson, K. Niihara and L. R. Wallenberg, Chem. Mater.,
2002, 14, 3686–3699.
17 G. Sedmak, S. Hocevar and J. Levec, J. Catal., 2003, 213,
135–150.
18 A. Tschope, J. Markmann, P. Zimmer and R. Birringer,
Chem. Mater., 2005, 17, 3935–3943.
19 Z. Wu, M. Li, J. Howe, H. M. Meyer and S. H. Overbury,
Langmuir, 2010, 26, 16595–16606.
¨
20 V. Shapovalov and H. Metiu, J. Catal., 2007, 245, 205–214.
21 P. G. Harrison, I. K. Ball, W. Azelee, W. Daniell and
D. Goldfarb, Chem. Mater., 2000, 12, 3715–3725.
22 P. Bera, K. R. Priolkar, P. R. Sarode, M. S. Hegde, S. Emura,
R. Kumashiro and N. P. Lalla, Chem. Mater., 2002, 14,
3591–3601.
Acknowledgements
23 D. Delimaris and T. Ioannides, Appl. Catal., B, 2009, 89,
295–302.
This work is financially supported by the National Natural
Science Foundation of China (21107106), the Strategic Priority
Research Program of the Chinese Academy of Sciences
(XDB05050500), the National Key Technology R&D Program
of the Ministry of Science and Technology (2011BAZ01960), the
National Basic Research Program of China (2010CB732300),
and the program of Innovation Team of the Bureau of Ningbo
Science and Technology (201182001). The authors acknowl-
edge Dr Lirong Zheng in Beijing Synchroton Radiation Facility
(BSRF, China) and Dr Dingliang Tang in Xiamen University for
the XAFS and XPS analysis, respectively.
24 P.-O. Larsson, A. Andersson, L. R. Wallenberg and
B. Svensson, J. Catal., 1996, 163, 279–293.
25 V. D. Arau´jo, J. D. A. Bellido, M. I. B. Bernardi, J. M. Assaf
and E. M. Assaf, Int. J. Hydrogen Energy, 2012, 37,
5498–5507.
26 P. Ratnasamy, D. Srinivas, C. V. V. Satyanarayana,
P. Manikandan, R. S. S. Kumaran, M. Sachin and V.
N. Shetti, J. Catal., 2004, 221, 455–465.
27 G. Avgouropoulos and T. Ioannides, Appl. Catal., A, 2003,
244, 155–167.
19654 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
´
´
´
´
28 U. Menon, H. Poelman, V. Bliznuk, V. V. Galvita,
D. Poelman and G. B. Marin, J. Catal., 2012, 295, 91–103.
29 S. M. Saqer, D. I. Kondarides and X. E. Verykios, Appl.
Catal., B, 2011, 103, 275–286.
55 A. Martınez-Arias, M. Fernandez-Garcıa, O. Galvez, J.
M. Coronado, J. A. Anderson, J. C. Conesa, J. Soria and
G. Munuera, J. Catal., 2000, 195, 207–216.
56 P. Bera, K. R. Priolkar, P. R. Sarode, M. S. Hegde, S. Emura,
R. Kumashiro and N. P. Lalla, Chem. Mater., 2002, 14,
3591–3601.
´
´
´
30 R. Lopez-Fonseca, J. I. Gutierrez-Oritz, M. A. Gutierrez-Oritz
´
and J. R. Gonzalez-Velasco, J. Catal., 2002, 209, 145–150.
31 Q. S. Huo, D. I. Margolese, U. Ciesla, P. Y. Feng, T. E. Gier,
P. Sieger, R. Leon, P. M. Petroff, F. Schu¨th and G. D. Stucky,
Nature, 1994, 368, 317–321.
57 W. J. Shan, W. J. Shen and C. Li, Chem. Mater., 2003, 15,
4761–4767.
58 K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R.
32 D. E. De Vos, M. Dams, B. F. Sels and P. A. Jacobs, Chem.
Rev., 2002, 102, 3615–3640.
´
A. Pierotti, J. Rouquerol and T. Siemieniewska, Pure Appl.
Chem., 1985, 57, 603–619.
33 Y. Ren, Z. Ma and P. G. Bruce, Chem. Soc. Rev., 2012, 41,
4909–4927.
59 C. Z. Yu, B. Z. Tian, J. Fan, G. D. Stucky and D. Y. Zhao, J.
Am. Chem. Soc., 2002, 124, 4556–4557.
34 H. Sirringhaus, T. Kawase, R. H. Friend, T. Shimoda,
M. Invasekaran, W. Wu and E. P. Woo, Science, 2000, 290,
2133–2137.
35 B. Z. Tian, X. Y. Liu, H. F. Yang, S. H. Xie, C. Z. Yu, B. Tu
and D. Y. Zhao, Adv. Mater., 2003, 15, 1370–1374.
36 H. O. Jacobs and G. M. Whitesides, Science, 2001, 291,
1763–1766.
37 C. He, Q. Li, P. Li, Y. F. Wang, X. Y. Zhang, J. Cheng and Z.
P. Hao, Chem. Eng. J., 2010, 162, 901–909.
38 T.-W. Kim, R. Ryoo, K. P. Gierszal, M. Jaroniec, L.
A. Solovyov, Y. Sakamoto and O. Terasaki, J. Mater.
Chem., 2005, 15, 1560–1571.
39 Y. Sakamoto, M. Kaneda, O. Terasaki, D. Y. Zhao, J.
M. Kim, G. D. Stucky, H. J. Shin and R. Ryoo, Nature, 2000,
408, 449–453.
40 C. Y. Ma, Z. Mu, C. He, P. Li, J. J. Li and Z. P. Hao, J.
Environ. Sci., 2011, 23, 2078–2086.
41 C. He, X. Y. Zhang, S. K. Gao, J. S. Chen and Z. P. Hao, J.
Ind. Eng. Chem., 2012, 18, 1598–1605.
42 K. K. Zhu, B. Yue, W. Z. Zhou and H. Y. He, Chem.
Commun., 2003, 98–99.
43 B. Z. Tian, X. Y. Liu, L. A. Solovyov, Z. Liu, H. F. Yang, Z.
D. Zhang, S. H. Xie, F. Q. Zhang, B. Tu, C. Z. Yu, O. Terasaki
and D. Y. Zhao, J. Am. Chem. Soc., 2004, 126, 865–875.
44 S. Kawi and M. W. Lai, Chem. Commun., 1998, 1407–1408.
45 B. Z. Tian, X. Y. Liu, C. Z. Yu, F. Gao, Q. Luo, S. H. Xie, B. Tu
and D. Y. Zhao, Chem. Commun., 2002, 1186–1187.
46 C. He, F. W. Zhang, L. Yue, X. S. Shang, J. S. Chen and Z.
P. Hao, Appl. Catal., B, 2012, 111–112, 46–57.
47 J. J. Li, X. Y. Xu, Z. Jiang, Z. P. Hao and C. Hu, Environ. Sci.
Technol., 2005, 39, 1319–1323.
48 C. He, P. Li, J. Cheng, H. L. Wang, J. J. Li, Q. Li and Z.
P. Hao, Appl. Catal., A, 2010, 382, 167–175.
49 C. Y. Ma, Z. Mu, J. J. Li, Y. G. Jin, J. Cheng, G. Q. Lu, Z.
P. Hao and S. Z. Qiao, J. Am. Chem. Soc., 2010, 132,
2608–2613.
60 W. Liu and M. Flytzani-Stephanopoulos, J. Catal., 1995,
153, 317–332.
61 B. Skårman, D. Grandjean, R. E. Benfield, A. Hinz,
A. Andersson and L. R. Wallenberg, J. Catal., 2002, 211,
119–133.
62 L. Qi, Q. Yu, Y. Dai, C. J. Tang, L. J. Liu, H. L. Zhang, F. Gao,
L. Dong and Y. Chen, Appl. Catal., B, 2012, 119–120,
308–320.
63 I. Platzman, R. Brener, H. Haick and R. Tannenbaum, J.
Phys. Chem. C, 2008, 112, 1101–1108.
64 F. Larachi, J. Pierre, A. Adnot and A. Bernis, Appl. Surf. Sci.,
2002, 195, 236–250.
65 Z.-Q. Zou, M. Meng and Y. Q. Zha, J. Phys. Chem. C, 2010,
114, 468–477.
66 M. F. Luo, J. M. Ma, J. Q. Lu, Y. P. Song and Y. J. Wang, J.
Catal., 2007, 246, 52–59.
67 J. Wang, D. Tsai and T. Huang, J. Catal., 2002, 208,
370–380.
68 G. R. Rao and H. R. Sahu, Proc.–Indian Acad. Sci., Chem.
Sci., 2001, 113, 651–658.
69 H. Praliaud, S. Michailenko, Z. Chajar and Z. M. Primet,
Appl. Catal., B, 1998, 16, 359–374.
70 R. Zhang, J. T. Miller and C. D. Baertsch, J. Catal., 2012,
294, 69–78.
71 J. E. Penner-Hahn, Coord. Chem. Rev., 2005, 249, 161–177.
´
´
72 D. Gamarra, G. Munuera, A. B. Hungrıa, M. Fernandez-
´
Garcıa, J. C. Conesa, P. A. Midgley, J. C. Hanson, J.
´
´
A. Rodrıguez and A. Martınez-Arias, J. Phys. Chem. C, 2007,
111, 11026–11038.
73 J. Y. Kim, J. A. Rodriguez, J. C. Hanson, A. I. Frenkel and P.
L. Lee, J. Am. Chem. Soc., 2003, 125, 10684–10692.
74 R. W. van den brink, R. Louw and P. Mulder, Appl. Catal., B,
2000, 25, 229–237.
75 F. J. Shi, F. Wang, H. X. Dai, J. X. Dai, J. G. Deng, Y. X. Liu,
G. M. Bai, K. M. Ji and C. T. Au, Appl. Catal., A, 2012, 433,
206–213.
76 X. Chen, Y. F. Shen, S. L. Suib and C. L. O’Young, J. Catal.,
2001, 197, 292–302.
77 Z. Popovic, Z. Dohcevic-Mitrovic, M. Konstantinovic,
M. Scepanovic and J. Raman, Spectrosc., 2007, 38, 750–755.
50 P. Li, C. He, J. Cheng, C. Y. Ma, B. J. Dou and Z. P. Hao,
Appl. Catal., B, 2011, 101, 570–579.
51 J. M. Kim, Y. Sakamoto, Y. K. Hwang, Y.-U. Kwon,
O. Terasaki, S.-E. Park and G. D. Stucky, J. Phys. Chem. B,
2002, 106, 2552–2558.
´
78 A. Gurbani, J. L. Ayastuy, M. P. Gonzalez-Marcos, J.
´
52 A. Gurbani, J. L. Ayastuy, M. P. Gonzalez-Marcos and M.
´
E. Herrero, J. M. Guil and M. A. Guttierrez-Ortiz, Int. J.
´
A. Gutierrez-Ortiz, Int. J. Hydrogen Energy, 2010, 35,
Hydrogen Energy, 2009, 34, 547–553.
79 W. J. Shan, Z. C. Feng, Z. L. Li, Z. Jing, W. J. Shen and
L. Can, J. Catal., 2004, 228, 206–217.
80 G. Marban and A. B. Fuertes, Appl. Catal., B, 2005, 57,
43–53.
11582–11590.
53 H. F. Yang and D. Y. Zhao, J. Mater. Chem., 2005, 15,
1217–1231.
54 P. Bera, S. T. Aruna, K. C. Patil and M. S. Hegde, J. Catal.,
1999, 186, 36–44.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 19639–19656 | 19655

View Article Online
Paper
RSC Advances
81 H. C. Liu, A. I. Kozlov, A. P. Kozlova, T. Shido, K. Asakura
and Y. Iwasawa, J. Catal., 1999, 185, 252–264.
82 V. Balcaen, R. Roelant, H. Poelman, D. Poelman and G.
B. Marin, Catal. Today, 2010, 157, 49–54.
83 U. Menon, V. V. Galvita and G. B. Martin, J. Catal., 2011,
283, 1–9.
85 W.-P. Dow, Y.-P. Wang and T.-J. Huang, Appl. Catal., A,
2000, 190, 25–34.
86 R. H. Nibbelke, M. A. J. Campman, J. H. B. J. Hoebink and
G. B. Marin, J. Catal., 1997, 171, 258–373.
87 J. Lichtenberger and M. D. Amiridis, J. Catal., 2004, 223,
296–308.
84 A. Martinez-Arias, M. Fernandez-Garcia, O. Galvez, J.
M. Coronado, J. A. Anderson, J. C. Conesa, J. Soria and
G. Munuera, J. Catal., 2000, 195, 207–216.
19656 | RSC Adv., 2013, 3, 19639–19656
This journal is ß The Royal Society of Chemistry 2013