Synthesis, characterization and catalytic performances of Cu- and Mn-containing ordered mesoporous carbons for the selective catalytic reduction of NO with NH3

Article abstract of DOI:10.1039/c4cy01221f

A series of copper- and manganese-containing ordered mesoporous carbons (OMCs) were synthesized for the low-temperature selective catalytic reduction (SCR) of NO with NH₃. The results demonstrated that NO conversion can be greatly improved by the addition of Cu or Mn. In addition, Cu and Mn had a synergetic effect and the optimal catalytic performance was obtained over the bimetallic catalyst Cu5Mn5-OMC. N₂ sorption, small-angle X-ray diffraction (XRD) and transmission electron microscopy (TEM) were used to confirm the ordered mesoporous structures. Wide-angle XRD results showed that in the bimetallic catalysts some Cu ions were substituted by Mn ions, resulting in lattice contraction by the formation of a solid solution. X-ray photoelectron spectroscopy (XPS) results indicated that the O-C=O groups and chemisorbed oxygen species increased with the addition of Cu or Mn. Temperature-programmed desorption of CO/CO₂ and NH₃ (CO/CO₂-TPD and NH₃-TPD) results verified that the acidity of the catalyst was enhanced by the addition of Cu and Mn and the bimetallic catalyst Cu5Mn5-OMC had the strongest acidity. Temperature-programmed reduction of H₂ (H₂-TPR) results showed that Cu5Mn5-OMC had the strongest oxidizing capacity. Based on the above analysis, a possible synergetic catalytic effect was proposed through the electron transfer between Cu and Mn ions.

Full text of DOI:10.1039/c4cy01221f

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: F. F. Cao, J. H.
Chen, C. L. Lyu, M. J. Ni, X. Gao and K. Cen, Catal. Sci. Technol., 2014, DOI: 10.1039/C4CY01221F.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/catalysis

Page 1 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
Synthesis, characterization and catalytic performances of Cu and Mn
containing ordered mesoporous carbon for the selective catalytic
reduction of NO with NH3
*
Feifei Cao ┼, Jinghuan Chen ┼, Changlei Lyu ╪, Mingjiang Ni ┼, Xiang Gao ┼, , Kefa Cen ┼
┼ State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou 310027,
Zhejiang, China
╪ Department of control science and engineering, Zhejiang University, Hangzhou 310027,
Zhejiang, China
Abstract: A series of copper and manganese containing ordered mesoporous carbons
(OMCs) were synthesized for the low temperature selective catalytic reduction (SCR)
of NO with NH3. The results demonstrated that NO conversion can be greatly
improved by the addition of Cu or Mn. In addition, Cu and Mn had a synergetic effect
and the optimal catalytic performance was obtained over the bimetallic catalyst
Cu5Mn5ꢀOMC. N2 sorption, smallꢀangle Xꢀray diffraction (XRD) and transmission
electron microscopy (TEM) were used to confirm the ordered mesoporous structures.
Wideꢀangle XRD results showed that in the bimetallic catalysts a part of Cu ions were
substituted by Mn ions resulting in the lattice contraction by the formation of solid
solution. Xꢀray photoelectron spectroscopy (XPS) results indicated that the OꢀC=O
groups and chemisorbed oxygen species increased with the addition of Cu or Mn.
Temperature programmed desorption of CO/CO2 and NH3 (CO/CO2ꢀTPD and
*
To whom correspondence should be addressed. Fax: (86)571ꢀ87951335. Eꢀmail: xgao1@
zju.edu.cn

Catalysis Science & Technology
Page 2 of 47
View Article Online
DOI: 10.1039/C4CY01221F
NH3ꢀTPD) results verified that the acidity of the catalyst was enhanced by the
addition of Cu and Mn and the bimetallic catalyst Cu5Mn5ꢀOMC had the strongest
acidity. Temperature programmed reduction of H2 (H2ꢀTPR) results showed that
Cu5Mn5ꢀOMC had the strongest oxidizing capacity. Based on the above analysis a
possible synergetic catalytic effect was proposed through the electron transfer
between Cu and Mn ions.
Keywords: Ordered mesoporous carbon (OMC), Cu and Mn, SCR, Synergetic effect.

Page 3 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
1. Introduction
Nitrogen oxides (NOx) are manꢀmade toxic gases formed during the combustion
of fuels. The main component is NO (>90%), and NO2 generally constitutes the most
of the remainder. They greatly contribute to the formation of photochemical smog,
acid rain, ozone depletion, and other air pollutants. The proven technology for
postꢀcombustion removal NO is the selective catalytic reduction (SCR) of NOx with
NH3.
4NO + 4NH3 + O2 → 4N2+ 6H2O
Currently, the most widely used catalysts in the SCR process are V2O5ꢀWO3/TiO2,
which have optimum reaction temperatures ranging within 350 – 450 °C. However,
some inevitable disadvantages still exist, such as the toxicity of VOx, the relatively
high operating temperature, the poisoning caused by high concentration of dust and
SO2 due to the SCR reactor located upstream before the dust removal or desulfurizer
device. Therefore, there is an urgent need for the development of highly active
catalysts for low temperature SCR which could be located downstream after the dust
removal or desulfurizer devices. Many catalysts were reported to be active for the
SCR at low temperatures, especially the carbonaceous materials due to their high
specific surface area and good antiꢀdeactivation, including activated carbon (AC),1ꢀ4
activated carbon fiber (ACF),5, 6 carbon nanotube (CNT),7, 8 and graphite,3 etc.
Owing to the wellꢀdeveloped structural ordering and wellꢀdefined pore size
distributions, ordered mesoporous carbon (OMC) has been widely applied in catalysis,
electrodes, and separations.9ꢀ11 To the best of our knowledge, OMC has not been used

Catalysis Science & Technology
Page 4 of 47
View Article Online
DOI: 10.1039/C4CY01221F
for SCR of NO with NH3. In our previous study OMC exhibited excellent NO
adsorption capacity compared with AC and its NO adsorption capacity could be
further improved by the introduction of cerium.12 It has been reported that the
adsorption of NO was important for the SCR process, and thus OMC may own
advantages over AC for the SCR reaction. Moreover, introducing active metal species
into OMC can effectively modify its property, such as FeꢀOMC for catalytic oxidation
of phenol,13 and CuꢀOMC for catalytic higher activity toward cinnamyl alcohol.14
According to literatures, transitional metal oxides have good low temperature SCR
activity, including MnOx,15,16 VOx,7 CuOx,3 FeOx,17 and so on. Moreover, bimetallic
catalysts usually owned superior property in catalytic performance, such as
MnCeOx18~21, MnCoOx22, CuCeOx23, etc. Hence, introducing transitional metal into
OMC would be an effective method to improve its catalytic performance.
In this study, a series of Cu and Mn containing OMCs were prepared using
solvent evaporationꢀinduced selfꢀassembly (EISA) method. Their characteristics were
investigated using N2 sorption, Xꢀray diffraction (XRD), Transmission electron
microscopy (TEM), Xꢀray photoelectron spectroscopy (XPS), temperature
programmed desorption of CO/CO2, NH3, and NO (CO/CO2ꢀTPD, NH3ꢀTPD and
NOꢀTPD), and temperature programmed reduction of H2 (H2ꢀTPR). Their catalytic
activities for SCR reaction were tested. It was found that Cu and Mn had a synergetic
effect on the NO conversion with NH3. A possible synergetic mechanism based on the
electron transfer between Cu and Mn ions was proposed.
2. Experimental

Page 5 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
2.1 Chemicals
Triblock copolymers Pluronic F127 (EO106ꢀPO70ꢀEO106, Mw = 12600) was
purchased from SigmaꢀAldrich Corp. The other chemicals were purchased from
Shanghai Chemical Corp. All the chemicals were used without any further
purification. Water used in the catalyst preparation was deionized.
2.2 Catalyst preparation
OMC was synthesized according to the procedure reported in our previous
study.12 In a typical process, phenol was mixed with formaldehyde to obtain resol
after the pH value of the mixture was adjusted to 7.0. A certain amount of F127 was
dissolved in ethanol and mixed with tetraethyl orthosilicate (TEOS), Cu nitrate, Mn
nitrate, and 25 wt.% resol ethanolic solution. Then the mixture was transferred into a
dish to evaporate ethanol after being stirred for 2 h. Afterward, it was dried at 100 °C
for 24 h followed by a calcination from room temperature to 850 °C under inert
atmosphere. The final products were obtained after washing and drying. The catalysts
are denoted as CuxMnyꢀOMC, where x and y represent the mass percentage of Cu and
Mn in OMC.
2.3 Characterization
Nitrogen sorption isotherms were obtained at liquid nitrogen temperature (77 K)
on a Micromeritics ASAP 2020 apparatus. Prior to N2 adsorption, the samples were
degassed at 250 °C for 3 h. The specific surface areas were calculated with the
BrunauerꢀEmmettꢀTeller (BET) equation, and pore size distributions were determined

Catalysis Science & Technology
Page 6 of 47
View Article Online
DOI: 10.1039/C4CY01221F
using the data of the desorption branch of the isotherms by the BarettꢀJoynerꢀHalenda
(BJH) method.
The power XRD patterns were recorded with Cu Kα radiation. The crystal phases
of the catalysts were analyzed using wideꢀangle XRD, which were collected from 10°
to 80° of the scattering angles (2θ) with a step length of 0.0167°. Crystalline phases
were identified by comparison with the reference data from International Center for
Diffraction Data (ICDD) files. Smallꢀangle XRD was also performed for 2θ between
0.5° and 10° to certify the ordered mesoporous structures.
TEM images were taken by a Tecnai electron microscope (SIRON) at 200 kV. A
suspension of the samples in ethanol was deposited on a grid and airꢀdried. Scanning
TEM (STEM) images were also taken to observe their morphology. Energy dispersive
Xꢀray spectroscopy (EDX) was used to ascertain the elemental contents.
XPS was recorded with a Thermo ESCALAB 250 electron spectrometer equipped
with an Al Kα Xꢀray source (hν=1486.6 eV) and a hemispherical electron analyzer
being operated at a constant pass energy (20 eV). The binding energy of C 1s which
have been adjusted to 284.6 eV was taken as the reference. According to literature24
the peaks were deconvoluted after subtraction of the Shirley background using a sum
of Lorentzian/Gaussian functions.
CO/CO2ꢀTPD was done to study the surface chemical properties of the catalysts.
A sample of 20 mg was packed into a quart reactor and prior to the measurement, the
sample was heated in a Ar flow (30 cm3/min) at 100 °C for 1 h. After cooling to 50 °C,
the temperature was raised to 900 °C at a rate of 10 °C /min under the same argon

Page 7 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
flow. CO and CO2 in the reactor outlet were detected using a quadrupolar mass
spectrometer. A thermal conductivity detector (TCD) was used for the
semiquantitative determination.
NH3ꢀTPD data were collected on an automated catalyst characterization system
(Micromeritics, model AutoChem II 2920). Each sample (20 mg) was pretreated
through heating at 200 °C in an ultraꢀhighꢀpure He stream (30 cm3/min) with 1 h.
After that the temperature was cooled to 100 °C and the samples were then saturated
with NH3 at a flow rate of 40 cm3/min for 1.5 h. Afterwards, the sample was flushed
with He for 1 h to remove the weakly bound NH3 followed by increasing the
temperature to 600 °C at a rate of 10 °C/min.
H2ꢀTPR of the various catalysts was performed on the same automated catalyst
characterization system with NH3ꢀTPD. The experiments were carried out at a heating
rate of 10 °C/min. The reactive gas composition was 10% H2 in argon with a flow rate
of 30 cm3/min. The amount of H2 consumed by the catalyst in a given temperature
range was calculated by the integration of the corresponding TCD signal intensities.
Prior to the analysis, 20 mg of the catalysts were pretreated at 200 °C for 1 h in
ultraꢀhighꢀpure helium (30 cm3/min) stream followed by increasing the temperature
from 100 to 800 °C. A mixture of isopropanol and liquid nitrogen was used in the
trapper to collect the formed water during the TPR experiment.
NOꢀTPD was tested to study the NO adsorption capacity and interaction between
the NO and the catalyst. A sample of 200 mg was packed into the reactor and heated
to 100 °C for 1 h under Ar. After that a gas mixture, containing 500 ppm NO, 5% O2

Catalysis Science & Technology
Page 8 of 47
View Article Online
DOI: 10.1039/C4CY01221F
and Ar to balance, was switched. When the concentration of NO was steady the NO
and O2 were turned off, and Ar was switched to purge the sample until the
concentration of NO was less than 5 ppm. Next, a TPD test was performed up to
400 °C at a heating rate of 10 °C/min.
2.4 Activity tests
Activity tests were carried out in a quartz flow tube reactor heated by a
temperatureꢀcontrolled furnace. Before being used, the samples were crushed and
sieved into 30 – 60 mesh particles. A 200 mg sample was packed into the reactor and
heated to 100 °C, followed by Ar purging for 60 min to purify the sample surface.
Then, a gas mixture, containing 500 ppm NO, 5% O2, and Ar to balance, was
switched. When a stable concentration level of NO was achieved, 500 ppm NH3 was
turned on. The temperature was raised by 50 or 25 °C to the next temperature level
until 325 °C. A total flow rate equal to 60 cm3/min, corresponding to a gas hourly
space velocity (GHSV) of 36,000 hꢀ1, was used at 1 atm throughout the process. The
SCR activity of the catalysts was expressed by the equation
Cin −Cout
XNO
=
×100%
Cin
Where Cin is the inlet concentration of NO, Cout is the outlet concentration of NO.
Each feedꢀgas flow rate was measured and controlled independently by a mass
flow controller. The outlet gas was measured on line by a gas analyzer (Gasmet
DXꢀ4000) and a quadrupole mass spectrometer. The sensitivity of each gas and the

Page 9 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
contribution of fragments were taken into consideration in the calculation of gas
concentrations.
3. Results
3.1 Catalytic performance
The catalytic performance of the prepared catalysts in NO reduction with NH3 was
evaluated between 100 and 325 °C and the results were shown in Fig 1. It can be seen
that the NO conversion over OMC was quite low, no more than 10% during the whole
process, which may result from the absence of mineral matter in OMC.4 Significant
enhancement in NO conversion was achieved with the addition of Cu or Mn. For
Cu0Mn10ꢀOMC, the NO conversion was nearly 20% at 100 °C and it increased
monotonically with temperature until a highest NO conversion (51%) was obtained at
275 °C. Further increasing the temperature led to a decrease of NO conversion.
However, Cu10Mn0ꢀOMC exhibited different characteristics with Cu0Mn10ꢀOMC.
At 100 °C the NO conversion over Cu10Mn0ꢀOMC was lower but it rapidly overtook
Cu0Mn10ꢀOMC at 200 °C. The highest NO conversion was 78% at 250 °C, which
was much higher than that over Cu0Mn10ꢀOMC. Overall, the NO conversion curve
for Cu10Mn0ꢀOMC was relatively sharper compared with Cu0Mn10ꢀOMC indicating
its narrow active temperature window for the SCR reaction. In order to broaden the
active temperature window and further improve the NO conversion, bimetallic
catalyst (Cu5Mn5ꢀOMC) was prepared expecting to utilize their respective
advantages. The NO conversion over Cu5Mn5ꢀOMC was illustrated in Fig. 1(a). It
can be seen that the catalytic performance of the bimetallic catalyst was significantly

Catalysis Science & Technology
Page 10 of 47
View Article Online
DOI: 10.1039/C4CY01221F
better than the monometallic catalysts. Compared with Cu0Mn10ꢀOMC the NO
conversion over Cu5Mn5ꢀOMC was greatly improved above 150 °C and its
maximum NO conversion reached 85% at 250 °C. In comparison with
Cu10Mn0ꢀOMC, the maximum NO conversion was not greatly improved but the
active temperature window was apparently broadened (T50, referring to the reaction
temperatures that corresponding to 50 % of NO conversion, was 200 – 325 °C
comparing with 220 – 270 °C for Cu10Mn0ꢀOMC). These results indicated that the
bimetallic catalyst possessed superior catalytic performance to monometallic catalyst
and Cu had a synergetic catalytic effect with Mn on the SCR reaction.
To investigate the optimum mass ratio of Cu to Mn in the bimetallic catalyst,
another two catalysts, Cu3Mn7ꢀOMC and Cu7Mn3ꢀOMC, were prepared to test their
catalytic performance. The NO conversion curves were also shown in Fig 1(a). It can
be seen that the NO conversion curve for Cu3Mn7ꢀOMC showed similar trend with
Cu0Mn10ꢀOMC. However, with the introduction of Cu the NO conversion for
Cu3Mn7ꢀOMC was improved except that at below 175 °C and above 300 C. This
may be ascribed to the role of Cu, high catalytic activity and narrow active
temperature window. However, it still displayed inferior catalytic performance to
Cu5Mn5ꢀOMC. For another catalyst, Cu7Mn3ꢀOMC, when the temperature was
below 300 °C, it exhibited comparable NO conversion with Cu5Mn5ꢀOMC and the
maximum conversion was even higher than Cu5Mn5ꢀOMC. However, the NO
conversion decreased dramatically after 300 °C. Considering the NO conversion
together with the active temperature window, the optimum mass ratio of Cu to Mn

Page 11 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
was selected to be 1/1 at these conditions.
In order to find the optimum metal loading, two more catalysts (Cu2Mn2ꢀOMC
and Cu10Mn10ꢀOMC) were prepared by the mass ratio of Cu to Mn equal to 1. The
variation of NO conversion with temperature was illustrated in Fig. 1(b). The NO
conversion over these catalysts showed similar trend with temperature, which
increased with temperature until a maximum was achieved and then decreased as a
result of ammonia oxidation. Among these catalysts, Cu5Mn5ꢀOMC showed the best
NO conversion even though Cu10Mn10 had more metal loading. Therefore,
Cu5Mn5ꢀOMC was selected as the optimal catalyst for the NH3ꢀSCR reaction at these
testing conditions.
N2O was a byproduct resulting from the nonselective oxidation of NH3. In order to
investigate the N2 selectivity of the catalysts, the outlet concentration of N2O was
measured during the SCR reaction. It was worth noting that the outlet concentrations
of N2O for all the catalysts were not higher than 50 ppm, which indicating excellent
N2 selectivity (more than 90%). The outlet concentrations of N2O for three
representative catalysts (Cu0Mn10ꢀOMC, Cu10Mn0ꢀOMC, and Cu5Mn5ꢀOMC) were
shown in Fig. 2. It can be seen that no N2O formed until 200 °C for all the catalysts.
Among the catalysts Cu0Mn10ꢀOMC had the lowest outlet concentration of N2O
indicating the optimum N2 selectivity, followed by Cu5Mn5ꢀOMC and
Cu10Mn0ꢀOMC. This may be related to the oxidizing capacity of catalysts. In order
to confirm this deduction, NH3 oxidation of the catalysts were performed and the
results were also shown in Fig. 2. It was found that the oxidation of NH3 followed by

Catalysis Science & Technology
Page 12 of 47
View Article Online
DOI: 10.1039/C4CY01221F
this order: Cu10Mn0ꢀOMC > Cu5Mn5ꢀOMC > Cu0Mn10ꢀOMC, which was
consistent with the outlet concentration of N2O. Therefore, the oxidizing capacity in
NH3 oxidation was related to the N2 selectivity.
In order to evaluate the stability of the catalysts, Cu5Mn5ꢀOMC was chosen to
test its catalytic performance at 250 °C for 24 h. It can be seen in Fig. 3 that the
activity of Cu5Mn5ꢀOMC was maintained unchanged during the test period. Zhang et
al. 16, 20 reported similar results in his research, in which MnOx was loaded on the
CNTs for SCR reaction. Moreover, the amount of N2O was around 20 ppm all the
time which indicated its excellent N2 selectivity. All of those demonstrated that
Cu5Mn5ꢀOMC had stable catalytic performance in SCR reaction.
3.2 EDX analysis
The constituents of the catalysts were determined by EDX analysis. Table 1 shows
that the catalysts were mainly composed by carbon and oxygen. The amount of
oxygen increased with the introduction of Cu and Mn, which may result from the
oxidation of carbon by the precursor nitrates during preparation.4 More importantly, it
should be noted that the content of Cu and Mn was coincided with the designed
values, indicating little metal loss in the preparation.
3.3 N2 sorption isotherms
The N2 sorption isotherms are shown in Fig. 4(a). It can be seen that all the
catalysts exhibited typical IV isotherms with wellꢀdefined H1ꢀtype hysteresis loops,
indicating the existence of mesoporous structures. For Cu0Mn10ꢀOMC, the hysteresis
loop located at the relative pressure from 0.45 to 0.7. With the increment of Cu the

Page 13 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
hysteresis loop shifted to the larger relative pressure and Cu10Mn0ꢀOMC located on
the highest relative pressure, which indicated that Cu10Mn0ꢀOMC owned the largest
pore diameter. This was also verified by the pore size distributions of the catalysts (as
shown in Fig. 4(b)). It can be also found in Fig. 4(b) that each sample owned a narrow
pore size distribution and almost no micropore was found for all the samples.
The textural parameters are listed in Table 1. OMC had a specific surface area
(SSA) of 1382 m2/g and total pore volume of 1.49 cm3/g. The values were close to
those reported by Peng et al,25 where similar preparation method was adopted. With
the addition of Cu and Mn, both SSA and total pore volume decreased, which may be
due to the slight poreꢀblocking that occurred on the introduction of metal.14 For the
metal loaded OMCs, Cu0Mn10ꢀOMC possessed the largest SSA and with the
increment of Cu the SSA became smaller. In contrast, the pore volume and average
pore diameter increased with the Cu contents.
3.4 XRD analysis
To verify the ordered mesoporous structure of the catalysts, smallꢀangle XRD
analysis was carried out. Fig. 5 shows the smallꢀangle XRD patterns of the catalysts.
All the catalysts except Cu0Mn10ꢀOMC had a strong peak between 0.5° and 1.5°,
confirming the ordered mesoporous structure.26 However, the peak intensity of the
metal loaded OMC, except for Cu5Mn5ꢀOMC, was weaker than that of OMC
indicating a decline in the ordered mesoporous structure.27 It can be found in the
embedded graph that Cu0Mn10ꢀOMC owned a peak at about 3°. This indicated that
the addition of Mn made the interplanar spacing smaller, which was consistent with

Catalysis Science & Technology
Page 14 of 47
View Article Online
DOI: 10.1039/C4CY01221F
the N2 sorption results that the average pore diameter of Cu0Mn10ꢀOMC was the
smallest.
The wideꢀangle XRD profiles of catalysts are shown in Fig. 6. All the profiles
exhibited a diffraction shoulder at 2θ of 15° to 23° and a broad diffraction peak at
~44°, which can be assigned to amorphous carbon.28 In addition to this, only very
broad and weak peaks of MnO2 (JCPDS 42ꢀ1316) were found for Cu0Mn10ꢀOMC
indicating the well dispersion of Mn oxides.29 However, these diffraction peaks
declined quickly with the decrease of Mn content. It can be seen that the diffraction
peaks representing Cu0 (at 43°, 50°, 74°, JCPDS 65ꢀ9743) and cuprite (Cu2+1O) (at
~36°, 73°, JCPDS 05ꢀ0667) were observed for the CuꢀOMCs. With the increase of Cu
content, the intensity of the diffraction peaks increased and the maximum peak
intensity was observed on Cu10Mn0ꢀOMC, which indicated that Cu10Mn0ꢀOMC had
the largest average size of crystallites.30 In addition to this, it can be found in the
embedded graph that in comparison with Cu10Mn0ꢀOMC the full width at half
maximum of the bands enlarged and the locations of peaks shifted to the higher angles
when Mn was introduced into CuꢀOMCs. The more the amount of Mn, the larger the
shift was. According to the Bragg equation, the lattice spacing of Cu0 in (1 1 1)
direction for Cu10Mn0ꢀOMC, Cu7Mn3ꢀOMC, Cu5Mn5ꢀOMC and Cu3Mn7ꢀOMC
was obtained, 2.091, 2.085, 2.085, and 2.083 nm, respectively. The lattice spacing of
Cu2+1O in (2 0 0) direction was also shortened from 2.137 to 2.127 nm. These
manifested the lattice contraction of the Cu oxides, which may result from the partial
substitution of Cu ion with Mn ion by the formation of solid solution with spinel

Page 15 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
structure.31 As the radius of Mn ion was smaller than Cu ion, the substitution of Cu
ion with Mn ion made the lattice contract. However, the diffraction peaks, ascribing to
the spinel structure, was not observed even in the enlarged XRD spectrum, which may
be due to its low content and the close position with the diffraction peaks of Cu
oxides.
3.5 TEM analysis
The TEM images of the catalysts are shown in Fig. 7. As reported in our previous
study12, OMC had wellꢀordered hexagonal mesopores, indicating a 2ꢀD hexagonal
mesostructure (p6mm).32 For Cu0Mn10ꢀOMC the ordered mesoporous structure was
well preserved and clear stripes can be observed on the TEM images. We can also find
that no agglomerate appeared for Cu0Mn10ꢀOMC in the TEM and STEM images.
This indicated that Mn oxides were highly dispersed on the OMC, which was
consistent with the XRD results. In the HighꢀResolution TEM (HRTEM) image of
Cu0Mn10ꢀOMC, we can find the lattice fringe of MnO2 in (110) direction. This was
also consistent with the wideꢀangle XRD result, which indicated the presence of
MnO2. For Cu10Mn0ꢀOMC, it can be seen that lots of dark blocks appeared and the
ordered mesoporous structure was not very clear. The dark blocks could be attributed
to Cu or Cu oxides as evidenced by the XRD results and they could lead to smaller
SSA. The STEM image of Cu10Mn0ꢀOMC showed that the particles were well
distributed on the OMC, similar with Liu’s report which loaded Cu into carbon
xerogels.33 Fig. 7(g) gave the HRTEM image of Cu10Mn0ꢀOMC. The lattice fringe
was attributed to Cu2O in (200) and (111) direction. The bimetallic catalysts were also

Catalysis Science & Technology
Page 16 of 47
View Article Online
DOI: 10.1039/C4CY01221F
observed by TEM and STEM. The image of Cu5Mn5ꢀOMC was displayed in Fig. 7.
It can be seen that the ordered hexagonal mesopores were formed and the
agglomerates was much less compared with Cu10Mn0ꢀOMC. The HRTEM image
also indicated the presence of Cu2O, which was consistent with the XRD result.
3.6 XPS analysis
XPS was recorded to obtain valuable information on the surface composition and
allow distinction of the different local atomic environments. The C 1s core level XP
spectra were resolved into four individual peaks representing graphitic carbon (CꢀC)
(BE =284.4 eV), carbon present in alcohol or ether groups (CꢀO) (BE = 285.1 – 286.0
eV), carboxyl or ester groups (OꢀC=O) (BE = 286.2 – 287.4 eV), and shakeꢀup
satellite peaks due to πꢀπ* transitions in the aromatic rings (BE = 289.0 – 290.2 eV).24
The results are summarized in Table 2. It can be seen that with the addition of Cu and
Mn the proportion of OꢀC=O groups increased and CꢀO groups decreased. The
OꢀC=O groups usually originated from carboxyl or ester groups. Due to the strongest
acidity of the carboxyl groups among the oxygen functional groups the stronger
acidity of the catalyst was formed, which was consistent with the reported literature.4
The highꢀresolution O 1s spectra were resolved into three peaks corresponding to the
C=O groups (BE = 530.4 – 531.2 eV), CꢀOH and/or CꢀOꢀC groups (BE = 531.6 –
532.6 eV), and chemisorbed oxygen and/or water (BE = 532.6 – 535.6 eV).24 Table 2
shows that most O atoms (60.4%) existed in the form of CꢀO for OMC and the
chemisorbed O only occupied a small part. With the introduction of Cu and Mn, the
proportion of chemisorbed O increased which was proved to be beneficial for NO

Page 17 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
adsorption in our previous study.12 Among the catalysts Cu5Mn5ꢀOMC had the
largest proportion of OꢀC=O group and chemisorbed O atoms which may result in the
strongest acidity and oxidizing capacity. The surface acidity and oxidizing capacity
were important for the NH3ꢀSCR reaction and thus Cu5Mn5ꢀOMC could exhibit the
optimal catalytic performance in NO conversion.
For the identification of the surface Mn oxide phases and the relative percentages
of Mn4+, Mn3+, Mn nitrate species, the overlapped Mn 2p peaks were deconvolved
into three individual peaks representing Mn4+ peak, Mn3+ peak, and Mn nitrate peaks
at 642.5 – 642.8 eV, 641.0 – 641.2 eV, and 644.6 – 647.9 eV.34 The relative
percentages of Mn were calculated by the area ratio of corresponding characteristic
peaks and listed in Table 3. Mn4+ was the main state for Cu0Mn10ꢀOMC. With the
addition of Cu, the ratio of Mn4+ to Mn3+ increased. It has been reported that MnO2
owned excellent catalytic performance due to its poor crystalline structure and
defective structural composition.31 Therefore, the increase of the ratio of Mn4+ to
Mn3+ may lead to the increase of NO conversion.
The Cu 2p core level XP spectra were resolved into two individual peaks
representing Cu2+ (BE =933.4 – 934.7 eV), and Cu+ (BE = 932.3 – 932.4 eV). The
fact that XRD results did not show Cu2+ phase, while XPS indicated the presence of
Cu2+ ions, suggesting that Cu2+ was present only on the surface of the catalysts and it
formed a thin amorphous outer shell.35 Previous studies36 have pointed out that the
SCR activity of Cuꢀcontaining catalysts is often interpreted in relation to a redox
cycle of Cu2+ to Cu+. Therefore, the presence of Cu2+ and Cu1+ may be beneficial for

Catalysis Science & Technology
Page 18 of 47
View Article Online
DOI: 10.1039/C4CY01221F
the NO conversion.
3.7 CO/CO2ꢀTPD analysis
CO/CO2ꢀTPD was employed to explore the composition of oxygen functional
groups at the surface. It has been reported that CO resulted from the groups where C
was bonded to one oxygen atom and CO2 resulted from that C was bonded to two
oxygen atoms.24 The profiles of CO evolved were similar and no obvious evolution
can be seen until 600 °C (not shown here). The amount of CO evolved during TPD
was shown in Table 4. It can be seen that the evolved amount for OMC was the most
and with the introduction of Cu and Mn the evolved amount decreased. This was
consistent with the XPS results, which demonstrated that CꢀO group decreased with
the addition of Cu and Mn. Fig. 8 shows the profiles of CO2 evolved during the TPD
process. The profile for OMC exhibited a strong peak at ~200 °C and another two
peaks at 400 and 650 °C, which can be attributed to the carboxylic groups, anhydride
groups and lactones based on literature.24 The intensity of the CO2 for metal loaded
OMC was stronger than that of OMC, suggesting that the surface oxygen functional
groups increased with the introduction of metal.4 This was also coincided with the
XPS result that OꢀC=O groups increased with the introduction of metal. As mentioned
above, the increase of OꢀC=O group would result in stronger acidity, which was
beneficial for the NH3 adsorption. In addition, the amount of CO2 evolved during
TPD was also shown in Table 4. It can be seen that Cu5Mn5ꢀOMC had the most
amount of evolved CO2, in agreement with the XPS results.
3.8 NH3ꢀTPD analysis

Page 19 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
The surface acidic sites distribution of the catalysts were determined with
NH3ꢀTPD. Fig. 9 shows the evolution behavior of NH3 during the TPD process. The
OMC showed relative small NH3 adsorption amount (not shown here). For
Cu0Mn10ꢀOMC the profile had a low temperature peak at 195 °C together with two
relative high temperature peaks at 303 and 394 °C. This indicated that there were at
least three kinds of adsorption sites on the surface of Cu0Mn10ꢀOMC.37 Compared
with Cu0Mn10ꢀOMC, the biggest difference for Cu10Mn0ꢀOMC was the absence of
the low temperature peak. When Cu and Mn were both introduced into OMC, the
behavior of NH3 evolved showed a mixed characteristics of Cu0Mn10ꢀOMC with
Cu10Mn0ꢀOMC. The evolved amount of NH3 was shown in Table 4. It can be seen
that Cu0Mn10ꢀOMC had larger value than Cu10Mn0ꢀOMC. This demonstrated that
Cu0Mn10ꢀOMC had stronger acidity than Cu10Mn0ꢀOMC, especially at low
temperature range. It is established in literature that the acid site distribution could be
broadened by the addition another metal.29 In the present study, acid site distribution
of Cu5Mn5ꢀOMC was broadened and it owned the most evolved amount of NH3
among the catalysts. This was consistent with the XPS and CO/CO2ꢀTPD results,
which indicated that Cu5Mn5ꢀOMC had the strongest acidity.
3.9 H2ꢀTPR analysis
H2ꢀTPR is a valid method to determine the redox performance of the catalyst. Fig.
10 shows the TPR curves of the catalysts. The OMC can hardly be reduced by H2,
while obviously H2 consumption can be seen over the metal loaded catalysts. The
Cu0Mn10ꢀOMC sample shows TPR patterns characterized by two peaks at 305 and

Catalysis Science & Technology
Page 20 of 47
View Article Online
DOI: 10.1039/C4CY01221F
467 °C, which could be interpreted as a reduction of MnO2 to Mn2O3, and Mn2O3 to
MnO, respectively.29 The H2ꢀTPR result for Cu10Mn0ꢀOMC showed a strong peak at
173 °C and two relatively weak peaks at 321 and 491 °C, which could be ascribed to
the reduction of CuO to Cu2O and Cu.38 When Cu and Mn was coexisted in the OMC,
the strong peak at 173 °C shifted to higher temperature and the two peaks at 321 and
491 °C shifted to lower temperatures. The amount of H2 consumed for the catalysts
were shown in Table 4. It can be seen that Cu10Mn0ꢀOMC had larger H2
consumption than Cu0Mn10ꢀOMC, indicating stronger oxidizing capacity of Cu. The
bimetallic catalysts exhibited excellent oxidizing capacity within 150 to 600 °C due to
the synergetic effect of Cu and Mn. It was worth noting that Cu5Mn5ꢀOMC had the
largest H2ꢀconsumption (even larger than Cu10Mn0ꢀOMC) which demonstrated that it
owned the best redox performance.
3.10
NOꢀTPD analysis
The NOꢀTPD patterns were shown in Fig. 11. For OMC the desorption profile of
NO showed two main peaks at 240 and 335 °C, which meant that at least two NO
adsorption species existed, i.e. monodentate nitrito and bridging monodentate
nitrito.12 With the addition of Mn or Cu the NO adsorption capacity improved (as
shown in Table 4) and the NO desorption peaks shifted to the lower temperatures.
This demonstrated that more adsorption sites for NO molecule were provided as the
introduction of Cu or Mn and the interaction between the NO molecule and the
catalyst became weaker due to the shift of the desorption peaks. In addition,
Cu0Mn10ꢀOMC had the largest NO adsorption capacity. Correspondingly, the

Page 21 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
temperatures of NO desorption peaks were lowest for Cu10Mn0ꢀOMC. The
difference in NO adsorption may result in the different behaviors in the SCR reaction.
4. Discussion
4.1 The physicochemical properties of the catalysts
From the N2 sorption results we can find that all the catalysts owned mesoporous
structures with narrow pore size distributions. With the addition of Cu and Mn, both
SSA and total pore volume decreased due to the slight poreꢀblocking.14 Compared
with Cu10Mn0ꢀOMC, Cu0Mn10ꢀOMC possessed larger SSA, which indicated that
the distribution of Mn oxides was more uniform than Cu. These were certified by the
wideꢀangle XRD and TEM analysis. In the wideꢀangle XRD analysis it can be seen
that Mn oxides were highly dispersed on the surface and metal agglomerate was only
found for the Cu containing catalysts in the TEM images. The smallꢀangle XRD
results demonstrated that all the catalysts except Cu0Mn10ꢀOMC had a strong peak
between 0.5° and 1.5°, confirming their ordered mesoporous structure.26
Cu0Mn10ꢀOMC owned a peak at about 3° indicating the smaller interplanar spacing,
consistent with the smallest pore diameter obtained from N2 sorption. In the TEM
images clear stripes can be observed demonstrating the well developed ordered
mesoporous structure. Together with the N2 sorption and smallꢀangle XRD results, we
can conclude that all the catalysts possessed the ordered mesoporous structure.
From the XPS results we can find that with the introduction of Cu and Mn the
proportion of OꢀC=O group and chemisorbed O increased. The increase of OꢀC=O
group usually resulted in stronger acidity and the chemisorbed O was proved to be

Catalysis Science & Technology
Page 22 of 47
View Article Online
DOI: 10.1039/C4CY01221F
beneficial for NO adsorption.12 These were both important for the NH3ꢀSCR reaction.
Among the catalysts Cu5Mn5ꢀOMC had the largest proportion of OꢀC=O group and
chemisorbed O atoms and thus it had the optimal catalytic performance in NH3ꢀSCR
reaction. CO/CO2ꢀTPD results showed that with the addition of Cu and Mn the
amount of CO2 evolved during TPD increased. This verified the XPS result that
OꢀC=O group increased with the introduction of Cu and Mn. Cu5Mn5ꢀOMC had the
most evolved amount of CO2 which also indicated that Cu5Mn5ꢀOMC had the
strongest acidity, in agreement with the XPS results. NH3ꢀTPD results demonstrated
that Cu0Mn10ꢀOMC had stronger acidity than Cu10Mn0ꢀOMC, especially at low
temperature range. And the acid site distribution was broadened for Cu5Mn5ꢀOMC.
This was consistent with the XPS and CO/CO2ꢀTPD results, which showed that
Cu5Mn5ꢀOMC had the strongest acidity. It can be seen from the H2ꢀTPR results that
Cu10Mn0ꢀOMC had stronger oxidizing capacity than Cu0Mn10ꢀOMC and the
bimetallic catalysts exhibited excellent oxidizing capacity due to the synergetic effect
of Cu and Mn. Likewise, Cu5Mn5ꢀOMC had the strongest oxidizing capacity,
consistent with the XPS results. The NOꢀTPD results showed that Cu0Mn10ꢀOMC
had larger NO adsorption capacity than Cu10Mn0ꢀOMC. Together with the NH3ꢀTPD
results it can be seen that Cu0Mn10ꢀOMC had superior adsorption property than
Cu10Mn0ꢀOMC. However, Cu0Mn10ꢀOMC did not exhibit satisfying catalytic
performance probably due to its shortage of oxidizing capacity, which was obtained
from H2ꢀTPR results. Therefore, the combination of Cu and Mn would generate the
synergetic effect in SCR reaction.

Page 23 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
Moreover, in the wideꢀangle XRD analysis we can find that with the addition of
Mn the diffraction peak of Cu oxides shifted to higher angles. This manifested the
lattice contraction of Cu oxides. This may result from the substitution of Cu ion with
Mn ion by the formation of solid solution with spinel structure. It has been reported
that Cu and Mn oxides would form the spinel structures when the calcination
temperature was above 500 °C.39ꢀ41 As the calcination temperature was 850 °C in the
preparation the spinel structures should be existed. Due to the smaller radius of Mn
ion, the replacement of Cu ion by Mn ion lead to the lattice of Cu oxides contracted.
In the XPS analysis Mn was divided into Mn4+, Mn3+ and Mn nitrate species. Due to
the introduction of Cu the proportion of Mn4+ increased. It may result from the right
shift of the equilibrium Cu2+ + Mn3+ ↔ Cu+ + Mn4+. This redox process played an
important role in the spinel structure and higher temperatures was helpful for the
equilibrium to shift to the right.41 As the calcination temperature was as high as 850
°C, more Cu+ and Mn4+ ions were created derived from the oxidation of Mn3+ by
Cu2+.41, 42
4.2 The promoting effect of Cu and Mn
It can be seen that the catalytic activity in NO reduction with NH3 for OMC was
below 10 % between 100 and 325 °C. However, significant enhancement in NO
conversion was achieved with the addition of Cu or Mn. It has been reported that the
SCR reaction over carbon may be controlled by two consecutive steps: the adsorption
of NH3 on acidic sites, followed by the adsorption of NO or NO2 in the neighborhood
of NH3 complexes.43 The adsorption of NH3 depended on the surface acidic sites of

Catalysis Science & Technology
Page 24 of 47
View Article Online
DOI: 10.1039/C4CY01221F
the catalyst. The XPS and CO/CO2ꢀTPD analysis showed that with the addition of Cu
and Mn the acidity of the catalyst was enhanced. On the other hand, the adsorption of
NO was related with the chemisorbed O atoms.12 The proportion of chemisorbed
oxygen was improved after the addition of Cu and Mn according to the XPS results.
The NOꢀTPD result also confirmed that with the addition of Cu and Mn the NO
adsorption capacity improved. Thence, the NO conversion over the monometallic
catalysts was significantly improved. For the monometallic catalysts, the valence
transition of metal oxides was beneficial for the dehydrogenation of NH3 to NH2. The
NH2 was absorbed on the surface and reacted with the adsorbed NO to form N2 and
H2O. The reduced active sites were recovered by the reoxidation of O2.44 The possible
reaction route over MꢀOMC was illustrated in Scheme 1. At first NH3 was adsorbed
by the acidic sites to form NH2, which was the intermediate for the SCR reaction.44
The electron transferred from NH3 to Mn+ and reduced Mn+ into M(nꢀ1)+. At the same
time O2 was adsorbed on the neighborhood and reacted with active free (Cf) sites to
form C(O) group. In the next step NO was adsorbed by the C(O) group and
monodentate nitrito (CꢀOꢀN=O) was formed.12 Then the adsorbed NH2 reacted with
CꢀOꢀN=O and generated N2 and H2O. Finally the reduced M(nꢀ1)+ was reꢀoxidized by
O2 and the catalytic reaction continued following the next cycle. The possible
mechanism was as follows:
Mn+ + NH3(g) + O2ꢀ → M(nꢀ1)+ + NH2(ad) + OHꢀ
Cf + O2(g) → C(O)
(1)
(2)
(3)
C(O) + NO → C(–O–N=O)

Page 25 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
C(–O–N=O) + NH2(ad) → N2 + H2O + Cf
Mn(nꢀ1)+ + O2(g) → Mn+ + O2ꢀ
OHꢀ → H2O + O2ꢀ
(4)
(5)
(6)
Scheme 1 The proposed mechanism of the SCR reaction over MꢀOMC.
For the bimetallic catalysts, especially Cu5Mn5ꢀOMC, the activity was further
improved in comparison with the monometallic catalysts. This indicated that Cu and
Mn may have a synergetic catalytic effect. It can be seen from the XPS results that
with the addition of Cu the proportion of Mn4+ increased which was reported to be
important in the catalytic performance.45, 46 Previous studies36 have pointed out that
the SCR activity of Cuꢀcontaining catalysts is often interpreted in relation to a redox
cycle of Cu2+ to Cu+. Therefore, the redox process Cu2+ + Mn3+ ↔ Cu+ + Mn4+ played
an important role in the spinel structure. From the NH3ꢀTPD results we can find that
the Cu0Mn10ꢀOMC owned stronger acidity than Cu10Mn0ꢀOMC, especially in the
low temperature region. The NOꢀTPD results indicated that Mn could provide more

Catalysis Science & Technology
Page 26 of 47
View Article Online
DOI: 10.1039/C4CY01221F
adsorption sites for NO molecules. Therefore, Mn may be mainly responsible for the
NH3 and NO adsorption. Correspondingly, the H2ꢀTPR results indicated that the
Cu10Mn0ꢀOMC had stronger oxidizing capacity than Cu0Mn10ꢀOMC. Therefore, the
combination of Cu and Mn could provide stronger acidity and oxidizing capacity. The
results that Cu5Mn5ꢀOMC possessed the most H2 consumption and NH3 evolved
amount certified this hypothesis. On the basis of the above analysis, we proposed a
possible synergetic catalytic effect (redox cycle) between Mn and Cu ions for low
temperature SCR reaction over Cu and Mn bimetallic catalysts. This based on the
electron transfer between Mn and Cu ions to maintain a dynamic equilibrium, as
depicted in Scheme 2. The possible mechanism was as follows.
Mn4+ + NH3(g) + O2ꢀ → Mn3+ + NH2(ad) + OHꢀ
Cu+ + O2(g) → Cu2+ + O(ad)
Cf + O(ad) → C(O)
(7)
(8)
(9)
(3)
(4)
(10)
(6)
C(O) + NO → C(–O–N=O)
C(–O–N=O) + NH2(ad) → N2 + H2O + Cf
Mn3+ + Cu2+ → Mn4+ + Cu+
OHꢀ → H2O + O2ꢀ
In Step 1, Mn4+ was responsible for the dehydrogenation of NH3 following
reaction Eq. (7). In the meantime Cu+ was actually activated into Cu2+ by donating an
electron to O2 to form adsorbed O atoms. In the next step the adsorbed NH2 reacted
with CꢀOꢀN=O at the neighborhood to generate N2 and H2O, which was similar with

Page 27 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
that over monometallic catalyst. Finally Cu2+ captured an electron from Mn3+ due to
its stronger oxidizing capacity and then they became Cu+ and Mn4+, respectively.42 In
the monometallic catalyst the reduced active sites were recovered by the reoxidation
of O2. With the addition of Cu Mn3+ was also reꢀoxidized by Cu2+ besides O2 through
reaction Eq. (10). And thus the SCR catalytic performance was further improved on
the bimetallic catalysts.
Scheme 2 The proposed mechanism of the SCR reaction over CuxMnyꢀOMC
catalysts and the synergetic catalytic effect between Mn and Cu ions.
5. Conclusions
A series of Cu and Mn containing ordered mesoporous carbons were prepared by
EISA method and their DeNOx property were studied. N2 sorption, smallꢀangle XRD
and TEM were used to confirm their structures. The results indicated that the ordered
and uniform structures were successfully synthesized for all the samples. Wideꢀangle

Catalysis Science & Technology
Page 28 of 47
View Article Online
DOI: 10.1039/C4CY01221F
XRD and TEM results showed that Mn oxides were highly distributed on the surface
and agglomerate was found for Cuꢀcontaining catalysts. Wideꢀangle XRD also
demonstrated that in the bimetallic catalysts a part of Cu ions were replaced by the
Mn ions resulting in the lattice contraction by the formation of solid solution. XPS
and CO/CO2ꢀTPD results indicated that with the introduction of Cu or Mn the surface
acidity was enhanced. NH3ꢀTPD and H2ꢀTPR were adopted to study their acidity and
oxidizing capacity. The results demonstrated that bimetallic catalyst Cu5Mn5ꢀOMC
possessed the stronger acidity and oxidizing capacity compared with the
monometallic catalysts. Therefore, Cu5Mn5ꢀOMC exhibited the optimal catalytic
activity on the NH3ꢀSCR reaction. The synergetic effect between Mn and Cu may
play an important role for the superior catalytic performance. A possible redox
cycleꢀbased synergetic catalytic effect was proposed through the electron transfer
between Mn and Cu ions.
Acknowledgements
This work is supported by the National Science Foundation for Distinguished
Young Scholars of China (No.51125025), Hiꢀtech Research and Development
Program of China (863 Program, No. 2013AA065401), the Key Innovation Team for
Science and Technology of Zhejiang Province (Grant No. 2011R50017), Major
Program of the Environmental Protection Bureau of Hangzhou City
(No. HZYXꢀHJꢀ12200GK), and the Shanghai Tongji Gao Tingyao Environmental
Science & Technology Development Foundation.

Page 29 of 47
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C4CY01221F
References
1. M. J. Lazaro, M. E. Galvez, C. Ruiz, R. Juan and R. Moliner, Appl. Catal., B, 2006, 68, 130ꢀ138.
2. X. Gao, S. Liu, Y. Zhang, X. Du, Z. Luo and K. Cen, Catal. Today, 2011, 175, 164ꢀ170.
3. Q. Li, H. Yang, Z. Ma and X. Zhang, Catal. Commun., 2012, 17, 8ꢀ12.
4. H. S. Teng, L. Y. Hsu and Y. C. Lai, Environ. Sci. Technol., 2001, 35, 2369ꢀ2374.
5. Y. Hou, Z. Huang and S. Guo, Catal. Commun., 2009, 10, 1538ꢀ1541.
6. G. Marban, R. Antuna and A. B. Fuertes, Appl. Catal., B, 2003, 41, 323ꢀ338.
7. B. Huang, R. Huang, D. Jin and D. Ye, Catal. Today, 2007, 126, 279ꢀ283.
8. X. Fan, F. Qiu, H. Yang, W. Tian, T. Hou and X. Zhang, Catal. Commun., 2011, 12, 1298ꢀ1301.
9. S. H. Joo, S. J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki and R. Ryoo, Nature, 2001, 412, 169ꢀ172.
10. Y. Meng, D. Gu, F. Zhang, Y. Shi, H. Yang, Z. Li, C. Yu, B. Tu and D. Zhao, Angew. Chem. Int. Ed.,
2005, 117, 7215ꢀ7221.
11. C. Liang and S. Dai, J. Am. Chem. Soc., 2006, 128, 5316ꢀ5317.
12. F. Cao, J. Chen, M. Ni, H. Song, G. Xiao, W. Wu, X. Gao and K. Cen, RSC Adv., 2014, 4,
16281ꢀ16289.
13. J. Li, J. Gu, H. Li, Y. Liang, Y. Hao, X. Sun and L. Wang, Micropor. Mesopor. Mate., 2010, 128,
144ꢀ149.
14. Z. Liu, Y. Yang, J. Mi, X. Tan and Y. Song, Catal. Commun., 2012, 21, 58ꢀ62.
15. X. Wang, Y. Zheng and J. Lin, Catal. Commun., 2013, 37, 96ꢀ99.
16. C. Fang, D. Zhang, S. Cai, L. Zhang, L. Huang, H. Li, P. Maitarad, L. Shi, R. Gao and J. Zhang,
Nanoscale, 2013, 5, 9199ꢀ8207.
17. R. Nedyalkova, S. Shwan, M. Skoglundh and L. Olsson, Appl. Catal., B, 2013, 138, 373ꢀ380.
18. H. Chang, X. Chen, J. Li, L. Ma, C. Wang, C. Liu, J.W. Schwank and J. Hao, Environ. Sci.
Technol., 2013, 47, 5294ꢀ5301.
19. P. Maitarad, D. Zhang, R. Gao, L. Shi, H. Li, L. Huang, T. Rungrotmongkol and J. Zhang, J. Phys.
Chem., C, 2013, 117, 9999ꢀ10006.
20. D. Zhang, L. Zhang, L, Shi, C. Fang, H. Li, R. Gao, L. Huang and J. Zhang, Nanoscale, 2013, 5,
1127ꢀ1136.
21. R. Gao, D. Zhang, P. Maitarad, L. Shi, T. Rungrotmongkol, H. Li, J. Zhang and W. Cao, J. Phys.