Cr-MnOx mixed-oxide catalysts for selective catalytic reduction of NOx with NH3 at low temperature

Article abstract of DOI:10.1016/j.jcat.2010.08.016

Cr-Mn mixed-oxide based catalysts were prepared for the low-temperature selective catalytic reduction of NO_x with ammonia in the presence of excess oxygen. It was found that the Cr(0.4)-MnO_x showed the highest activity and yielded 98.5% NO_x conversion at 120 °C. XRD, TPR and Raman data results suggested that a crystalline phase of CrMn _1.5O₄ was present in the Cr-MnO_x catalysts, which contained the active species. XPS results of fresh, used and regenerated Cr(0.4)-MnO_x catalysts illustrated clearly the presence of Mn ²⁺, Mn³⁺, Mn⁴⁺ and Cr²⁺, Cr ³⁺, Cr⁵⁺ oxidation states. Efficient electron transfer between Cr and Mn in the crystal of CrMn_1.5O₄ was thought to be the reason for the high activity and long lifetime of the Cr(0.4)-MnO _x catalyst. In addition, the SCR activity was gradually suppressed in the presence of SO₂, while such an effect was shown to be reversible after switching off the SO₂ injection.

Full text of DOI:10.1016/j.jcat.2010.08.016

Journal of Catalysis 276 (2010) 56–65
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Cr–MnOx mixed-oxide catalysts for selective catalytic reduction of NOx with NH3
at low temperature
a
b,
Zhihang Chen a, Qing Yang a, Hua Li a, Xuehui Li a, , Lefu Wang , Shik Chi Tsang
⇑
⇑⇑
a School of Chemistry and Chemical Engineering, The Guangdong Province Laboratory of Green Chemical Technology, South China University of Technology,
Guangzhou 510640, PR China
b Wolfson Catalysis Centre, Inorganic Chemistry Laboratory, University of Oxford, Oxford OX1 3QR, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Cr–Mn mixed-oxide based catalysts were prepared for the low-temperature selective catalytic reduction
of NOx with ammonia in the presence of excess oxygen. It was found that the Cr(0.4)–MnOx showed the
highest activity and yielded 98.5% NOx conversion at 120 °C. XRD, TPR and Raman data results suggested
that a crystalline phase of CrMn1.5O4 was present in the Cr–MnOx catalysts, which contained the active
species. XPS results of fresh, used and regenerated Cr(0.4)–MnOx catalysts illustrated clearly the presence
of Mn2+, Mn3+, Mn4+ and Cr2+, Cr3+, Cr5+ oxidation states. Efficient electron transfer between Cr and Mn in
the crystal of CrMn1.5O4 was thought to be the reason for the high activity and long lifetime of the
Cr(0.4)–MnOx catalyst. In addition, the SCR activity was gradually suppressed in the presence of SO2,
while such an effect was shown to be reversible after switching off the SO2 injection.
Received 1 May 2010
Revised 30 July 2010
Accepted 28 August 2010
Keywords:
Cr–Mn mixed oxide
Low temperature
Selective catalytic reduction
Nitrogen oxides
CrMn1.5O4
Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction
and As2O3) and SO2. Thus, the development of a highly active cat-
alyst for low-temperature SCR (<200 °C) that works downstream of
Nitrogen oxides (NO, NO2 and N2O) resulting from fossil fuel
combustion have been a major source of air pollution. These oxides
have caused environmental problems such as photochemical smog,
acid rain, ozone depletion and greenhouse effects. At present,
about 49% of NOx from transportation and 46% from power plants
are recorded [1]. Several methods have been used for the elimina-
tion of NOx, such as selective catalytic reduction (SCR) [2–4], cata-
lytic decomposition [5,6] and plasma catalysis [7,8]. SCR of NOx
with NH3 is an effective technique and has been commercialized
in the post-treatment of flue gases of power plants. In this process,
NOx from flue gases is reduced by ammonia in excess oxygen to
molecular nitrogen and water:
the electrostatic precipitator and desulfurization offers enormous
potential, which could also lead to improved economics for the
SCR process.
Some transition metal-oxides and mixed-oxide catalysts have
been investigated in the low-temperature SCR reaction including
Co3O4 [11], MnOx [12], MnOx/TiO2 [13], CrOx/TiO2 [14], Fe–Mn/
TiO2 [15], CuOx–MnOx [16] and MnOx–CeO2 [17]. Of all the cata-
lysts screened, Mn-containing catalysts exhibited relatively high
activity for the conversion of NOx. An early report suggested that
MnOx–CeO2 mixed oxides gave greater than 90% of NO conversion
at 100 °C [18]. Sreekanth et al. [19] compared the NO conversions
of Cr/TiO2, Mn/TiO2, Mn–Cu/TiO2 and Mn–Cr/TiO2, and their results
showed that Mn/TiO2 gave the best activity, of about 90% of NO
conversion, at 120 °C, whereas TiO2 supported Mn–Cr/TiO2 showed
a low activity of about 60% NO conversion at 120 °C at a low space
velocity ([NO] = [NH3] = 2000 ppm, GHSV = 8000 hꢁ1). In many
cases, the activity of redox mixed catalysts is highly dependent
on the precise molar ratio used and their method of preparation.
In this paper, an extensive study of Cr–MnOx mixed oxide for
low-temperature NOx reduction with NH3 is reported. The identifi-
cation of the present Cr–MnOx was mainly achieved through the
initial screening of 120 mixed oxides prepared by the solid phase
synthesis. This combinatorial approach of blending two elements
from the given list of Mg, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Sr, Zr,
Mo, Ba, W and Bi in searching active bimetallic mixed oxides was
4NO þ 4NH3 þ O2 ! 4N2 þ 6H2O;
D
Gꢀ298 ¼ ꢁ1627 kJ=mol
ð1Þ
A typical commercial catalyst for SCR of NOx is V2O5–WO3-
(MoO3)/TiO2 [9,10]. This catalyst is operated at a temperature
range of 300–400 °C, and the SCR reactor is located upstream from
the particle removal or desulfurizer device to avoid reheating of
the flue gas. However, this renders the catalyst more susceptible
to deactivation from high concentration of dust (e.g., K2O, CaO
⇑
Corresponding author. Fax: +86 20 8711 4707.
Corresponding author. Fax: +44 186 528 2610.
E-mail addresses: cexhli@scut.edu.cn (X. Li), edman.tsang@chem.ox.ac.uk (S. Chi
⇑⇑
Tsang).
0021-9517/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2010.08.016

Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
57
rather empirical and no extensive comparison between them at
different testing conditions was carried out. It would be ideal to
identify the guiding rules for the selection of active mixed redox
oxides. However, the effects of synergy between two or more metal
oxides at a close proximity are rather complex and are still not yet
clear from the literature. Nevertheless, we noted that the Cr–MnOx
mixed oxides show exceptional SCR activity in our initial evalua-
tion. In this paper, we report various Cr–MnOx phases that were
prepared by the citric acid method and tested for low-temperature
SCR activity. The mixed-oxide phase, CrMn1.5O4, was clearly found
to be active for low-temperature SCR. The surface area, crystalline
phases, ratios of chromium and manganese ions and the promoting
effect of CrMn1.5O4 were attributed to different catalytic activities
measured for the Cr–MnOx catalysts. The electron transfer between
chromium and manganese with relevance to the mechanism for
low-temperature SCR process is discussed.
(I = 15 mA, U = 14 kV) and Al K
a
radiation (1486.6 eV) were used.
The sample powders before or after testing were transferred and
pelletized with a diameter of 0.2 mm under an inert atmosphere
without exposure to air. They were then examined by the XPS
equipment at room temperature. Intensities were estimated from
the integration of each peak, after smoothing, subtraction of the
L-shaped background, and fitting the experimental curve to a com-
bination of Lorentzian and Gaussian lines of variable proportion.
All binding energies were referenced to the C 1s line at 284.6 eV.
Binding energy values were measured with a precision of 0.3 eV.
Raman spectra of materials were taken with a Raman spectrom-
eter (LabRam HR, Horiba Jobin Yvon Inc.) equipped with Olympus
BX-41 microscope (objective ꢂ 100) and TE-cooled CCD detector
(Andor), in a backscattering configuration using He–Ne laser
(632.8 nm excitation line) with power of 5–20 mV at a sample
and spectral resolution of 0.8 cmꢁ1
.
2.3. Catalytic activity measurement
2. Experimental
The SCR activity measurement was carried out in a fixed-bed
quartz reactor (i.d. = 8 mm). The typical reactant gas composition
was as follows: 1000 ppm NO, 1000 ppm NH3, 3% O2, and balance
N2 or He (for selectivity experiments). About 3.2 g catalyst (60–
100 mesh) was used for each run. Under ambient conditions, the to-
tal flow rate was 860 mL minꢁ1 and the gas hourly space velocity
(GHSV) was 30,000 hꢁ1. The compositions of the feed gases and
the effluent streams were monitored continuously using on-line
sensors with emission monitors: gas analyzer (SWG-300, MRU, Ger-
many) for NO, NO2 and SO2. Without catalyst, it was found that the
NO conversion to NO2 was less than 1%. Thus, the envisaged effect of
NO2 on the SCR activity at such low concentration was not signifi-
cant. The effluent streams were also analyzed by a gas chromato-
graph (4890D, HP, USA) at 60 °C with 5 Å molecular sieve column
for N2 and Porapak Q column for N2O. All data were collected from
10 min to 30 min (steady state) at the chosen temperature. From
the concentration of the gases at steady state, the conversion and
N2 selectivity are calculated according to the following equations:
2.1. Catalyst preparation
Chromium nitrate, manganese acetate and citric acid were
mixed in designated ratios. The mole ratio of citric acid to the me-
tal components (the total moles of chromium and manganese) was
set at 1.0. The resulting mixture was stirred at room temperature
for 1.0 h. The solution was dried at 120 °C for 12 h, resulting in a
porous, foam-like solid. The foam-like precursor was calcined at
the desired temperature for 3.0 h in air in a temperature pro-
grammed muffle furnace (3–550, Vulcan, USA). Finally, the samples
were crushed and sieved to 60–100 mesh. Pure chromium oxide
(CrOx) and manganese oxide (MnOx) were also prepared using
the same procedure; CrOx–MnOx was prepared by physically mix-
ing the pure CrOx and MnOx with an equal molar ratio of Cr to Mn
and then calcined at 650 °C for 3.0 h. The mixed oxide was denoted
as Cr(y)–MnOx(z), where y represented the mole ratio of Cr/
(Cr + Mn) and z denoted the calcination temperature in °C, e.g.,
Cr(0.4)–MnOx (650).
½NOxꢃin ꢁ ½NOxꢃout
NOx conversion ð%Þ ¼
N2 selectivity ð%Þ ¼
ꢂ 100%
ð2Þ
ð3Þ
½NOxꢃin
2.2. Catalyst characterization
½N2ꢃout
½N2ꢃout þ ½N2Oꢃout
ꢂ 100%
A Micromeritics ASAP 2010 micropore-size analyzer was used
to measure the N2 adsorption isotherms of the samples at liquid
N2 temperature (ꢁ196 °C). The specific surface area was deter-
mined from the linear portion of the BET plot. The pore-size distri-
bution was calculated from the desorption branch of the N2
adsorption isotherm using the Barrett–Joyner–Halenda (BJH) for-
mula. Prior to the surface area and pore-size distribution measure-
ments, the samples were degassed in vacuum at 300 °C for 24 h.
X-ray diffraction (XRD) patterns were obtained using a Rigaku
where [NOx] = [NO] + [NO2], and the subscripts in and out indicated
the inlet and outlet concentration at steady state, respectively. Be-
cause the reaction was carried out at low temperature, it was
important to ensure that the decrease in NO was not caused by
the adsorption of NO by catalyst material. In each experiment, cat-
alyst was purged until the NO concentration reached the expected
gas concentration (1000 ppm).
D/MAX-3A Auto X-ray diffractometer with Cu
Ka radiation
2.4. Normal temperature and pressure plasma reoxidation [20]
(k = 1.5418 Å). Intensity data were collected over a 2h range of 5–
85° with a 0.05° step size and a counting time of 1 s per point.
The XRD phases were identified by comparison with the reference
data from International Center for Diffraction Data (ICDD) files. A
scanning electron microscope (JSM-7401F, JEOL, Japan) was used
to obtain surface information on the catalysts.
The TPR experiments were carried out on a Micromeritics Auto-
Chem 2920 chemisorption analyser. A 30 mL minꢁ1 (NTP) gas flow
of 10% H2 in Ar was passed over approximately 100 mg samples in
a quartz reactor through a cold trap to the detector. The reduction
temperature was linearly raised at 10 °C minꢁ1 from 30 to 950 °C.
X-ray photoelectron spectroscopy (XPS) analyses were per-
formed using a Quantum-2000 Scanning ESCA Microprobe (Physi-
A normal pressure electrical glow discharge was used to acti-
vate oxygen on catalytic surface in a stream of flowing pure oxy-
gen. The discharge was excited by a 10 MHz RF generator in a
silica glass tube with external sleeve electrodes. Typical experi-
mental conditions were a temperature of 25 °C, a pure oxygen flow
rate at 50 mL minꢁ1 and residence time of 1.2 s, and a 6 h duration
of treatment.
2.5. NO oxidation to NO2
The experiment of NO oxidation to NO2 was performed in the
fixed-bed quartz reactor. Conversion at each temperature was
determined using 3.2 g of sample, and 30 min after, steady state
cal Electronics) with
a hemispherical detector operating at
constant pass energy (PE = 46.95 eV). An X-ray source at 210 W

58
Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
conditions were reached. The reactant gas composition was as fol-
lows: 1000 ppm NO, 3% O2, with the balance N2; the total flow rate
was 860 mLꢄminꢁ1 (ambient conditions). The NO concentration
was continually monitored by the above gas analyzer. NO conver-
sion to NO2 was obtained by Eq. (4):
½NO2ꢃout ꢁ ½NO2ꢃin
NO conversionto NO2 ð%Þ ¼
ꢂ 100%
ð4Þ
½NOꢃin
3. Results and discussion
3.1. Low-temperature activity of CrOx, MnOx, CrOx–MnOx and Cr(0.5)–
MnOx
The performances of CrOx, MnOx, Cr(0.5)–MnOx catalysts pre-
pared by the citric acid method and mechanically mixed CrOx–
MnOx catalyst for SCR of NOx with NH3 in the presence of excess
oxygen are shown in Fig. 1. Pure MnOx catalyst exhibited a well-
defined catalytic activity at the relatively high space velocity of
30,000 hꢁ1, which matched well with a previous report [21]. Pure
CrOx showed negligible activity, while the activity of the physical
mixture, CrOx–MnOx, was found between CrOx and MnOx. Notably,
a higher NOx conversion was obtained using the mixed Cr(0.5)–
MnOx (650) catalyst. This observation clearly implies that the
inclusion of Cr into the MnOx lattice in the mixed oxide (citric acid
method), rather than the physical mixing, is the determining factor
in the significant gain in activity of the Cr(0.5)–MnOx catalysts.
Fig. 2 shows the XRD patterns of the above CrOx, MnOx, CrOx–
MnOx and Cr–MnOx catalysts. CrOx and MnOx sample give sharp
XRD peaks (Fig. 2a and b) representing Cr2O3 (ICDD PDF #84-
1616 2h = 33.6°, 54.9° and 36.3°) and Mn3O4 phases (ICDD PDF
#89-4837 2h = 36.1°, 59.9° and 32.3°), respectively. The XRD pat-
tern of CrOx–MnOx (Fig. 2c) is consistent with the combination of
Cr2O3 and Mn2O3 phases (ICDD PDF #89-4836 2h = 32.9°, 55.1°
and 65.8°), while a totally new and dominant phase, CrMn1.5O4
(Fig. 2d, ICDD PDF #71-0982 2h = 35.1°, 61.9° and 29.8°), appears
in the Cr(0.5)–MnOx catalyst. Hence, the superior high activity of
the mixed oxide suggested that the generated CrMn1.5O4 phase
may be crucially important in the Cr(0.5)–MnOx catalyst compared
to CrOx, MnOx and CrOx–MnOx.
Fig. 2. XRD patterns of CrOx, MnOx, CrOx–MnOx and Cr–MnOx catalysts: (a) CrOx; (b)
MnOx; (c) CrOx–MnOx; (d) Cr(0.5)–MnOx.
a new crystalline phase [22]. Fig. 3 shows the XRD patterns of
Cr(0.5)–MnOx calcined at different temperatures. It is clear that
the XRD peaks of all catalysts calcined at various temperatures
could all be plausibly attributed to CrMn1.5O4 and the intensities
of the XRD peaks were sharper with increasing calcination temper-
ature. The SEM images of Cr(0.5)–MnOx catalysts calcined at differ-
ent temperatures are also shown in Fig. 4. All the above
phenomena match with the increasing crystallinity of the materi-
als (Fig. 3). Based on the XRD patterns and SEM images, the crystal-
linity of the materials was found to be significantly improved when
the calcination temperature was raised from 550 to 700 °C.
The effect of calcination temperature on the activities of
Cr(0.5)–MnOx mixed-oxide catalysts is shown in Fig. 5. It shows
clearly that the SCR activity depends on calcination temperature.
NOx conversions reached nearly 100% on every catalyst at above
140 °C. It is noted that the Cr(0.5)–MnOx calcined at 650 °C gave
the optimum activity at lower temperature. The slight decrease
in activity at 700 °C was attributed to lower surface area of the
crystalline material due to sintering (Table 1). It was also found
that the BET surface areas of the samples continuously decreased
upon the increase in calcination temperature, while the conversion
first increased before it was dropped to lower values. It should be
noted that the BET surface area is not the only parameter affecting
activity but also the phase composition. Thus, there must be a com-
plex interplay between surface area and phase composition affect-
3.2. Effect of calcination temperature
For the mixed oxides, the calcination temperature is important
to provide activation energy to reorganize the mixed materials into
Fig. 1. SCR activity of the CrOx, MnOx, CrOx–MnOx and Cr(0.5)–MnOx. Reaction
conditions: [NO] = [NH3] = 1000 ppm, [O2] = 3%, GHSV = 30,000 hꢁ1, catalysts cal-
cined at 650 °C.
Fig. 3. XRD patterns of the Cr(0.5)–MnOx mixed-oxide catalysts calcined at
different temperatures: (a) 550 °C; (b) 600 °C; (c) 650 °C; (d) 700 °C.

Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
59
Fig. 4. SEM images of Cr(0.5)–MnOx calcined at different temperatures: (a) Cr(0.5)–MnOx(550); (b) Cr(0.5)–MnOx(600); (c) Cr(0.5)–MnOx(650); (d) Cr(0.5)–MnOx(700).
3.3. Effect of the composition of Cr–MnOx catalyst
A series of Cr–MnOx catalysts with different molar ratios of Cr/
(Cr + Mn) were prepared, calcined at 650 °C and their catalytic per-
formances investigated (Table 2). The data show that NOx conver-
sion was close to 100% for temperatures above 160 °C for each Cr–
MnOx catalyst. Below 160 °C, there are significant differences in
activities among these catalysts: catalytic activity increases with
increasing chromium content and the optimum catalytic activity
peaked at Cr(0.4)–MnOx, in which the molar ratio of Cr to Mn is
consistent with the stoichiometry of CrMn1.5O4 (Cr:Mn = 2:3). At
this ratio, we obtain 80.9% and 98.5% conversions of NOx at 100
and 120 °C, respectively, at a space velocity of 30,000 hꢁ1. Any fur-
ther increase in Cr content results in a decline in activity. Fig. 6
shows the XRD patterns of these Cr–MnOx catalysts. Cr(0.1)–MnOx
could be considered as mixed oxides of Mn3O4 and MnO (ICDD PDF
#89-4835 2h = 40.6°, 58.7° and 34.9°), with Mn3O4 as the major
Fig. 5. SCR activity of Cr(0.5)–MnOx catalyst calcined at different temperatures:
component. With increasing Cr content, the XRD peak of the
CrMn1.5O4 phase increased at the expense of the Mn3O4 and MnO
phases. For Cr(0.4)–MnOx, the CrMn1.5O4 phase was dominant with
MnO phase also present (Fig. 6d). The Cr(0.5)–MnOx catalyst gave
the same sharp XRD peaks of the CrMn1.5O4 phase (Fig. 6e) as that
of Cr(0.4)–MnOx, but without any evidence of a MnO phase. At this
Cr:Mn ratio, we conclude that all the manganese atoms were in the
form of CrMn1.5O4. At the same time, no chromic oxide phase was
detected by our XRD. From the theoretical viewpoint, the content
of the CrMn1.5O4 phase in the Cr(0.4)–MnOx catalyst would be
higher than that of Cr(0.5)–MnOx, as judged by the stoichiometry
of CrMn1.5O4. By inspection of the surface properties of these cata-
lysts (Table 3), the formation of CrMn1.5O4 seemed to be associated
with an increase in the surface area of the materials. When the mo-
lar ratio of Cr/(Cr + Mn) reached the stoichiometry of the CrMn1.5O4
phase, the pore volume and average pore diameter appeared to de-
crease. The SEM micrographs of Cr–MnOx catalysts (Fig. 7) clearly
reflected the phase change with the increasing content of Cr. For
Reaction conditions: [NO] = [NH3] = 1000 ppm, [O2] = 3%, GHSV = 30,000 hꢁ1
.
Table 1
Effect of calcination temperature on surface area and pore characterization.
Entry Catalysts
BET surface
area (m2 gꢁ1
Pore volume
Average pore
diameter (nm)
)
(cm3 gꢁ1
0.046
)
1
2
3
4
Cr(0.5)–
MnOx (550)
Cr(0.5)–
MnOx (600)
Cr(0.5)–
MnOx (650)
Cr(0.5)–
65.3
60.6
56.1
40.9
3.90
0.045
0.049
0.045
3.83
3.78
3.65
MnOx (700)
ing the present overall activity trend with respect to calcination
temperature.

60
Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
Table 2
NOx conversions of various Cr–MnOx catalysts doped at differing Cr levels.a
Catalysts
NOx conversion (%)
80 °C
100 °C
120 °C
140 °C
160 °C
180 °C
200 °C
220 °C
Cr(0.1)–MnOx
Cr(0.2)–MnOx
Cr(0.3)–MnOx
Cr(0.4)–MnOx
Cr(0.5)–MnOx
31.8
35.1
45.9
48.5
46.3
52.7
59.7
74.8
80.9
77.6
79.1
88.0
97.1
98.5
97.8
94.6
98.4
98.8
99.2
98.8
98.5
98.8
98.9
99.5
98.9
98.7
99.0
99.1
99.7
99.0
98.8
99.0
99.2
99.7
99.1
98.8
99.0
99.2
99.7
99.1
a
Reaction conditions: [NO] = [NH3] = 1000 ppm, [O2] = 3%, GHSV=30,000 hꢁ1
very high (NOx conversion at 99%) and still remained at 91.7% with
100% N2 selectivity after 500 h of a continuous lifetime test. This
catalyst can be reactivated easily, as NOx conversion was recovered
to the same level comparable to a fresh sample, with the selectivity
still maintained at 100% after regeneration. This clearly suggests
that the Cr(0.4)–MnOx displays a satisfactory degree of stability
and regeneration properties in our laboratory scale work.
It is difficult to remove all of the SOx after any desulfurization
process, and thus, traces of SO2 may still exist (usually about
100 ppm) and affect the denitrogen catalysis downstream. The
SO2 resistance of the Cr(0.4)–MnOx(650) catalysts was therefore ta-
ken into consideration in comparison with that of MnOx(650)
(Fig. 8). It indicates that the NOx conversion on Cr(0.4)–MnOx de-
creases slowly, losing 15% of its activity in 4 h when 100 ppm
SO2 was added to the reaction gas stream under 120 °C with a total
space velocity of 30,000 hꢁ1. Nevertheless, it still shows a very high
level of activity for NOx conversion. Interestingly, the activity was
rapidly restored to 96% when the SO2 flow was stopped. However,
the NOx conversion on MnOx decreases sharply, and most of its
activity (about 84%) is lost in 4 h only 50% of its original activity
could be restored, while SO2 was stopped. It can be concluded that
this new Cr(0.4)–MnOx catalyst maintains a good de-NOx activity
in a high sulfur content gas stream. Although the sulfur-tolerance
of the Cr(0.4)–MnOx catalyst needs to be improved, its low-tem-
perature SCR activity was much higher than the MnOx catalyst in
the presence of SO2. It should be cautious to note that these exper-
iments were conducted in the short time exposure of SO2. The ef-
fects of continuous feeding SO2 at prolonged period of time were
not studied. In addition, the present results were obtained from
using our small-scale laboratory testing and the industrial applica-
bility of our new catalysts is not yet known at this stage.
Fig. 6. XRD patterns of the Cr–MnOx catalysts doped at a range of Cr content: (a)
Cr(0.1)–MnOx; (b) Cr(0.2)–MnOx; (c) Cr(0.3)–MnOx; (d) Cr(0.4)–MnOx; (e) Cr(0.5)–
MnOx.
Table 3
Surface area and pore characterization of catalysts.
Entry Catalysts
BET surface
area (m2 gꢁ1
Pore volume
Average pore
diameter (nm)
)
(cm3 gꢁ1
)
1
2
3
CrOx (650)
MnOx (650)
CrOx–MnOx
(650)
29.9
19.2
22.6
0.127
0.112
0.108
14.1
21.2
16.9
4
5
6
7
8
Cr(0.1)–
MnOx (650)
Cr(0.2)–
MnOx (650)
Cr(0.3)–
MnOx (650)
Cr(0.4)–
60.4
66.5
86.8
70.4
56.1
0.097
0.097
0.082
0.053
0.049
11.0
13.0
7.80
4.52
3.78
3.5. H2-TPR analysis
MnOx (650)
Cr(0.5)–
The H2-TPR patterns of the catalysts are shown in Fig. 9. CrOx is
reduced at 233 °C due to the reduction of Cr2O3 to CrO [23,24];
MnOx shows two stages of reduction at 187 and 439 °C, respec-
tively, and Mn2O3 reduction to Mn3O4 followed by Mn3O4 reduc-
tion to MnO [25]. The TPR profile of CrOx–MnOx is that of the
combination of two physically mixed oxides, but interaction be-
tween CrOx and MnOx is also evident, as the reduction temperature
of MnOx in CrOx–MnOx is clearly shifted to lower temperature and
that of CrOx in CrOx–MnOx to higher temperature. Both Cr(0.4)–
MnOx and Cr(0.5)–MnOx mixed-oxide catalysts, on the other hand,
show nearly the same reduction behaviors, which are quite differ-
ent from those of pure oxides. With comparison with pure oxides
and combined with the fact that the reduction peak around
240 °C increases with the increasing Cr content, peak temperatures
around 240 and 340 °C can be attributed to the reduction of Cr–O
bonds and Mn–O bonds in CrMn1.5O4 phases, respectively. Never-
theless, the reduction of the mixed-oxide phase was clearly facili-
tated by the significant peak shift; for example, the reduction
temperature of Cr–O bonds is a little higher than that of CrOx, while
MnOx (650)
example, Cr(0.1)–MnOx was composed mainly of small particles of
an amorphous phase (Fig. 7a), while a more crystalline phase ap-
peared in the Cr(0.2)–MnOx; at this point, the microcrystalline
component increased and became dominant with the increase in
Cr, which also matched well with the results of XRD characteriza-
tion (Fig. 6).
3.4. Lifetime study and SO2 resistance of Cr(0.4)–MnOx
As the above, Cr(0.4)–MnOx catalyst showed very good NOx SCR
activity, its lifetime and regeneration properties were investigated.
The catalyst was first operated for 500 h and then generated by
normal temperature and pressure plasma. Experimental results
(Fig. 8) showed that the initial SCR activity of Cr(0.4)–MnOx was

Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
61
Fig. 7. SEM images of the catalysts: (a) Cr(0.1)–MnOx(650); (b) Cr(0.2)–MnOx(650); (c) Cr(0.3)–MnOx(650); (d) Cr(0.4)–MnOx(650).
Fig. 8. Lifetime and sulfur-tolerance study of Cr(0.4)–MnOx catalyst at 120 °C.
Reaction conditions: [NO] = [NH3] = 1000 ppm, [O2] = 3%, [SO2] = 100 ppm,
GHSV = 30,000 hꢁ1
.
the reduction temperature of Mn–O bonds is at least 100 °C lower
than that of MnOx.
According to the ICDD, Mn3O4 has a body-centered tetragonal
structure, Cr2O3 is hexagonal and CrMn1.5O4 is face-centered cubic.
The crystalline structures of Mn3O4, MnO, Cr2O3 and CrMn1.5O4 are
shown in Fig. 10, and the lengths of all metal–oxygen bonds are
compiled in Table 4. There are three different types of Mn–O bonds
in Mn3O4 (MnOx), with bond lengths of 2.0142, 1.9475 and
2.2922 Å. Two types of Cr–O bond exist in the Cr2O3 (CrOx) with
bond lengths of 2.0367 and 1.9458 Å. As for the unit cell of
CrMn1.5O4, we can see that metal atoms (Cr or Mn) are connected
through oxygen atoms in the form of Cr(Mn)–O–Mn in the
CrMn1.5O4 lattice. The interatomic connections are mainly in the
Fig. 9. TPR profiles of the catalyst samples.
form of Cr–O–Mn and Mn–O–Mn, the Cr atom is connected tetra-
hedrally to four oxygen atoms, while the Mn atom is connected
octahedrally to six oxygen atoms. It is noted that in the CrMn1.5O4
lattice, the length of the Mn–O bond (2.3509 Å) is greater than that
of the Mn–O bonds in MnOx, but that the bond length of the Cr–O
bond (1.4686 Å) is less than those of the Cr–O bonds in CrOx.

62
Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
Fig. 10. Structures of Mn3O4, MnO, Cr2O3, CrMn1.5O4 crystalline and CrMn1.5O4 unit cell: (A) Mn3O4; (B) MnO; (C) Cr2O3; (D) CrMn1.5O4; (E) CrMn1.5O4 unit cell. Purple ball:
Mn; Red ball: O; blue ball: Cr. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).
It had been established in the literature that a shift in the peak
position of the reduction temperature can be attributed to many
factors, such as change in particle size, lattice oxygen mobility,
Table 4
M–O bond lengths in the manganese–chromium–oxide system.
Crystal phase
Mn3O4
Bond
Bond length (Å)
phase composition or structural defects [26]. The change in reduc-
tion temperature also reflects the variation in reduction potential
of oxidized metallic species in CrMn1.5O4 by comparison with those
of corresponding monometallic oxide catalysts. The above TPR pat-
terns clearly reflect the changes in the chemical environments of
Mn and Cr (i.e. bond energy change); for example, the reduction
temperature of manganese in CrMn1.5O4 is lower than that of pure
MnOx owing to the longer Mn–O distance, and similarly the in-
crease in reduction temperature of chromium with respect to that
of CrOx reflects the shortening of the Cr–O bond.
Mn–O
Mn–O
Mn–O
2.2922
1.9475
2.0142
MnO
Mn–O
2.2215
Cr2O3
Cr–O
Cr–O
2.0367
1.9458
CrMn1.5O4
Mn–O
Cr–O
2.3509
1.4686
3.6. Analysis of Raman spectra
The Raman spectra of pure CrOx, MnOx and Cr–MnOx catalysts
are shown in Fig. 11, and their detailed Raman shifts are listed in
Table 5. The spectrum of CrOx shows three Raman bands peaked
at 301.5, 336.9 and 533.6 cmꢁ1 (Fig. 11a). These bands are attrib-
uted to Cr2O3 [27–29]. For MnOx, the broad Raman shift peak at
642.7 cmꢁ1 (Fig. 11b) is characteristic of Mn3O4 [30,31] and repre-
sents the single Mn–O bond vibration. These results were in agree-
ment with pure chromium and magnesium oxide phases detected
by XRD (Fig. 2). Although the characteristic Raman band of the
Mn–O bond in MnO lies at 537 cmꢁ1 [32], it may not be detected
by this technique, as MnO can be easily transformed to Mn3O4 un-
der a laser beam [33].
It can be seen from Fig. 11 that the Raman peak at around 540
and 640 cmꢁ1 increases in intensity with increasing Cr content.
Hence, these bands can be assigned to the Raman shift of Cr–O
and Mn–O bond in the form of Cr–O–Mn, respectively, as the
CrMn1.5O4 phase increases with the increasing of Cr content proved
by the XRD data (Fig. 6) and SEM observations (Fig. 7). These peaks
are quite different from those of pure CrOx and MnOx (533.6 and
642.7 cmꢁ1, respectively). The changes in crystal structure and in
the bond lengths are responsible for the shifts of the Raman peaks
[34]. For example, not only the peaks at 301.5, and 336.9 cmꢁ1
(attributed to CrOx) disappeared, but also another sharp Raman
band due to CrOx (located at 533.6 cmꢁ1) was replaced by a flat
peak around 539–549 cmꢁ1. This wavenumber shift could be
attributed to a decrease in the bond length of Cr–O in CrMn1.5O4
relative to those of CrOx (Table 4); Accordingly, the Raman shift
of the Mn–O bond in the mixed oxide is found at lower wavenum-
ber than that of Mn–O in MnOx as the length of Mn–O in CrMn1.5O4
is longer (than that in MnOx; Table 4). When the Cr content ex-
ceeded the stoichiometry of the CrMn1.5O4 phase (for Cr(0.5)–
MnOx, Table 5, entry 7), the Raman spectra reflected the mixture
Fig. 11. Raman spectra of catalysts. (a) CrOx; (b) MnOx; (c) Cr(0.1)–MnOx; (d)
Cr(0.2)–MnOx; (e) Cr(0.3)–MnOx; (f) Cr(0.4)–MnOx; (g) Cr(0.5)–MnOx. Crude line:
original data; Smooth line: fitting line.

Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
63
Table 5
this phase is perfect and stable with a large number of Cr(Mn)–
O–Mn bonds (Fig. 10); therefore, we could draw a conclusion that
the maxima at 539.3 and 641.7 cmꢁ1 can be assigned as Raman
bands of Cr–O and Mn–O bonds in the form of Cr(Mn)–O–Mn
respectively. Furthermore, bands of Cr6+ in chromium oxide were
confirmed to be in the range 800–1000 cmꢁ1 [35,36], which were
not observed in our patterns. It is well known that there are many
other parameters (sample homogeneity, size of crystalline phase,
purity, etc.) that affect the Raman shift value. It should be cautious
that there is no simple linearity relationship with respect to
increasing Cr content.
Raman shift data of oxide catalysts.
Entry
Catalysts
Raman shift (cmꢁ1
Cr–O
)
Mn–O
1
2
3
4
5
6
7
CrOx
MnOx
533.6
–
–
642.7
646.1
640.3
640.1
641.7
644.6
Cr(0.1)–MnOx(650)
Cr(0.2)–MnOx(650)
Cr(0.3)–MnOx(650)
Cr(0.4)–MnOx(650)
Cr(0.5)–MnOx(650)
549.1
544.2
541.1
539.3
532.5
of Cr–O bonds (in the form of CrOx) and Cr–O–Mn (in the form of
CrMn1.5O4). Hence, the variations of the structure and chemical
state of the mixed catalyst are readily shown by this Raman inves-
tigation, which also matches well with the above characterization
and theoretical considerations. For Cr(0.4)–MnOx, the molar ratio
of Cr to Mn is equal to the stoichiometry of CrMn1.5O4 phase, and
3.7. Oxidation activities of NO to NO2 on CrMn1.5O4
From above results, the most active catalyst for low-tempera-
ture NOx reduction with NH3 should be the CrMn1.5O4 phase with
suitable surface properties. The in-depth investigation of the oxi-
dation activities of NO to NO2 on catalysts doped with various ra-
tios of Cr/(Cr + Mn) at 80–160 °C are shown in Fig. 12. With
increase of chromium in the Cr–MnOx catalysts, the NO oxidation
activity improved significantly and reached its maximum with
Cr(0.4)–MnOx. The direct catalytic reaction of NO and NH3 to N2
is thought to be difficult and it is well known that ammonium ni-
trite decomposes quickly to give N2 below 100 °C [37]. The forma-
tion of NH4NO2 from NO, O2 and NH3 requires a prior oxidation of
NO to NO2 by O2. A highly active catalyst for the oxidation of NO to
NO2 at low temperature, which also gave enhanced SCR activities,
has been reported [38,39]. We propose that the beneficial effect of
chromium incorporation in Mn as CrMn1.5O4 is because this phase
could oxidize NO to NO2 at low temperatures, thereby enhancing
the SCR activity. In addition, it is interesting to observe that the
oxidation of NO to NO2 conversion (Fig. 12) is at least 10 times low-
er than the NO conversion during SCR process (Table 2). It is now
widely accepted that the SCR efficiency can be significantly im-
proved with increasing NO2 amount due to a faster reaction rate
between NO2 and adsorbed NH3 [40]. It is also believed that in
the SCR process, the presence of the adsorbed NH3 can continu-
ously mob up the adsorbed NO2, in situ formed from NO oxidation
on catalytic surface. This would dramatically speed up the NO to
NO2 conversion when compared to the pure NO oxidation.
Fig. 12. Oxidation activity of NO to NO2 on Cr–MnOx catalysts. Reaction conditions:
[NO] = 1000 ppm, [O2] = 3%, and GHSV = 30,000 hꢁ1
.
Fig. 13. XPS spectra for (A) Mn 2p, (B) Cr 2p, and (C) O 1s of the Cr(0.4)–MnOx catalysts. (a) Fresh catalyst; (b) used catalyst; (c) regenerated catalyst.

64
Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
3.8. XPS analysis and a possible redox reaction over CrMn1.5O4
The binding energies (eV) of Mn, Cr and O of Cr(0.4)–MnOx be-
fore testing (fresh) and after testing (used) and plasma regenera-
tion (regenerated) are shown in Table 7. It was found that the
relative surface concentration of the three species did not change
much, but that their relative oxidation states varied significantly.
After 500 h of SCR reaction, the concentrations of relatively lower
valence state chromium ions (Cr2+ and Cr3+) increased at the ex-
pense of higher valence state chromium (Cr5+), but the higher va-
lence manganese state (Mn4+) increased, while Mn3+ decreased.
Plasma treatment of the spent catalysts has the advantage of not
changing the structure and morphological properties. It can be
seen that the relative high surface concentrations of Mn3+ and
Cr5+ recovered matching with the recovery in the activity of
Cr(0.4)–MnOx. So, the above partial loss in activity of catalyst after
500 h of life test might be due to the decline in Mn3+ and Cr5+ sur-
face concentrations. Hence, it can be concluded that both Mn3+ and
Cr5+ must play an important role in this SCR process.
To investigate the surface chemical states of the most active
catalyst, the XPS spectra of the Mn 2p3/2, Cr 2p3/2 and O 1s in fresh,
used (catalyst quenched from 500 h operation) and regenerated
Cr(0.4)–MnOx(650) catalyst (catalyst operated 500 h and then trea-
ted by plasma before quenched for analysis) were obtained, as
shown in Fig. 13. Two main peaks due to Mn 2p3/2 (peak at
643 eV) and Mn 2p1/2 (peak at 653 eV) were observed (Fig. 13A).
The measured binding energies of Mn 2p3/2 in Cr(0.4)–MnOx(650)
catalyst (640ꢅ645 eV) were slightly higher than those reported
for MnO, Mn2O3 and MnO2 [41], which shows a nice distinction
in the change in chemical environment between the CrMn1.5O4
phase and single MnOx, and also fits well with the result discussed
above. By performing a peak-fitting deconvolution, the Mn 2p3/2
spectra can be separated into three peaks: 640.4–640.5, 641.9–
642.2 and 644.5–644.8 eV, respectively. There are quite close to
the previously reported Mn2p3/2 binding energies of Mn2+, Mn3+
and Mn4+ in MnOx–CeO2 mixed oxides with 640.3–640.7, 641.6–
642.3 and 643.2–644.5 eV, respectively [42]. The Cr 2p3/2 of
Cr(0.4)–MnOx(650) catalyst was separated into two peaks by the
same peak-fitting deconvolution technique. The ‘‘low valence” CrLV
occurred at about 576 eV and the ‘‘high valence” CrHV was charac-
terized by an average binding energy of about 578.3–578.5 eV. The
former peaks can be assigned to Cr2+ (575.6–575.7 eV) and Cr3+
(576.6–576.7 eV) [43,44]. The second peak (578.3–578.5 eV)
should be assigned to a higher valence chromium state, perhaps
Cr5+ or Cr6+. It has been reported that binding energy of Cr5+ has
a value in the range 578.0–578.8 eV [45,46], and hence, peaks with
high oxidation state in the range of 578.3–578.5 eV could be as-
signed to Cr5+ (Fig. 13B). It has also been reported that compounds
containing Cr6+ exhibited a higher XPS band energy, at 579–580 eV
[47,48], which we did not observe. Together with the above Raman
shift analysis (Fig. 11), we conclude that the chromium oxidation
state in the Cr(0.4)–MnOx(650) catalyst is mainly Cr5+ rather than
To summarize, we observe (1) changes in oxidation states of Mn
and Cr after 500 h of operation and also after plasma generation;
Table 7
Binding energies (eV) of core electrons of Cr(0.4)–MnOx and the percent of differential
valence state.a
XPS spectra
Element
valence
Binding energies (percent of valence
state, %)
Fresh
catalyst
Used
catalyst
Regenerated
catalyst
Cr 2p (eV)
concentration (%)
Cr2+
Cr3+
Cr5+
575.6
(13.9)
576.7
(42.2)
578.4
(43.9)
575.7
(16.8)
576.7
(49.9)
578.3
(33.3)
575.7 (19.7)
576.6 (38.0)
578.5 (42.3)
Mn 2p (eV)
concentration (%)
Mn2+
Mn3+
Mn4+
640.4
(14.6)
641.9
(54.2)
644.6
(31.2)
640.5
(15.5)
641.9
(46.9)
644.5
(37.6)
640.5 (16.0)
642.2 (60.6)
644.8 (23.4)
Cr6+
.
Table 6 lists the atom percentage of catalysts determined by
XPS. The value of Cr/(Cr + Mn) characterized by XPS was higher
than the overall molar ratio of Cr(0.1)–MnOx and Cr(0.2)–MnOx
and similar to Cr(0.2)–MnOx, and Cr(0.3)–MnOx, but lower than
overall molar ratio for Cr(0.4)–MnOx and Cr(0.5)–MnOx. Notably,
the concentration of manganese on the surface of the Cr(0.4)–
O 1s (eV)
concentration (%)
O2ꢁ
529.8
(71.0)
531.6
(29.0)
529.8
(73.9)
531.7
(26.1)
529.9 (72.2)
531.8 (27.8)
OHꢁ/CO23ꢁ
MnOx catalyst reached
a constant value. The formation of
a
Surface concentration of different Mn, Cr and O states are in parenthesis.
CrMn1.5O4 would account for this. As shown in Fig. 13C, an asym-
metric peak was observed in the XPS spectra of O 1s for both fresh,
used and regenerated samples. The peak at 529.6–530.0 eV corre-
sponds to lattice oxygen (O2ꢁ
) whereas the one at 531.3–
531.7 eV corresponds to several O1s states assigned to the sur-
face-adsorbed oxygen such as O22ꢁ or O-, in the form of hydroxyl,
OHꢁ and carbonate, CO23ꢁ [49]. Values of Cr/(Cr + Mn) on the sur-
face of used and regenerated catalysts are higher than that of fresh
catalyst, which implies that chromic oxide tends to migrate to the
surface.
Table 6
Atom percentage of catalysts determined by XPS.
Catalyst
Cr (%)
Mn (%)
O (%)
Cr/(Cr + Mn)
Cr(0.1)–MnOx
Cr(0.2)–MnOx
Cr(0.3)–MnOx
Cr(0.4)–MnOx
Cr(0.5)–MnOx
Cr(0.4)–MnOx (used)
Cr(0.4)–MnOx (regenerated)
3.5
3.6
4.6
10.9
12.8
14.2
13.7
21.0
13.6
11.4
20.3
17.2
21.0
18.6
75.5
82.8
84.0
68.8
70.0
64.8
66.7
0.14
0.21
0.29
0.35
0.43
0.40
0.42
Scheme 1. A redox catalytic cycle of the low-temperature SCR reactions over
Cr(0.4)–MnOx catalysts.

Z. Chen et al. / Journal of Catalysis 276 (2010) 56–65
65
[3] Y. Li, H. Cheng, D. Li, Y. Qin, Y. Xie, S. Wang, Chem. Commun. 12 (2008) 1470.
[4] X. Li, P. Zhu, F. Wang, L. Wang, S.C. Tsang, J. Phys. Chem. C 112 (2008) 3376.
[5] P.J. Smeets, M.H. Groothaert, R.M. Teeffelen, H. Leeman, E.J.M. Hensen, R.A.
Schoonheydt, J. Catal. 245 (2007) 358.
[6] J. Zhu, D. Xiao, J. Li, X. Yang, Y. Wu, J. Mol. Catal. A 234 (2005) 99.
[7] J.V. Durme, J. Dewulf, C. Leys, H.V. Langenhove, Appl. Catal. B 78 (2008) 324.
[8] J.L. Hueso, J. Cotrino, A. Caballero, J.P. Espinós, A.R. González-Elipe, J. Catal. 247
(2007) 288.
[9] G.L. Bauerle, S.C. Wu, K. Nobe, Ind. Eng. Chem. Prod. Res. Dev. 14 (1975) 268.
[10] V.M. Mastikhin, V.V. Terkikh, O.B. Lapina, O.B. Filimonova, M. Seial, H.
Knözinger, J. Catal. 156 (1995) 1.
[11] R. Ke, J. Li, X. Liang, J. Hao, Catal. Commun. 8 (2007) 2096.
[12] M. Kang, E.D. Park, J.M. Kim, J.E. Yie, Appl. Catal. A 327 (2007) 261.
[13] Z. Wu, B. Jiang, Y. Liu, Appl. Catal. B 79 (2008) 347.
(2) apparent gain of electrons by Cr and loss by Mn in the quenched
samples (3) that the concentration of O2ꢁ changes little (Table 6)
after 500 h of operation and regeneration. We therefore propose
electron transfer between Cr and Mn ions. For the CrMn1.5O4 phase,
most of the Cr and Mn species are interconnected in the form of
Cr–O–Mn through oxygen bridges, facilitating electron transfer. A
redox electron transfer between different ions (from Mn3+ and
Cr5+) through the oxygen bridge (Eq. (5)) is probably similar to
the electron transfer between copper and manganese in the CuM-
n2O4 system, previously described [16,20,50]. Thus, we propose a
redox cycle for low-temperature SCR over this new type of cata-
lysts as depicted in Scheme 1.
[14] P.G. Smirniotis, D.A. Peña, B.S. Uphade, Angew. Chem. Int. Ed. 40 (2001) 2479.
[15] G. Qi, R.T. Yang, Appl. Catal. B 44 (2003) 217.
[16] M. Kang, E.D. Park, J.M. Kim, J.E. Yie, Catal. Today 111 (2006) 236.
[17] X. Tang, Y. Li, X. Huang, Y. Xu, H. Zhu, J. Wang, W. Shen, Appl. Catal. B 62 (2006)
265.
Cr5þ þ 2Mn3þ ! Cr3þ þ 2Mn4þ
ð5Þ
[18] G. Qi, R.T. Yang, Chem. Commun. 7 (2003) 848.
[19] P.M. Sreekanth, D.A. Peña, P.G. Smirniotis, Ind. Eng. Chem. Res. 45 (2006) 6444.
4. Conclusions
ˇ
[20] S. Veprek, D.L. Cocke, S. Kehl, H.R. Oswald, J. Catal. 100 (1986) 250.
[21] X.L. Tang, J.M. Hao, W.G. Xu, J.H. Li, Chin. J. Catal. 27 (2006) 843.
[22] D.A. Peña, B.S. Uphade, P.G. Smirniotis, J. Catal. 221 (2004) 421.
[23] A. Hakuli, M.E. Harlin, L.B. Backman, A.O.I. Krause, J. Catal. 184 (1999) 349.
Novel Cr–Mn mixed oxides with high activity for the low-tem-
perature SCR of NOx with ammonia in the presence of oxygen were
developed. Greater than 98.5% of NOx conversion was obtained
over the Cr(0.4)–MnOx (650) catalyst, prepared by the citric acid
´
[24] B. Grzybowska, J. Słoczynski, R. Grabowski, K. Wcisło, A. Kozłowska, J. Stoch, J.
Zielin´ ski, J. Catal. 178 (1998) 687.
[25] S. PalDey, S. Gedevanishvili, W. Zhang, F. Rasouli, Appl. Catal. B 56 (2005) 241.
[26] S. Bhatia, J. Beltramini, D.D. Do, Catal. Today 7 (1990) 309.
[27] A.Y. Kuznetsov, J.S. Almeida, L. Dubrovinsky, R. Ahuja, S.K. Kwon, I. Kantor, N.
Guignot, J. Appl. Phys. 99 (2006) 053909.
[28] D. Stanoi, G. Socol, C. Grigorescu, F. Guinneton, O. Monnereau, L. Tortet, T.
Zhang, I.N. Mihailescu, Mater. Sci. Eng. B 118 (2005) 74.
[29] H.C. Barshilia, N. Selvakumar, K.S. Rajam, J. Appl. Phys. 103 (2008) 023507.
[30] M. Ludvigsson, J. Lindgren, J. Tegenfeldt, J. Mater. Chem. 11 (2001) 1269.
[31] F. Kapteijn, A.D. Langeveld, J.A. Moulijn, A. Andreïni, M.A. Vuurman, A.M.
Turek, J. Jehng, I.E. Wachs, J. Catal. 150 (1994) 94.
[32] H.-H. Chou, H.Y. Fan, Phys. Rev. B 13 (1976) 3924.
[33] F. Buciuman, F. Patcas, R. Craciun, D.R.T. Zahn, Phys. Chem. Chem. Phys. 1
(1999) 185.
[34] H. Wan, D. Li, H. Zhu, Y. Zhang, L. Dong, Y. Hu, B. Liu, K. Sun, L. Dong, Y. Chen, J.
Colloid Interface Sci. 326 (2008) 28.
[35] S. Gómez, L.J. Garces, J. Villegas, R. Ghosh, O. Giraldo, S.L. Suib, J. Catal. 233
(2005) 60.
method at 120 °C, under flow conditions of GHSV = 30,000 hꢁ1
.
From the XRD patterns, the spinel CrMn1.5O4 was found to be pres-
ent in active Cr–Mn mixed-oxides catalysts. The TPR profiles are
revealed for the first time the reduction process of CrMn1.5O4 and
the existence of the CrMn1.5O4 phase clearly depressed the reduc-
tion temperature of manganese oxides. Raman peaks at around
539.3 and 641.7 cmꢁ1 can be assigned to the characteristic Raman
bands of the CrMn1.5O4 lattice phase, a new phase to the best of our
knowledge. Better oxidation of NO could be ascribed to the forma-
tion of CrMn1.5O4, with confirmation of electron transfer between
chromium and manganese by XPS. The addition of SO2 in the fee-
der gas could exert an adverse effect on the NOx conversion, but
does not seem to create permanent structural alteration to the
Mn species (totally reversible sulfur poisoning), presumably due
to the existence of high redox potential pairs of Cr and Mn in the
structure, permitting efficient regeneration.
[36] Y. Han, F. Chen, Z. Zhong, K. Ramesh, L. Chen, E. Widjaja, J. Phys. Chem. B 110
(2006) 24450.
[37] F. Notoya, C. Su, E. Sasako, S. Nojima, Ind. Eng. Chem. Res. 40 (2001) 3732.
[38] M. Koebel, M. Elsener, G. Madia, Ind. Eng. Chem. Res. 40 (2001) 52.
[39] R.Q. Long, R.T. Yang, J. Catal. 198 (2001) 20.
[40] R.Q. Long, R.T. Yang, J. Catal. 207 (2002) 224.
[41] B.R. Strohmeier, D.M. Hercules, J. Phys. Chem. 88 (1984) 4922.
[42] G. Qi, R.T. Yang, J. Phys. Chem. B 108 (2004) 15738.
[43] B. Liu, H. Nakatani, M. Terano, J. Mol. Catal. A 184 (2002) 387.
[44] J. Sainio, M. Eriksson, J. Lahtinen, Surf. Sci. 532–535 (2003) 396.
[45] A. Trunschke, D.L. Hoang, J. Radnik, H. Lieske, J. Catal. 191 (2000) 456.
[46] N. Russo, D. Fino, G. Saracco, V. Specchia, J. Catal. 229 (2005) 459.
[47] G. Karamullaoglu, T. Dogu, Ind. Eng. Chem. Res. 46 (2007) 7079.
[48] D.L. Hoang, S. Farage, A. Dittmar, A. Trunschke, H. Lieske, A. Martin, Catal. Lett.
112 (2006) 173.
Acknowledgment
The National Natural Science Foundation of China (20876063) is
gratefully acknowledged for the financial support of this work.
References
[49] S. Ponce, M.A. Peña, J.L.G. Fierro, Appl. Catal. B 24 (2000) 193.
[50] F.C. Buciuman, F. Patcas, T. Hahn, Chem. Eng. Process. 38 (1999) 563.
[1] H. Schneider, U. Scharf, A. Wokaun, A. Baiker, J. Catal. 147 (1994) 545.
[2] A. Tomita, T. Yoshii, S. Teranishi, M. Nagao, T. Hibino, J. Catal. 247 (2007) 137.