Performance of co-doped Mn-Ce catalysts supported on cordierite for low concentration chlorobenzene oxidation

Article abstract of DOI:10.1016/j.apcata.2016.11.013

The catalytic activity for low concentration chlorobenzene oxidation was measured with Mn-Ce/cordierite and Mn-Co-Ce/cordierite catalysts with different mole ratios, which were prepared by a sol-gel method and characterized using XRD, BET, SEM Raman, H₂-TPR and XPS. The results demonstrated that part of the manganese and cobalt could be incorporated into the lattice of CeO₂ to form a solid solution phase. Among all of the catalysts synthesized, Mn₈Co₁Ce₁/cordierite presented the best activity and stability. When the concentration of chlorobenzene was 500 ppm, and the GHSV was 15000 h^?1, the complete combustion temperature (T_90%) of chlorobenzene was 325 °C. In addition, there was almost no change in the conversion of chlorobenzene during the long-term reaction at 350 °C. These results were primarily attributed to the synergistic effect of ceria, manganese and cobalt, which can promote the formation of more lattice defects, more oxygen vacancies and smaller crystallite sizes.

Full text of DOI:10.1016/j.apcata.2016.11.013

Applied Catalysis A: General 530 (2017) 21–29
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Performance of co-doped Mn-Ce catalysts supported on cordierite for
low concentration chlorobenzene oxidation
Jiawei Kana,b, Lei Denga,b, Bing Lia,b, Qiong Huangd, Shemin Zhua,c, Shubao Shena,b
,
Yingwen Chena,b,∗
a State Key Laboratory of Materials-Oriented Chemical Engineering, Nanjing Tech University, Nanjing 210009, China
b College of Biotechnology and Pharmaceutical Engineering, Nanjing Tech University, Nanjing 210009, China
c College of Materials Science and Engineering, Nanjing Tech University, Nanjing 210009, China
d Jiangsu Engineering Technology Research Center of Environmental Cleaning Materials, Nanjing University of Information Science and Technology, Nanjing
210044, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 August 2016
Received in revised form 3 November 2016
Accepted 10 November 2016
Available online 11 November 2016
The catalytic activity for low concentration chlorobenzene oxidation was measured with Mn-
Ce/cordierite and Mn-Co-Ce/cordierite catalysts with different mole ratios, which were prepared by a
sol-gel method and characterized using XRD, BET, SEM Raman, H2-TPR and XPS. The results demon-
strated that part of the manganese and cobalt could be incorporated into the lattice of CeO2 to form a solid
solution phase. Among all of the catalysts synthesized, Mn8Co1Ce1/cordierite presented the best activity
and stability. When the concentration of chlorobenzene was 500 ppm, and the GHSV was 15000 h−1, the
complete combustion temperature (T90%) of chlorobenzene was 325 ◦C. In addition, there was almost no
change in the conversion of chlorobenzene during the long-term reaction at 350 ◦C. These results were
primarily attributed to the synergistic effect of ceria, manganese and cobalt, which can promote the
formation of more lattice defects, more oxygen vacancies and smaller crystallite sizes.
Keywords:
Chlorobenzene
Mn-Ce/cordierite catalysts
Mn-Co-Ce/cordierite catalysts
Synergistic effect
Lattice defect
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
model for CVOCs because it is a precursor or intermediate product
of polychlorinated wastes, such as PCDD/PCDF [5,6].
In recent years, volatile organic compounds (VOCs) have
attracted considerable attention because they contribute the most
to air pollution, such as chemical smog and atmospheric haze,
which seriously affects the environment and the health of people.
Chlorinated volatile organic compounds (CVOCs), a type of VOCs,
are a special group due to their acute toxicity and persistency; they
are released to the atmosphere from the petrochemical, medical,
mechanical and painting industries [1,2]. Compared with other VOC
elimination techniques, catalytic combustion has been considered
as a promising technology owing to its high conversion efficiency
(at temperatures between 250 and 550 ◦C) and especially its lim-
ited secondary pollution and low energy consumption compared to
thermal combustion [3,4]. Chlorobenzene (CB) is a chlorinated con-
taminant from industrial processes that is frequently used as the
Generally, there are three categories of catalysts used for the oxi-
dation of CVOCs: noble metals [7–9], transition metals [10–12] and
zeolites [13–15]. Although noble metal catalysts show higher activ-
ity, they are unable to resist chlorine poisoning during the catalytic
combustion process [16,17]. Moreover, the high value and low
reserves of precious metal elements can limit their practical appli-
cation. The high activity that zeolites present is attributed to their
acidic properties; however, this catalyst system would generate
polychlorinated by-products that would result in secondary pol-
lution. Compared with the above-mentioned catalysts, transition
metal oxides have better thermal stability, a strong toxic resistance
and low cost, and they have been explored and used extensively
[18]. Among transition metals, Ce-containing transition metal com-
posite oxides have been considered as potential catalysts for CB
oxidation due to their high oxygen storage capacity, abundant oxy-
gen vacancies and strong redox property from the valence change
of Ce3+/Ce+4 [19,20]. However, the active sites of CeO2-based cat-
alysts are easily occupied due to the strong adsorption of HCl or
Cl2 produced from the degradation of CB. Despite the poor sta-
bility, some Ce-containing composite oxides have been explored
by many researchers, and some of these composite oxides showed
∗
Corresponding author at: State Key Laboratory of Materials−Oriented Chemi-
cal Engineering, College of Biotechnology and Pharmaceutical Engineering, Nanjing
Tech University, No. 5 XinMoFan Road, Nanjing 210009, China.
E-mail address: ywchen@njtech.edu.cn (Y. Chen).
http://dx.doi.org/10.1016/j.apcata.2016.11.013
0926-860X/© 2016 Elsevier B.V. All rights reserved.

22
J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
higher catalytic efficiency and stability, such as Mn-Ce [21], V-Ce [5]
and Cr-Ce [22] catalysts. Moreover, Chen et al. [23] discussed that
the Co-Cu-Mn mixed-oxide catalyst exhibited high catalytic activ-
ity in the oxidation of VOC binary mixtures and single components,
and the catalyst showed excellent stability during the 550 h long-
term reaction. The Co3O4/La-CeO2 catalyst was found to provide
more adsorbed species and lattice oxygen species in the combus-
tion of toluene and a mixture of toluene and dioxygen due to the
interaction between the Co3+/Co2+ and Ce4+/Ce3+ couples, which
resulted in an enhancement of the catalytic activity and the reaction
rate [24]. He et al. [25] reported that the nanostructured meso-
porous CuO-MnOx-CeO2 catalyst with a Cu/Mn atomic ratio of 1/1
presented the best catalytic efficiency of CB by providing of large
amounts of oxygen vacancies. However, the active component Co-
doped Mn-Ce for the catalytic combustion of CB was not discussed.
with a scanning rate of 5◦/min. The nitrogen adsorption and des-
orption isotherms were measured at −196.8 ◦C on a Tristar3020 M
Micromeritics apparatus. The samples were treated in vacuum at
80 ◦C for 8 h before the measurements. The specific surface area was
calculated using the BET model, and the pore size distribution and
average pore diameter were derived from the desorption branch of
the N2 isotherm using the Barrett-Joyner-Halenda (BJH) method.
Scanning electron microscopy (SEM) images were obtained on
a JSM-5900 instrument after the samples were pre-coated with
gold. Raman spectra were obtained on a DXR532 spectrometer at
ambient temperature and under moisture-free conditions. Raman
signals were produced with the 532 nm emission line from an Ar+
ion laser and a spot size of approximately 1 mm. The acquisition
time was varied according to the intensity of the Raman scattering,
and the wavenumbers obtained from the spectra were accurate to
within 0.1 cm−1. The H2-temperature programmed reduction (H2-
TPR) experiments were performed on an automated chemisorption
analyzer (Quantachrome Instruments) using 100 mg of the catalyst
under flowing (70 mL/min) hydrogen (10 vol.%) and nitrogen (90
vol.%). The temperature of the catalysts was increased at a rate
of 10 ◦C/min. A thermal conductivity detector was used to mon-
itor the hydrogen consumed during the TPR experiment. Before
the analysis, the sample was purged with helium (70 mL/min)
at 50 ◦C for 0.5 h. The X-ray photoelectron spectroscopy (XPS)
measurements were performed using a PHI-5000C ESCA system
(PerkinElmer) with Mg-Ka radiation (hv = 1253.6 eV) or Al-Ka radi-
ation (hv = 1486.6 eV) as the excitation source. In general, the X-ray
anode was operated at 250 W, and a high voltage was maintained
at 14.0 kV with a detection angle of 54◦. The sample was pressed
directly into a self-supported disk (10 × 10 mm), mounted on a
sample holder, and then transferred into the analysis chamber.
Meanwhile, the radius of Co3+ is between that of Mn3+ and Ce4+
,
which could lead to the formation of a ternary solid solution cat-
alyst. The doping of Co ions probably promotes the formation of
more lattice defects and oxygen vacancies and improves the oxy-
gen mobility. Therefore, Co-doped Mn-Ce/cordierite catalysts may
exhibit a high catalytic activity for chlorobenzene oxidation and
remove the Cl adsorbed on the surface of the catalysts. To enable
the industrial use and practical application of catalysts for removing
CB, supported non-noble catalysts, especially the Mn-Co-Ce cata-
lysts, should be investigated. Among a variety of catalyst supports,
cordierite is usually recognized as an excellent carrier by virtue of
its pore structure and high temperature resistance; however, it is
has seldom been reported as a catalyst support for the catalytic
combustion of CB. Hence, Co-doped Mn-Ce/cordierite catalysts are
worth researching in the catalytic combustion of chlorobenzene.
Consequently, the Co-doped Mn-Ce mixed-oxide catalysts sup-
ported on cordierite with different molar ratios of Mn-Ce and
Co-Mn-Ce were synthesized through the sol-gel method and were
tested in the catalytic combustion of low concentration CB. Charac-
terization of the supported catalysts was performed by XRD, BET,
SEM, Raman, H2-TPR and XPS.
2.3. Evaluation of catalytic activity
2. Experimental procedures
CB oxidation was carried out in a fixed-bed flow reactor with a
quartz tube with an inner diameter of 10 mm under atmospheric
pressure. Two grams of the catalyst was placed in the middle of a
quartz reactor erected in the center of a tubular furnace. The CB
stream was fed into the reaction system by bubbling air into the
liquid CB solution at 55 ◦C, and then it was mixed with another air
stream. To precisely simulate the relatively low concentration of
CVOCs from industrial processes, the feed flow through the reac-
tor was set with a CB concentration of 500 ppm and a gas hourly
space velocity (GHSV) of 15000 h−1 [27]. The reaction temperature
was controlled with a thermocouple in the range of 150–400 ◦C. In
addition, the pipeline was heated to 55 ◦C and kept at that tem-
perature to decrease the condensation and adsorption of CB on
the inner surface. The inlet and outlet gases were analyzed after
stepwise changes in the reaction temperature using an on-line
gas chromatograph (GC2014, Shimadzu Corp) equipped with a FID
detector and a Restek RTX-1 column for the quantitative analy-
sis of CB. The quality of CO and CO2 was analyzed by the on-line
gas chromatograph (GC2014, Shimadzu Corp) with a TCD detec-
tor. In addition, no organic polychlorinated benzene was produced
during the catalytic combustion of CB, according to the observa-
tion of the outlet chromatographic peak that only displayed the
CB peak. The concentration of both HCl and Cl2 was measured
by bubbling the exhaust gases through a 0.0125 mol·L−1 NaOH
solution. The Cl2 concentration in the bubbled solution was deter-
mined by chemical titration with ferrous ammonium sulfate (FAS)
using N,N-diethyl-p-phenylenediamine (DPD) as the indicator [28],
and the Cl−1 concentration was tested by using an ion-selective
2.1. Catalyst preparation
The cordierite support was modified by our team by boiling in
10 wt% nitric acid for 2 h, and then, it was dried at 80 ◦C for 2 h and
finally calcined under air atmosphere at 500 ◦C for 2 h [26].
The Co-doped Mn-Ce catalysts supported on cordierite were
manufactured using a sol-gel method, as follows: an aqueous solu-
tion containing Ce(NO3)3·6H2O (SCRC, 99%), Mn(NO3)2 (50 wt%
solution), Co(NO3)2·6H2O (SCRC, 99%) and citric acid (SCRC, 99%,
n(citric acid): n(Ce + Mn + Co) = 1:6) was gradually heated to 60 ◦C,
and then 4 g of pretreated cordierite was dipped into that solu-
tion and subsequently maintained at that temperature for 5 h under
continuous stirring to produce a sol. After the sol was aged for 12 h,
the compound was dried overnight at 110 ◦C and then calcined in
air at 500 ◦C for 5 h. Catalysts with 20 1% loading were obtained,
which included different atomic ratios (Mn: Co: Ce = 3:1:1, 8:1:1
and 2:2:1) based on the Mn-Ce/cordierite catalyst that was pre-
pared by the same method (Mn: Ce = 1:1, 2:1, 4:1). To increase the
readability, x, y and 1 in MnxCoyCe1/cordierite represent the molar
ratios of Mn, Co and Ce, respectively, in this paper.
2.2. Catalyst characterization
Powder X-ray diffraction (XRD) patterns were recorded on
an AXS D8 advance diffractometer equipped with a graphite
monochromator and Cu-Ka radiation in the 2␪ range from 5−80◦

J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
23
electrode [29]. The catalytic activity of CB and the CO2 selectivity
are calculated by the following equations:
small amount of organic byproducts, and then the organic byprod-
ucts are further oxidized by the active oxygen species to produce
CO2, H2O and HCl. A small amount of HCl may be attached to the
surface of the catalysts, and through this, some active sites may
be occupied. As the reaction temperature increased, the adsorbed
Cl could be removed by the combination of the Deacon Reaction
(4HCl + O2 → 2Cl2 + 2H2O) and active oxygen species. Finally, the
consumed oxygen species were replenished by the gas-phase oxy-
gen adsorbed on the oxygen vacancies. This cycle of oxygen species
was critical for the rapid elimination of Cl on the active sites.
C1Qsn1-C2Qsn2
ꢀ1
=
× 100%
(a)
C1Qsn1
where ␩ is the conversion efficiency, C1 and C2 (mg/m3) are the
1
inlet and outlet concentrations of chlorobenzene and Qsn1 and Qsn2
(m3/h) are the inlet and outlet fluxes of air, respectively.
nCO
2
SCO
=
× 100%
)
(b)
2
6 · n
(
CB
where nCO and nCB are moles of CO2 produced in the outlet gas and
3.2. Analysis of the products
the moles2of CB feed in the inlet gas, respectively.
With regards to FID/TCD detection, the Mn-Ce/cordierite and
Mn-Co-Ce/cordierite catalysts presented no polychlorinated by-
products and a selectivity of at least 99.6% to carbon oxides (more
than 99% to CO2 and trace CO). Moreover, the amount of the Cl
element of the outlet stream was 12%–15% lower than that of the
inlet stream, and the selectivity to the Cl2 produced during the cat-
alytic combustion of CB was less than 2%; the results indicated that
a small amount of chlorinated metals was formed on the surface of
the catalysts.
3. Results and discussion
3.1. Catalytic activity for CB combustion
The conversion of CB over different types of supported cat-
alysts synthesized is presented in Fig. 1. As shown in Fig. 1, the
Mn8Co1Ce1/cordierite catalyst exhibited the best activity under
the same reaction conditions, and almost 90% conversion of CB
was obtained when the temperature reached 325 ◦C. Compared
with the T90% (the temperature at which a 90% conversion is
attained) of the other catalysts, it appears that the efficiency of
all supported catalysts were as follows: Mn8Co1Ce1/cordierite
3.3. Long-term reaction of the catalysts
For dealing with CVOCs produced by catalytic combustion from
> Mn3Co1Ce1/cordierite > Mn2Co2Ce1/cordierite > Mn4Ce1/cordierite industrial processes, most catalysts were easily poisoned because
> Mn2Ce1/cordierite > Mn1Ce1/cordierite. The performance of the
representative Ce-based catalysts has been investigated for the
CB oxidation. MnOx-CeO2 [21], MnOx/CeO2-NPS [30], CuO/CeO2
[31] and Mn-Ce-Mg/Al2O3 [32] have been investigated for the
catalytic combustion of CVOCs. Ma et al. [31] reported that meso-
porous CuO/CeO2 bimetal oxides were very active for the catalytic
destruction of 1,2-dichlorobenzene. Wu et al. [32] showed that
Mn-Ce-Mg/Al2O3 catalysts exhibited high activity, good selectivity
and desired stability for the catalytic oxidation of chlorobenzene.
It is well known that the unsupported catalysts are unsuitable for
practical applications because of their high cost and low service life.
However, the system of Mn-Ce catalysts supported on cordierite
has rarely been studied; in addition, it did not show better activ-
ity than the aforementioned unsupported catalysts, and the CB
conversion is less than 90% at 350 ◦C over the Mn4Ce1/cordierite
catalyst presented in this paper. It can be observed in Fig. 1 that
Mn4Ce1/cordierite presents a higher catalytic activity than the
other two catalysts, and more than 90% of the CB is not converted
until 375 ◦C. The results in this paper indicate that the activity
of the catalyst is significantly improved with the addition of a
small amount of Co [33–35]. Therefore, Mn8Co1Ce1/cordierite is a
promising ternary oxide catalyst for CB treatment.
The oxidation mechanism of chlorobenzene over Mn-Ce/H-
ZSM5 and V2O5/TiO2 catalysts has been studied by Sun et al. [36]
and Lichtenberger et al. [37], and the researchers proposed that the
nucleophilic and electrophilic substitutions occurred towards CB
combustion. It is known that the energy of C Cl bonds is weaker
than that of C H bonds, and therefore, the C Cl bonds are more
likely to be attacked by nucleophiles. First, dissociative adsorp-
tion on the active cobalt and manganese sites occurred via Cl
abstraction in the molecular combustion of chlorobenzene. The dis-
sociated Cl was absorbed over the surface of the Mn4Ce1/cordierite
and Mn8Co1Ce1/cordierite catalysts. At low reaction temperatures
(<200 ◦C), chlorobenzene adsorption was the main control step, and
the adsorbed Cl could not be removed until the dynamic equilib-
rium of adsorbed and desorbed Cl species was achieved due to the
absence of active oxygen species or the synergistic effect of cobalt,
manganese and cerium, which was clarified by XPS. In the process
above, the partial desorption of Cl may lead to the formation of a
chlorine typically occupied the active sites through adsorption on
the surface of the catalysts [38]. To enable the commercial use
of the supported catalysts, the stability of the catalysts is a crit-
ical factor. Li et al. [39] tested the stability of Mn-Ti catalysts
with different additives for chlorobenzene combustion during a
long-term reaction of 30 h, and the Sn-Mn-Ti catalyst showed bet-
ter stability. Ma et al. [40] examined the stability of CrOx/Al2O3
catalysts for dichloromethane oxidation, which was measured at
300 ◦C during a long-term reaction of 16 h. Therefore, Fig. 2 presents
the results of the stability test at 350 ◦C for CB combustion, and
the variation in activity was observed for 24 h at that tempera-
ture. As shown in Fig. 2, the Mn4Ce1/cordierite catalyst presented
only a 86% initial conversion of CB, and the conversion tended to
reduce during the 24 h reaction at 350 ◦C, decreasing from 86%
to 82% in the continuous stream. At the same temperature, all
the Mn-Co-Ce/cordierite catalysts exhibited a catalytic efficiency
of more than 90% for the catalytic combustion of CB. Compar-
ing the durability of the four catalysts, the downward trend in
the activity of the Mn-Co-Ce/cordierite catalysts was smoother
than that of the Mn4Ce1/cordierite catalyst, which indicated that
the addition of cobalt to Mn-Ce/cordierite enhanced the chlo-
rine tolerance owing to the increased mobility of active oxygen
species on the surface, which was shown in Section 3.6. In addi-
tion, the Mn8Co1Ce1/cordierite catalyst showed the initial activity
and the best resistance to chlorine poisoning, without a notice-
able deactivation during the entire testing phase. The stability
of the Mn-Co-Ce/cordierite catalysts decreased as the amount of
cobalt ions increased. Therefore, the Mn8Co1Ce1/cordierite catalyst
was the optimal three-component solid solution in the Mn-Co-
Ce/cordierite catalyst system for CB combustion.
3.4. XRD analysis
XRD patterns of the Mn-Ce/cordierite and Mn-Co-Ce/cordierite
catalysts are displayed in Fig. 3. From the XRD spectrum, it is
observed that the diffraction peak corresponding to the cordierite
support is relatively strong because of its high content. Therefore,
the relatively weak diffraction peaks of the active components are
observed in Fig. 3. According to the XRD profiles in Fig. 3(A), the

24
J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
Fig. 1. Catalytic activity for chlorobenzene destruction over the catalysts; the concentration of chlorobenzene: 500 ppm, GHSV = 15000 h−1
.
diffraction peaks (at 28.4, 33.0, 47.4 and 56.2◦) of CeO2 with a
cubic fluorite-like structure appeared over the Mn-Ce/cordierite
catalysts, but these peak positions shifted to the left compared to
the standard peak (JCPDS No. 81-0792) [25], and the peak intensity
changed slightly as the amount of Mn increased. It is speculated that
the crystal lattice of CeO2 is distorted by the effect of ion doping,
and the Ce species with a low content are uniformly distributed on
cordierite. For all catalysts with different contents of manganese, no
diffraction peak corresponding to manganese oxides are observed,
which indicates that the manganese element may be incorporated
into the fluorite lattice to form the solid solution since the ionic
radius of Mn3+ (0.066 nm) is smaller than that of Ce4+ (0.094 nm),
or it may scatter homogeneously on the surface of CeO2 [41]. For
the Mn-Co-Ce/cordierite catalysts, some diffraction peaks of Co3O4
crystallites at 19.0, 36.8 and 68.6◦ are observed in Fig. 3(B), and the
diffraction peak intensity of CeO2, especially that at 33.0◦, clearly
decreased. This phenomenon implies that CeO2 may be covered
with highly dispersed cobalt oxides, or part of the cobalt ions also
enter into the crystal lattice of CeO2 to form a ternary solid solution
[12]. Subsequently, the Mn-Ce/cordierite modified by the doping
of a small amount of cobalt was beneficial to catalytic combus-
tion of CB due to synergistic effects including the variable valence,
the oxygen transfer rate and the presence of more active sites. The
diffraction peaks of Mn8Co1Ce1/cordierite had the broadest of full
width at half maximum, especially at 36.8◦, which indicated the
formation of more lattice defects and a smaller crystallite size [20].
Fig. 2. The stability tests of the Mn-Co-Ce/cordierite catalysts and the Mn4Ce1/cordierite catalyst for chlorobenzene oxidation during a sustained reaction of 24 h.

J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
25
Fig. 3. (A) XRD patterns of the Mn-Ce/cordierite catalysts and cordierite (a: cordierite, b: Mn:Ce = 1, c: Mn:Ce = 2, and d: Mn:Ce = 1) (B) XRD patterns of the Mn-Co-Ce/cordierite
catalysts and the Mn4Ce1/cordierite catalyst. (d: Mn:Ce = 4 e: Mn:Co:Ce = 3:1:1, f: Mn:Co:Ce = 2:2:1, and g: Mn:Co:Ce = 8:1:1).
Table 1
Meanwhile, the Mn8Co1Ce1/cordierite catalyst showed the supe-
Textural properties of the Mn-Ce/cordierite and Mn-Co-Ce/cordierite catalysts with
different molar ratios.
rior catalytic performance for chlorobenzene combustion, which
was in full accordance with the results above of the catalytic activ-
ity.
Catalysts
SBET (m2/g)
DV (cm3/g)
DP (nm)
cordierite
5.0
0.005
0.043
0.052
0.073
0.035
0.042
0.040
3.4
Mn1Ce1/cordierite
Mn2Ce1/cordierite
Mn4Ce1/cordierite
Mn3Co1Ce1/cordierite
Mn2Co2Ce1/cordierite
Mn8Co1Ce1/cordierite
15.0
16.1
20.4
14.6
16.9
17.0
11.6
12.9
14.3
9.6
10.1
9.4
3.5. N2 adsorption/desorption
Table 1 lists the specific surface area (SBET), total pore volume
(DV), and average pore size (DP) of the cordierite and synthetic
supported catalysts. The surface areas of all catalysts were much
higher than that of pure cordierite (5.0 m2/g). The SBET value of
the Mn-Ce/cordierite catalyst increased with the increase in the
manganese content, which indicated that incorporation of a large
amount of MnOx into CeO2 could promote the dispersion of the cat-
alyst, to some extent. Regardless of if cobalt ions partially replace
manganese or cerium ions, after a comparison of the two catalyst
systems in Table 1, the SBET of the Mn-Co-Ce/cordierite catalysts
decreased slightly compared with that of the Mn4Ce1/cordierite
catalyst. For chlorobenzene oxidation, although the Cl species from
chlorobenzene were easily absorbed on the surface of the catalysts,
they could be rapidly removed by catalysts with a large amount of
active surface oxygen. Consequently, this did not mean that the cat-
alyst with the largest surface area had the highest catalytic activity.

26
J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
Fig. 4. SEM images of the Mn8Co1Ce1/cordierite and Mn4Ce1/cordierite catalysts (A, C: Mn8Co1Ce1/cordierite, B, D: Mn4Ce1/cordierite).
The Mn8Co1Ce1/cordierite catalyst had a relatively large specific
surface area, which resulted in a high catalytic efficiency in the
degradation of chlorobenzene, as shown in Fig. 1.
Mn-Co-Ce solid solutions, oxygen vacancies and lattice defects.
Moreover, the significant peak offset of the Mn8Co1Ce1/cordierite
catalyst indicates the formation of more lattice defects in this
catalyst. In addition, the amount of oxygen vacancies over the
Mn4Ce1/cordierite and Mn-Co-Ce/cordierite catalysts was evalu-
ated by a Gaussian deconvolution of the region at approximately
435 cm−1 [43]. It can be observed that the Mn8Co1Ce1/cordierite
catalyst presented the highest vacancy concentration.
3.6. Scanning electron microscopy
The surface morphological properties of the Mn4Ce1/cordierite
and Mn8Co1Ce1/cordierite catalysts were investigated by SEM, as
shown in Fig. 4. From Fig. 4, it can be observed that the catalysts
have morphological structures that are ball-like nanoclusters with
asymmetrical sizes. However, the Mn8Co1Ce1/cordierite catalyst
exhibited a smaller average pore size than the Mn4Ce1/cordierite
catalyst. As observed in the SEM images, the cordierite was homo-
geneously covered with active components in Fig. 4(A, C), which
indicated that the cobalt dopant can increase the dispersion of the
catalyst and prevent the aggregation of manganese and cerium
metal particles. The differences above indirectly reflect the syn-
ergistic effect of Mn, Co and Ce, which was proposed in Section
3.4.
3.8. H2-TPR analysis
The H2-TPR data for the Mn4Ce1/cordierite and Mn-Co-
Ce/cordierite catalysts are shown in Fig. 6. The Mn4Ce1/cordierite
catalyst exhibits three strong reduction peaks at approximately
363, 444 and 591 ◦C. The first two peaks are associated with a two-
step reduction of MnO2 and Mn2O3, respectively [44,45]. The third
peak is related to the reduction of the surface Ce+4 ions [46]. With
regards to the reduction peak area corresponding to Mn species of
the Mn-Co-Ce system, the area of the Mn8Co1Ce1/cordierite cata-
lyst was larger than that of others, which indicated that more H2
consumption on the Mn8Co1Ce1/cordierite catalyst was required;
this demonstrated that the active component has a smaller size
and higher dispersion. Compared with the reduction features
of the Mn4Ce1/cordierite catalyst, the reduction temperatures of
the Mn-Co-Ce/cordierite composite metal oxides shifted to lower
temperatures in the temperature range of 300–450 ◦C, and the off-
set increased as the doping amount of cobalt increased. For the
Mn2Co2Ce1/cordierite (reduction peaks at approximately 330, 390
and 566 ◦C) and Mn3Co1Ce1/cordierite catalysts (reduction peaks at
approximately 352, 400 and 570 ◦C), their reduction temperatures
were lower than those of the Mn4Ce1/cordierite catalyst, which
indicated that the surface oxygen of ceria can be easily reduced.
3.7. Raman spectra
The Raman spectra of the CeO2/cordierite, Mn4Ce1/cordierite
and Mn-Co-Ce/cordierite catalysts are shown in Fig. 5. As observed
in Fig. 5, the Raman spectra of the CeO2/cordierite catalyst has a
principal peak at 455 cm−1 and a weak peak at 252 cm−1, which
may be attributed to the strong F2g mode of the fluorite phase and
the second-order transverse acoustic mode, respectively [42]. On
the other hand, the Mn4Ce1/cordierite and Mn-Co-Ce/cordierite
catalysts exhibited a single peak at ∼ 430–435 cm−1 that shifted
to a lower wavelength compared with the F2g band of the
CeO2/cordierite catalyst due to the formation of the Mn-Ce-O and

J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
27
Fig. 5. Raman spectra of the synthesized catalysts (CeO2/cordierite, Mn4Ce1/cordierite, and Mn-Co-Ce/cordierite).
However, the former present the Co3O4 phase: the reduction tem-
perature and peak shape at 330 ◦C [47,48], which indicates that
excessive cobalt ions can significantly change the structure of man-
ganese and hinder the contact between CB and the active phase.
Although the content of manganese is the same as that in the
Mn4Ce1/cordierite catalyst, the reduction temperature of Mn2O3
in the Mn8Co1Ce1/cordierite catalyst shifted to the left by approx-
imately 24 ◦C. In addition, the reduction of the surface oxygen of
ceria occurred at a higher temperature due to the decrease of a
small amount of ceria or the strong interaction between CeO2 and
the manganese species [49,50]. It can be speculated that the addi-
tion of small amounts of cobalt ions promotes the incorporation of
manganese into the CeO2 lattice and changes the structure/phase
of the manganese oxides, which can improve the mobility of active
oxygen species on the surface, as active oxygen species are easily
adsorbed on chlorobenzene and released to the reaction system by
the synergistic effect of ceria, manganese and cobalt with variable
valence states.
3.9. XPS characterization
To investigate the surface states and composition of the
Mn4Ce1/cordierite and Mn8Co1Ce1/cordierite catalysts, Fig.
7
shows XPS spectra of the O 1s, Mn 2p, Co 2p, and Ce 3d electronic
levels in the catalysts above. The experimental and fitted Ce 3d
spectra of the synthesized catalysts are shown in Fig. 7(A). Four
Fig. 6. TPR profiles of the Mn-Co-Ce/cordierite catalysts and the Mn4Ce1/cordierite catalyst.

28
J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
Fig. 7. XPS spectra of Ce 3d(A), Mn 2p(B), O 1s(C), and Co 2p(D) for the Mn4Ce/cordierite and Mn8Co1Ce1/cordierite catalysts.
peaks denoted as u’, uo, v’, and vo arise from the Ce3+ contribu-
tion, while Ce4+ is characterized by six other peaks. The existence
of Ce3+ in the Mn4Ce1/cordierite and Mn8Co1Ce1/cordierite cat-
alysts indicates the formation of an oxygen vacancy, which can
boost the oxygen transfer rate during chlorobenzene oxidation
[51,52]. The fraction of Ce3+ ions with respect to the total cerium
of the Mn4Ce1/cordierite and Mn8Co1Ce1/cordierite catalysts is
calculated as 17.8% and 20.7%, respectively, according to the mea-
surement of the peak areas. Apparently, the peaks of the Ce3+ and
Ce4+ ions experience some changes in their position and intensity
upon cobalt doping, which can be attributed to the mixing with
the Co 2p band [53,54]. The reducibility of ceria does not dimin-
ish the fraction of Ce3+ ions that are already present because the
additional electrons in Ce3+ occupy localized cerium f-states that
do not interact with each other. Fig. 7(B) shows the XPS spectra of
Mn 2p. The binding energies of Mn 2p3/2 and Mn 2p1/2 are simi-
lar to those reported in the literature [55]. The component in the
528.9–529.3 is assigned to the lattice oxygen (O2−, denoted as O␣),
and the binding energies at ∼530.9–531.3 and ∼532.1–532.4 are
attributed to surface-adsorbed oxygen (O2−, O22−or O− denoted as
O␤) and chemisorbed water, and dissociated oxygen and hydroxyl
species (O␥) [56]. Compared to the Mn4Ce1/cordierite catalyst,
the binding energy of O␣ over Mn8Co1Ce1/cordierite increases
due to the “Co ←− O” electron-transfer process, which indicates
that the Co-doped Mn-Ce/cordierite promotes the formation of a
ternary solid solution. Additionally, the concentration of O␤ clearly
increases from 48% to 60% (estimated through the integration of the
peak areas), indicating that the catalytic activity can be improved
and the Cl adsorbed on the surface of the catalysts can be removed
quickly due to the higher mobility of O␤ compared to that of O␣.
Therefore, the addition of small amounts of cobalt ions to Mn-
Ce/cordierite plays an important role in improving the catalytic
activity and Cl resistance.
The Co 2p XPS spectra for the Mn8Co1Ce1/cordierite samples
are shown in Fig. 7(D). The original spectra are very disordered
due to the low content of cobalt and interference of other species.
The Co 2p spectra present two spin-orbit doublets and two shake-
up satellite peaks. The binding energies at ca.779.6–781.3 and
ca.794.3–795.9 are assigned to the two spin-orbit doublets, which
can be attributed to Co 2p3/2 and Co 2p1/2; the other peaks are
assigned to the satellite peak [57,58]. The Co 2p3/2 is deconvo-
luted into two components at BE = 779.6 and 781.3 eV; the former
is attributed to the Co3+ species, whereas the latter is attributed to
the Co2+ species. The Co 2p1/2 peak is also deconvoluted into two
Mn4Ce1/cordierite at BE = 651.6 and 640.7 eV is attributed to Mn3+
,
and that at BE = 653.1 and 642.1 eV is attributed to Mn4+ ions. On the
other hand, the component in the Mn8Co1Ce1/cordierite at 652.7
and 641.1 eV is assigned to Mn3+, and that at BE = 654.1 and 642.4 eV
is assigned to Mn4+ ions. When cobalt ions are mixed into the
Mn-Ce/cordierite catalyst, the Mn4+/Mn3+ ratio clearly decreases.
It is speculated that more oxygen vacancies and lattice defects are
formed by the addition of cobalt ions.
The
O 1s XPS spectra for the Mn4Ce1/cordierite and
Mn8Co1Ce1/cordierite catalysts are deconvoluted into three con-
tributions, as observed in Fig. 7(C). The binding energy (BE) at ca.

J. Kan et al. / Applied Catalysis A: General 530 (2017) 21–29
29
components at BE = 794.3 and 795.9 eV, which correspond to the
Co3+ and Co2+ species, respectively.
[14] Q. Dai, S. Bai, X. Wang, G. Lu, J. Porous Mater. 21 (2014) 1041–1049.
[15] M. Taralunga, J. Mijoin, P. Magnoux, Appl. Catal. B: Environ. 60 (2005)
163–171.
[16] W. Xingyi, K. Qian, L. Dao, Appl. Catal. B: Environ. 86 (2009) 166–175.
[17] P. Topka, R. Delaigle, L. Kaluzˇa, E.M. Gaigneaux, Catal. Today 253 (2015)
172–177.
4. Conclusions
[18] S. Cao, H. Wang, F. Yu, M. Shi, S. Chen, X. Weng, Y. Liu, Z. Wu, J. Colloid
Interface Sci. 463 (2016) 233–241.
[19] H. Huang, Q. Dai, X. Wang, Appl. Catal. B: Environ. 158–159 (2014) 96–105.
[20] C. He, Y. Yu, J. Shi, Q. Shen, J. Chen, H. Liu, Mater. Chem. Phys. 157 (2015)
87–100.
[21] X. Wang, Q. Kang, D. Li, Catal. Commun. 9 (2008) 2158–2162.
[22] P. Yang, Z. Shi, S. Yang, R. Zhou, Chem. Eng. Sci. 126 (2015) 361–369.
[23] H. Chen, H. Zhang, Y. Yan, Chem. Eng. Sci. 111 (2014) 313–323.
[24] D.M. Gómez, V.V. Galvita, J.M. Gatica, H. Vidal, G.B. Marin, Phys. Chem. Chem.
Phys. 16 (2014) 11447.
In this work, Mn-Ce/cordierite and Mn-Co-Ce/cordierite cat-
alysts with different molar ratios were prepared by the sol-gel
method and used in the catalytic combustion of low concentration
chlorobenzene. Through characterization analysis by XRD, SEM,
BET, H2-TPR, Raman, and XPS, all catalysts presented morphological
structures that are ball-like nanoclusters with asymmetrical sizes.
A significant amount of manganese and small amounts of cobalt
ions were incorporated into the lattice of ceria, which resulted in
the formation of crystal defects and an improvement in the mobility
of active oxygen species on the surface. The Mn8Co1Ce1/cordierite
catalyst presented a higher activity and stability than the other cat-
alysts, and at least 90% chlorobenzene can be removed at 325 ◦C; in
addition, the catalytic efficiency remained almost unchanged dur-
ing a sustained reaction for 24 h at 350 ◦C. That high performance
was ascribed to the synergistic effect of ceria, manganese and cobalt
with variable valence states, which can promote the formation of
more lattice defects and a smaller crystallite size. Moreover, the
Mn8Co1Ce1/cordierite catalyst should have a widespread appli-
cation in the catalytic combustion of chlorinated volatile organic
compounds.
[25] C. He, Y. Yu, Q. Shen, J. Chen, N. Qiao, Appl. Surf. Sci. 297 (2014) 59–69.
[26] Q. Huang, Z. Zhang, Y. Chen, S. Zhu, S. Shen, J. Chem. Eng. Jpn. 43 (2010)
413–420.
[27] Q. Dai, S. Bai, X. Wang, G. Lu, Appl. Catal. B: Environ. 129 (2013) 580–588.
[28] J.R. Gonzalez-Velasco, A. Aranzabal, R. Lopez-Fonseca, R. Ferret, J.A.
Gonzalez-Marcos, Appl. Catal. B: Environ. 24 (2000) 33–43.
[29] R. Lopez-Fonseca, A. Aranzabal, J.I. Gutierrez-Ortiz, J.I. Alvarez-Uriarte, J.R.
Gonzalez-Velasco, Appl. Catal. B: Environ. 30 (2001) 303–313.
[30] C.W.F.H. Pei Zhao, RSC Adv. (2014) 45665–45672.
[31] X. Ma, X. Feng, X. He, H. Guo, L. Lv, J. Guo, H. Cao, T. Zhou, Micropor. Mesopor.
Mater. 158 (2012) 214–218.
[32] M. Wu, X. Wang, Q. Dai, D. Li, Catal. Commun. 11 (2010) 1022–1025.
[33] P.J. Jodłowski, R.J. Je˛drzejczyk, A. Rogulska, A. Wach, P. Kus´trowski, M. Sitarz,
T. Łojewski, A. Kołodziej, J. Łojewska, Spectrochim. Acta Part A 131 (2014)
696–701.
[34] J. Chen, W. Shi, J. Li, Catal. Today 175 (2011) 216–222.
[35] J. Cheng, H. Wang, Z. Hao, S. Wang, Catal. Commun. 9 (2008) 690–695.
[36] P. Sun, W. Wang, X. Dai, X. Weng, Z. Wu, Appl. Catal. B: Environ. 198 (2016)
389–397.
[37] J. Lichtenberger, M.D. Amiridis, J. Catal. 223 (2004) 296–308.
[38] X. Wang, L. Ran, Y. Dai, Y. Lu, Q. Dai, J. Colloid Interface Sci. 426 (2014)
324–332.
[39] J. Li, P. Zhao, S. Liu, Appl. Catal. A: Gen. 482 (2014) 363–369.
[40] R. Ma, P. Hu, L. Jin, Y. Wang, J. Lu, M. Luo, Catal. Today 175 (2011) 598–602.
[41] Y. She, Q. Zheng, L. Li, Y. Zhan, C. Chen, Y. Zheng, X. Lin, Int. J. Hydrogen Energ.
34 (2009) 8929–8936.
[42] W.H. Weber, K.C. Hass, J.R. Mcbride, Phys. Rev. B 48 (1993) 178–185.
[43] A. Gurbani, J.L. Ayastuy, M.P. Gonzalez-Marcos, M.A. Gutierrez-Ortiz, Int. J.
Hydrogen Energ. 35 (2010) 11582–11590.
[44] J. Trawczyn´ ski, B. Bielak, W. Mis´ta, Appl. Catal. B: Environ. 55 (2005) 277–285.
[45] B. Li, Q. Huang, X.K. Yan, X.L. Xu, Y. Qiu, B. Yang, Y.W. Chen, S.M. Zhu, S.B.
Shen, J. Ind. Eng. Chem. 20 (2014) 2359–2363.
Acknowledgments
This research is financially supported by the Project of Science
and Technology Department of Jiangsu Province (BE2016769), the
Natural Science Foundation of China (No. 51172107), the Natural
Science Foundation of the Jiangsu Higher Education Institutions of
China (No. 14KJB430014), the Open fund by Jiangsu Engineering
Technology Research Center of Environmental Cleaning Materials
(KFK1503) and A Project Funded by the Priority Academic Program
Development of Jiangsu Higher Education Institutions (PAPD).
[46] X. Tang, Y. Xu, W. Shen, Chem. Eng. J. 144 (2008) 175–180.
[47] T. Cai, H. Huang, W. Deng, Q. Dai, W. Liu, X. Wang, Appl. Catal. B: Environ.
166–167 (2015) 393–405.
References
[48] M.H. Kim, K. Choo, Catal. Commun. 8 (2007) 462–466.
[49] J.L. Ayastuy, E. Fernández-Puertas, M.P. González-Marcos, M.A.
Gutiérrez-Ortiz, Int. J. Hydrogen Energy 37 (2012) 7385–7397.
[50] P. Zhao, C. Wang, F. He, S. Liu, RSC Adv. 4 (2014) 45665–45672.
[51] C. He, Y. Yu, L. Yue, N. Qiao, J.I.L.Q. Shen, W. Yu, J. Chen, Z. Hao, Appl. Catal. B:
Environ. 147 (2014) 156–166.
[1] P. Yang, S. Yang, Z. Shi, Z. Meng, R. Zhou, Appl. Catal. B: Environ. 162 (2015)
227–235.
[2] A. Sre˛bowata, R. Baran, D. Łomot, D. Lisovytskiy, T. Onfroy, S. Dzwigaj, Appl.
Catal. B: Environ. 147 (2014) 208–220.
[3] H. Sedjame, C. Fontaine, G. Lafaye, J. Barbier Jr, Appl. Catal. B: Environ. 144
(2014) 233–242.
´
[52] J. Gonzalez-Prior, R. Lopez-Fonseca, J.I. Gutierrez-Ortiz, B. de Rivas, Appl.
Catal. B: Environ. 199 (2016) 384–393.
[4] A. Aznárez, R. Delaigle, P. Eloy, E.M. Gaigneaux, S.A. Korili, A. Gil, Catal. Today
246 (2015) 15–27.
[53] C. Gionco, M.C. Paganini, S. Agnoli, A.E. Reeder, E. Giamello, J. Mater. Chem. A
1 (2013) 10918–10926.
[54] W. Deng, Q. Dai, Y. Lao, B. Shi, X. Wang, Appl. Catal. B: Environ. 181 (2016)
848–861.
[55] Y. Zhang-Steenwinkel, J. Beckers, A. Bliek, Appl. Catal. A: Gen. 235 (2002)
79–92.
[56] C. He, X. Liu, J. Shi, C. Ma, H. Pan, G. Li, J. Colloid Interface Sci. 454 (2015)
216–225.
[5] H. Huang, Y. Gu, J. Zhao, X. Wang, J. Catal. 326 (2015) 54–68.
[6] V.H. Vu, J. Belkouch, A. Ould-Dris, B. Taouk, J. Hazard. Mater. 169 (2009)
758–765.
[7] Q. Dai, S. Bai, J. Wang, M. Li, X. Wang, G. Lu, Appl. Catal. B: Environ. 142–143
(2013) 222–233.
[8] E. Finocchio, G. Ramis, G. Busca, Catal. Today 169 (2011) 3–9.
[9] J.M. Giraudon, A. Elhachimi, G. Leclercq, Appl. Catal. B: Environ. 84 (2008)
251–261.
[57] J. Gonzalez-Prior, R. Lopez-Fonseca, J.I. Gutierrez-Ortiz, B. de Rivas, Appl.
Catal. B: Environ. 199 (2016) 384–393.
[10] W. Deng, Q. Dai, Y. Lao, B. Shi, X. Wang, Appl. Catal. B: Environ. 181 (2016)
848–861.
[58] T. Cai, H. Huang, W. Deng, Q. Dai, W. Liu, X. Wang, Appl. Catal. B: Environ. 166
(2015) 393–405.
[11] S. Zuo, M. Ding, J. Tong, L. Feng, C. Qi, Appl. Clay Sci. 105–106 (2015) 118–123.
[12] C. He, B. Xu, J. Shi, N. Qiao, Z. Hao, J. Zhao, Fuel Process. Technol. 130 (2015)
179–187.
[13] L. Zhang, S. Liu, G. Wang, J. Zhang, React. Kinet. Mech. Catal. 112 (2014)
249–265.