Reaction kinetics and mechanism of benzene combustion over the NiMnO3/CeO2/Cordierite catalyst

Article abstract of DOI:10.1016/j.molcata.2016.01.023

The kinetics of the catalytic combustion of benzene at different concentrations over the NiMnO₃/CeO₂/Cordierite catalyst were investigated to gain more insight into the catalytic reaction mechanism. A kinetic study was performed in a packed-bed tubular reactor under different conditions. The catalytic combustion kinetics were modeled using a Power-Law model and the Mars-Van Krevelen model. The results showed that the Mars-Van Krevelen kinetic model provided a significantly better fit to explain the catalytic combustion kinetics of benzene over the NiMnO₃/CeO₂/Cordierite catalyst. The results showed that the reduction reaction occurred more easily than did the oxidation reaction on the surface of the catalyst. Moreover, the values of the pre-exponential factor for the reduction steps (7.84 × 10¹¹ s^-1) is higher than those of the oxidation steps (1.04 × 10⁹ s^-1), indicating that the more is the effective collision times between the activated molecules, and the easier for chemical reaction to occur, and the degree and speed are more intense and rapid. Therefore, it can be concluded that the catalyzed surface oxidation reaction is the control step of catalytic benzene combustion. Based on this analysis of the experimental results and the assumptions of the Mars-Van Krevelen model, it was determined that the catalytic combustion of benzene over the NiMnO₃/CeO₂/Cordierite catalyst obeys the Mars-Van Krevelen mechanism. The catalytic combustion reaction occurred by the interaction between the benzene molecules and the active sites of the NiMnO₃/CeO₂/Cordierite catalyst. The catalytic oxidation of benzene involves a catalytic redox cycle of adsorption, deoxidation, desorption, oxygen supply and regeneration.

Full text of DOI:10.1016/j.molcata.2016.01.023

Accepted Manuscript
Title: Reaction kinetics and mechanism of benzene
combustion over NiMnO3/CeO2/Cordierite catalyst
Author: Bing Li Yingwen Chen Lin Li Jiawei Kan Shuo He
Bo Yang Shubao Shen Shemin Zhu
PII:
S1381-1169(16)30023-1
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.01.023
MOLCAA 9755
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
8-12-2015
20-1-2016
20-1-2016
Pleasecitethisarticleas:BingLi, YingwenChen, LinLi, JiaweiKan, ShuoHe, BoYang,
Shubao Shen, Shemin Zhu, Reaction kinetics and mechanism of benzene combustion
over NiMnO3/CeO2/Cordierite catalyst, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.01.023
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Reaction kinetics and mechanism of benzene combustion
over NiMnO3/CeO2/Cordierite catalyst
Bing Li a,b, Yingwen Chen a,b,*, Lin Li a,b, Jiawei Kan a,b Shuo Hea,b , Bo Yang a,c,
Shubao Shen a,b,, Shemin Zhu a,c
a State Key Laboratory of Materials-Oriented Chemical Engineering, Nanjing Tech
University, Nanjing 210009, China
b College of Life Science and Pharmaceutical Engineering, Nanjing Tech University,
Nanjing 210009, China,
c College of Materials Science and Engineering, Nanjing Tech University,
Nanjing 210009, China
*Corresponding author: Prof. Yingwen Chen, Ph.D,
Address: State Key Laboratory of Materials-Oriented Chemical Engineering,
College of Life Science and Pharmaceutical Engineering, Nanjing Tech University,
No.5 Xinmofan Road, Nanjing 210009, P. R. China.
E–mai：yingwenchen_nj@163.com Tel: +86 25 58139922
Fax: +86 25
83587326

Graphical Abstract
The catalytic combustion mechanism of benzene over
the NiMnO3/CeO2/Cordierite catalyst
Abstract:
The kinetics of the catalytic combustion of benzene at different concentrations
over the NiMnO3/CeO2/Cordierite catalyst were investigated to gain more insight into
the catalytic reaction mechanism. A kinetic study was carried out in a packed-bed
tubular reactor under different conditions. The catalytic combustion kinetics were
modeled using a Power-Law model and the Mars-Van Krevelen model. The results
showed that the Mars-Van Krevelen kinetic model provided a significantly better fit to
explain
the
catalytic
combustion
kinetics
of
benzene
over
the
NiMnO3/CeO2/Cordierite catalyst. The results showed that the reduction reaction
occurred more easily than did the oxidation reaction on the surface of the catalyst.
Moreover, the values of the pre-exponential factor for the reduction steps(7.84×1011s-1)
are higher than those of the oxidation steps(1.04×109s-1), indicating that the density of
the active centers in the reduction position is higher than the density of those in the
oxidation position; therefore, it can be concluded that the catalyzed surface oxidation
reaction is the control step of catalytic benzene combustion. Based on this analysis of
the experimental results and the assumptions of the Mars-Van Krevelen model, it was
determined
that
the
catalytic
combustion
of
benzene
over
the
NiMnO3/CeO2/Cordierite catalyst obeys the Mars-Van Krevelen mechanism. The
catalytic combustion reaction occurred by the interaction between the benzene
molecules and the active sites of the NiMnO3/CeO2/Cordierite catalyst. The catalytic
oxidation of benzene involves a catalytic redox cycle of adsorption, deoxidation,
desorption, oxygen supply and regeneration.
Keywords: kinetic study ; catalyst mechanism; NiMnO3/CeO2/Cordierite catalyst

1. Introduction
Volatile organic compounds (VOCs) emitted from many industrial processes are an
important class of air pollutants[1-2]. These pollutants can cause many environmental
problems such as toxic emissions, precursors of ozone and photochemical smog
formation[3-4]. Reducing VOC emissions is a very important environmental issue for
human health. VOCs are usually eliminated by thermal oxidation, catalytic oxidation,
adsorption and absorption. Among these removal methods, catalytic combustion has
been recognized as one of the most promising techniques for VOC removal due to its
practical applications in pollutant abatement, and this technology has been determined
to be more environmentally friendly than conventional flame combustion because it is
more versatile and economical for dealing with organic emissions. Moreover, catalytic
incineration, which operates at relatively low temperatures and in controlled
[5-7]
conditions, does not emit undesirable by-products, such as dioxins and NOx
.
In general, there are two major types of efficient catalysts for total VOC
oxidation, supported noble metals and transition metal oxides. Although the former
are highly active at relatively low temperatures, their application is limited due to the
high price of precious metals and problems related to sintering and volatility[8-9]. In
recent years, research efforts in this field have been oriented towards developing new
catalytic materials with low manufacturing cost that arecapable of high activity at
moderate temperatures[10-12]. We investigated the catalytic performance of
Ni-Mn/CeO2/Cordierite catalysts for benzene combustion in a previous work, which
demonstrated that Ni-Mn mixed oxide demonstrates higher activity than the pure
oxides. When the Ni/Mn molar ratio was 1:1, the temperature required for the
complete conversion of benzene over the catalyst was even lower than that of a
supported Pt catalyst[13]. The XRD and TPR results supported our findings, as the
catalyst with the highest activity had a perovskite crystal structure and higher redox
activity.
Apart from the catalyst composition, another aspect that must be taken into
account for catalytic combustion is the kinetic model, which represents a key element
in predicting the reaction rate of VOC combustion over the catalyst[14-16]. Based on

these results, a reaction mechanism is proposed that can be used as the theoretical
basis for the design of high activity catalysts, providing more insight into the behavior
of the catalysts. The catalytic combustion kinetics of VOCs as single oxides or simple
mixed oxides over various catalysts have been presented in the literature[17-19]
.
However, studies of kinetic models devoted to obtaining perovskite composite oxides
are scarce. Therefore, the search for a suitable kinetic model that offers greater insight
into the catalytic combustion reaction mechanism of VOCs over the
NiMnO3/CeO2/Cordierite catalyst is a worthwhile effort.
Based on previous research, a kinetic study was carried out in a packed-bed
tubular reactor under differential conditions. The catalytic combustion kinetics were
investigated by using Power-Law and Mars-Van Krevelen models. The aim of this
work is to investigate the catalytic combustion reaction kinetics and mechanism of
benzene over the NiMnO3/CeO2/Cordierite catalysts. Emphasis has been placed on
deriving kinetic expressions describing the rate of this reaction. To the best of our
knowledge, this article is the first report on the kinetics and mechanism of the
complete oxidation of benzene over the perovskite-type NiMnO3/CeO2/Cordierite
catalyst.
2. Materials and methods
2.1 Catalyst preparation
The details of the preparation of the NiMnO3/CeO2 catalysts, as well as the
structural characteristics, have been described in a previous work[13]. Namely, the
CeO2/Cordierite supports were prepared using a wet impregnation method with an
aqueous Ce(NO3)3·6H2O solution to obtain a 20 wt% CeO2 loading. Subsequently, the
material was dried overnight at 100°C and calcined at 500°C for 5 h. The
NiMnO3/CeO2/Cordierite catalysts were prepared using a sol–gel method. An aqueous
solution containing Ni(NO3)2·6H2O, Mn(NO3)2 and citric acid (n(citric
acid):n(Ni):n(Mn) = 1:2.5:2.5) was gradually heated to 80°C and stirred at this
temperature for 4 h to form the sol. The CeO2/Cordierite supports were immersed in
the sol for 2 h. The samples were subsequently dried overnight at 105°C and calcined
for 7 h at 500°C.

2.2 Catalyst characterization
The H2-temperature programmed reduction (H2-TPR) experiments were carried
out using 20 mg of catalyst under a mixture of hydrogen (5%) and nitrogen (95%)
flowing at 30 mL/min. The temperature of the catalysts was increased at 10ºC/min. A
thermal conductivity detector was used to monitor the consumption of hydrogen
during the H2-TPR experiment.
The X-ray photoelectron spectroscopy (XPS) experiments were carried out on an
RBD upgraded PHI-5000C ESCA system (Perkin Elmer) with Mg Ka radiation (hv
=1253.6 eV) or Al Ka radiation (hv = 1486.6 eV). In general, the X-ray anode was run
at 250 W and the high voltage was kept at 14.0 kV with a detection angle at 54°. The
pass energy was fixed at 23.5, 46.95 or 93.90 eV to ensure sufficient resolution and
sensitivity. The base pressure of the analyzer chamber was approximately 5×10-8 Pa.
The sample was directly pressed into a self-supported disk (10×10 mm),mounted on a
sample holder, and then transferred into the analyzer chamber. The O core level
binding energies were determined, referencing to the binding energy of adventitious C
(284.6 eV), which gives an accuracy of 0.1 eV. The data analysis was carried out with
the RBD Auger Scan 3.21 software provided by RBD Enterprises and the XPS
Peak4.1 program provided by Raymund W.M. Kwok (The Chinese University of
Hong Kong, China). Astandard Shirley background and Gaussian (Y%)–Lorentzian
(X%) were used for each component.
2.3 Experimental Setup
The experimental runs for the kinetic study were carried out in a quartz tubular
fixed bed reactor (Φ=8 mm) operated at atmospheric pressure. The reactive mixture
was composed of nitrogen, oxygen and liquid benzene. The molar fraction of oxygen
was kept at 21% to simulate the gas concentration. The benzene was fed by means of
a syringe pump, which allowed for variable reactant flows and therefore variable
benzene concentration. The reactor wall and quartz particles had no catalytic effect on
the benzene combustion. For the catalyst activity test, the catalyst was crushed into
different meshes, and 200 mg of catalyst was diluted with quartz particles and loaded
into the fixed-bed reactor. In these conditions, the benzene conversion into CO2 was

below 10%, which allowed a differential analysis of the kinetic data.
The benzene concentration in the gas stream was determined by an on-line gas
chromatograph (GC-2014, Shimadzu Corp) equipped with FID(Restek Rtx-1), while
the CO2 was analyzed by a GC with a TCD(SP-3420). The GC signal was calibrated
based on mixtures of known composition. For benzene combustion over the catalyst,
no byproduct was detected by either of the GCs’detectors. Furthermore, the
repeatability of the experimental data was verified in several experiments with the
same benzene conversion under the same experimental conditions.
The benzene conversion (xbenzene) and the reaction rate (rbenzene) for the
differential reactor were calculated as follows
Cbenzene,in C
xbenzene

benzene,out *100%
(1)
(2)
Cbenzene,in
xbenzene
Wcat / Fbenzene,in
- rbenzene 
where Cbenzene, in and Cbenzenen, out are the inlet and outlet concentration of benzene,
respectively; Fbenzene, in is the inlet molar flow rate of the benzene, and Wcat is the
weight of the catalyst used in the reactor.
2.4 Effect of mass transfer limitations
Mass transfer limitations were checked by conducting several experiments at
reaction temperatures in the range of 200 to 245oC.
2.4.1 External mass transfer limitations
If the mass transfer rate between the gas and the catalyst’s outer surface is slower
than that of other reaction processes, the external diffusion is the rate-controlling step
of the whole reaction. Therefore, eliminating the influence of external diffusion by
increasing the speed of air flow is necessary to accelerate the external diffusion.
As the air flow velocity increases, the retention layer on the two-phase interface
becomes thinner, so the diffusion resistance decreases and the diffusion coefficient
increases, thereby increasing the external diffusion rate. To eliminate the influence of

external diffusion, the conventional approach is to set the appropriate gas hourly
space velocity (GHSV). For this approach, the external mass transfer effect is studied
by carrying out catalytic combustion experiments on benzene at six different GHSVs
(12000 h-1, 13000 h-1, 14000 h-1, 15000 h-1, 16000 h-1, 17000 h-1) over a catalytic
reactor based on a structured catalyst with a bed height of 1 cm.
2.4.2 Internal mass transfer limitations
When the diffusion velocity in the microscopic voids between particlesis slower
than that of other processes, the internal diffusion is the rate-controlling step of the
whole reaction. To eliminate the influence of internal diffusion, it is necessary to
make the catalyst particle size small enough to fully expose the inner surface of the
catalyst so that the concentration of the reactants inside the material is similar to that
on the outer surface; therefore, internal diffusion resistance can be ignored. To
estimate the internal diffusion effect, catalytic combustion experiments are carried out
on benzene in air over catalysts with different mean particle sizes (20-100 meshes).
The particle size of the catalyst is controlled using sieves that allow the selection of a
particle size range. Four different ranges have been selected: 20-40 meshes, 40-60
meshes, 60-80 meshes and 80-100 meshes.
In the gas-solid heterogeneous reaction, the catalytic oxidization removal process
includes external mass transfer through the external film of the catalyst, internal mass
transfer, and surface reaction. To study the intrinsic reaction kinetics, external and
internal mass transfer effects should be eliminated to ensure the surface reaction is the
rate-determining step. Mass transfer limitations were tested by conducting catalytic
combustion experiments on the reaction.
3. Results and discussion
3.1 Effect of mass transfer limitations
3.1.1 External mass transfer limitations
Fig. 1 shows that the external diffusion limitations were evaluated by performing
several experiments with six different GHSVs ranging from12000 to 17000 h-1. As
observed in Fig. 1, the profiles of the benzene conversions in this range are essentially

identical, indicating that the catalytic system was operated in the absence of an
external diffusional limitation. A possible explanation is that the catalyst has an
entirely open macroscopic structure offering a large void volume. Therefore, the
GHSV was then set to 15000 h-1 for the catalytic combustion kinetics experiments to
avoid any external mass transfer limitations.
3.1.2 Internal mass transfer limitations
Fig. 2 shows that the internal diffusion effect was estimated over the catalyst
with different mean particle sizes (20-40 meshes, 40-60 meshes, 60-80 meshes and
80-100 meshes). As shown in Fig. 2, when the particle size of the catalyst was 60-80
and 80-100 meshes, the conversion rate of benzene was basically the same, which
indicated that the internal diffusion effect was eliminated when the catalyst particles
were 60-100 meshes. In theory, the chemical regime requires a minimal Thiele
modulus and therefore a smaller particle size. Therefore, only catalysts with particle
sizes between 80-100 meshes were used for the rest of the kinetic study to limit the
internal mass transfer.
3.2 Catalytic combustion performance
The performance of catalytic benzene combustion over the catalytic structured reactor
based on the NiMnO3/CeO2/Cord catalyst was also investigated using different inlet
concentrations of benzene (1.595-6.38×10-8 mol/cm3) at a constant GHSV of 15000
h-1. Table 1 shows that some differences in conversion can be observed under different
inlet concentrations of benzene. from the table also indicates that the conversion rate
of benzene increased at higher reaction temperatures. This is mainly because when the
temperature increases, the benzene molecules move faster, thereby increasing the
number of collisions per unit time. Furthermore, increasing temperature can also
increase the energy of the single benzene molecules, thereby increasing the
percentage of activated benzene molecules. Table 1 also shows that the conversion
rate of benzene decreases as the inlet concentration of benzene increases, which is

because, when the concentration of benzene increases, the number of benzene
molecules per unit volume increases. The active site of benzene per unit catalyst is
maintained constant, which increases the load per unit catalyst, resulting in a decrease
in the catalytic activity of benzene. Using the data in Table 1 and in Eq. 2, the reaction
rates for the catalytic oxidation of benzene were obtained, and the results are listed in
Table 2. We can draw from Table 2 that increasing both the inlet concentration and the
reaction temperature can raise the benzene reaction rate.
3.3 The kinetic models of the reaction
Modeling the kinetics of the catalytic reaction involves multiple variables. In addition
to reactant concentration, the reaction temperature should also be included if
considering different operating conditions. Several possible mechanisms are proposed
considering the different reaction paths. In the current study, the experimental data
were fitted to the available single variable kinetic models to evaluate the mechanism
of the reaction. Next, the reaction process and mechanism considering multiple
variables was deduced, and a multivariable kinetic model considering the benzene
concentration and temperature was developed[20]
.
3.3.1 Power-Law model
The simplest kinetic equation is the Power-law model that describes the rate as a
function of a rate constant and the reactant concentration. This model is very useful
for a quick comparison of the kinetic behavior of different catalysts[21]
.
m
dC
- r

b e n z en ke Cn C
b e n z e n e
b
e
n
z
eOn2 e
dt
(3)
For the catalytic combustion of benzene in gas, the oxygen concentration in the
gas stream is always much larger than that of benzene: therefore, the above expression
is simplified to:

dCbenzene
dt
- rbenzene 
 kCn
b
e n
(4)
where (-rbenzene) is the benzene oxidation rate; Cbenzene and C O2 are the benzene
and oxygen concentration in the gas stream, respectively; k is the surface reaction rate
constant according to the Power-Law model, and n and m is the kinetic orders with
respect to benzene and oxygen, respectively. The apparent surface reaction rate
constant (k) follows the Arrhenius equation and thus connects the reaction
temperature with the reaction rate[22]. The Arrheniuse quation can be expressed as:
- EA
k  k0exp(
)
(5)
RT
where EA is the activation energy (kJ/mol); R is the gas constant (8.314×10-3
kJ/mol K); k0 is the pre-exponential factor, and T is the reaction temperature (K).
The plot of ln-rbenzene against lnCbenzenefor benzene combustion over the
NiMnO3/CeO2 catalyst is shown in Fig.3. According to Eq. 4, both the slope and
intercepts of those lines relate to the reaction order n and the reaction rate constant k,
as shown in Fig. 3. The linear regression correlation coefficient R2is relatively high,
with R2 values above 0.95, except when the temperature is 245°C. The reaction rate
constant increases with high erreaction temperatures, in accordance with the
Arrhenius equation. Fig. 4 shows the Arrhenius equation fitting of the reaction rate
constants for the Power-Law model, calculated with the kinetic data from Fig. 3. The
Power-Law model provided a good fit for the kinetic data with a relatively high R2
value (0.964). Table 3 shows the factor k0 =1.436×107s-1 and the activation energy
EA=115.79 kJ/mol, which were calculated using Eqs. 4 and 5, respectively. Table 3
also shows that the values of kinetic order n does increase with higher temperatures
but not at a constant rate; so, the Power-Law model cannot explain the variation of the
apparent kinetic order n simply with the reaction temperature. Therefore, a more
elaborate model derived from mechanistic considerations is needed to account for the
observed behavior of the catalyst for benzene combustion. To build an exact kinetic
model, an understanding of the reactions at the molecular level seems to be necessary.

However, the full elucidation of the reaction mechanism for heterogeneous reactions
is always difficult. Thus, calculating the rate equation may require an empirical
method. Though the exact reaction mechanism is not well understood, it has been
suggested that benzene oxidation may involve lattice oxygen or surface-adsorbed
oxygen for metal oxides catalysts[21]. Based on reasonable assumptions, several
kinetic equations derived from mechanistic considerations have been developed, such
as the Mars-Van Krevelen model.
3.3.2 Mars-Van Krevelen Model
The Mars-Van Krevelen Model (MVK) is a two-step redox reaction model that has
been used to explain oxidation reactions rates for heterogeneous oxide catalysts. The
model assumes reaction occurs when a reactant molecule interacts with an
oxygen-rich portion of the catalyst. The oxygen in the catalyst can be either
chemisorbed or lattice oxygen. A particular portion of the catalyst surface is
alternately reduced and oxidized[19]
.
The equation representing the MVK model has been used by several authors
[23-25],suggesting a general equation for the oxidation of hydrocarbons in this case,
benzene (Eq. 6).
k k
C
C
O2 b e n z e bn ee n z e On2 e
- r

b e n z e n ke C kb e n z e bn ee n z e
C
(6)
O2 O2
Or equivalently,
1

k C
1
kb e n zCe bn ee n z e
(7)


- r
b e n z e n e O2 O2
where (-rbenzene) is the benzene oxidation rate; kbenzene is rate constant of benzene
oxidation; kO2 is rate constant of catalyst re-oxidation; Cbenzeneis the benzene
concentration in the gas stream;CO2 is the O2 concentration in the gas stream; and γ is
the stoichiometric coefficient of oxygen in the oxidation of benzene (C6H6 + 7.5O2 →

6CO2 + 3H2O), which is equal to 7.5.
For catalytic combustion of benzene in the gas stream, the oxygen concentration
is always much larger than that of the benzene, so the γ/kO2CO2 term can be assumed
to be constant A. Therefore, Eq.7 expression can be simplified to:
1
1
 A
(8)
- r
kb e n zCe bn ee n z e
b e n z e n e
The plot of 1/-rbenzeneas a function of 1/Cbenzenefor benzene combustion over the
NiMnO3/CeO2 catalyst is shown in Fig.5. According to Eq. 8, both the slope and
intercepts of those lines related to the rate constant of benzene oxidation (kbenzene) and
rate constant of the re-oxidation of the catalyst (kO2) have been calculated from these
data, as shown in Table 4. It is clear from Table 4 that the values of kbenzeneand kO2
increase regularly as the reaction temperature increases, indicating that the variation
of the reaction rate constants kbenzene and kO2 with different reaction temperatures obey
an empirical Arrhenius equation. The results of fitting the MVK model reaction rate
constants to the Arrhenius equation are shown in Fig. 6 and Table 3, which shows that
the reaction activation energies for the reduction and oxidation step are 69.79 and
103.9kJ/mol, respectively. Because of its lower activation energy, the reduction
reaction occurred more easily than the oxidation reaction on the surface of the catalyst.
Moreover, the value of the pre-exponential factor for the reduction (7.84×1011 s-1) is
more the factor for the oxidation (1.04×109s-1 , indicating that the density of the active
)
center of reduction position is higher than that of the oxidation position, so it can be
concluded that the catalyst surface oxidation reaction was the rate-controlling step of
catalytic combustion of benzene. Furthermore, the correlation coefficient R2 for the
rate constant fittings are relatively high (0.995 and 0.996). The comparison between
the experimental and modeled data are presented in Fig. 7, where the experimental
values of conversion (xexp) are plotted as a function of the calculated conversion by
the MVK model (xcalcMVK). The results presented in Fig.7 demonstrate the good
agreement between the calculated and experimental results. It is clear that the

predicted reaction rates agree well with the experimental values, giving a relatively
high correlation coefficient R2(0.998). In the literature, if a correlation coefficient R2
is greater than 0.9, the model can be considered appropriate. Therefore, the above
results clearly demonstrate that the Mars-Van Krevelen kinetic model can be used to
describe the catalytic combustion kinetics of benzene.
3.4 Reaction Mechanism
In the model equations tested, the MVK model shows the best fit to the intrinsic
kinetic experimental data, which means that the reaction mechanism can be predicted
accurately by this model. In our previous work[13], the H2-TPR profiles of NiMnO3
perovskite in Fig. S1 showed that these peaks appear at a lower temperature than
those associated with the reduction of pure NiO and MnOx, mainly because the
reduction of NiMnO3 perovskite proceeds in two main steps. During the first step,
most of the manganese cations are reduced to Mn2+, generating a solid solution of
NiO in MnO. Subsequently, the Ni2+cations are reduced in the MnO matrix. The next
step generates the final product: a mixture of MnO and metallic nickel or Ni+. At the
same time, the CeO2 catalyst coating can rely on redox states (Ce3+-Ce4+) and high
oxygen storage capacity to provide lattice oxygen for Mn2+ and Ni+, thus promoting
oxygen mobility. Fig. 8 shows the O 1s XPS spectra of the NiMnO3/CeO2/Cordierite
catalyst. In Fig. 8, the O 1s spectra was deconvoluted into three peaks at 529.4,
531.0and 532.3eV,indicating thesurface lattice oxygen (denoted as OⅢ), surface
adsorbed oxygen and some oxygen-containing groups such as OH−(denoted as OⅡ),
and oxygen in surface adsorbed water molecules (denoted as OⅠ), respectively. The
ratio of the surface-adsorbed oxygen to the lattice oxygen (OⅡ/OⅢ) can be used to
describe the proportion of different oxygen species[26]; a lower OⅡ/OⅢvalue implies
the presence of more active vacant oxygen species on the catalyst lattice, meaning that
it can be the richest source of active lattice oxygen species for the oxidation reaction.
If the lattice oxygen participates in hydrocarbon combustion, a Mars-Van Krevelen
mechanism may be claimed[27-28,21], that is, by the involvement of lattice oxygen,
whereby the created oxygen vacancy is subsequently replenished by oxygen from the

gas phase. Thus, on the basis of the above analysis of the experimental results and the
assumptions of the Mars-Van Krevelen model, the catalytic combustion of benzene
over the NiMnO3/CeO2/Cordierite catalyst was considered to obey the Mars-Van
Krevelen mechanism, as shown in Fig. 9.
From Fig. 9, we can see that the catalytic oxidation reaction is divided into two
steps: (1) first, the benzene molecules are adsorbed on the oxidized active sites of the
NiMnO3 perovskite composite oxide, and the dehydrogenation oxidation of benzene
molecules forms intermediate reactive oxygen species; at the same time, the active
oxygen species in CeO2migrate to the NiMnO3 perovskite composite oxide to
complement its consumption of reactive oxygen species and form oxygen vacancies.
The intermediate reactive oxygen species are further oxidized to CO2 and H2O, and
the active sites of the catalyst are reduced, forming reduced active sites. Second, O2
molecules are adsorbed on the surface of the CeO2, filling the oxygen vacancies, and
the reduced active sites are re-oxidized to form oxidized active sites, thereby forming
a cycle of active oxygen consumption and supply. The catalytic oxidation of benzene
molecules thus completes a catalytic redox cycle of adsorption, deoxidation,
desorption, oxygen supply and regeneration.
Through the reaction kinetics and mechanistic analysis, the catalytic oxidation of
benzene molecules under the action of the catalyst is shown to complete a redox
catalytic cycle to reduce the activation energy of the reaction; this changes the original
chemical reaction pathway and facilitates the oxidation of benzene. According to the
Arrhenius equation, decreasing the activation energy and increasing the
pre-exponential factor can improve the reaction rate of benzene combustion, which
can be used as the theoretical basis for the design of a high activity catalyst. The
surface of the gas-solid reaction catalyst is not uniform; so, under the reaction
conditions, only some special sites have adsorption, deoxidation, desorption, oxygen
supply and regeneration abilities, which causes a catalytic effect. It is well known that
the structure of the catalytic activity center determines the ability of the activated
reactant. Changing the active component the electronic properties ofthe catalyst can
change the structure of the active site, which can improve the activity of the active

center. The increase in the reaction rate is due to the further reduction of the reaction’s
activation energy. At this point, the following two situations may decrease the
activation energy: (1) there is no change in the reaction pathwaybuta decrease in the
activation energy of the rate-controlling step decreasesthe total activation energy, as
shown in Fig. 10(A). This situation applies when the activated structure of the catalyst
is close to the central structure (for example, the active component is the same as the
transition element).In the second situation, (2) there is a change in the reaction
pathway that reduces the apparent activation energy of the reaction, as shown in Fig.
10(B).This situation applies when the structure of the active center of the catalyst is
large. At the same time, we can also improve the pre-exponential factor to accelerate
the rate of catalytic reaction. By increasing the surface area of the catalyst or the
density of active centerson the catalyst surface, we can improve the number of active
centers permass unit of catalyst to achieve the above purpose.
The above analysis provides good theoretical support for further improving the
design of the catalyst as a way to improve its performance. In future studies, we will
again modify the NiMnO3 perovskite; the A-site (Ni2+) and/or B-site (Mn4+) cations
can be substituted by other cations without changing the structure. In this way, the
oxidation states of the B cation and/or the amounts of non-stoichiometric oxygen can
be modified, thus generating associated structural defects. Such defects (typically
oxygen vacancies) can be beneficial from a catalytic point of view because their
presence could favor the chemisorption of reactants or other relevant catalytic steps
(oxygen transport, electron transfer processes, etc.)[29-30]. At the same time, the
preparation method of the CeO2 coating will be optimized so that the crystalline
CeO2can be distributed more evenly and dispersed on the surface of the cordierite
carrier, thus increasing the surface area of the cordierite carrier. We will also add
electronic-type catalysts(such as Zr, Ti) to the CeO2 crystal to form a
Ce-Zr-O2/Ce-Ti-O2 solid solution structure, which can inhibit the sintering of the
catalyst. (Catalyst sintering will lead to a rapid decrease in the surface area, thereby
reducing the number of active centers.) That is, to make the pre-exponential factor,
can be kept constant in the process of using the catalyst. Using the above method, the

rate coefficient of the reaction may be further improved in the future.
Conclusion
The kinetics of the catalytic combustion of benzene over the
NiMnO3/CeO2/Cordierite catalyst were studied using the Power-Law kinetic model
and the Mars-Van Krevelen kinetic model. The Power-Law model cannot explain the
change of the apparent reactant order n with the reaction temperature. The Mars-Van
Krevelen kinetic model gave a better fit to explain the catalytic combustion kinetics.
This showed that the reduction reaction occurred more easily than did the oxidation
reaction on the surface of the catalyst. Moreover, the value of the pre-exponential
factor for the reduction steps(7.84×1011s-1) is larger than that for the oxidation
steps(1.04×109s-1), indicating that the density of active centers in the reduction
position is higher than that in the oxidation position. Thus, it can be concluded that
the catalyst surface oxidation reaction was the rate-controlling step for the catalytic
combustion of benzene. At the same time, the R2 correlation coefficients for the rate
constant fitting are relatively high (0.995 and 0.996), and the predicted reaction rates
agree well with the experimental values (R2= 0.998). Based on the experimental
results and the assumptions of the Mars-Van Krevelen model, it was concluded that
the catalytic combustion of benzene over the NiMnO3/CeO2/Cordierite catalyst obeys
the Mars-Van Krevelen mechanism. The reaction kinetics and mechanistic analysis
showed that the catalytic oxidation of benzene molecules under the action of the
catalyst completes a catalytic redox cycle to reduce the activation energy of the
reaction, which in turn changes the original chemical reaction path way to facilitate
the oxidation of benzene. The reduction in the activation energy and the increase in
the pre-exponential factor in the Arrhenius equation can improve the reaction rate of
benzene, and these principles can be used as a theoretical basis for the design of high
activity catalysts.
Acknowledgments
This work was financially supported by the Natural Science Foundation of China (No.
51172107 and No. 51272105), the National Key Technology R&D Program of China
(No. 2012BAE01B03-3), and the Natural Science Foundation of the Jiangsu Higher
Education Institutions of China (No. 14KJB430014).

References
[1] G. Pecchi, M.G. Jiliberto, E.J. Delgado, et al. Effect of B-site cation on the
catalytic activity of La1-xCaxBO3 (B = Fe, Ni) perovskite-typeoxides for toluene
combustion, J ChemTechnol Biot, 86 (2011) 1067-1073 .
[2] N. Rezlescu, E.Rezlescu, P.D. Popa. Some nanograined ferrites and perovskites
for catalytic combustion of acetone at low temperature, Ceram Int, 41 (2015)
4430-4437.
[3] G.l. Zhou, X.L. He, S. Liu, et al. Phenyl VOCs catalytic combustion on supported
CoMn/AC oxide, J Ind Eng Chem, 21 (2015) 932-941
[4] D. Delimaris, T. Ioannides. VOC oxidation over CuO-CeO2 catalysts prepared by
a combustion method, Applied Catalysis B: Environmental, 89 (2009) 295-302.
[5] D.F. Wu, W. Jiang, J.C. Zhou. Effect of drying and calcination on the toluene
combustion activity of a monolithic CuMnAg/γ-Al2O3/cordierite catalyst, J Chem
Technol Biot, 85 (2010) 569-576.
[6] M. Piumetti, D.Fino, N.Russo. Volatile Organic Compound (VOC) Emissions
From Soil And Litter Samples, Appl Cata B, 163 (2015) 277-287.
[7] T. Garcia, B. Solsona, D.M. Murph. Deep oxidation of light alkanes over
titania-supported palladium/vanadium catalysts, J Catal 229 (2005) 1-11 .
[8] Z. Mu, J.J. Li, M.H. Duan. Catalytic combustion of benzene on Co/CeO2/SBA-15
and Co/SBA-15 catalysts, Catal Commun, 9 (2008) 1874-1877.
[9] D.F. Wu, W.B. Li, R. Gao. Comparison of themethods for preparing a cordierite
monolith-supported Cu-Mnmixed-oxide catalyst, J Chem Technol Biote 89 (2014)
1559-1564 .
[10]X.J Zhang, H.J. Li, Y. Li, et al. Structural Properties and Catalytic Activity of
Sr-Substituted LaFeO3 Perovskite, Chinese J Catal 33 (2012) 1109-1104 .
[11]J.G. Deng, L. Zhang, H.X. Dai, et al. Hydrothermally fabricated single-crystalline
strontium-substituted lanthanum manganite microcubes for the catalytic
combustion of toluene, J Mol Catal A:Chem 299 (2009) 60-67.
[12]P. Yang, Z.N. Shi, S.S. Yang, et al. High catalytic performances of CeO2–CrOx
catalysts for chlorinated VOCs elimination, Chem Eng Sci, 126 (2015) 361-369.

[13]B. Li, Q. Huang , X.K. Yan, et al. Low-temperature catalytic combustion of
benzene over Ni–Mn/CeO2/cordierite catalysts, J Ind Eng Chem, 20 (2014)
2359-2363.
[14]J. Bedia, J.M. Rosas, J. Rodríguez-Mirasol, et al. Pd supported on meso-porous
activated carbonswith high oxidation resistance as catalysts for tolueneoxidation,
Appl Catal. B: Environ. 94 (2010) 8-18.
[15]V.R. Choudhary, G.M. Deshmukh, D.P. Mishra. Kinetics of the complete
com-bustion of dilute propane and toluene over Iron-doped ZrO2 catalyst, Energ
Fuel, 19 (2005) 54-63.
[16]S. Benard , A. Baylet, P. Vernoux, et al. Kinetics of the propene oxidation over a
Pt/alumina catalyst, Cata Commun, 36 (2013) 63–66.
[17]Y.F. Han, L.W. Chen, K. Ramesh. Kinetic and spectroscopic study of methane
combustion over α-Mn2O3 nanocrystal catalysts, J Catal, 253 (2008) 261-268.
[18]C.I. Meyer, A. Borgna, A. Monzón. Kinetic study of trichloroethylene
combustion on exchanged zeolites catalysts, J Hazard Mater, 190 (2011) 903-908.
[19]A.Z. Abdullah, M.Z.A. Bakar, S. Bhatia. A Kinetic Study of Catalytic
Combustion of Ethyl Acetate and Benzene in Air Stream over Cr-ZSM-5 Catalyst,
Ind Eng Chem Res, 42 (2003) 6059-6067.
[20]J.J. Pei, X. Han, Y. Lu. Performance and kinetics of catalytic oxidation of
formaldehyde over copper manganese oxide catalyst, Build Environ, 84 (2015)
134-141.
[21]C.Q. Hu. Catalytic combustion kinetics of acetone and toluene over
Cu0.13Ce0.87Oy catalyst, Chem Eng J, 168 (2011) 1185-1192.
[22]P. Hurtado, S. Ordonez, H. Sastre, et al. Development of a kinetic model for the
oxidation of methane over Pd/Al2O3 at dry and wet conditions, Appl Catal
Environ, 51 (2004) 229-238.
[23]C.M. Manta, G. Bozga , G. Bercaru, et al. Kinetics of o-xylene combustion over a
Pt/alumina catalyst, Chem Eng Technol, 35 (2012) 2147-2154.
[24]H. Li, H. Chen, M.F. Yao. Reaction Kinetics of Ethylene Combustion in a Carbon
Dioxide Stream over a Cu-Mn-O Hopcalite Catalyst in Low Temperature Range,

Ind Eng Chem Res, 52 (2013) 686-691.
[25]H.B. Chen, Y. Yan, Y. Shao. Catalytic Combustion Kinetics of Isopropanol over
Novel Porous Microfibrous-Structured ZSM-5 Coating/PSSF Catalyst, AIChE J,
61 (2015) 620-630.
[26]C.H. Zhang, W.C. Hua, C. Wang. The effect of A-site substitution by
Sr, Mg and Ce on the catalytic performance of LaMnO3 catalysts for
the oxidation of vinyl chloride emission, Appl Catal B: Environ, 134-135 (2013)
310-315.
[27]V.R. Choudary, G.M. Deshmukh. Kinetics of the complete combustion of dilute
propane and methyl ethyl ketone over Cr-doped ZrO2 catalyst, Chem Eng Sci, 60
(2005) 1575-1581.
[28]G. Arzamendi, R. Ferrero, A.R. Pierna. Kinetics of Methyl Ethyl Ketone
Combustion in Air at Low Concentrations over a Commercial Pt/Al2O3 Catalyst,
Ind Eng Chem Res, 46 (2007) 9037-9044.
[29]A.A. Leontiou, A.K. Ladavos, T.V. Bakas, et al. Reverse uptake of oxygen from
La1-xSrx (Fe3+/Fe4+)O3±δ perovskite-type mixed oxides (x = 0.00, 0.15, 0.30, 0.40,
0.60, 0.70, 0.80, 0.90), Appl. Catal. A, 241 (2003) 143-154.
[30]K. Rida, A. Benabbas, F. Bouremmad, et al. Surface properties and catalytic
performance of La1-xSrxCrO3 perovskite-type oxides for CO and C3H6
combustion, Catal Commun. 7 (2006) 963-968.

100
80
60
40
20
0
120000
130000
140000
150000
160000
170000
GHSV/ h -1
Fig. 1 Effect of external diffusion on conversion

100
80
60
40
20
0
100-80
80-60
60-40
40-20
particle size/ mesh
Fig. 2 Effect of internal diffusion on conversion

Fig. 3 Linear fitting for catalytic benzene combustion based on the
Power-Law model

Fig. 4 Reaction activation energy for catalytic benzene combustion based on the
Power-Law model
(Calculated from the Arrhenius equation by means of linear regression)

1.0
0.8
0.6
0.4
0.2
0.0
200 oC (1/-rbenzene) = 0.469/Cbenzene+0.00678*108 R2= 0.996
215 oC (1/-rbenzene) = 0.283/Cbenzene+0.00283*108 R2= 0.995
230 oC (1/-rbenzene) = 0.183/Cbenzene+0.0015*108
R2= 0.975
245 oC (1/-rbenzene) = 0.101/Cbenzene+0.00069*108 R2= 0.987
0.1
0.2
0.3
0.4
0.5
0.6
0.7
(Cbenzene-1)*108(cm3/mol)
Fig. 5 Linear fitting for catalytic benzene combustion based on the MVK model

Fig. 6 Reaction activation energy for catalytic benzene combustion based on the
MVK model
(Calculated from the Arrhenius equation by means of linear regression)

10
8
R2=0.998
6
4
2
2
4
6
8
10
Experimental cnversion Xexp,n
Fig. 7 Comparison between experimental and calculated conversions By using
the MVK model

Fig. 8 The O 1s XPS spectra of the NiMnO3/CeO2/Cordierite catalyst

Fig. 9 The catalytic combustion mechanism of benzene over the
NiMnO3/CeO2/Cordierite catalyst

Fig. 10 Different catalysts (1 and 2) may cause the same reduction in the
activation energy for two different reasons
(A) No change in the reaction pathway; (B) change in the reaction pathway