Catalytic self-sustained combustion of toluene and reaction pathway over CuxMn1-xCe0.75Zr0.25/TiO2 catalysts

Article abstract of DOI:10.1016/j.apcata.2018.10.034

The catalytic self-sustained combustion of toluene over Cu_xMn_1-xCe_0.75Zr_0.25/TiO₂ catalysts (x = 1, 0.5, 0) was studied in a microscale combustor. The catalytic activity is determined not only by the temperature with toluene conversion of 90% (T_co), but also by the corresponding lean-combustion limits. It was found that the self-sustained combustion was achieved successfully over the catalysts, with the toluene concentrations of 0.35 con.%, 0.50 con.% and 1.00 con.%. According to T_co the activity decreased in the order of CuCe_0.75Zr_0.25/TiO₂ (234 °C)＞Cu_0.5Mn_0.5Ce_0.75Zr_0.25/TiO₂ (238 °C) ＞MnCe_0.75Zr_0.25/TiO₂ (284 °C). The excellent activity of CuCe_0.75Zr_0.25/TiO₂ catalyst could be ascribed to the high content of active lattice oxygen, good low-temperature reducibility and homogeneous dispersion, combined with the results of H₂-TPR, O₂-TPD and XPS. Furthermore, the results of in-situ DRIFT and temperature-programmed oxidation (TPO) of toluene in N₂ atmosphere suggested that toluene was adsorbed on the catalyst surface by removing α-H from methyl and combining with active oxygen specie to form benzyl group, which was finally oxidized into CO₂ and H₂O with lattice oxygen.

Full text of DOI:10.1016/j.apcata.2018.10.034

AppliedCatalysisA,General569(2019)66–74
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Catalytic self-sustained combustion of toluene and reaction pathway over
CuxMn1-xCe0.75Zr0.25/TiO2 catalysts
Chenchen Zhaoa, Qinglan Haoa, Qing Zhanga, Ningna Yana, Jiaren Liua, Baojuan Doua,⁎
,
Feng Binb,⁎
a Tianjin University of Science & Technology, Tianjin 300457, China
b Institute of Mechanics, Chinese Academy of Science, Beijing 100190, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Toluene
Self-sustained combustion
Ce-Zr based catalyst
Lean-combustion limit
Reaction pathway
The catalytic self-sustained combustion of toluene over CuxMn1-xCe0.75Zr0.25/TiO2 catalysts (x = 1, 0.5, 0) was
studied in a microscale combustor. The catalytic activity is determined not only by the temperature with toluene
conversion of 90% (Tco), but also by the corresponding lean-combustion limits. It was found that the self-sus-
tained combustion was achieved successfully over the catalysts, with the toluene concentrations of 0.35 con.%,
0.50 con.% and 1.00 con.%. According to Tco, the activity decreased in the order of CuCe0.75Zr0.25/TiO2 (234 °C)
＞Cu0.5Mn0.5Ce0.75Zr0.25/TiO2 (238 °C) ＞MnCe0.75Zr0.25/TiO2 (284 °C). The excellent activity of CuCe0.75Zr0.25
/
TiO2 catalyst could be ascribed to the high content of active lattice oxygen, good low-temperature reducibility
and homogeneous dispersion, combined with the results of H2-TPR, O2-TPD and XPS. Furthermore, the results of
in-situ DRIFT and temperature-programmed oxidation (TPO) of toluene in N2 atmosphere suggested that toluene
was adsorbed on the catalyst surface by removing α-H from methyl and combining with active oxygen specie to
form benzyl group, which was finally oxidized into CO2 and H2O with lattice oxygen.
1. Introduction
Catalytic self-sustained combustion of toluene in a small-scale
combustor is flameless combustion with a small amount of fuels, which
Toluene is one of the major contributors to VOCs which are mainly
released by industries processes. It is hazardous to environment and
even carcinogenic for both human and animal [1]. Abundant global
technologies based on recovery and destruction have been developed to
meet the strict environment regulations of toluene emission limitation.
To accelerate the miniaturization of energy systems and increase the
reaction rate, the catalytic combustion in microscale tube, named
“catalytic self-sustained combustion”, has been developed recently [2].
The technology of stabilizing combustion is an economical strategy for
complete VOCs degradation without heat input. In this process, only a
small amount of heat is need for the ignition, and the heat released from
the reaction is enough to maintain the combustion without any other
energy input [3]. Yang [2] and Scarpa [4] found that alkanes including
methane, propane and n-octane could achieve self-sustained combus-
tion successively over Pt-based catalyst supported ZSM-5 or γ-Al2O3
with relatively lower concentration. In addition, the self-sustained
combustion can be obtained by CO → CO2 exothermic reaction over the
CuCe/ZSM-5 and CuCe0.75Zr0.25/TiO2 catalysts in our previous studies
[5,6].
are used to generate the local hot spots on the catalyst surface, leading
to the successful self-sustained combustion. The challenge of the tech-
nology is to design a proper catalyst to enhance the stability of micro-
combustion. Copper-based and manganese-based catalysts are con-
sidered as the excellent catalysts for VOCs removal due to their variety
of electronic structures and capability of regeneration [7–11]. Copper
doping could significantly enhance the catalytic activity to achieve
complete VOCs degradation by extracting the lattice oxygen from CuO
[8,9]. The MnOx catalysts for VOCs oxidation have been considered as
the better alternative than the expensive noble metal, owing to the
coexistence of mixed valance states from −3 to +7 and low toxicity
[1,10,11]. Some investigators have tried to prepare bimetallic catalysts
to study their high activity through the synergistic effect between the
two phases [12–14]. For instance, Li et al. [12] found that CuMn(y)Ox/
γ-Al2O3 exhibited higher catalytic activity (T90 = 229 ℃) compared
with CuOx/γ-Al2O3, MnOx/γ-Al2O3 for toluene oxidation, resulting from
the formation of the Cu1.5Mn1.5O4 spinel structure in CuMn(y)Ox/γ-
Al2O3 catalyst. However, the pure Cu-based or Mn-based catalysts are
still away from arrival of satisfied catalytic performance to VOCs
⁎ Corresponding authors.
E-mail addresses: bjdou@tust.edu.cn (B. Dou), binfeng@imech.ac.cn (F. Bin).
https://doi.org/10.1016/j.apcata.2018.10.034
Received 27 August 2018; Received in revised form 11 October 2018; Accepted 27 October 2018
Availableonline29October2018
0926-860X/©2018ElsevierB.V.Allrightsreserved.

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
combustion due to the poor oxygen storage-release capacity and in-
stability of catalysts. CeO2 has the unique properties due to the plenty
of oxygen vacancies associated with strong interactions with metals,
and the adding of CeO2 can further improve the catalytic performance
[5,9,15–17]. The thermal stability of the catalyst can be enhanced
through the zirconium addition [15–17]. In our previous study, we
have investigated the influence of Cu/Ce molar ratio of CuCe/ZSM-5
catalysts [15] and the Ce/Zr molar ratio of CuCeZr/ZSM-5 catalysts
[16] for the catalytic activity, respectively. And we found that the
CuCe0.75Zr0.25Ox mixed oxides exhibited excellent catalytic activity in
the toluene catalytic oxidation owing to the formation of Cu-Ce-Zr-O
solid solution, which improved the redox capability and oxygen storage
capacity [15,16], could be favorable for toluene self-sustained com-
bustion.
Many studies have focused on the self-sustained combustion reac-
tion with low-carbon organic compounds, such as CO and CnH2n-2
(n≤4), however, little statistical evidences related to catalytic self-
sustained combustion of aromatic compounds have not been reported
yet. In this paper, to acquire the performance of toluene self-sustained
combustion over different catalysts, the lean-combustion limit of self-
sustained combustion and the efficiency of toluene degradation were
studied. CuxMn1-xCe0.75Zr0.25/TiO2 catalysts (x = 1, 0.5, 0) were pre-
pared by the impregnation method, and the physical and structural
properties were characterized by H2-TPR, O2-TPD and XPS.
Furthermore, the in-situ DRIFT experiment was carried out to discuss
the possible reaction pathways for toluene catalytic combustion, and
the role of lattice oxygen was also proposed based on the results of
programmed-temperature oxidation in N2 atmosphere.
amount of catalyst was placed in a pure oxygen flow (50 mL/min) for
1 h at 500 ℃. Subsequently, the catalyst was heated from room tem-
perature to 1000 °C at a rate of 10 °C/min under the Ar atmosphere.
TCD was employed to continuously monitor the consumption of
oxygen. X-ray photoelectron spectra (XPS) was measured with a Perkin-
Elmer PHI-1600 ESCA spectrometer. The X-ray source was an Mg anode
target (350 eV). The binding energy values of each species were cor-
rected with an internal standard of C1 s (Eb = 284.8 eV).
2.3. In-situ DRIFT
The in-situ diffuse reflectance Fourier transform (DRIFT) result of
the toluene combustion over CuCZ/T catalyst was analyzed by using a
Bruker Tensor 27 spectrometer equipped with MCT detector. Prior to all
measurements, the CuCZ/T catalyst was pretreated in the high-purity
N2 atmosphere (50 mL/min) at 300 °C for 0.5 h. The toluene gas was
bubbled into the reactor by oxygen flow (30 mL/min), and the corre-
sponding sample spectrum was recorded until the catalyst was satu-
rated by the toluene adsorption. DRIFTS were probed with the tem-
perature interval of 10 ℃ from 50 ℃ to 400 ℃ under O2 and toluene
gas.
2.4. Catalytic activity measurement
Toluene catalytic self-sustained combustion experiment was carried
out in a micro-reactor made of quartz tube (Ø = 6.0 mm × 1.0 mm,
L = 200 mm). The experimental set-up of toluene catalytic self-sus-
tained combustion includes continuous flow gas supplying systems,
catalytic self-sustained combustion reactor and gaseous analytical sys-
tems (Fig. 1). The flow rates of toluene and air were controlled by mass
flow controllers, which are shown as MFC1 (F1) and MFC2 (F2) in
Fig. 1, respectively, with a full-scale measurement accuracy of 1%.
The toluene concentration was controlled by adjusting the flow rate of
F1 (10∼50 mL/min) and F2 (150∼190 mL/min) while keeping the
total flow rate constant (200 mL/min).
The reactor filled with 200 mg catalyst was first heated for ignition,
then the heat supply was turned off to observe the combustion running
auto-thermally. During the reaction process, the temperature of reactor
wall was measured by using an infrared thermography (T640, FLIR,
USA). The judgment of catalytic self-sustained combustion in the ex-
periment is based on the variation of the wall temperature, which is
2. Experimental specifications
2.1. Catalyst preparation
A series of CuxMn1-xCe0.75Zr0.25/TiO2 catalysts (x = 1, 0.5, 0) with
4.0 wt.% CuxMn1-xOy supported on TiO2 (supplied by the Evonik
Industries AG), were prepared by the incipient impregnation method.
The appropriate amount of Cu(NO3)2·xH2O, Mn(NO3)2·2H2O, Ce
(NO3)3·6H2O and Zr(NO3)4·5H2O were well dissolved in deionized
water before being added dropwise to nanometer titanium dioxide. The
resulting emulsions were dried at room temperature until complete
evaporation of the water, then staved at 105 °C for 12 h and calcined in
air at 550 °C for 4 h. The molar ratio of CuxMn1-x: Ce: Zr was 4: 3: 1, and
the CuxMn1-x: (Ce + Zr) ratio was 1: 1, which are the optimal ratio from
our previous works [15,16].
Finally, all the catalysts were sieved in a size of 20–30 meshes for
testing. The obtained catalysts were named as CuCe0.75Zr0.25/TiO2
(CuCZ/T),
Cu0.5Mn0.5Ce0.75Zr0.25/TiO2
(CuMnCZ/T)
and
MnCe0.75Zr0.25/TiO2 (MnCZ/T).
2.2. Catalyst characterization
The Powder X-ray diffraction (XRD) data was performed on XRD-
6100 (Shimadzu, Japan) equipped with X-ray diffractometer using Cu-
Kα (λ = 1.540 Å, 40 kV). The measurements were conducted in the 2Ɵ
range from 5° to 80° with a scanning rate of 2°/min. Hydrogen tem-
perature programmed reduction (H2-TPR) was measured with a PCA-
140 instrument (Bolider). To eliminate contaminants, 200 mg sample
was pretreated at 500 °C for 1 h in argon (50 mL/min). After the tem-
perature was cooled to 50 ℃, the TPR analysis were performed in 5 vol.
% H2/Ar (50 mL/min) at the temperature programmed (10 °C/min) in
the range of 50–800 °C. The reduction degree (Rd) of the catalysts was
calculated by the ratio of the hydrogen consumption of CuMnOx and
theoretical content (4%) in the catalysts. Temperature-programmed
desorption of O2 (O2-TPD) data was also collected with the same in-
strument. To remove moisture, a cold trap was introduced before the
released gas entered the thermal conductivity detector (TCD). The same
Fig. 1. Schematic of the experimental set-up for toluene self-sustained com-
bustion.
67

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
Table 1
The ignition, light-off and complete combustion of toluene catalytic combus-
tion.
Catalyst
C7H8 concentration (%)
0.35
0.50
1.00
Tig
Tlo
Tco
Tig
Tlo
Tco
Tig
Tlo
Tco
CuCZT
CuMnCZ/T
MnCZ/T
187
196
230
224
228
273
234
238
284
180
184
229
226
230
276
234
238
288
190
216
237
228
233
285
236
243
296
Tig: the ignition temperature required for 10% conversion of toluene at the
heating process.
T
lo: the light-off temperature which 50% toluene conversion is achieved at the
heating process.
Tco: the complete temperature that the conversion of toluene is 90% at the
heating process.
catalyst activity [4,5].
Fig. 2. Lean-combustion limits of toluene self-sustained combustion over
Fig. 3 illustrates the toluene conversion in the fixed bed with the
different lean-combustion limit, and the corresponding Tig, Tlo and Tco
are listed in Table 1. In Fig. 3(A), for instance, when the toluene con-
centration at 0.35 con. %, Tco of CuCZ/T, CuMnCZ/T and MnCZ/T
catalysts are 234 °C, 238 °C and 284 °C, respectively. In addition, the
heat lines of toluene conversion shift toward higher temperature with
the increase of toluene concentration and manganese content in cata-
lysts. It is worth noting that the toluene conversion over CuCZ/T re-
mains at 100% even if the temperature drops below 50 ℃ during the
cooling process, which is evident that the heat released from reaction
can maintain the stable combustion. Comparing the results of lean-
combustion limit with the Tco of different catalysts, it is implied that the
catalytic activity of CuCZ/T is the best, with the decrease order of
CuCZ/T ＞ CuMnCZ/T ＞ MnCZ/T. The phenomenon observed is dif-
ferent with Xin et al [9] in Introduction, and the synergistic effect be-
tween Cu and Mn species was failed to show in CuMnCZ/T catalyst used
in this work. A possible reason for better activity of CuCZ/T is due to
the strong interaction between copper oxide and titanium dioxide.
Highly dispersed copper oxide is converted to low-valent copper ions by
Cu2++Ti3+→Cu++Ti4+. The existence of Cu+ could be more bene-
ficial for surface adsorbed oxide ion to trap electrons through the fol-
CuCZ/T, CuMnCZ/T and MnCZ/T catalystsc.
higher than the room temperature with a range of
2 °C in 10 min.
The activity of the catalyst can be estimated in terms of the lean-com-
bustion limit (the lowest concentration required for toluene self-sus-
tained combustion over the catalyst) in the combustion. Meanwhile, it
can also be measured by the toluene conversion.
3. Results and discussion
3.1. Lean-combustion limits and catalytic activities of toluene self-sustained
combustion
The lean-combustion limits with different catalysts for toluene self-
sustained combustion in the micro-reactor were investigated by pro-
gressively decreasing the toluene concentration (described in Section
2.4) until the process was quenched, and the results are shown in Fig. 2.
It can be seen from the figure that the lean-combustion limits over
CuCZ/T, CuMnCZ/T and MnCZ/T are 0.35%, 0.50%, and 1.00%, re-
spectively. The lean-combustion limit is in direct proportional to the
lowing process: Cu++Oads→Oads−+Cu2+
, which promotes the
Fig. 3. Activity of toluene catalytic combustion with different lean-combustion limit of 0.35 con. % (A), 0.50 con. % (B), 1.00 con. % (C) and the recorded
combustion image (D) and the corresponding temperature distribution (E).
68

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
Fig. 4. H2-TPR patterns of CuCZ/T, CuMnCZ/T and MnCZ/T catalysts.
external diffusion and mass transfer. In the third step (self-sustained
combustion stage), the combustion reaction rate is mainly controlled by
the toluene and O2 diffusion rather than temperature on the catalyst
surface [19]. In cooling process, the heat transfer between the sur-
rounding and the wall of the combustion reactor is based mainly on
radiation heat transfer, and the nature convection is minor. The lower
lean-combustion limit of toluene over CuCZ/T catalyst, which generates
less heat in exothermic reaction, is favorable to minimize the tem-
perature gradient between the reactor wall and the surroundings.
Consequently, it can allow the stability of micro-combustion based on
the heat accumulated by the reaction.
Table 2
The relative intensity of the reduction peaks on the H2-TPR curves.
Catalysts
H2 consumptions (μmol·g−1
)
α peak
β peak
γ peak
Total
Rd (%)
CuCZ/T
CuMnCZ/T
MnCZ/T
25.7
36.5
29.3
0.4
0.5
0.4
98.2
104.5
45.9
0.8
0.9
0.6
70.6
50.3
41.5
0.5
0.6
0.6
194.5
191.3
116.7
1.7
2.0
1.6
38.9
53.1
41.4
∗The degree of the uncertainty was calculated by equations (1) in the sup-
porting information.
formation of radicals with strong oxidizing properties, resulting in an
improvement of the activity of the catalyst [18]. Another reason might
be that there was low content of CuMnOx (4%) in the process of catalyst
preparation, and CuOx species are prior to form the Cu-Ce-Zr-O solid
solution with Ce and Zr species detected by H2-TPR, thus limiting the
Mn species interacting with CuOx to form the Cu-Mn spinel.
3.2. Catalyst characterization
3.2.1. XRD analysis of catalysts
The XRD patterns of CuCZ/T, CMCZ/T and MnCZ/T catalysts were
shown in Fig. S1 (supporting information). All the reflections exhibited
in the samples are in agreement with the typical diffraction patterns of
TiO2 (P25). No additional peaks were observed in the patterns in-
dicating that the presence of highly dispersed species of metal oxides
supported on TiO2 or the possible formation of amorphous or nano-
crystalline of metals which are too small to be detected by XRD.
Therefore it could not be deduced with the cell parameters. There, the
In the toluene self-sustained catalytic combustion, the continuous
combustion can be achieved by the heat released from the reaction, and
no external heating is required. The catalyst with lower activity needs
higher toluene concentration for affording higher ignition temperature
and achieving intense exothermic reaction rapidly. Take CuCZ/T for
instance (Fig. 3A), the heating process of catalytic self-sustained com-
bustion can be divided into three steps. The first step is described as an
induction process with which the toluene conversion increases slowly at
a lower temperature (< Tig = 187 ℃). Here, the consumed reactants
can be quickly supplemented by the internal diffusion, as embodied in
the temperature of the catalyst bed near the setting temperature. When
the sufficient power is generated from reaction exothermicity in com-
parison to the heat dissipation in the process, the toluene conversion is
fast enough to ignite, occurring at local hot points of catalyst surface.
The accumulation of heat induces a steep increase of the local tem-
perature and reactant rates, then the flame spreads rapidly in the cat-
alyst bed to enter the second step－－transient light-off period
(< Tig–Tlo (224 ℃)). Here toluene oxidation rate is increasing to lead a
thermochemical runaway reaction, which consequently expedites the
content of CuxMn1-xO2-δ, CeO2 and ZrO2, which were calculated before
the preparation of the catalysts, are 4.0%, 8.1% and 5.8%, respectively.
3.2.2. H2-TPR analysis of catalysts
The redox properties of CuCZ/T, CuMnCZ/T and MnCZ/T catalysts
investigated by H2-TPR are displayed in Fig. 4 (The TPR patterns of
pure CuO, MnO2 and CeZrOx are presented in Fig. S2). In Fig. S2, the
reduction of pure CuO is observed at 150–350 °C and characterized by
two overlapping peaks at 240 and 264 °C. The peak at lower tempera-
ture is attributed to the reduction of highly dispersed CuOx on the
surface of catalyst, including the isolated Cu2+ and the small-scale
copper clusters with two dimensions and three dimensions structures.
And another peak represents the reduction of Cu+ species [20]. For
69

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
Fig. 5. O2-TPD patterns of CuCZ/T, CuMnCZ/T and MnCZ/T catalysts.
because of its low free energy to form the MnOx [13]. The three peaks
observed for MnCZ/T at 300–500 ℃ is similar to the pure MnO2. It is
divided into the three peaks, labeled α (200–400 ℃), β (250–500 ℃)
and γ (300–550 ℃), are related to the reductions of MnO2, Mn2O3 and
the Mn3O4, respectively [12].
Table 3
The relative intensity of the desorption peaks on the O2-TPD curves.
catalysts
O2 consumption (μmol·g−1
)
adsorbed oxygen
surface lattice
The corresponding H2 consumptions and the Rd of CuMnOx are
listed in Table 2. It is suggested that the reduction of Mn4+ → Mn3+
might occur at the lower temperature to increase the H2 consumption of
β peak in CuMnCZ/T catalyst [22]. Addition of Mn to CuCZ/T slightly
modified the redox ability of catalyst and promoted the reduction at
lower temperature. That can be explained by the fact that an amount of
MnOx doped with CuO can form the lattice defects and oxygen va-
cancies, and consequently enhance the transfer of oxygen species to
promote each other’s reduction ability [23]. The total hydrogen con-
α peak
β peak
γ peak
δ peak
φ peak
CuCZ/T
CuMnCZ/T
MnCZ/T
8.0
4.0
0.3
0.1
6.0
5.0
0.2
0.2
10
1.0
6.0
0.4
0.2
0.2
1.0
0.5
1.0
0.4
0.2
0.2
0.7
–
–
0.4
12
0.4
13
0.4
pure MnO2, there is only a broad hydrogen consumption below 500 °C
related to the reduction of MnO2. The hydrogen consumption of CeZrOx
are shown a single reduction peak at 400–700 °C attributed to the
partial reduction of Ce4+ to Ce3+ at the surface and bulk of the Ce-Zr-O
solid solution [21], is not detected in the TPR profiles of catalysts
(Fig. 4).
sumption of the CuMnCZ/T(191.3
on the CuCZ/T(194.5
1.7 μmol·g−1), which are higher than that of
MnCZ/T (116.7
2.0 μmol·g−1) is similar to that
1.6 μmol·g−1). The reason probably is due to that
the more content of easier reduction Mn4+ ions of CuMnCZ/T catalyst
in comparison with MnCZ/T [24], which were detected by XPS. It is
also consistent with the performance of catalytic activity to some ex-
tent.
As shown in Fig. 4, the reduction peaks for all the catalysts can be
divided into three peaks, labeled as α, β and γ. The reduction of CuO in
CuCZ/T catalyst takes place at the lower temperature than the pure
CuO species, including the main peak (β) at the range of 150–270 ℃
and a smaller peak (γ) at 200–400 ℃. The addition peak (α) at the
lowest temperature about 170 °C can be ascribed to the reduction of the
CuOx species that interact with ceria [13]. The dispersed CuOx species
could incorporate into the cubic vacant sites of ceria and form a five-
coordinated surface structure with the capping oxygen, and this un-
steady structure can promote the reduction of the CuOx species and the
surface oxygen [22]. For CuMnCZ/T catalyst, the attributions of three
peaks are similar to that of CuCZ/T, including the α peak (50–350 ℃), β
peak (150–250 ℃) and γ peak (200–300 ℃). According to the literature,
it is attributes to the reduction of the CuOx and MnOx [20]. But it is
difficult to distinguish the peak corresponding to copper or manganese
species because the peaks were overlapping with each other. However,
MnCZ/T is reduced difficultly in comparation with other catalysts
From the Table 2, the results showed that the Rd of catalysts with
the decrease order of CuMnCZ/T (53.1%) ＞ MnCZ/T (41.4%) ＞
CuCZ/T (38.9%). In this work, the Tig and Tco of toluene self-sustained
combustion are below 300 ℃. However, manganese starts to be reduced
when the temperature is higher than 300 ℃. Probably it is the main
reason that catalysts with manganese has the higher reduction degree
and the lower activity. Hence, the catalytic activity is mainly related to
the reduction temperature of catalysts, and is little affected by the re-
duction degree.
3.2.3. O2-TPD spectra of the catalysts
The O2-TPD profiles of the CuCZ/T, CuMnCZ/T and MnCZ/T sam-
ples are shown in Fig. 5, and the relative intensity of the desorption
70

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
Fig. 6. XPS patterns of Cu 2p (A), Mn 2p (B), Ce 3d (C), and O 1 s (D) for CuCZ/T, CuMnCZ/T and MnCZ/T catalysts.
Table 4
The surface species of CuCZ/T, CuMnCZ/T and MnCZ/T catalysts.
Caralyst
(Cu + Mn)/Ti%
Cu+ /Cu2+
Ce3+/Ce4+
Oxygen Species (%)
Oad
Ola
Ola/Oto
CuCZ/T
CuMnCZ/T
MnCZ/T
2.00
2.58
1.60
0.06
0.07
0.09
0.31
0.26
–
0.02
0.01
0.35
0.31
0.25
0.02
0.02
0.01
11
32
14
1
1
1
76
67
85
2
1
2
87.36
67.67
85.54
0.10
0.33
0.27
Ola=Ola,r+Ola,c, Oto=Oad+Ola = 99%.
peaks on the O2-TPD curves are listed in Table 3. The catalysts exhibit
similar spectra and can be distinguishably classified into two types of
oxygen species, namely, the adsorbed oxygen located at surface va-
cancies below 550 ℃ and the surface lattice oxygen at the higher
temperature between 700 ℃ and 900 ℃. The first broad peak can be
divided into three characteristic peaks (denoted as α, β and γ peak)
differentiated by the strength of the catalyst surface, which are attrib-
oxide solid solution with fluorite structure, thereby strengthens the
capacity of the active center of oxygen [21].
3.2.4. XPS measurements
XPS is used to characterize the Cu, Mn, Ce and O oxidation state for
CuCZ/T, CuMnCZ/T and MnCZ/T catalysts, and the results are depicted
in Fig.6. The Cu 2p spectrum for the catalysts represents two main
peaks of Cu 2p3/2 (933.6 eV) and Cu 2p1/2 (953.7 eV) in Fig. 6A. After
the peak deconvolution, the coexistence of Cu+/Cu2O and Cu2+/CuO
are well evidenced by the main peaks at 932.6 eV with a shoulder peak
observed at 934.3 eV. The shake-up satellites of Cu at about 942.2 eV
are assigned to the divalent copper [27]. The Mn 2p spectra exhibits
two contributions of Mn 2p3/2 and Mn2p1/2 located at 642.2 eV and
653.6 eV in Fig. 5B, corresponding to Mn3+ and Mn4+ ions, respec-
tively [12].
In Fig. 6C, the Ce 3d spectra is distinguished into Ce 3d5/2 and Ce
3d3/2 (labeled as u and v, respectively) performing the Ce3+/Ce4+
electron-pair. The four intense peaks at u (898.1 eV), v (916.6 eV), u3
(882.3 eV) and v3 (900.9 eV) with the two weak peaks at u1 (888.9 eV)
and v1 (908.3 eV), can be attributed to the different Ce 4f electron
configurations in the final states of the Ce4+ species. The u2 and v2
peaks presented at 904.7 eV and 884.1 eV belong to the Ce3+ species
uted to the weakly adsorbed oxygen species, O2 and O− species
−
formed by the adsorbed O2 [25]. The δ and φ peaks are assigned to
lattice oxygen species (O2−) [26]. The lattice oxygen is considered as a
major participator in deep oxidation of toluene, which is deeply dis-
cussed in Section 3.3. Compared with CuMnCZ/T (0.5 μmol g−1) and
MnCZ/T (1.0 μmol g−1), CuCZ/T has the highest content of lattice
oxygen (1.7 μmol g−1), indicating a largest amount of active oxygen
centers for the catalyst. The higher concentration of Ce3+ of CuCZ/T
catalyst, which was detected by XPS, can promote the migration of the
lattice oxygen from the bulk to the surface of catalyst. Sequentially,
more active oxygen centers to facilitate the mobility of the lattice
oxygen [25]. Obviously, the desorption peak of δ for CuMnCZ/T shifts
slightly to lower region in comparison with CuCZ/T and MnCZ/T. The
phenomenon can be explained by the fact that the presence of the ap-
propriate amount of Mn enhances the formation of Cu-Mn-Ce-O ternary
71

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
observed that the contents of (Cu + Mn)/Ti for the obtained catalysts
are less than the theoretical value (loading 4.0 wt%), which confirms
that the Cu and/or Mn ions are introduced into CeO2-ZrO2 lattice and
the lattice imperfections are formed. The higher Cu+/Cu2+ ratio in
CuCZ/T (0.31) compared to CuMnCZ/T (0.26) provides the evidence of
the interaction between divalent copper ions and titanium ions. Con-
sequently, the cerium species with lower oxidation states via the redox
equilibrium of Cu++Ce4+↔ Cu2++Ce3+ [5,28], improve the ratio of
Ce3+/Ce4+ (0.35) in CuCZ/T, which is higher than that in CuMnCZ/T
(0.31). The presence of more Cu+ species in the catalysts can promote
the catalytic performance for the oxidation reaction. The replacement
of Ce4+ from Ce3+ can form the defect structure, resulting in the
oxygen vacancies and unsaturated chemical bonds on the catalyst sur-
face, and bringing the transfer of more chemisorbed oxygen species in
ceria into lattice oxides for charge balance [28]. Ola species is the main
participator in the process of toluene oxidation according to MVK re-
action mechanism. Consequently, the highest Ola ratio is up to 87.36%
in CuCZ/T compared with 67.67% in CuMnCZ/T catalyst, indicating
the lattice oxygen species relative concentration increases with the
content of copper. The result is confirmed by O2-TPD.
3.3. Reaction pathway of toluene catalytic combustion
Fig. 7. The in-situ DRIFT spectra of toluene oxidation over CuCZ/T catalyst. (A)
The wavenumber range of 1000 ∼ 4000 cm−1. (B) The wavenumber range of
The catalytic combustion of toluene can generate some intermediate
products before complete oxidation into CO2 and H2O. In our previous
work, the analysis of intermediate species combined with the Mars–van
Krevelen (MVK) mechanism could explain the oxidation performances
of ethyl acetate and toluene on CuCZ/T catalyst to some extent [17].
However, the specific reaction pathways including the adsorption and
oxidation of toluene are still not clear.
In this study, the in-situ DRIFT was used to further analyse the re-
action process of toluene catalytic combustion over the CuCZ/T cata-
lyst, and the results are shown in Fig. 7. The samples were saturated
under the toluene-oxygen atmosphere at the room temperature, until
the changeless intensity of the characteristic IR bands associated with
toluene, including the bands at 3100–2800 cm−1 (the stretching vi-
bration of CeH groups), and the bands at 1610–1380 cm−1 (the typical
aromatic ring vibration) [29]. It is worth noting that the bands at 2934
and 2886 cm−1 are assigned to the methylene rather than methyl, be-
cause the symmetric and asymmetric CeH stretching vibration of the
CH3 group is usually observed in the 2970–2950 cm−1 region [10]. The
appearance of weak bands at 1153 and 1082 cm−1 are probably
1000 ∼ 2000 cm−1
.
[16], which is due to the entrance of copper/manganese ions into the
CeO2 lattice causing lattice imperfection and generating trivalent
cerium ions [27].
The O 1 s spectra of the three catalysts is clearly resolved into three
peaks by deconvolution (Fig. 6D), proving the presence of three che-
mical environments for oxygen atoms. The value of 529.6 eV is gen-
erally accepted as being a feature of the characteristic lattice oxygen
(Ola,m) bonded to copper, manganese, cerium and zirconium cations.
The second one at higher binding energies (532.1 eV), with lower in-
tensity, is the characteristic of the regular lattice oxygen (Ola,t) of TiO2,
and the shoulder peak at 533.3 eV indicates the presence of chemi-
sorbed oxygen (Oad) [27].
The related description for the XPS data is listed in Table 4, in-
cluding the surface species of (Cu + Mn)/Ti, Cu+/Cu2+, Ce3+/Ce4+
and atomic ratio of the CuCZ/T, CuMnCZ/T and MnCZ/T catalysts. It is
Fig. 8. The toluene combuston over CuCZ/T catalyst in N2 (160 ℃).
72

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
Fig. 9. The reaction pathway of toluene combustion.
ascribed to a CeO vibration mode [10]. These results suggest that to-
luene is adsorbed on the CuCZ/T surface via the cleavage of the α-H
bond in the methyl group of toluene to form the benzyl species, which
are subsequently combined with the active oxygen on the catalyst
surface through the formation of benzoyl oxide species (C6H5-CH2-O)
[10,30]. In addition, the positions of these bands do not shift to lower
frequency compared with the spectra of liquid- or gaseous toluene. This
phenomenon also testifies that the toluene is adsorbed on CuCZ/T
surface via the −CH3 rather than the aromatic ring, because the in-
teraction between the aromatic ring of toluene and the surface of TiO2
can lead to an obvious blue shift by forming π-complex [10]. The ne-
gative band at 3671 cm−1 and the band at 3593 cm−1 are caused by the
surface hydroxyl group interacting with the methyl group of toluene
[31], and it confirmsthe methyl group of toluene is adsorbed on catalyst
once again.
Under low temperature conditions, hydroxyl groups adsorbed on
the catalyst surface occupy the active sites, which hinders the futher
oxidation of toluene. In Fig. 7A, it can be seen that when the band at
3593 cm−1 disappears, the signals of CO2 (2368 and 2341 cm−1) [5]
increase rapidly with the decreasing of the toluene signals. In addition,
the new bands appearing at 1301 and 1239 cm−1 (Fig. 6B) are attrib-
uted to the methylene group and the CeO vibration mode of benzyl
alcohol [29]. With the temperature increasing, the occurrence of the
strong peaks at 1585, 1518 and 1409 cm−1 resulting from the carbox-
ylate group suggests the production of benzoic acid [32]. Furthermore,
the weak bands at 1695 and 1648 cm−1 designated to aldehydic species
are also tracked. And the weak bands at 1798 and 1737 cm−1 imply the
appearance of anhydride species [25]. The band intensities of benzoic
acid, aldehydic and anhydride species continuously increase with the
increasing of temperature. Noticeably, with the reaction temperature
higher than 350 ℃, the intensities of CO2 bands are significantly higher
than the others. According to the integral of the band areas at 400 ℃,
the formation of CO2 has a approximately ratio of 98%, which de-
monstrates that CO2 is the dominant product of toluene catalytic oxi-
dation with traces of benzoic acid, aldehydic and anhydride species as
by-products.
with the previous result found by Sun et al. [10]. Furthermore, the
amount of lattice oxygen consumed in the reaction by integral shows a
good qualitative agreement with the peak area of lattice oxygen in-
vestigated by O2-TPD.
To sum up, combining with our previous study [17], the model
established on the MVK reaction mechanism provides a good fit for the
toluene combustion over CuCZ/T. Based on the redox process, there are
two steps in the MVK model. Firstly, adsorbed toluene reacts with the
lattice oxygen on the catalyst surface to produce the oxidation products,
resulting in the reduction metal oxides. Secondaly, the reduced metal
oxides are reoxidized by the gas phase oxygen. As schematically shown
in Fig. 9, toluene is adsorbed on the active oxygen species on the cat-
alyst surface, which occurs in the initiation by removing the α-H of
methyl to form the benzoyl oxide species (C6H5eCH2eO) (dark blue
line), and then the reaction moves into subsequent oxidation steps
(green line). Toluene is firstly converted to benzyl alcohol by lattice
oxygen of the catalyst, then benzyl alcohol is further oxidized to ben-
zoic acid, benzaldehyde and anhydride species on the surface of the
catalyst, which are ultimately mineralized to CO2 and H2O to release
into the atomsphere. In this case, the CuCZ/T catalyst seems to be a
reservoir of oxygen species to sustain oxidation, and the Cu-Ce-O2-x
solid solution and CeO2 can provide the significant oxygen vacancy.
The lattice oxygen species are consumed for the oxidation of toluene
and form oxygen vacancy, which can be replenished easily by the gas-
phase oxygen activated on the catalyst surface (rose red line). This re-
sults can be explained by the coexistence of the mixed-valent Cu and
the facile transform between Cu2+ and Cu+, as shown from XPS results.
And the presence of coupled Cu2+/Cu+ sites with Ce4+/Ce3+ sites can
enhance the oxygen mobility throughout the bulk and the surface
through electron exchange with cerium ions.
4. Conclusions
The catalytic self-sustained combustion of toluene over the CuCZ/
T、CuMnCZ/T and MnCZ/T catalysts was investigated. The lean-com-
bustion limits of toluene self-sustained combustion decrease with the
improvement of the intrinsic catalyst activity. The results demonstrate
that CuCZ/T is more efficient than others for toluene self-sustained
combustion based on the lower Tig, Tco and the lean-combustion limits.
The performance enhancement of CuCZ/T is probably attributed to the
homogeneous morphology, the high redox properties and the strong
oxygen storage/release ability, because of the greater quantities of
monovalent copper ions and lattice oxygen.
To investigate whether the lattice oxygen participates in the com-
bustion of toluene or not, the experiment of the toluene combustion
over CuCZ/T catalyst (0.8 g) at 160 ℃ (the temperature at which CO2
can be tracked firstly through GC) in N2 atmosphere was carried out,
and the results are depicted in Fig. 8. The sample was first pretreated
with N2 for 0.5 h at 400 ℃ to eliminate the effects of surface oxygen
species. Although the reactants contain no gas-phase oxygen, CO2 was
still detected by GC, which proves that lattice oxygen is in the in-
volvement of the deep catalytic oxidation of toluene, in accordance
In-situ DRIFT results indicate the reaction pathway of the toluene
catalytic oxidation. The active oxygen species are the adsorption sites
73

C. Zhao et al.
AppliedCatalysisA,General569(2019)66–74
for the toluene oxidation. Toluene is adsorbed on the catalyst surface by
abstracting a-H of the methyl group and forming benzoyl intermediates.
These intermediates are oxidized successively into aldehydic, benzoate
species, which are ultimately mineralized into CO2 and H2O. by the
lattice oxygen in the catalyst.
[9] U. Menon, V.V. Galvita, G.B. Marin, J. Catal. 283 (2011) 1–9.
[10] H. Sun, Z. Liu, S. Chen, X. Quan, Chem. Eng. J. 270 (2015) 58–65.
[11] P. Liu, H. He, G. Wei, D. Liu, X. Liang, T. Chen, J. Zhu, R. Zhu, Microporous
Mesoporous Mater. 239 (2017) 101–110.
[12] X. Li, L. Wang, Q. Xia, Z. Liu, Z. Li, Catal. Commun. 14 (2011) 11–19.
[13] X. Guo, J. Li, R. Zhou, Fuel 163 (2016) 56–64.
[14] H.C. Genuino, S. Dharmarathna, E.C. Njagi, M.C. Mei, S.L. Suib, J. Phys. Chem. C
116 (2012) 969–976.
Acknowledgments
[15] Q. Yang, Q. Hao, N. Yan, R. Zhao, C. Zhao, Q. Zhang, B. Dou, F. Bin, J. Fuel Chem.
Tech. 45 (2017) 1401–1408.
[16] S. Li, Q. Hao, R. Zhao, D. Liu, H. Duan, B. Dou, Chem. Eng. J. 285 (2016) 536–543.
[17] B. Dou, S. Li, D. Liu, R. Zhao, J. Liu, Q. Hao, F. Bin, RSC Adv. 6 (2016) 53852-
538553859.
This work was accomplished under the support of National Nature
Science Foundation of China (No 51776216) and Research Centre of
Modern Analytic Technology in Tianjin University of Science ＆
Technology.
[18] S. Hao, Y. Yin, Z. Ma, Y. Qing, F. He, Xi. Qi, J. Hebei Univ. (Nat. Sci. Ed. 25 (05)
(2005) 486–494.
[19] Y. Wang, Research on Carbon Monoxide Self-sustained Combustion Based on CuCe1-
xZrxOy Series Catalyst, Xi’an: Xi’an Jiaotong University, 2016.
[20] X. Guo, J. Li, R. Zhou, Fuel 163 (2016) 056–064.
[21] L. Li, F. Jing, J. Yan, J. Jing, W. Chu, Catal. Today 297 (2017) 167–172.
[22] L. Liu, Basic Research of Catalytic Reduction of Nn by The Catalytic Reduction of
Copper Oxide Nano-Catalyst in Cerium and Zirconia, Nanjing university, Nanjing,
2010.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2018.10.034.
[23] H. Lu, X. Kong, H. Huang, Y. Zhou, Y. Chen, J. Environ. Sci: China 32 (2015)
102–107.
References
[24] F. Hu, J. Chen, S. Zhao, K. Si, H. Song, J. Li, Appl. Catal. A Gen. 540 (2017) 57–67.
[25] V.P. Santos, M.F.R. Pereira, J.J.M. Órfão, J.L. Figueiredo, Appl. Catal. B: Environ.
99 (2010) 353–363.
[1] S.C. Kim, W.G. Shim, Appl. Catal. B: Environ. 98 (2010) 180–185.
[2] W. Yang, Y. Wang, J. Zhou, Z. Wang, K. Cen, Chem. Eng. Sci. 158 (2017) 30–36.
[3] K. Bijjula, D.G. Vlachos, Proc. Combust. Inst. 33 (2011) 1801–1807.
[4] A. Scarpa, R. Pirone, G. Russo, D.G. Vlachos, Combust. Flame 156 (2009) 947–953.
[5] F. Bin, X. Wei, B. Li, K.S. Hui, Appl. Catal. B: Environ. 162 (2015) 282–288.
[6] F. Bin, X. Wei, T. Li, D. Liu, Q. Hao, B. Dou, Proc. Combust. Inst. 36 (2017)
4193–4200.
[7] Z. Huo, Transition Metal Oxides Nanostructures: Controlled Synthesis, Assembly
and Their Properties, Tsinghua University, Beijing, 2009.
[8] N. Viswanadham, S.K. Saxena, A.H. Al-Muhtaseb, Mater. Today Chem. 3 (2017)
[26] C. Deng, Q. Huang, X. Zhu, Q. Hu, W. Su, J. Qian, L. Dong, B. Li, M. Fan, C. Liang,
Appl. Surf. Sci. 389 (2016) 1033–1049.
[27] R. Kang, X. Wei, F. Bin, Z. Wang, Q. Hao, B. Dou, Appl. Catal. A Gen. (2018),
https://doi.org/10.1016/j.apcata.2018.07.026.
[28] B. Dou, G. Lv, C. Wang, Q. Hao, K. Hui, Chem. Eng. J. 270 (2015) 549–556.
[29] J. Li, H. Na, X. Zeng, T. Zhu, Z. Liu, Appl. Surf. Sci. 311 (2014) 690–696.
[30] Y. Liu, H. Huang, X. Li, S. Sun, D. Zhao, J. Saf. Environ. 5 (2005) 14–16.
[31] F. Rainone, D.A. Bulushev, L.K. Minsker, A. Renken, Phys. Chem. Chem. Phys. 5
(2003) 4445–4449.
[32] S. Besselmann, E. Löffler, M. Muhler, J. Mol. Catal. A Chem. 162 (2000) 401–411.
37–48.
74