Enhanced performance of low Pt loading amount on Pt-CeO2 catalysts prepared by adsorption method for catalytic ozonation of toluene

Article abstract of DOI:10.1016/j.apcata.2021.118342

The supported catalyst of Pt-CeO₂/BEA with low loading amount (0.07 wt%) of Pt nanoparticles presented competitive performance towards catalytic ozonation of toluene. Characterization results demonstrated that the improved activity could be attributed to the enhanced Pt-CeO₂ strong metal-support interaction induced by the formation of much smaller, more uniform and dispersive Pt nanoparticles through the adsorption method. Furthermore, Pt/CeO₂ and Pt/SiO₂ were employed to elucidate the intrinsic reasons for the activity improvement. Various techniques such as XPS, Raman, O₂-TPD, etc. were performed to characterize the physicochemical properties. Results showed that the anchoring effect for Pt nanoparticles by the Pt-O-Ce bonds, greater amount of oxygen vacancies and more generated peroxide species are potentially responsible for the improvement of catalytic performance. In-situ Raman confirmed the participation of peroxide species in toluene oxidation, which were previously formed during the ozone decomposition. Plausible reaction pathways for the involving reactions on catalyst surface were also proposed.

Full text of DOI:10.1016/j.apcata.2021.118342

Applied Catalysis A, General 625 (2021) 118342
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Enhanced performance of low Pt loading amount on Pt-CeO catalysts
2
prepared by adsorption method for catalytic ozonation of toluene
a
,
1
a
,
1
a
,
*
a
a
a
Jialing Wang , Xuefeng Shi , Longwen Chen , Huanyi Li , Mengqi Mao , Guangyi Zhang ,
c
a
a
a,b,**
Hui Yi , Mingli Fu , Daiqi Ye , Junliang Wu
a
School of Environment and Energy, South China University of Technology, Guangzhou 510006, China
Advanced Plasma Catalysis Engineering Laboratory for China Petrochemical Industry, Changzhou 213000, China
Liuzhou Institute of Technology, Liuzhou 545000, China
b
c
A R T I C L E I N F O
A B S T R A C T
Keywords:
The supported catalyst of Pt-CeO /BEA with low loading amount (0.07 wt%) of Pt nanoparticles presented
2
Catalytic ozonation
competitive performance towards catalytic ozonation of toluene. Characterization results demonstrated that the
2
Pt-CeO catalysts
improved activity could be attributed to the enhanced Pt-CeO
formation of much smaller, more uniform and dispersive Pt nanoparticles through the adsorption method.
Furthermore, Pt/CeO and Pt/SiO were employed to elucidate the intrinsic reasons for the activity improve-
ment. Various techniques such as XPS, Raman, O -TPD, etc. were performed to characterize the physicochemical
2
strong metal-support interaction induced by the
Particle adsorption method
Peroxide species
2
2
In-situ Raman
2
properties. Results showed that the anchoring effect for Pt nanoparticles by the Pt-O-Ce bonds, greater amount of
oxygen vacancies and more generated peroxide species are potentially responsible for the improvement of cat-
alytic performance. In-situ Raman confirmed the participation of peroxide species in toluene oxidation, which
were previously formed during the ozone decomposition. Plausible reaction pathways for the involving reactions
on catalyst surface were also proposed.
1
. Introduction
(activated carbon, zeolite, γ-Al
2 3
O , etc.) to increase the number of active
sites and improve the gas-solid interactions [6]. Based on the literature
Volatile organic compounds (VOCs) are an important pollutant that
[6–8], many catalysts display a moderate catalytic activity. However,
can seriously affect the quality of the atmospheric environment and are
directly harmful to humans by causing short- and long-term adverse
health effects [1]. In recent years, VOCs control technologies such as
adsorption, catalytic combustion, non-thermal plasma degradation, etc.
are widely investigated [2]. Among them, catalytic ozonation of VOCs,
also named ozone-catalytic oxidation of VOCs that can effectively and
the low mineralization rate, undesired by-product formation and O
3
residues significantly hinder their practical application [9,10]. Zhu et al.
[11] have conducted catalytic ozonation of benzene on AgMn/HZSM-5
catalysts at room temperature. The results showed that the CO
2
selec-
tivity was lower than 60% even though benzene and O achieved the
3
highest conversion. Similar results have also been observed by other
authors [12–15]. Hui et al. [15] have prepared a series of MCM-41
catalysts modified by different transition metal oxides (Co and Cu) for
simultaneously decompose VOCs and O
3 2 2
to the harmless CO , H O and
O
2
under mild and environmentally friendly conditions, has long been
the subject of intense studies [3]. This method can be applied not only
for the treatment of industrial organic waste gas but also for indoor air
purification, representing a potential VOCs control technology [4,5].
At present, the most common catalysts for the catalytic ozonation of
2
catalytic ozonation of toluene. At 22 ℃, the CO selectivity reached only
40% in the Co-modified MCM-41, which, however, represented the best
activity. Therefore, it is still of great necessity to develop novel catalysts
with enhanced performance.
VOCs are transition metal-based catalysts, such as MnO
x
, FeO
x
, CoO
x
,
Due to the high oxygen storage capacity and the possibility of readily
switching ability between Ce3 and Ce , together with rich oxygen
+
4+
etc., which are generally dispersed on support with high surface area
*
Corresponding author.
** Corresponding author at: School of Environment and Energy, South China University of Technology, Guangzhou 510006, China.
E-mail addresses: chenlongwen@scut.edu.cn (L. Chen), ppjl@scut.edu.cn (J. Wu).
1
These authors contributed equally to this work.
https://doi.org/10.1016/j.apcata.2021.118342
Received 6 May 2021; Received in revised form 3 August 2021; Accepted 30 August 2021
Available online 2 September 2021
0
926-860X/© 2021 Elsevier B.V. All rights reserved.

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
vacancy concentration, CeO
2
exhibits special catalytic properties [16].
conversion and CO
selectivity for the catalytic ozonation of VOCs by using the CeO . Zhang
et al. [17] have degraded formaldehyde and O over CeO and the re-
sults showed that the CO yield could achieve 97% with an O conver-
sion of approximately 90% at room temperature. The high performance
of CeO is associated with its surface oxygen vacancies, which are
favorable for O adsorption and decomposition to provide active oxygen
species (AOS). As a reducible material, CeO can forcefully bind the
Afterward, the slurry was dried at 110 ℃ overnight and calcined at 550
℃ for 5 h with a heating rate of 1 ℃/min.
Many researches demonstrate the relatively high O
3
2
2
Pt-CeO
method, denoted as Pt-CeO
cess of Pt NPs was identical to that described previously, i.e., adding a
certain amount of CeO /BEA into the Pt NPs solution under ultrasonic
treatment for 1 h. After drying at 80 ℃ overnight and calcining at 550 ℃
for 5 h, the Pt-CeO /BEA-P catalyst was obtained.
Meanwhile, another Pt-CeO /BEA catalyst was synthesized via the
incipient wetness impregnation method for comparison, denoted as Pt-
CeO /BEA-IM. A certain amount of CeO /BEA was added to the H PtCl
2
/BEA catalyst was prepared by the Pt NPs adsorption
3
2
2
/BEA-P. The preparation and loading pro-
2
3
2
2
3
2
2
2
highly dispersed metal particles with oxygen linkages and generate
active sites through the strong metal-support interaction (SMSI),
involving electron transfer and formation of metal-O-Ce bonds on the
2
2
2
6
solution under ultrasonic treatment. After drying overnight, the solid
interface of metal and CeO
catalytically active components and are generally combined with CeO
to improve catalytic activity and stability [19]. Various preparation
methods have been applied to prepare the Pt-CeO catalyst by tuning the
SMSI between Pt NPs and CeO . Among them, Pt NPs adsorption method
2
[18]. Pt nanoparticles (NPs) are promising
product was calcined at 550 ℃ for 5 h.
2
2.1.3. Preparation of Pt/CeO
CeO was obtained from Ce(NO
furnace at 550 ℃ for 5 h. Commercial SiO
Factory, Tianjin, China) was selected as the contrast support. Both Pt/
CeO and Pt/SiO were synthesized through the Pt NPs adsorption
method. To better elucidating the Pt-CeO SMSI, the loading amount of
Pt NPs was moderately increased to approximately 0.4 wt%. Typically, a
certain amount of CeO or SiO was added to the Pt NPs solution. After
2
and Pt/SiO
2
2
2
3
)
3
.6 H
2
O by calcined in the muffle
2
(Fuchen Chemical Reagent
2
is intensively reported owing to the operability, repeatability and
controllability of particle size and loading amount. He et al. [20] and
Sun et al. [21] have prepared supported noble metal catalysts of
2
2
2
2
Pt-Mn/SiO and Pd/Ce-SBA15, respectively, through the adsorption
method. The results have demonstrated an improved catalytic activity
due to the SMSI. Nevertheless, the loading amount of noble metal NPs
for supported catalysts in the literature is usually approximately 1 wt%
or more, which breaches the cost-effectiveness and restricts their in-
dustrial application [22]. Reducing the amount of precious metals while
maintaining or even enhancing the catalytic performance is still a great
challenge for scientific research.
2
2
ultrasonic treatment for 1 h, the mixed slurry was dried at 80 ℃ over-
night and finally calcined at 350 ℃ for 5 h.
2.2. Catalyst characterization
Physicochemical properties of the catalysts were analyzed through
In the present study, a cost-effective Pt-CeO
2
/BEA catalyst was pre-
various techniques, such as XRD, N
2
adsorption/desorption, SEM,
pared by the Pt NPs adsorption method with a Pt loading amount lower
in order of magnitude than those reported in the literature. The sample
showed encouraging performance on the catalytic ozonation of toluene
HRTEM, XPS, H -TPR, O -TPD, etc. Detailed experimental procedures
2
2
can be seen in Supporting Information (SI).
with a remarkably increased CO
2
selectivity. Furthermore, Pt/CeO
2
and
2.3. In-situ Raman spectroscopy
Pt/SiO
Pt-CeO
HRTEM, H
2
were also synthesized to elucidate the promotion effect of the
-
2
SMSI. Various characterization techniques such as XRD,
Because of the covalency of the O-O bond in superoxide (O ) and
2
peroxide species (O 2-), their stretching vibration has a high response in
2
2
-TPR, O -TPD, XPS, in-situ Raman, etc. were used to inves-
2
tigate the formation of the Pt-O-Ce bonds, variation of oxygen vacancy
concentration and the type and role of AOS. The goal of this work is to
develop highly efficient supported catalysts by lowering the loading
amount of precious metal while enhancing the catalytic performance.
the visible Raman spectra, which were thus employed to characterize
these two important AOS. In this case, the wavelength of the excitation
-
1
light source was 532 nm and the scanning range was 700–1400 cm .
The in-situ Raman experiment process was designed as shown in Fig. S1.
To remove adsorbed water from the surface, all the samples were pre-
2
. Experimental
treated at 150 ℃ for 1 h under N
signals of the sample in N atmosphere after cooling to 30 ℃; (2)
switching the gas to 21% O /N for O adsorption and recording the
signals after blowing for 30 min; (3) switching the gas to 4900 ppm O
(O ) and recording the signals after reaction for 30 min; (4)
+N
switching the gas to N to remove the weakly adsorbed surface oxygen
species and recording the signals after blowing for 30 min; (5) switching
2
flow. (1) Recording the initial Raman
2
2
2
.1. Catalyst preparation
2
2
2
3
/
.1.1. Preparation of Pt NPs
2
2
All the involved chemicals are analytical grade without any further
2
purification. Pt NPs were synthesized by the glycol reduction method, as
described elsewhere [23]. Briefly, 2.5 mL H PtCl solution (20 g/L) and
20 mg PVP (k29-32, M
=58000) were dissolved in 25 mL glycol in a
2
the gas to 940 ppm toluene/N and recording the signals after reaction
2
6
2
w
for 30 min
round-bottom flask, and a yellow solution was obtained. Under mild
stirring, the yellow solution was heated to 120 ℃ for 1 h, until the color
changed to dark-brown. Then, 60 mL acetone was slowly added to the
dark-brown solution to extract Pt NPs. After cooling down to room
temperature, Pt NPs were collected by centrifugation and washed with
n-hexane several times. Finally, the obtained Pt NPs were dispersed into
ethanol for subsequent utilization.
2.4. Catalytic performance evaluation
25 mg catalyst (40–60 mesh) was placed into a quartz tube (8 mm
inner diameter and 500 mm length) with quartz wool packed at both
ends of the catalyst bed. Before each experiment, the catalyst was
purged with N
molecules. After cooling to room temperature, a gas mixture of (O
) or (toluene, O and N ) passed through the reactor while increasing
the temperature. The concentrations of toluene, CO and CO were
analyzed by an online gas chromatograph (GC-9560, China) equipped
with two flame ionization detectors (FID); O concentration was recor-
ded by an O analyzer (2B, 106-M, USA). O decomposition (XO ),
toluene conversion (XToluene), CO selectivity (SCO ), carbon balance
(Bcarbon) and CO yield (YCO ) were calculated according to the equations
2
at 120 ℃ for 30 min to remove the adsorbed water
3
and
2
.1.2. Preparation of Pt-CeO
CeO /BEA was prepared by the incipient wetness impregnation
method with β-zeolite (BEA) as support. Typically, a certain amount of
Ce(NO .6H O was dissolved in deionized water. After stirring for 10
2
/BEA
N
2
3
2
2
2
3
)
3
2
3
min, a corresponding quantity of BEA, which was pre-calcined to
remove the adsorbed water, was added to the above solution to achieve
a loading amount of about 1.5 wt% for the Ce. The obtained mixed slurry
was then treated by ultrasonic for 1 h with a frequency of 60 kHz.
3
3
3
2
2
2
2
listed in SI. Besides, experiments of thermal catalytic oxidation of
2

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
toluene (denoted as catalyst-T) were also carried out. Experimental
3. Results and discussion
conditions were the same as those for the catalytic ozonation of toluene,
with the only difference of the absence of O
3
in the inlet gas mixture.
3.1. BEA supported catalysts
3
.1.1. Catalytic performance evaluation
Catalytic performance for the Pt-CeO
with Pt loading amount of approximately 0.06 wt% was evaluated, as
shown in Fig. 1. Toluene conversion, CO selectivity, carbon balance and
2
conversion increased with temperature for both catalysts. Pt-CeO /
2
2
/BEA-IM and Pt-CeO /BEA-P
2
O
3
Fig. 1. Catalytic ozonation of toluene at different temperatures (a) toluene conversion, (b) CO
2
3
selectivity, (c) carbon balance, (d) O conversion, (e) effect of water
vapor on toluene conversion and (f) reaction stability with time for toluene conversion over Pt-CeO
2
/BEA-P catalyst. (50 mg of catalyst, flow rate for the simulated
dry air of 100 mL/min, toluene concentration of 35 ppm, O
3
concentration of 350 ppm).
3

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
BEA-P presented better catalytic performance with a CO
2
selectivity and
3.1.2. Catalyst characterization
Specific surface area (SSA), pore diameter and volume of the cata-
lysts were determined by N adsorption/desorption isotherms, as pre-
O
3
conversion reaching 100% at 90 ℃ and 70 ℃, respectively, while
those for Pt-CeO /BEA-IM were only 75% and 87% at the corresponding
temperature. In addition, the CO selectivity of CeO /BEA and Pt-CeO
2
2
2
2
2
/
sented in Table 1 and Fig. S2. Compared with pure BEA, SSA for the
2
BEA-IM showed a downward trend after 70 ℃, which could be possibly
ascribed to the accumulation of by-products on these catalysts. In fact, in
order to obtain the selectivity data of each temperature, the holding time
supported catalysts decreased to about 555 m /g and the difference
between each other is negligible. This proves the successful loading of Pt
NPs and CeO
fluence of Pt NPs and CeO
2
2
on the BEA. For the pore diameter distribution, the in-
of the test was long (> 2 h). Since the CO
may lead to the accumulation of organic by-products. This phenomenon
further highlighted the outstanding catalytic activity of the Pt-CeO
BEA-P, which displayed special performance for the CO selectivity due
to the ability for the decomposition of by-products. Carbon balance for
Pt-CeO /BEA-P was also higher, especially at lower temperatures. As
evidenced in Fig. 2, the better catalytic performance could be primarily
due to the higher O conversion in the absence of toluene [14]. More-
over, the BEA support could display a certain O decomposition ability
2
selectivity was not 100%, it
on the structure of BEA could be ignored,
mainly due to their low loading amount, high dispersion and large SSA
of the BEA. These results were further testified by XRD (Fig. S3), in
which peaks related to the Pt and Ce species are not observed and the
crystal structure of BEA could be well preserved after the loading of Pt
2
/
2
2
2
NPs and CeO .
The morphology of the catalysts was analyzed by TEM and HRTEM.
As seen from Fig. 3, Pt NPs with different sizes ranging from 4 to 30 nm
3
3
are dispersed on the surface of Pt-CeO
relatively uniform in shape with an average size of about 5 nm and well
distributed on the Pt-CeO /BEA-P. From HRTEM images in Fig. 3(b) and
(e), lattice fringes of 0.23 and 0.31 nm can be observed, corresponding
to the (111) crystal facets of Pt and CeO , respectively. It should be
pointed out that Pt NPs seem to randomly distribute on the surface of Pt-
CeO /BEA-IM; however, they are selectively dispersed around CeO in
the Pt-CeO /BEA-P, which can increase the contact possibility between
each other. From these observations, one can conclude that the size of Pt
NPs and the presence of contact interface between Pt NPs and CeO are
2
/BEA-IM. Whereas, Pt NPs are
owing to the presence of acid sites with Al element [24], which further
contributed to the catalytic activity. Although the actual Pt loading
2
amount of Pt-CeO
2
/BEA-P is much lower than that in our previous study
[
25], its catalytic performance is fairly encouraging and comparable to
2
those reported in the literature, as summarized in Table S1. The
competitive results could be possibly attributed to the enhanced SMSI
2
2
between Pt NPs and CeO
1]. The enhanced SMSI creates a higher amount of oxygen vacancies
and thus provides more active sites for O adsorption and decomposition
26]. During this process, electrons from the O atoms are transferred to
2
induced by the Pt NPs adsorption method [20,
2
2
3
2
[
closely related to the preparation method. Compared with the tradi-
tional impregnation method, Pt NPs are much smaller, more uniform
and dispersive through the adsorption method. Zhang et al. [21] have
synthesized Pd/Ce-SBA15 catalysts using the electrostatic adsorption
method and impregnation method. The authors found that Pd NPs were
4
+
3+
Ce ions by reducing the Ce
to Ce . Oxygen vacancies are then
into AOS.
Subsequently, the highly reactive AOS can oxidize the adsorbed toluene
into intermediate products and finally to CO when toluene is present in
generated, acting as active sites to decompose the adsorbed O
3
2
the gaseous phase [27]. Moreover, consecutive runs under different
relative humidity (RH) and long-term catalytic performance of toluene
highly deposited around CeO
static adsorption method, which exhibited enhanced SMSI and better
catalytic activity. He et al. [20] have prepared Pt-Mn/SiO catalyst by
the Pt NPs adsorption method. Characterization results showed that Pt
NPs closely contacted with Mn (222) crystal facets and interacted
with them to form Pt-O-Mn bonds, which facilitated the electron transfer
2
in the sample prepared by the electro-
oxidation over the best performing catalyst Pt-CeO
2
/BEA-P were con-
2
ducted to investigate the stability of the catalyst. It could be found in
Fig. 1(e) that toluene conversion still remained above 99% under 30%
and 50% RH. When the RH increased to 97%, water vapor exhibited a
negative effect on the catalytic activity. As seen in Fig. 1(f), no signifi-
cant decrease in toluene conversion was observed during 12 h test under
dry air. Even though under extreme conditions such as 97% RH, the
catalyst could still show a relatively stable performance. These results
demonstrate the good reaction stability of the catalyst.
2 3
O
0
2+
from Mn to Pt, ultimately increased the proportion of Pt /Pt on the
surface and then improved the catalyst activity and stability. Therefore,
the close contact between Pt NPs and CeO
Pt-CeO SMSI, which could significantly affect the catalytic
performance.
The information on surface coordination and oxidation state of CeO
2
is proved to enhance the
2
2
was studied by UV-Vis diffuse reflectance spectroscopy and the results
are shown in Fig. 4(a). In the spectra, absorption peaks at about 250 and
3
+
2-
4+
2
90 nm are attributed to the electron transfer of Ce ← O and Ce
←
2
-
O , respectively [28]. Among them, the peak intensity at 250 nm is the
strongest for Pt-CeO /BEA-P, followed by Pt-CeO /BEA-IM and CeO /-
BEA, signifying that the loading of Pt NPs can effectively increase the
2
2
2
3
+
Ce
amount of Ce3 . This may be ascribed to the greater contact possibility
and the larger interface between Pt NPs and CeO , resulting in more
conducive for the electron transfer and hence an enhanced SMSI. A
proportion and thus the Pt-CeO
2
/BEA-P possesses the largest
+
2
higher Ce3 proportion usually means more oxygen vacancies that
+
facilitate the O decomposition [19]. Besides, toluene-TPD was used to
3
determine the toluene adsorption capacity for the catalysts. As exhibited
in Fig. 4(b), a toluene desorption peak appears in the temperature range
of 50–230 ℃ for all the samples. The temperatures of toluene desorption
for Pt-CeO
34.1 ℃ and 136.2 ℃, respectively, while that for Pt-CeO
much higher (147.5 ℃). This result indicates that toluene has the
strongest binding force with the surface of Pt-CeO /BEA-P, which ben-
2
/BEA-IM and CeO
2
/BEA are rather closed, which are
1
2
/BEA-P is
2
efits the adsorption of toluene and therefore facilitates the degradation.
It should be pointed out that due to such low loading amount of Pt
NPs and CeO
oxygen vacancies and AOS was obtained by XPS, H
analysis, etc., as shown in Fig. S5 and Fig. S6. To further elucidate the
2
on the BEA, little information about the chemical states,
Fig. 2. O
3
decomposition performance for the catalysts. (30 ℃, 50 mg of
2
-TPR and O -TPD
2
catalyst, flow rate for simulated dry air of 100 mL/min, O
of 350 ppm).
3
concentration
4

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
Table 1
Textural and structural properties of the catalysts.
a
2
a
3
a
3
b
b
c
Sample
SSA (m /g)
V
total (cm /g)
V
micropore (cm /g)
Pt loading amount (wt%)
Ce amount (wt%)
Average size of Pt NPs (nm)
BEA
611.5
557.3
550.7
555.2
0.41
0.39
0.38
0.38
0.18
0.17
0.16
0.17
/
/
/
CeO
2
/BEA
/
1.5
1.6
1.7
/
Pt-CeO
Pt-CeO
2
/BEA-IM
/BEA-P
0.06
0.07
11.6 ± 0.5
5.2 ± 0.4
2
a
2
Determined by N adsorption/desorption isotherms.
b
c
Determined by ICP-OES.
Determined by TEM.
2 2
Fig. 3. TEM, HRTEM images of the catalysts and the distribution of Pt NPs diameter (a, b and c) Pt-CeO /BEA-IM, (d, e and f) Pt-CeO /BEA-P.
Fig. 4. (a) UV–Vis diffuse reflectance spectra and (b) toluene-TPD profiles of the catalysts.
5

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
intrinsic reasons for the performance improvement related to the Pt-
alone or the interaction between Pt NPs and CeO
support was evaluated with Pt/SiO , of which SiO
inert support. Since toluene and O conversions were lower than 5% for
SiO whatever the temperature was, the catalytic activity displayed by
Pt/SiO could be roughly regarded to that of Pt NPs. At 110 ℃, toluene
and O conversions for Pt/SiO were 10.2% and 43.3%, respectively,
confirming the effect of Pt NPs for the reaction. When a comparison is
carried out between Pt/SiO and Pt/CeO , the catalytic performance of
the latter is much better. For the performance improvement in Pt/CeO
individual Pt NPs play a role rather limited, while the interaction, spe-
cifically, the Pt-CeO SMSI is predominant. Concretely, the enhanced Pt-
CeO SMSI induced by the Pt NPs adsorption method can increase the
concentration of oxygen vacancies in CeO , which are favorable for O
adsorption and decomposition, giving rise to the improvement of
toluene oxidation [19]. The better catalytic performance of Pt/CeO
2
, the influence of the
CeO
adsorption method, was employed for subsequent investigation and
compared with Pt/SiO that is characterized by the inert support.
2
SMSI, Pt/CeO
2
sample without BEA, also prepared by the Pt NPs
2
2
is well-known as an
3
2
2
2
3
2
3
3
2 2
.2. Catalysts of Pt/CeO and Pt/SiO
2
2
.2.1. Catalytic performance evaluation
2
,
Fig. 5 depicts the results of catalytic ozonation of toluene for Pt/CeO
2
and Pt/SiO
thermal catalytic oxidation of toluene by CeO
carried out, corresponding to the curves of CeO
Fig. 5(a), respectively. When O was absent in the inlet, the maximum
toluene conversion for CeO -T and Pt/CeO -T was only 1% and 2%,
2
. To exclude the effect of thermal catalysis, experiments of
and Pt/CeO were also
-T and Pt/CeO -T in
2
2
2
2
2
2
2
3
3
2
2
2
respectively, indicating that the effect of thermal catalysis could be
could be also attributed to the anchoring effect for Pt NPs through the
formation of Pt-O-Ce bonds, in which electronic interaction between
ignored within the temperature range of 30–110 ℃. After introducing
the O
that of Pt/CeO
yield and O conversion of Pt/CeO
respectively. This result confirms the promotion effect of Pt NPs and the
pivotal role of O in the enhancement of toluene oxidation. O decom-
position ability in the absence of toluene was also tested (Fig. 5(d)). Pt/
CeO presented a higher O decomposition ability with a stable O
conversion of approximate 90% during the test of 150 min at 30 ℃,
while that of CeO decreased to 80%. This result is in accordance with
3
, toluene conversion significantly increased for both samples, and
4
+
2-
2+
2-
Ce , O and Pt facilitates the redox exchanges through electron
2
was much superior. At 80 ℃, toluene conversion, CO
were 92.3%, 92.1% and 99.9%,
2
2
+
4+
2+
transfer via Pt -O -Ce -Pt linkage [29,30]:
3
2
Ce4 -O -Pt + e ↔ Ce4+-O -Pt ↔ Ce -O -Pt ↔ Ce -O -Pt + e
+
2-
2+
2-
+
3+ 2-
2+
4+ 2-
2+
3
3
In any case, such electron transfer is inexistent in Pt/SiO
2
. To sum up,
the performance improvement in Pt/CeO
2
is mainly attributed to the
2
3
3
accessibility of the Ce4 /Ce redox couple with Pt /Pt through the
+
3+
2+
0
oxygen vacancies, resulting in lowering the reaction temperature. A
2
more detailed explanation of the improvement effect by Pt-CeO
will be discussed in the following sections.
2
SMSI
the catalytic activity for toluene oxidation.
To clarify whether the improvement effect is attributed to the Pt NPs
Fig. 5. Catalytic ozonation of toluene at different temperatures (a) toluene conversion, (b) CO
2
yield, (c) O
3
3
conversion and (d) O conversion versus time at 30 ℃.
(
50 mg of catalyst, flow rate for the simulated dry air of 100 mL/min, toluene concentration of 250 ppm, O
3
concentration of 2500 ppm).
6

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
3
3
.2.2. Catalyst characterization
Table 2
Structural parameters of the prepared materials.
.2.2.1. Structural and morphological properties. XRD was employed to
Sample
Specific
Pore
Pt loading
Lattice
parameter
(nm)
Average
a
b
c
evaluate the nature of the crystalline phases. From Fig. 6(a), charac-
teristic peaks of the face-centered cubic fluorite structure of ceria (PDF
surface
diameter
(nm)
amount
size of Pt
a
2
d
area (m /
g)
(wt%)
NPs (nm)
3
4-0394) are observed for both CeO
crystal structure of CeO is not affected by the addition of Pt NPs. No
diffraction peaks related to the Pt (40 and 33 , PDF 65-2868) and PtO
2 2
and Pt/CeO , indicating that the
CeO
Pt/
2
64.8
69.9
12.2
12.1
/
0.5414
0.5412
/
2
◦
◦
0.43
3.6 ± 0.3
x
CeO
Pt/
SiO2
2
◦ ◦ ◦ ◦
species (Pt
984) are displayed, which is attributed to the low loading amount,
small particle size and high dispersion of Pt NPs on the CeO surface. As
determined from the diffraction peak (111), the lattice parameter of Pt/
CeO is slightly smaller than that of pure CeO (Table 2). This decreased
value certifies the incorporation of partial Pt (rPt2+= 0.074 nm) or
2
O in 40 and 33 , PDF 65-5066; PtO in 30 and 32 , PDF 65-
/
/
0.42
/
3.4 ± 0.3
6
2
a
b
c
2
Determined by N adsorption/desorption isotherms.
Determined by ICP-OES.
Determined by Jade software.
Determined by TEM.
2
2
2
+
d
4
+
Pt (rPt4+= 0.076 nm) species into the fluorite-like lattice of CeO
Ce4+= 0.094 nm), resulting in the shrinkage of the cell to form a
defective structure and also the formation of Pt-O-Ce bonds due to the
Pt-CeO SMSI [19]. The formation of Pt-O-Ce bonds can suppress the
agglomeration of Pt NPs since the highly dispersed Pt oxide on the CeO
2
(
r
splitting caused by the redistribution of the entire energy spectrum [32].
Six peaks corresponding to three pairs of spin-orbit doublets noted as
2
U’’’(916.1 eV), V’’’(897.6 eV), U’’(906.9 eV), V’’(888.5 eV),
U
2
(
900.3 eV) and V(881.9 eV), are assigned to the characteristic peaks of
surface under oxidizing conditions is considered to be more stable owing
to the greater bond energy of Pt-O-Ce bonds [31]. The SSA and pore
diameter are also presented in Table 2 and Fig. 6(b). After loading Pt
4+
Ce oxide; while the other four peaks due to two pairs of doublets,
0
0
noted as U’(901.9 eV), V’(883.9 eV), U (898.6 eV) and V (880.6 eV)
can be identified as Ce oxide, which is usually served as an indicator
3+
2
NPs, the SSA slightly increases from 64.8 to 69.9 m /g, which may be
for the existence of oxygen vacancy in CeO
2
[19]. As summarized in
Table 3, the ratio of Ce /(Ce +Ce ) in CeO and Pt/CeO is 31.3%
and 44.2%, respectively, demonstrating a higher oxygen vacancy con-
centration in the Pt/CeO due to the presence of Pt NPs.
ascribed to the loading of Pt NPs on the catalyst surface. Besides, the
hysteresis loop located at the relative pressure of 0.7–0.9 in adsorp-
tion/desorption isotherms signifies the presence of a large mesoporous
structure and the preservation of the pore structure in the samples.
The morphology of the catalysts and the Pt NPs size distribution are
illustrated in Fig. 7. As seen in the images, Pt NPs are uniformly
3+
3+
4+
2
2
2
Regarding the surface oxygen species, O 1s spectra could be decon-
voluted into three characteristic peaks at (528.7–528.9 eV),
(
530.5–530.7 eV) and (533.5–533.7 eV), which are assigned to the
distributed on CeO
2
and SiO
2
surfaces. In HRTEM images, lattice fringes
2- 2- -
surface lattice oxygen OLat (O ), surface adsorbed oxygen OSur (O
2 2
,O )
of 0.21 nm can be observed, which correspond to the (111) crystal facets
of Pt NPs. From Fig. 7(c) and (f), the average particle size of Pt NPs is
formed due to the oxygen vacancy and other probably surface adsorbed
2-
oxygen species OAds (CO
ratio of OSur/OLat could be used to evaluate the surface oxygen vacancy
concentration [30] and the results are presented in Table 3. Pt/CeO
exhibits a higher amount of surface oxygen vacancy with the OSur/OLat
ratio of 86.9% while that for CeO is 77.3%, confirming the beneficial
3
, -OH), respectively [29,33]. In general, the
(
3.6 ± 0.3) nm and (3.4 ± 0.3) nm for Pt/CeO
2 2
and Pt/SiO , respec-
tively, indicating that different support has little effect on the particle
size distribution through the Pt NPs adsorption method. The mentioned
structural and morphological properties testify the low loading amount,
2
2
uniform particle size and high dispersion of Pt NPs on both Pt/CeO
Pt/SiO
2
and
effect of Pt NPs for the generation of oxygen vacancies. Notably, on
account of the low loading amount of Pt NPs, the detection accuracy for
Pt species is precluded by XPS and the Pt 4f spectra are not presented in
this study.
2
.
3
.2.2.2. Chemical states. XPS was performed to obtain further insight
into the composition and chemical states of the surface elements. Fig. 8
presents the XPS spectra of Ce 3d and O 1s for CeO and Pt/CeO . Two
, namely Ce3 and Ce . According to
Partial Pt species could probably get incorporated into the CeO
matrix due to the highly ionic nature of CeO compared to SiO
Generally, solid solution in form of Ce1ꢀ x 2ꢀ x (M = Pt , Pd , etc.)
is one of the possible approaches for the incorporation of ions in CeO
2
2
2
2
.
2
+
4+
valence states mainly exist in CeO
2
2+
4+
2+
x
M O
the literature, Ce 3d spectra can be fitted with ten components that take
into account the spin-orbit splitting of Ce 3d5/2, Ce 3d3/2 and other
2
,
2+
since the ionic radii of Pt (80 pm) is close to that of Ce (97 pm).
2 2 2
Fig. 6. (a) XRD patterns and (b) N -adsorption/desorption isotherms of CeO and Pt/CeO .
7

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
2 2
Fig. 7. TEM, HRTEM images of the catalysts and the distribution of Pt NPs diameter on (a, b and c) CeO and (d, e and f) SiO .
2 2
Fig. 8. XPS spectra of (a) Ce 3d and (b) O 1s for CeO and Pt/CeO .
Therefore, O -Ce -O -Pt -O2- type of linkage on CeO
2
-
4+ 2-
2+
surface could
2
Table 3
be formed. The decreased lattice parameter calculated in XRD possibly
XPS, Raman and O
2
-TPD analysis of the catalysts.
2+
4+
2
suggests the ionic substitution of Pt ions for Ce in CeO . Because of
3
+
+
3+
ꢀ
(ad) + Oꢀ (ad)
Sample
Ce / (Ce
+
O
O
Sur
/
I
D
/
O
2
(μmol/
the lower valent ionic substitution, oxygen vacancies are thus created to
4
Ce
)
Lat
IF2g
g)
2+
maintain the charge neutrality. Substitution of Pt in the CeO
2
matrix
CeO
Pt/
2
31.3%
44.2%
77.3%
86.9%
0.31
0.91
0.486
0.576
could further enhance the oxygen storage capacity due to the presence of
longer bonds [34]. Metiu et al. [34] have demonstrated that substitution
CeO
2
2
of Pt in CeO surface weakened the bonds of oxygen atoms in the
neighborhood as evidenced by lowering the formation energy of oxygen
vacancies, which become much easier to be generated.
8

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
3
.2.2.3. Oxygen vacancy. The structural information of the surface can
be obtained by visible Raman spectroscopy, of which the results for
CeO and Pt/CeO are shown in Fig. 9(a). Raman peaks at 258, 460, 598
amount of H
2
consumption corresponding to this peak is 29.9 mol/g,
μ
which is greater than the theoretical value for the reduction of PtO
x
are Pt4 ). This extra
+
2
2
(22.1
μ
mol/g, assuming that Pt species in Pt/CeO
2
-
1
and 1179 cm are observed for both samples, corresponding to the
second-order transverse acoustic mode (2TA), the symmetric stretching
vibrations mode (F2 g), the band of defect-induced mode (D) and the
second-order longitudinal optical mode (2LO), respectively [35]. The
asymmetry of the main peak at 460 cm-1 and the appearance of other
weak peaks are caused by the presence of oxygen vacancy defects and
2
oxygen species might come from the surface and/or subsurface of CeO ,
resulting in the reduction of Ce4 adjacent to the Pt species. This phe-
+
nomenon could be ascribed to the spillover effect induced by the
2
Pt-CeO SMSI [31]. Meanwhile, reduction peaks of surface and subsur-
face oxygen in the Pt/CeO
360 ℃), suggesting that a small amount of Pt, either Pt2 or Pt , can
2
shift to a lower temperature (204 ℃ and
+
0
lattice strain of CeO
2
, which facilitate the adsorption, activation and
presents other peaks
activate the surface Ce-O bonds and promote the surface reduction of
migration of oxygen species [36]. Besides, Pt/CeO
2
CeO
bonds strongly bounded with Pt species. Summarily, the lattice oxygen
on the CeO surface and/or subsurface could be activated by Pt NPs,
2
[41]. This result is probably caused by the relaxation of Ce-O
-
1
at 558 and 665 cm , belonging to the vibration of Pt-O-Ce and Pt-O
bands, respectively [35]. This result directly confirms the formation of
2
Pt-O-Ce bonds in the Pt-CeO
2
interface due to the Pt-CeO
2
SMSI.
leading to an increase of the lattice oxygen mobility and the higher
structural defects [19,42].
Surface oxygen vacancies have been proved to play a key role in
catalytic ozonation reactions, which mainly participate in the trans-
formation of AOS on the surface, as well as the adsorption and decom-
2
The type of AOS was analyzed by O -TPD, which could also reflect
information about the oxygen vacancy concentration to some extent
based on the amount of desorbed oxygen [43,44]. As depicted in Fig. 10
(b), four main desorption peaks are observed in the range of 60–700 ℃
for both samples. Desorption peak at 120 ℃ belongs to the physically
3
position of O [37]. UV Raman spectroscopy that is sensitive to the
defective features was used to detect the surface oxygen vacancies. As
-
1
depicted in Fig. 9(b), Raman peaks at 417, 459, 592 and 1179 cm are
observed for both catalysts. The last three peaks are assigned to the
corresponding mode as described in visible Raman spectra; the first peak
adsorbed oxygen O2(ad), other three peaks between 180 and 500 ℃ are
ꢀ
assigned to the chemically adsorbed oxygen species such as O
2
(ad) and
-
1
ꢀ
at 417 cm also appeared in other studies, but its attribution is still
unclear [38]. Specifically, the relative intensity ratio of D peak to F2 g
O
(ad) that adsorb on oxygen vacancies, and the desorption peak for
2
ꢀ
lattice oxygen O (lat) is generally located at temperatures higher than
500 ℃ [29]. After loading Pt NPs, the intensity of desorption peaks
belonging to the chemically adsorbed oxygen is remarkably enhanced,
among which the enhancement for the peak at 378 ℃ is the most sig-
nificant. By integrating the curves, the desorption amount of chemically
peak (I
oxygen vacancies in CeO
Pt/CeO (0.91) is much greater than that of CeO
higher oxygen vacancy concentration in the Pt/CeO
D
/IF2g) can be employed to characterize the relative amount of
[19]. As listed in Table 3, the I /IF2g value of
(0.31), meaning a
. Combined with
2
D
2
2
2
the XPS results, it can be inferred a positive correlation between the
catalytic performance and the amount of oxygen vacancies.
adsorbed oxygen could be calculated, as listed in Table 3. Pt/CeO
2
ꢀ ꢀ
possesses a higher amount of O
2
(ad) and O (ad), indicating the presence
of more surface oxygen vacancies [41]. This result is in good agreement
with those of XPS and Raman.
3
.2.2.4. Active oxygen species. Information about AOS was obtained
through H -TPR and the results are shown in Fig. 10(a). According to the
literature, the reduction process of CeO could be divided into three
2
3.2.3. Evolution of active oxygen species
2
temperature ranges of 250–400 ℃, 400–600 ℃ and 600–1000 ℃,
attributing to the reduction of surface oxygen, subsurface oxygen and
bulk lattice oxygen, respectively [22,39]. Two peaks appearing at
In the case of CeO
2
-based catalysts, surface coordinatively unsatu-
rated metal cations and oxygen anion vacancies could participate in the
activation of reactants [45]. Specifically, adsorbed O
3
2-
might result in the
-
3
59 ℃ and 458 ℃ in CeO
2
represent the reduction of surface and sub-
formation of superoxide (O ) and/or peroxide (O ) species on defect
2
2
surface oxygen, respectively. After loading Pt NPs, three peaks can be
sites, which are postulated to be kinetically important intermediates
during the O decomposition process [37]. Of particular note is that
observed at 84 ℃, 204 ℃ and 360 ℃, among which the first peak is
3
attributed to the reduction of surface PtO
metallic Pt could be oxidized to PtO by lattice oxygen from CeO
than the O in the gaseous phase. As a result, Pt NPs in the Pt/CeO
x
. Previous studies state that
rather
act as
atomic oxygen is also an intermediate, which, however, is in transient
state and eventually transforms to the above oxygen species [46]. To
x
2
investigate the intrinsic relationship between the O
3
decomposition
2
2
3+
4
+
the oxygen atom acceptor, which decreases the ratio of Ce /Ce and
thus increases the oxygen vacancy concentration [40]. Besides, the
performance and the catalytic ozonation activity, in-situ Raman was
-
2
2-
2
conducted to detect the O
and O
3
generated from O decomposition
Fig. 9. Raman spectra of CeO
2
2
and Pt/CeO (a) visible Raman and (b) UV Raman.
9

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
Fig. 10. (a) H
2
2 2 2
-TPR and (b) O -TPD profiles of CeO and Pt/CeO .
-
in the presence or absence of toluene. In return, the response of O
2
and
Fig. 12. In this experiment, the samples that were pretreated under O
were firstly swept by N to remove the weakly adsorbed oxygen species
(spectra marked with “N
3
2
-
O
2
can be used to verify the amount of oxygen vacancies on the catalyst
surface [45].
The curves marked with “initial” in Fig. 11(a) and (b) represent the
Raman spectra for catalysts after pretreatment. In the range of
2
-
2
2
”). Notably, the peak corresponding to the O
-
1
at 1139 cm disappears, and the intensity of the peaks ascribed to the
2
-
at 833, 860 and 890 cm-1 decreases in this step. This result dem-
O
2
-
1
2-
7
00–1400 cm , both catalysts possess three peaks at 1179, 1050 and
33 cm , belonging to the second-order longitudinal optical mode
onstrates the higher stability of O
2
on the catalyst surface. Afterward,
-
1
8
toluene was introduced into the system (spectra marked with
2
-
(
2LO) of ceria, trace nitrate or phosphate species in ceria, and O
2
“Toluene”). Toluene adsorption peaks appear at 822, 1005, 1155 and
-
1
2-
adsorbed on defect sites, respectively [47]. After introduction of O
2
, the
1290 cm , while the intensity of the peaks for the O
2
at 860 and
intensity of the peak at 833 cm-1 is significantly increased and a weak
890 cm weakens. From this result, one could deduce that O
consumed during the toluene oxidation. Besides, the structure signal of
-1
2-
was
2
shoulder peak appears at 860 cm-1 in the spectra marked with “O
both samples. Subsequently, O was replaced by O . In the spectra
marked with “O
2
” for
-
1
2
3
2
CeO at 1179 cm seems to disappear and be replaced by that of toluene
-
1
-1
3
”, the intensity of the peaks at 833 and 860 cm are
at 1155 cm . This could be probably attributed to the more sensitive
-
1
further increased, and a new shoulder peak comes out at 1139 cm for
both catalysts. Notably, a new shoulder peak at 890 cm-1 is particularly
detection for the toluene signals in Raman spectroscopy. Similarly, Mao
2-
et al. [49] have confirmed that the O
2
formed during O
3
decomposition
observed in the Pt/CeO
2
. Based on the previous studies, the peaks at
were pivotal reactive species for the catalytic ozonation of VOCs.
Compared to CeO , the Pt/CeO possesses a higher amount of oxygen
vacancies that are beneficial for O adsorption and decomposition, and
thus generation of more O
2-. The presence of more O
facilitate the toluene oxidation and the Pt/CeO exhibits enhanced
catalytic activity as a consequence.
-
1
2-
8
33, 860 and 890 cm are assigned to the O
2
adsorbed on defect sites
2
2
-
1
with different aggregation states [48]. Whereas, the peak at 1139 cm is
3
-
2
2-
2-
2
assigned to the adsorbed O
in the O atmosphere, Raman spectra were deconvoluted through peak
fitting. By using the peak area of 2LO mode (1179 cm ) as the basis to
. To compare the amount of O
2
generated
2
is proved to
3
2
-
1
2
-
normalize other peaks, the relative content of O
seen in Fig. 11(c), the relative contents of the three different O
Pt/CeO are 41.4%, 23.3%, and 17.1%, which are significantly higher
than those in CeO
on the Pt/CeO surface in the presence of O
In-situ Raman spectroscopy in the presence of toluene is displayed in
2
could be obtained. As
2
-
2
in
3
.3. Proposed reaction pathways
2
2
-
2
. These results highlight that more O
2
can be formed
A general mechanism of O
been established by Oyama et al. [50]: adsorbed O
form an O and an atomic oxygen (O*), which subsequently reacts with
3
decomposition on catalyst surface has
2
3
.
3
firstly dissociates to
2
2 2 2 2 3
Fig. 11. In-situ Raman spectroscopy of (a) CeO , (b) Pt/CeO and (c) relative content of peroxide species on CeO and Pt/CeO under O flow.
1
0

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
Fig. 12. In-situ Raman spectroscopy of (a) CeO
2
2 2
and (b) Pt/CeO under N and toluene flow.
another O
final step involves the decomposition of the O
our study, in-situ Raman in the presence of toluene allows identifying
3
to generate a gaseous O
2
and an adsorbed oxygen (O
2
*); the
ꢀ
+ Oꢀ2
O
3
+ O →O
2
(R2)
(R3)
(R4)
2
* to form a gaseous O
2
. In
ꢀ
2ꢀ
O + e→O
2
-
2
2
the participation of O
2
that is preliminarily formed during O
3
decomposition for the catalytic ozonation of toluene. In light of the
above results and discussion, plausible reaction pathways based on the
Langmuir-Hinshelwood mechanism for such kind of reaction on the
∗
C
C
7
7
H
H
8
+ S
a
→C
7
H
8
∗
+ (Oꢀ + O 22 ꢀ … )→by_products →CO
∗
8
2
+ H
2
O
(R5)
Pt-CeO
2
/BEA surface are proposed as the following:
+ Oꢀ
where SOv is assigned to active sites such as oxygen vacancies and acid
sides for O adsorption and decomposition, S represents active sites for
O
3
+ S v
O
+ e→O
2
(R1)
3
a
2
Fig. 13. Plausible reaction pathways for the catalytic ozonation of toluene on Pt-CeO /BEA.
1
1

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
toluene adsorption and activation. From Section 3.2.1, Pt/CeO
2
hardly
Appendix A. Supporting information
reacts with toluene in the absence of O
major role of Pt/CeO is to provide adsorption sites for the toluene. As
described in Fig. 13, after introducing O , the O molecule chemically
adsorbs on the CeO surface by inserting an O atom into the oxygen
3
below 110 ℃. In this case, the
2
Supplementary data associated with this article can be found in the
online version at doi:10.1016/j.apcata.2021.118342.
3
3
2
vacancy site, which is an electron donor and transfers an electron to this
References
-
O atom. A charged oxygen species (O ) is thus formed in this oxygen
[
[
1] K. Rumchev, J. Spickett, M. Bulsara, M. Philips, S. Stick, Thorax 59 (2004)
vacancy and an O
another O molecule reacts with the abovementioned O to generate a
gaseous O
electron to produce an O
2
molecule desorbs in the gas phase (R1). Then,
7
46–751.
-
3
2] B. Wang, X. Xu, W. Xu, N. Wang, H. Xiao, Y. Sun, H. Huang, L. Yu, M. Fu, J. Wu,
-
2
molecule and an O
2
(R2), in which the latter could accept an
L. Chen, D. Ye, Catal. Surv. Asia. 22 (2018) 73–94.
2
-
-
2-
[3] M. Kim, E. Park, J. Jurng, Powder Technol. 325 (2018) 368–372.
2
(R3). The formation of O
2
and O
2
was
2
-
[4] L. Chen, M. Ondarts, J. Outin, Y. Gonthier, E. Gonze, J. Environ. Sci. 74 (2018)
confirmed by in-situ Raman in Section 3.2.3. Subsequently, the O
2
5
8–70.
[5] L. Chen, E. Gonze, M. Ondarts, J. Outin, Y. Gonthier, Sep. Purif. Technol. 247
2020), 116964.
[
[
participates in the oxidation of toluene, which preliminarily adsorbs on
active sites of the catalyst surface (R4), to form oxygen-containing
(
6] H. Einaga, S. Futamura, React. Kinet. Catal. Lett. 81 (2004) 121–128.
7] F. Lin, Z. Wang, Z. Zhang, Y. He, Y. Zhu, J. Shao, D. Yuan, G. Chen, K. Cen, Chem.
Eng. J. 382 (2020), 123030.
byproducts and finally the harmless CO
2 2
and H O (R5). Summarily,
O
3
decomposition is the key step that depends on the amount of oxygen
2
-
vacancies and the O
ozonation of toluene.
2
is an important participant in the catalytic
[8] X. Li, J. Ma, H. He, J. Environ. Sci. 94 (2020) 14–31.
[
9] H. Einaga, Y. Teraoka, A. Ogata, J. Catal. 305 (2013) 227–237.
[
[
10] H. Einaga, N. Maeda, Y. Nagai, Catal. Sci. Technol. 5 (2015) 3147–3158.
11] Y. Liu, X. Li, J. Liu, C. Shi, A. Zhu, Chin. J. Catal. 35 (2014) 1465–1474.
4
. Conclusions
In this study, Pt-CeO
[12] H. Huang, W. Huang, Y. Xu, X. Ye, M. Wu, Q. Shao, G. Ou, Z. Peng, J. Shi, J. Chen,
Q. Feng, Y. Zan, H. Huang, P. Hu, Catal. Today 258 (2015) 627–633.
[
[
13] J. Li, H. Na, X. Zeng, T. Zhu, Z. Liu, Appl. Surf. Sci. 311 (2014) 690–696.
14] H.-H. Kim, M. Sugasawa, H. Hirata, Y. Teramoto, K. Kosuge, N. Negishi, A. Ogata,
Plasma Chem. Plasma Process 33 (2013) 1083–1098.
2
/BEA catalysts with a Pt loading amount lower
in order of magnitude than those reported in the literature were pre-
pared by the Pt NPs adsorption method and displayed encouraging
[
15] M. Li, K.N. Hui, K.S. Hui, S.K. Lee, Y.R. Cho, H. Lee, W. Zhou, S. Cho, C.Y.H. Chao,
Y. Li, Appl. Catal. B Environ. 107 (2011) 245–252.
catalytic performance toward catalytic ozonation of toluene. Pt/CeO
and Pt/SiO also prepared by the same method were employed to
2
[
16] A. Trovarelli, Catal. Rev. 38 (1996) 439–520.
2
[17] Y. Zhang, M. Chen, Z. Zhang, Z. Jiang, W. Shangguan, H. Einaga, Catal. Today 327
(
2019) 323–333.
elucidate the intrinsic reasons for the performance improvement. Based
on the catalytic behaviors and structural characterization, several con-
[
18] A.M. G a¨ nzler, M. Casapu, P. Vernoux, S. Loridant, F.J. CadeteSantosAires,
T. Epicier, B. Betz, R. Hoyer, J.-D. Grunwaldt, Catal. Angew. Chem. Int. Ed. 56
(2017) 13078–13082.
clusions can be drawn: (1) the enhanced Pt-CeO
BEA-P is induced by the formation of much smaller, more uniform and
dispersive Pt NPs through the adsorption method; (2) O decomposition
, which is an important
2 2
SMSI in the Pt-CeO /
[
[
19] R.S. Peng, S.J. Li, X.B. Sun, Q.M. Ren, L.M. Chen, M.L. Fu, J.L. Wu, D.Q. Ye, Appl.
Catal. B Environ. 220 (2018) 462–470.
3
20] C. He, Z. Jiang, M. Ma, X. Zhang, M. Douthwaite, J.-W. Shi, Z. Hao, ACS Catal. 8
(2018) 4213–4229.
2
-
process determines the generation of O
2
participant in the catalytic ozonation of toluene; (3) the anchoring effect
for Pt NPs by the Pt-O-Ce bonds, the greater amount of oxygen vacancies
[21] J. Sun, J. Zhang, H. Fu, H. Wan, Y. Wan, X. Qu, Z. Xu, D. Yin, S. Zheng, Appl. Catal.
B Environ. 229 (2018) 32–40.
[
22] J. Lee, Y. Ryou, X. Chan, T.J. Kim, D.H. Kim, J. Phys. Chem. C 120 (2016)
2
2
-
and more generated O
are potentially responsible for the performance
2
5870–25879.
improvement. What is more important, the presented catalysts with such
a low loading amount of Pt NPs exhibit the facility of preparation, cost-
efficiency and competitive catalytic activity, demonstrating the poten-
[23] S. Mo, Q. Zhang, M. Zhang, Q. Zhang, J. Li, M. Fu, J. Wu, P. Chen, D. Ye, Nanoscale
Horiz. 4 (2019) 1425–1433.
[
24] N. Brodu, M.-H. Manero, C. Andriantsiferana, J.-S. Pic, H. Vald ´e s, Chem. Eng. J.
2
31 (2013) 281–286.
[25] H. Xiao, J. Wu, X. Wang, J. Wang, S. Mo, M. Fu, L. Chen, D. Ye, Mol. Catal. 460
3
tial practical application for the simultaneous treatment of O and VOCs.
(
2018) 7–15.
[
[
26] E. Rezaei, J. Soltan, N. Chen, J. Lin, Chem. Eng. J. 214 (2013) 219–228.
27] G. Zhu, J. Zhu, W. Jiang, Z. Zhang, J. Wang, Y. Zhu, Q. Zhang, Appl. Catal. B
Environ. 209 (2017) 729–737.
CRediT authorship contribution statement
[
[
[
28] G. Ranga Rao, H.R. Sahu, J. Chem. Sci. 113 (2001) 651–658.
29] Z. Hu, X. Liu, D. Meng, Y. Guo, Y. Guo, G. Lu, ACS Catal. 6 (2016) 2265–2279.
30] X. Chen, X. Chen, E. Yu, S. Cai, H. Jia, J. Chen, P. Liang, Chem. Eng. J. 344 (2018)
469–479.
Jialing Wang: Conceptualization, Methodology, Investigation,
Formal analysis, Xuefeng Shi: Methodology, Investigation, Resources,
Longwen Chen: Validation, Writing – Original Draft, Writing – Review
[
[
31] Y. Nagai, T. Hirabayashi, K. Dohmae, N. Takagi, T. Minami, H. Shinjoh,
S. Matsumoto, J. Catal. 242 (2006) 103–109.
&
Editing, Huanyi Li: Resources, Mengqi Mao: Resources, Guangyi
Zhang: Resources, Hui Yi: Resources, Mingli Fu: Supervision, Daiqi Ye:
Supervision, Junliang Wu: Supervision, Conceptualization, Writing –
Review & Editing.
32] G.-F. Zhou, J. Ma, S. Bai, L. Wang, Y. Guo, Rare Metals 39 (2020) 800–805.
[33] L. Chen, T. Li, J. Zhang, J. Wang, P. Chen, M. Fu, J. Wu, D. Ye, ACS Appl. Mater.
Interfaces 13 (2021) 21436–21449.
[
34] W. Tang, Z. Hu, M. Wang, G.D. Stucky, H. Metiu, E.W. McFarland, J. Catal. 273
(
2010) 125–137.
Declaration of Competing Interest
[35] W.H. Weber, K.C. Hass, J.R. Mcbride, Phys. Rev. B 48 (1993) 178–185.
[
36] S. Damyanova, B. Pawelec, K. Arishtirova, M.V.M. Huerta, J.L.G. Fierro, Appl.
Catal. B Environ. 89 (2009) 149–159.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[
37] X. Wang, J. Wu, J. Wang, H. Xiao, B. Chen, R. Peng, M. Fu, L. Chen, D. Ye, W. Wen,
Chem. Eng. J. 369 (2019) 233–244.
[38] S. Agarwal, X. Zhu, E.J.M. Hensen, L. Lefferts, B.L. Mojet, J. Phys. Chem. C 118
(
2014) 4131–4142.
[
[
39] M. Sun, J. Liu, C. Song, Y. Ogata, H. Rao, X. Zhao, H. Xu, Y. Chen, ACS Appl. Mater.
Interfaces 11 (2019) 23102–23111.
Acknowledgments
40] W. Lin, A.A. Herzing, C.J. Kiely, I.E. Wachs, J. Phys. Chem. C 112 (2008)
5
942–5951.
This work was supported by the National Natural Science Foundation
of China (51678245, 52000075, 51578245), National Key Research and
Development Program (2016YFC0204200), Natural Science Foundation
of Guangdong Province (2020A1515010929), Science and Technology
[
[
41] Y. Gao, W. Wang, S. Chang, W. Huang, ChemCatChem 5 (2013) 3610–3620.
42] H. Huang, X. Ye, W. Huang, J. Chen, Y. Xu, M. Wu, Q. Shao, Z. Peng, G. Ou, J. Shi,
X. Feng, Q. Feng, H. Huang, P. Hu, D.Y.C. Leung, Chem. Eng. J. 264 (2015) 24–31.
43] L. Chen, J. Zhang, J. Wang, P. Chen, M. Fu, J. Wu, D. Ye, J. Hazard. Mater. 420
[
[
[
(
2021), 126604.
Program
of
Guangdong
and
Guangzhou
(201804020026,
44] H. Xu, Z. Zhang, J. Liu, C.-L. Do-Thanh, H. Chen, S. Xu, Q. Lin, Y. Jiao, J. Wang,
Y. Wang, Y. Chen, S. Dai, Nat. Commun. 11 (2020) 3908.
45] M. Daniel, S. Loridant, J. Raman Spectrosc. 43 (2012) 1312–1319.
2
015A020215010, 2015B020236002), Fundamental Research Funds
for the Central Universities (D2192010) and Young Teacher’s Ability
Improvement Project of Guangxi Universities (2018KY0882).
1
2

J. Wang et al.
Applied Catalysis A, General 625 (2021) 118342
[
[
46] W. Yang, Y. Zhu, F. You, L. Yan, Y. Ma, C. Lu, P. Gao, Q. Hao, W. Li, Appl. Catal. B
Environ. 233 (2018) 184–193.
[48] V.V. Pushkarev, V.I. Kovalchuk, J.L. d’Itri, J. Phys. Chem. B 108 (2004)
5341–5348.
47] Z. Wu, M. Li, J. Howe, H.M. Meyer, S.H. Overbury, Langmuir 26 (2010)
[49] L. Mao, Z. Chen, X. Wu, X. Tang, S. Yao, X. Zhang, B. Jiang, J. Han, Z. Wu, H. Lu,
T. Nozaki, J. Hazard. Mater. 347 (2018) 150–159.
1
6595–16606.
[
50] W. Li, G.V. Gibbs, S.T. Oyama, J. Am. Chem. Soc. 120 (1998) 9041–9046.
1
3