Effect of molybdenum on the structure and performance of V2O5/TiO2–SiO2–MoO3 catalysts for the oxidative degradation of o-chlorotoluene

Article abstract of DOI:10.1016/j.apcata.2020.117496

V/TiO₂–SiO₂–MoO₃ (TSM) catalysts exhibit enhanced redox capabilities, but the role of each species in the mixed oxide support is not well understood. Herein, the physicochemical characteristics of molybdenum oxide in TSM were investigated using X-ray diffraction, X-ray photoelectron spectroscopy, and X-ray absorption fine structure spectroscopy, and the performance of V/TSM catalysts for the oxidative decomposition of o-chlorotoluene was examined. The V/TSM catalysts exhibited superior performances in this decomposition reaction at low temperatures. The molybdenum species were highly dispersed as Mo⁶⁺, replacing Ti⁴⁺ in the anatase TiO₂ framework and forming a solid solution. The redox capability of the vanadium species in the catalyst was enhanced by electron transfer from the TSM support to vanadium, likely due to the coexistence of Mo⁶⁺ and Ti⁴⁺ in the TSM solid solution, which has a unique structure. These results are expected to contribute to enhancing the abatement of dioxins at low temperatures.

Full text of DOI:10.1016/j.apcata.2020.117496

Applied Catalysis A, General 595 (2020) 117496
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Effect of molybdenum on the structure and performance of V O /
2
5
TiO –SiO –MoO catalysts for the oxidative degradation of o-chlorotoluene
2
2
3
a,
b
a
c,d
Ryoji Kuma *, Tomoyuki Kitano , Takuya Tsujiguchi , Tsunehiro Tanaka
a
Catalysts & Green Energy Materials Research Department, Nippon Shokubai Co. Ltd., 992-1 Aza-Nishioki, Okinohama, Aboshi-ku, Himeji, Hyogo, 671-1292, Japan
Analysis Technology Center, Nippon Shokubai Co. Ltd., 5-8, Nishi Otabi-cho, Suita, Osaka, 564-0034, Japan
Department of Molecular Engineering, Graduate School of Engineering, Kyoto University, Kyoto, 615-8510, Japan
Elements Strategy Initiative for Catalysts and Batteries (ESICB), Kyoto University, Kyoto, 615-8510, Japan
b
c
d
A R T I C L E I N F O
A B S T R A C T
3
V/TiO –SiO –MoO (TSM) catalysts exhibit enhanced redox capabilities, but the role of each species in the
Keywords:
/TiO
2
2
V
2
O
5
2
–SiO
2
–MoO
3
mixed oxide support is not well understood. Herein, the physicochemical characteristics of molybdenum oxide in
TSM were investigated using X-ray diffraction, X-ray photoelectron spectroscopy, and X-ray absorption fine
structure spectroscopy, and the performance of V/TSM catalysts for the oxidative decomposition of o-chlor-
otoluene was examined. The V/TSM catalysts exhibited superior performances in this decomposition reaction at
o-Chlorotoluene
XAFS
6
+
4+
low temperatures. The molybdenum species were highly dispersed as Mo , replacing Ti in the anatase TiO
2
framework and forming a solid solution. The redox capability of the vanadium species in the catalyst was en-
hanced by electron transfer from the TSM support to vanadium, likely due to the coexistence of Mo6+ and Ti4+
in the TSM solid solution, which has a unique structure. These results are expected to contribute to enhancing
the abatement of dioxins at low temperatures.
1. Introduction
Nevertheless, few reports exist concerning the catalytic decomposition
of dioxins at temperatures lower than 200 °C. Although WO
has conventionally been added to promote the activity and SO
ability of V /TiO catalysts [7,8,10,12], these catalysts require
3
or MoO
3
Flue gases from municipal waste incineration processes commonly
contain harmful NO gases (NO, NO ) and toxic dioxins, such as poly-
2
dur-
x
2
2
O
5
2
chlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans. As
temperatures higher than 250 °C to realize satisfactory dioxin decom-
such, the catalytic degradation of these pollutants has been intensively
position efficiencies.
investigated, with the decomposition of NO
carried out industrially over selective catalytic reduction (SCR) cata-
lysts such as V /TiO [1–6]. As V /TiO catalysts are also capable
of decomposing dioxins, they have been applied for the simultaneous
abatement of NO and dioxins discharged from municipal waste in-
cinerators [7–15]. Generally, an electrostatic precipitator or a fabric
filter is used for such processes (Fig. 1), the latter of which is known to
collect fly ash more effectively [16]. However, owing to the low heat
resistance of fabric filters, the gas temperature should be lower than
x
and NH
3
being commonly
In this context, we previously developed a V
(V/TSM) catalyst with high activity for low-temperature NH
2
O
5
/TiO
2
–SiO
2 3
–MoO
3
-SCR [17].
2
O
5
2
2
O
5
2
The mixed oxide support material of this catalyst, TSM, was prepared
via a coprecipitation method. As the activity of the V/TSM catalyst was
found to be enhanced by its redox capability, the activity would be
expected to be improved for the oxidative decomposition of dioxins.
However, the role of the TSM support material and the mechanism of
redox enhancement require clarification to allow optimization of its
composition to obtain a superior catalytic performance.
x
1
50 °C in most cases. Subsequently, the outlet gas from the fabric filter
requires reheating to a higher temperature to achieve the catalytic
decomposition of both NO and dioxins. Although a higher catalytic
Thus, we herein report our study into the physicochemical char-
acteristics of molybdenum oxide in V/TSM catalysts. Catalytic activity
tests were carried out using o-chlorotoluene as a substitute for dioxins
due to the high toxicity of dioxins and the similar decomposition be-
havior of o-chlorotoluene [18]. The structure–activity relationships for
the V/TSM catalysts were investigated using Mo K-edge X-ray absorp-
tion near-edge structure (XANES) and X-ray absorption fine structure
x
activity can be achieved at higher temperatures, a large amount of
energy is required for the reheating process.
Highly active catalysts are therefore necessary for low-temperature
dioxin decomposition and in particular for the fabric filter process.
⁎
Corresponding author.
E-mail address: ryoji_kuma@shokubai.co.jp (R. Kuma).
https://doi.org/10.1016/j.apcata.2020.117496
Received 5 November 2019; Received in revised form 26 February 2020; Accepted 29 February 2020
Available online 29 February 2020
0926-860X/ © 2020 Elsevier B.V. All rights reserved.

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
Fig. 1. Catalytic dioxin decomposition process in municipal waste incineration plants.
(
EXAFS) spectroscopy. XANES spectroscopy provides information re-
fixed on double-sided carbon tape on an Al stub. The spectra were re-
−
6
garding the local geometry around the absorbing atom and its oxidation
state, whereas EXAFS spectroscopy is a powerful technique for ob-
taining direct structural information.
corded at 25 °C under a pressure of < 1.33 × 10
Pa using an AlKα
radiation source with a beam diameter of 100 μm and a pass energy of
140 eV. The binding energies were calibrated to the C1s peak.
The V K-edge XANES measurements were performed at the BL-9C
beamline of the Photon Factory (KEK-PF). The sample masses were
calculated to adjust the height of edge jump and mixed with boron
nitride (catalyst sample:boron nitride = 1:11) before being pressed into
an in situ cell to form a self-supporting disc with a diameter of 7 mm.
Nitrogen gas was fed into the in situ cell and the data were collected in
transmission mode using a Si(1 1 1) monochromator at ambient tem-
perature (25 °C) or at 200 °C. The XANES spectra were analyzed with
the ATHENA software.
2
. Experimental
2
.1. Catalyst preparation
TSM mixed oxides were prepared by a coprecipitation method, as
described in our previous report [17], using titanium sulfate solution
70 g/L TiO , Tayca Corp.), silica sol (30 % SiO , Nissan Chemical
Corp.), ammonium molybdate (Taiyo Koko Co., Ltd.). TiO –SiO (TS)
and TiO were prepared similarly. The SiO content in all TS and TSM
samples was fixed to 7 wt%.
For comparison, MoO /TS (Mo/TS) and MoO
were prepared by an incipient wetness method. Ammonium molybdate
1.36 g) was mixed with monoethanol amine (0.55 g, Nippon Shokubai)
and deionized water (5 g) at approximately 25 °C, and then the TS or
TiO powders (10 g) were impregnated with this molybdenum solution
(
2
2
2
2
The Mo K-edge XAFS measurements were carried out at the BL01B1
beamline of SPring-8 (Japan Synchrotron Radiation Research Institute)
in transmission mode. A double-crystal Si(1 1 1) monochromator was
used. The sample masses were calculated to adjust the height of edge
jump and mixed with boron nitride (catalyst sample:boron ni-
tride = 2:1). The Ti K-edge XAFS measurements were performed at the
BL-9C beamline of the KEK-PF using a Si(1 1 1) monochromator. The
samples were mixed with boron nitride at a sample:boron nitride ratio
of 1:3. For the Mo and Ti K-edge XAFS measurements, each samples
mixed with boron nitride was pressed to form a self-supporting disc
with a diameter of 10 mm, and the XAFS spectra were recorded at
ambient temperature (25 °C) and pressure. The Mo and Ti K-edge XAFS
data were analyzed using the REX2000 software (Rigaku Co.).
The V, Mo and Ti K-edge XANES spectra were normalized to the
edge jump, and the first derivative of the near-edge region was em-
2
2
3
3 2 2
/TiO (Mo/TiO )
(
2
and mixed well. The conditions and procedure for drying and calcina-
tion were the same as those employed for the TSM samples. The no-
menclature used for the various samples is “XMo/TiO
2
”, “XMo/TS”, and
“
XTSM”, where X represents the molybdenum content calculated as the
weight percentage of MoO
between 5 and to 30 wt%, and those in Mo/TS and Mo/TiO
3
3
. The molybdenum content in TSM ranged
were 10 or
0 wt%. Honeycomb-shaped vanadium catalysts were prepared in ac-
2
cordance with our previous report [17]. Ammonium metavanadate was
mixed with an aqueous solution of monoethanol amine and oxalic acid
at approximately 25 °C. The TSM powders and the vanadium solution
obtained above were mixed and kneaded under the continuous addition
of water. The resultant blend was fed into an extrusion molding device
and molded into a honeycomb monolith. After drying at 60 °C and then
calcining at 450 °C for 5 h in air, a honeycomb catalyst with an outer
diameter of 50 mm and a pitch of 3.2 mm was obtained.
3
ployed for determination of the edge absorption energy. The k -
weighted EXAFS oscillations of the Mo and Ti K-edges were extracted
3
using spline smoothing. Subsequently, the filtered k -weighted χ(k) was
−
1
Fourier transformed into R space (k range: 3–13 Å ) using a Hanning
function window [19].
2.3. Catalytic activity measurements
V
2
O
5
–MoO
V10Mo/TiO ) were also prepared by the same method, with ammo-
nium molybdate added to the mixed aqueous solution of ammonium
metavanadate, monoethanol amine, and oxalic acid. V /TiO –SiO
V/TS) was prepared by the same method as V10Mo/TS, except that no
ammonium molybdate was added. The loading amount of vanadium on
all the catalysts in the present study was fixed at 8 wt% V
3
/TiO
2
–SiO
2
(V10Mo/TS)
and
2 5 3 2
V O –MoO /TiO
(
2
Oxidative degradation tests using o-chlorotoluene were conducted
in a U-shaped tubular stainless-steel reactor with an inner diameter of
.8 cm. The honeycomb catalyst to be tested was cut into 25 cells and
loaded into the reactor. A reaction gas mixture consisting of 100 ppm o-
chlorotoluene, 10 % O , and 10 % H O by volume and balanced with N
was supplied to the reactor. o-Chlorotoluene and H O were supplied
using two glass saturators controlled at 31 °C for o-chlorotoluene and
2
O
5
2
2
3
(
2
2
2
2 5
O .
2
2
.2. Characterization of the catalysts
3
4
2
6 °C for H O. The reaction gas flow rate was 0.6 Nm /h and the gas
−1
hourly space velocity GHSV was 6000 h . The reaction was carried out
overnight (∼15 h) at each temperature, and the o-chlorotoluene con-
centrations before and after the reaction were analyzed by gas chro-
matography using a Shimadzu GC-14B instrument equipped with a
Chromosorb column controlled at 140 °C. The concentrations were
analyzed 3–5 times every 20 min, and after confirming the value to be
almost constant, the average value was used for decomposition rate
The X-ray diffraction (XRD) measurements of the support materials
and catalysts were carried out using a Rigaku SmartLab diffractometer
with a CuK radiation source under ambient conditions. The specific
surface areas of the samples were evaluated by the
Brunauer–Emmett–Teller (BET) method using a Mountech Macsorb
α
Model 1210 apparatus. The samples were degassed in flowing N
00 °C for 60 min.
X-ray photoelectron spectroscopy (XPS) spectra were obtained using
a ULVAC PHI Quantera SXM instrument. The sample powders were
2
at
2
calculations. The CO and CO
2
concentrations in the gas were measured
using a Horiba PG-250 gas analyzer.
2

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
Table 1
BET specific surface areas of the support materials and V–Mo catalysts.
2
Sample
Composition
BET surface area / m /g
TiO
TS
2
TiO
93 wt% TiO
2
(anatase)
–7 wt% SiO
82
163
2
2
30Mo/TiO
30Mo/TS
10Mo/TiO
10Mo/TS
2
2
30 wt% MoO
30 wt% MoO
10 wt% MoO
10 wt% MoO
3
3
3
3
/TiO
/TS
/TiO
/TS
2
2
55
102
64
124
30TSM
20TSM
10TSM
63 wt% TiO
MoO
73 wt% TiO
MoO
83 wt% TiO
MoO
2
2
2
–7 wt% SiO
–7 wt% SiO
–7 wt% SiO
2
2
2
–30 wt%
–20 wt%
–10 wt%
115
122
135
3
3
3
V10Mo/TiO
V10Mo/TS
V/10TSM
2
8 wt% V
8 wt% V
8 wt% V
2
2
2
O
5
O
5
O
5
–10 wt% MoO
–10 wt% MoO
/10TSM
3
3
/TiO
/TS
2
60
111
101
Fig. 3. o-Chlorotoluene decomposition over V/TSM as a function of MoO
3
loading. The reaction temperature is 200 °C and all other conditions are as
described for Fig. 2.
3. Results
3.1. Characterization of the supports and catalysts
supported on TSM containing different molybdenum contents, where
3
the MoO = 0 wt% data point corresponds to the performance of the V/
Table 1 summarizes the BET specific surface areas of the support
TS catalyst. Notably, the catalytic activity is sharply enhanced upon the
addition of 5 wt% molybdenum by the coprecipitation method. A gra-
dual increase in catalytic activity is observed up to 10 wt%, but at
higher molybdenum loadings, the activity decreases once again,
thereby indicating that excess molybdenum inhibits the reactivity. Al-
though the main active species in the o-chlorotoluene decomposition
reaction is vanadium, it is clear that the activity strongly depends on
the addition method and the amount of molybdenum.
materials with various compositions and the corresponding V–Mo
containing catalysts. TS and TSM were found to exhibit noticeably high
2
specific surface areas (163 and 135 m /g, respectively) compared to
2
TiO
2
(82 m /g), which was attributed to the presence of SiO
2
[3]. The
surface areas of TS and 10TSM decreased when vanadium and/or mo-
2
lybdenum were added, but remained > 100 m /g.
3.2. o-Chlorotoluene oxidation activity
3.3. XRD
The results of the o-chlorotoluene oxidation reaction over three
types of V–Mo catalysts are shown in Fig. 2. In the catalyst outlet gas,
no organic compounds were detected with the exception of unreacted o-
chlorotoluene, and it was confirmed that reacted o-chlorotoluene was
converted to CO by checking the carbon balance. The V/10TSM cat-
2
alyst showed a significantly higher activity than the other two catalysts,
especially at low temperatures.
The XRD patterns of the catalysts and the support materials are
shown in Fig. 4. In all samples, diffraction peaks ascribable to anatase
TiO
TiO
2
(JCPDS: 00-021-1272) are observed. The XRD patterns of 30Mo/
and 30Mo/TS display strong peaks assigned to α-MoO (JCPDS:
2
3
0
0-005-0508), and weak peaks at the same positions are observed for
0Mo/TiO and 10Mo/TS (inset, Fig. 4). In contrast, the crystal phases
and V (JCPDS: 00-041-1426) are not detectable for the
V10Mo/TiO and V10Mo/TS catalysts. These observations suggest that
the structure of molybdenum oxide is modified by interactions with
vanadium in the V10Mo/TiO and V10Mo/TS catalysts.
1
2
Fig. 3 shows the performance of the catalysts with 8 wt% vanadium
of α-MoO
3
2 5
O
2
2
Furthermore, the XRD patterns of TSM display peaks ascribable to
Fig. 2. o-Chlorotoluene decomposition over V/10TSM ( ), V10Mo/TS ( ), and
V10Mo/TiO
2
( ). Reaction conditions: 100 ppm o-chlorotoluene, 10 vol% O
2
,
Fig. 4. (a) XRD patterns of the Mo-containing catalysts, and (b) enlarged data in
the region of 26–29°. Symbols: MoO (◆); TiO (○).
3 2
−
1
1
0 vol% H O, N
2
2
balance; GHSV: 6000 h
.
3

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
Fig. 5. Enlarged XRD patterns of the 10Mo/TS and TSM samples. The peak
2
maxima for these samples (solid blue lines) and anatase TiO (JCPDS No. 00-
021-1272, dotted red line) are indicated (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this
article).
Fig. 6. (a) Mo3d and (b) Ti2p XPS spectra of the Mo-containing samples, nor-
anatase TiO
oxide, even at a MoO
2
but do not display peaks corresponding to molybdenum
content of 30 wt% (30TSM). These results imply
malized to the intensity of Ti2p2/3 for comparison of the Mo/Ti ratio.
3
that the molybdenum species are highly dispersed within the TSM
support.
Table 3
Surface atomic compositions and Mo/Ti ratios estimated by XPS.
As shown in Fig. 5, the diffraction peak observed at 2θ = 48° for
Surface atomic composition / %
Mo/Ti
1
0Mo/TS is consistent with that assigned to the (2 0 0) plane of anatase
2
TiO . A peak shift to lower diffraction angles is observed for TSM, with
Mo
Ti
Si
O
the extent of the shift increasing at higher molybdenum contents. The
spacing between the (2 0 0) planes in the atomic lattice of 30TSM can
be calculated as 1.900 Å (Table 2), which is clearly greater than that of
10Mo/TiO
10Mo/TS
10TSM
2
8.0
6.8
4.6
28.2
25.5
27.0
–
4.0
5.8
63.8
63.8
62.6
0.29
0.27
0.17
2
anatase TiO (1.892 Å).
The surface compositions of these samples calculated from the XPS
peak areas are summarized in Table 3. The surface atomic ratio of Mo/
Ti for 10TSM is lower than those for 10Mo/TiO and 10Mo/TS, thereby
indicating that the surface molybdenum density on TSM is lower than
those on the other supports, which suggests that molybdenum is in-
corporated within the particles.
3
.4. XPS
2
The Mo-containing samples were characterized by XPS to in-
vestigate the properties of the surface molybdenum species (Fig. 6). The
binding energies for Mo3d (232.6 eV) and Ti2p (458.7 eV) were iden-
tical in all samples, indicating that the majority of molybdenum and
titanium species on the surface are present as Mo and Ti . The XPS
Mo3d peaks of 10Mo/TiO , 10Mo/TS and 10TSM, which are almost
identical in shape, are broadened at lower binding energies compared
to that of MoO . This broadening can be attributed to the presence of
another type of molybdenum species or to electron transfer between
molybdenum and the support material [20]. These results indicate that
the electronic features of the surface molybdenum species on 10TSM
6
+
4+
3.5. V K-edge XANES spectroscopy
2
The V K-edge XANES spectra of the three vanadium catalysts as well
as the V O and V O reference samples are presented in Fig. 7. For all
2 4 2 5
samples, pre-edge peaks are observed at approximately 5468 eV. These
peaks are observable due to the 3d–4p orbital mixing of vanadium,
3
which is induced by the distortion of VO
structure [21–23]. The pre-edge areas and main-edge positions are
summarized in Table 4. As indicated, V displays a larger pre-edge
area than V , suggesting a distorted square-pyramidal structure
around the vanadium atom in V [21–25]. The XANES spectra col-
lected at 25 °C (thin line) and 200 °C (bold line) are identical for V
and V . In contrast, the pre-edge and main-edge positions of the three
x 6
from the octahedral VO
2
are similar to those on 10Mo/TiO and 10Mo/TS.
2 5
O
Table 2
XRD peak positions and interplanar distances of the (2 0 0) plane of anatase
TiO
2 4
O
2
.
2 5
O
2 4
O
2θ / °
Interplanar distance / Å
2 5
O
anatase TiO
2
(JCPDS)
48.05
48.04
47.98
47.92
47.88
1.892
1.894
1.896
1.898
1.900
catalysts shift to lower photon energies after heating to 200 °C, ac-
companied by a decrease in the pre-edge area. These results suggest
that vanadium oxide is reduced to a lower oxidation state owing to the
desorption of surface oxygen, and thus the morphology around the
vanadium atom is altered. A greater shift is observed in the main-edge
1
1
2
3
0Mo/TS
0TSM
0TSM
0TSM
4

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
Fig. 7. Normalized V K-edge XANES spectra of the V-containing catalysts and
reference samples collected at 25 °C (thin lines) and 200 °C (bold lines) under
flowing N .
2
Table 4
Fig. 9. Pseudo-Voigt fitting result for the normalized Mo K-edge XANES spec-
trum of MoO .
3
Pre-edge peak areas and main edge positions in the V K-edge XANES spectra.
Pre-edge peak area Main-edge position / eV
2
5 °C
200 °C
25 °C
200 °C
and MoO
peaks, which can be observed due to the d–p orbital hybridization
caused by the distortion of the octahedral MoO structure [26–29]. For
a quantitative analysis, the ATHENA software was used to fit the XANES
spectra in this region with a pseudo-Voigt function in combination with
an error function (Fig. 9). The absorption edge energies and pre-edge
3 2
). All samples with the exception of MoO exhibit pre-edge
V
V
2
O
O
4
0.96
1.73
1.81
1.79
1.74
0.99
1.76
1.75
1.77
1.65
5478.6
5480.0
5480.1
5480.1
5480.2
5478.5
5480.1
5479.7
5479.7
5479.5
6
2
5
V10Mo/TiO
V10Mo/TS
V/10TSM
2
3
peak positions are the same as those of crystalline MoO , indicating that
molybdenum is present as Mo6+, regardless of the type of support, Mo
loading amount, preparation method, or the presence of vanadium.
position of V/10TSM than in those of V10Mo/TiO
2
and V10Mo/TS,
indicating that the surface oxygen around vanadium is more easily
desorbed over the V/10TSM catalyst. In other words, the reducibility of
vanadium oxide is enhanced when supported on TSM.
Furthermore, 30Mo/TiO
tensities to MoO , which has distorted octahedral symmetry [26]
magnified spectra in Fig. 10). While 10Mo/TiO and V10Mo/TiO
exhibit lower pre-edge intensities, 10Mo/TS demonstrates a higher pre-
edge intensity than MoO . It has been reported that MoO /Al cat-
2
and 30Mo/TS display similar pre-edge in-
3
(
2
2
3
3
2 3
O
3.6. Mo K-edge XANES spectroscopy
alysts exhibit intense pre-edge peaks, indicating the presence of a
monomeric molybdenum oxide species [28]. This species seems to be
Fig. 8 displays the Mo K-edge XANES spectra of the molybdenum-
containing samples and the molybdenum reference compounds (MoO
highly dispersed owing to the large specific surface area of the Al
support. Therefore, the increased pre-edge intensity observed for
0Mo/TS can be attributed to the large specific surface area of the TS
2 3
O
2
1
support, and it is likely that some molybdenum oxides on 10Mo/TS are
present as monomeric species. In contrast, 30TSM demonstrates a lower
pre-edge intensity than 10Mo/TiO , and the intensity decreases as the
2
molybdenum content decreases. According to the XRD results (Fig. 4),
molybdenum atoms in TSM are more highly dispersed than those in
2
Mo/TiO and Mo/TS; however, monomeric molybdenum oxide species
are not formed in the TSM samples, suggesting that the structure
around molybdenum is less distorted.
The pre-edge intensity of the V10Mo/TS catalyst is clearly lower
than that of 10Mo/TS, indicating the presence of an interaction be-
tween vanadium and molybdenum on V10Mo/TS. In the case of TSM,
the pre-edges of 10TSM and V/10TSM are almost identical, indicating
that the local symmetry around molybdenum in TSM is not affected by
the presence of vanadium.
3.7. Mo K-edge EXAFS spectroscopy
3
The k -weighted EXAFS spectra of the Mo K-edge and the corre-
sponding Fourier transforms (radial structure function; RSF) of 30Mo/
TiO and 30Mo/TS are shown in Fig. 11, where the data of crystalline
MoO and MoO are shown for reference. Intense peaks at 0.8–2.0 Å
assigned to Mo–O are observed for all samples [28–31]. Furthermore,
0Mo/TiO and 30Mo/TS exhibit EXAFS oscillations similar to those of
crystalline MoO , with a predominant peak at 3.0–4.0 Å owing to
2
2
3
Fig. 8. Normalized Mo K-edge XANES spectra of the Mo-containing catalysts
and reference samples.
3
2
3
5

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
Fig. 10. Enlarged pre-edge regions in the normalized Mo K-edge XANES spectra. The values in parentheses are the pre-edge peak areas obtained by pseudo-Voigt
fitting.
3
3
Fig. 11. (a) k -weighted Mo K-edge EXAFS spectra, and (b) corresponding
Fig. 12. (a) k -weighted Mo K-edge EXAFS spectra, and (b) corresponding
Fourier transforms of the Mo-containing catalysts and reference samples.
Fourier transforms of the samples containing 10 wt% molybdenum oxide.
Mo–Mo interactions. The majority of molybdenum oxides on these
Table 5
samples are found to be present as crystalline MoO
agreement with the XRD results.
The EXAFS oscillations and RSF of the samples containing 10 wt%
molybdenum oxides either with or without vanadium and supported on
3
, which is in good
Positions and intensities of peaks A and B in the RSF in Figs. 12 and 13.
Peak A
Peak B
Position
Intensity
Position
Intensity
TiO
0Mo/TiO
MoO , and peaks are observed at approximately 2.4 and 3.0 Å. Al-
though small peaks corresponding to MoO are observed in the XRD
patterns (Fig. 4), the majority of molybdenum oxides have different
structures to crystalline MoO . According to the literature concerning
the EXAFS spectra of analogous MoO /TiO catalysts [32], the peaks
2
or TS are presented in Fig. 12. The EXAFS spectra and RSF of
1
0Mo/TiO
V10Mo/TiO
0Mo/TS
V10Mo/TS
2
2.42
2.38
2.42
2.42
1.61
1.51
1.25
1.54
3.07
2.99
3.03
2.99
1.79
1.25
1.10
1.38
1
2
and 10Mo/TS are quite different from those of crystalline
2
3
1
3
30TSM
20TSM
10TSM
2.42
2.45
2.53
2.53
1.34
1.73
2.29
2.55
2.95
2.95
2.88
2.84
1.63
1.65
2.22
2.23
3
3
2
observed at 2.0–3.5 Å for samples with lower molybdenum loadings are
V/10TSM
ascribable to Mo–Mo or Mo–Ti interactions.
The positions and intensities of the peaks observed in the 2.0–3.5 Å
region of the RSF in Fig. 12 are summarized in Table 5. The amplitudes
of peaks A and B in the RSF of 10Mo/TS are lower than those in the RSF
of vanadium. This change indicates the presence of interactions be-
tween vanadium and molybdenum, as suggested by the XRD and
XANES results. The morphology around molybdenum seems to be al-
tered by these interactions.
The EXAFS oscillations and RSF for TSM and V/10TSM are shown in
Fig. 13. The EXAFS spectrum and RSF of 30TSM are similar to those of
10Mo/TS (Fig. 12), with peaks being observed close to 2.4 and 3.0 Å in
the RSF. In accordance with the XRD results, it is confirmed that
crystalline MoO3 is not formed in 30TSM.
2
of 10Mo/TiO , indicating a lower crystallinity and/or the presence of
monomeric molybdenum species, as suggested by the XRD (Fig. 4) and
XANES results (Fig. 10). Mensch et al. suggested that higher peaks close
to 3 Å in the RSF of Mo K-edge EXAFS spectra reflect the polymerization
of molybdenum species [28], thereby supporting the present results.
The Mo K-edge EXAFS spectra and the RSF of V10Mo/TiO and
2
V10Mo/TS are similar to those of the samples without vanadium. A
slight shift to shorter distances is observed for peak B after the addition
Furthermore, changes in the EXAFS oscillations and RSF are
6

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
2
different from those of TiO .
4. Discussion
4.1. Structural characteristics of molybdenum in V/TSM catalysts
V/TSM catalysts, which are highly active for NH
exhibit superior activity for the oxidative decomposition of o-chlor-
otoluene, especially at low temperatures (Fig. 2). The V10Mo/TiO
3
-SCR [17], also
2
catalyst showed the lowest activity, probably owing to its small surface
area (Table 1). In contrast, the V/10TSM catalyst exhibited a higher
activity than the V10Mo/TS catalyst, which had a higher surface area,
thereby suggesting that the catalytic activity is affected by factors other
than the surface area.
The V K-edge XANES spectra (Fig. 7) revealed that vanadium sup-
ported on TSM exhibits enhanced reducibility. Chang et al. suggested
that the redox capability of vanadium is important for the oxidative
decomposition of 1,2-dichlorobenzene [33]. Similarly, the redox cap-
ability of the vanadium species may contribute to the higher activity of
V/TSM catalysts.
The major difference between V/10TSM and V10Mo/TS is the
method for adding molybdenum. Therefore, the physicochemical
properties of molybdenum in the V/TSM catalyst were investigated, and
the XRD results showed that molybdenum is dispersed in the TSM
3
Fig. 13. (a) k -weighted Mo K-edge EXAFS spectra, and (b) corresponding
Fourier transforms of the TSM and V/TSM samples.
observed as the molybdenum content decreases, and 10TSM shows a
characteristic spectrum that is clearly different from those of MoO
MoO , Mo/TiO , and Mo/TS. The positions and intensities of peaks A
and B in Fig. 13 are shown in Table 5. Shifts to longer distances for peak
A and to shorter distances for peak B are observed, correlating to the
molybdenum contents in the TSM sample. In addition, the amplitudes
of peaks A and B increase with a decrease in the molybdenum content.
V/10TSM demonstrates an EXAFS spectrum similar to that of
2
,
3
2
3
support with a structure different from crystalline MoO (Fig. 4). The
observed shift of the XRD peaks toward lower angles indicates an ex-
pansion of the interplanar distance brought about by the presence of
molybdenum added by the coprecipitation method.
Moreover, as demonstrated by XPS, molybdenum is incorporated
into the bulk structure of the TSM support, and no obvious differences
are observed in the electronic behavior of the molybdenum species on
the surface (Fig. 6). Mo K-edge XANES spectroscopy revealed that the
oxidation state of molybdenum in V/10TSM is 6+ and that the octa-
hedral structure around molybdenum is less distorted than in V10Mo/
1
0TSM, but peak B is shifted to a lower distance in the RSF. These
results indicate that interactions between vanadium and molybdenum
occur in the V/10TSM catalyst, similar to those in the other two V–Mo
containing catalysts. However, the structure of 10TSM was mostly
maintained following the addition of vanadium.
3
.8. Ti K-edge EXAFS spectroscopy
2
TS and V10Mo/TiO (Fig. 10). These results suggest that the electronic
behavior of the surface molybdenum species has almost no influence on
the catalytic activity for o-chlorotoluene decomposition.
In addition, the RSF data obtained from the Mo K-edge EXAFS
spectra established that the Mo–Mo or Mo–Ti distances in the TSM
3 2
support are different from those in crystalline MoO , Mo/TiO , and Mo/
TS (Figs. 11–13). In the RSF of 10TSM, the peaks A and B were located
close to each other and the peak amplitudes were higher than those in
other samples. These results imply that the structure of the V/10TSM
catalyst is different from those of the other samples, and that 10TSM
mostly consists of molybdenum oxide species with similar structures.
We attempted a curve-fitting analysis on the second shell con-
tributions in the Mo K-edge EXAFS spectra. However, we could not
obtain good fitting for the TSM-type samples because in this case, the
second shell consists of more than two scatterings and we could not find
an appropriate structural model for these peaks. Moreover, evaluation
of the inverse Fourier transforms of the second shell did not reveal any
significant difference between the TSM samples and 10Mo/TS, due to
overlapping of plural EXAFS oscillations.
3
The k -weighted EXAFS spectra of the Ti K-edge and the corre-
sponding RSF of TiO
edge EXAFS data for 10TSM and 10Mo/TiO
support material for 10Mo/TiO has an anatase structure (confirmed by
XRD, data not shown). The EXAFS oscillations in the Mo K-edge spec-
trum of 10TSM resemble those in the Ti K-edge EXAFS spectra of
anatase TiO
tangle, Fig. 14a), suggesting that the Mo–Mo or Mo–Ti distances in
0TSM are very similar to the Ti–Ti distances in anatase TiO . This
similarity is reflected in the RSF in the R range of 2–3 Å, whereas the
2
and 10TSM are shown in Fig. 14, with the Mo K-
2
. The TiO used as the
2
2
−
1
2
and 10TSM in the k range of 7–11.5 Å
(dotted rec-
1
2
2
EXAFS spectrum and RSF of 10Mo/TiO in these ranges are obviously
4
+
In contrast, according to past studies [34,35], Ti
(0.068 nm) in
the lattice can be replaced by Mo6+ (0.062 nm), because they have
similar ionic radii. Thus, molybdenum in the TSM support can be
homogeneously incorporated into the TS framework as Mo , forming
a solid solution. Furthermore, the RSF of the Ti K-edge and Mo K-edge
EXAFS spectra (Fig. 14) correspond to the second-shell contributions,
and indicate that the Mo–Mo or Mo–Ti distances are very similar to the
6
+
2
Ti–Ti distances in anatase TiO , strongly supporting the formation of a
TSM solid solution.
3
Fig. 14. (a) k -weighted Ti and Mo K-edge EXAFS spectra, and (b) corre-
2
sponding Fourier transforms of 10TSM and TiO -based samples.
7

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
4.2. Influence of molybdenum oxides on the performance of V/TSM
Mo6+ and Ti4+, and thus the redox capability of the vanadium species
supported on TSM was enhanced. However, the extent of electron
catalysts
2
transfer depended not on the total distortion of the anatase TiO fra-
The XRD and Mo K-edge XAFS analyses revealed the presence of
mework, but on the local morphologies around the molybdenum in
TSM. Thus, the superior performance of the V/10TSM catalyst for the
low-temperature decomposition of o-chlorotoluene was attributed to
the high dispersion of Mo6+ species in the 10TSM solid solution, which
has a unique structure around molybdenum. This detailed physico-
chemical investigation provided insight into the structure of the TSM
solid solution, and the enhanced redox capability of V/TSM was re-
vealed to be effective for the low temperature oxidative decomposition
of o-chlorotoluene. However, as the behaviors of the vanadium species
and the support material during catalytic reactions have yet to be
elucidated, future work in our group will focus on these points to
contribute to the development of novel catalysts for improved NOx and
dioxin abatement.
V–Mo interactions in the V–Mo-containing catalysts, and it was found
that the structural changes around molybdenum after the addition of
vanadium were not remarkable for 10TSM (Figs. 10 and 13). However,
the changes in the V K-edge XANES spectra after heating to 200 °C
2
under flowing N were most evident for V/10TSM (Fig. 7). These
findings suggest that the superior catalytic activity of V/10TSM is the
result of electronic interactions, mainly from the TSM support toward
vanadium.
2
Tsai et al. reported that Mo-doped TiO nanoparticles form a solid
solution and that electron transfer from the support material to the
active component is induced by oxygen vacancies generated in the
nanoparticles [36]. It should be noted that the morphology around
2
molybdenum in these Mo-doped TiO nanoparticles was found to be
tetrahedral, which is different from the octahedral structure of mo-
lybdenum oxide in the TSM support in the present study. This differ-
ence in structure is presumably ascribable to different preparation
methods. However, similar to the case of Mo-doped TiO nanoparticles,
2
we believe that the electron transfer process is related to oxygen va-
cancies generated in the TSM solid solution.
CRediT authorship contribution statement
Ryoji Kuma: Conceptualization, Methodology, Formal analysis,
Investigation, Resources, Writing - original draft, Writing - review &
editing, Project administration. Tomoyuki Kitano: Formal analysis,
Investigation, Resources. Takuya Tsujiguchi: Investigation, Resources.
Tsunehiro Tanaka: Writing - review & editing, Supervision.
While 30TSM exhibited a larger interplanar distance than 10TSM
(
(
Fig. 5), V/10TSM demonstrated a higher activity than V/30TSM
Fig. 3). These results indicate that the extent of electron transfer de-
Declaration of Competing Interest
The authors have no competing interests to declare.
Acknowledgements
2
pends not on the total extent of distortion in the anatase TiO frame-
work, but on the local morphologies around molybdenum in the TSM
support. Thus, electron transfer from the TSM support to vanadium is
found to be enhanced when octahedrally coordinated molybdenum
with a less distorted structure is well dispersed in the TSM support, in
4
+
6+
which Ti
is replaced by Mo . The characteristics and unique
We acknowledge Dr. K. Kimijima (High Energy Accelerator
Research Organization, KEK) and Professor K. Okumura (Kogakuin
University) for helpful and valuable assistance in the XAFS experiments.
Publication permission and funding from Nippon Shokubai Co. Ltd. is
also gratefully acknowledged.
structure of the TSM support contribute to enhancing the redox cap-
ability of vanadium in the V/TSM catalyst, resulting in a remarkable
low-temperature o-chlorotoluene decomposition activity.
In our previous study [17], we reported that the V/TSM catalyst
3
exhibits an improved NH -SCR activity, which seemed to be related to
the redox properties of V/TSM. The results in the present study are
consistent with this hypothesis and enabled us to confirm it to be cor-
rect, with additional evidence for the role of the support material.
Furthermore, it was found that the enhanced redox capability of V/TSM
is also effective for the low temperature oxidative decomposition of o-
chlorotoluene. As a result, the simultaneous decomposition of both NOx
and dioxins can be achieved with a lower catalyst volume.
References
[1] L. Lietti, J. Svachula, P. Forzatti, G. Busca, G. Ramis, P. Bregani, Catal. Today 17
(1993) 131–139, https://doi.org/10.1016/0920-5861(93)80016-T.
[
2] G. Busca, L. Lietti, G. Ramis, F. Berti, Appl. Catal. B Environ. 18 (1998) 1–36,
https://doi.org/10.1016/S0926-3373(98)00040-X.
[3] M. Kobayashi, R. Kuma, S. Masaki, N. Sugishima, Appl. Catal. B Environ. 60 (2005)
73–179, https://doi.org/10.1016/j.apcatb.2005.02.030.
4] J.-M. Herrmann, J. Disdier, Catal. Today 56 (2000) 389–401, https://doi.org/10.
016/S0920-5861(99)00299-0.
1
[
1
5. Conclusions
[5] I. Giakoumelou, C. Fountzoula, C. Kordulis, S. Boghosian, J. Catal. 239 (2006) 1–12,
https://doi.org/10.1016/j.jcat.2006.01.019.
[
[
[
[
6] S.T. Choo, Y.G. Lee, I.-S. Nam, S.-W. Ham, J.-B. Lee, Appl. Catal. A Gen. 200 (2000)
We herein reported our investigation into the physicochemical
characteristics of molybdenum oxide in V/TiO –SiO –MoO (TSM)
2 2 3
1
77–188, https://doi.org/10.1016/S0926-860X(00)00636-0.
7] S. Krishnamoorthy, J.P. Baker, M.D. Amiridis, Catal. Today 40 (1998) 39–46,
https://doi.org/10.1016/S0920-5861(97)00117-X.
8] S. Krishnamoorthy, M.D. Amiridis, Catal. Today 51 (1999) 203–214, https://doi.
org/10.1016/S0920-5861(99)00045-0.
9] M.A. Larrubia, G. Busca, Appl. Catal. B Environ. 39 (2002) 343–352, https://doi.
org/10.1016/S0926-3373(02)00116-9.
catalysts using X-ray diffraction (XRD), X-ray photoelectron spectro-
scopy (XPS), and X-ray absorption fine structure spectroscopy (XAFS),
and their performance for the oxidative decomposition of o-chlor-
otoluene was examined. The V/TSM catalysts showed superior perfor-
mance for the low-temperature oxidative decomposition of o-chlor-
otoluene, especially in the case of the TSM support containing 10 wt%
molybdenum oxide. The vanadium species supported on 10TSM ex-
hibited a higher reducibility, and this improved redox capability of
vanadium is likely responsible for the enhanced catalytic activity. XRD,
XPS, and Mo K-edge XAFS analyses revealed that the molybdenum
present in the TSM support is octahedrally coordinated, with a less
[
[
[
[
[
[
10] E. Finocchio, G. Busca, M. Notaro, Appl. Catal. B Environ. 62 (2006) 12–20, https://
doi.org/10.1016/j.apcatb.2005.06.010.
11] D.P. Debecker, R. Delaigle, K. Bouchmella, P. Eloy, E.M. Gaigneaux, P.H. Mutin,
Catal. Today 157 (2010) 125–130, https://doi.org/10.1016/j.cattod.2010.02.010.
12] C. Gannoun, R. Delaigle, D.P. Debecker, P. Eloy, A. Ghorbel, E.M. Gaigneaux, Appl.
Catal. A Gen. 447–448 (2012) 1–6, https://doi.org/10.1016/j.apcata.2012.08.034.
13] M. Yu, W. Li, X. Li, X. Lin, T. Chen, J. Yan, Chemosphere 156 (2016) 383–391,
https://doi.org/10.1016/j.chemosphere.2016.05.011.
14] C.E. Hetrick, J. Lichtenberger, M.D. Amiridis, Appl. Catal. B Environ. 77 (2008)
2
55–263, https://doi.org/10.1016/j.apcatb.2007.07.022.
3
distorted structure than in the case of crystalline MoO , and is also
15] G. Busca, M. Baldi, C. Pistarino, J.M. Gallardo Amores, V. Sanchez Escribano,
E. Finocchio, G. Romezzano, F. Bregani, G.P. Toledo, Catal. Today 53 (1999)
25–533, https://doi.org/10.1016/S0920-5861(99)00140-6.
[16] M. Hiraoka, T. Fujii, K. Kashiwabara, K. Ieyama, M. Kondo, Chemosphere 23 (1991)
439–1444, https://doi.org/10.1016/0045-6535(91)90168-D.
highly dispersed inside the TSM framework, thereby indicating the
formation of a solid solution. Preparation of the TSM supports via a
5
4
+
coprecipitation method appeared to result in the Ti
within the TiO
2
1
structure being replaced by Mo6+. Electron transfer from the TSM
support to the vanadium species was enhanced by the coexistence of
[17] M. Kobayashi, R. Kuma, A. Morita, Catal. Lett. 112 (2006) 37–44, https://doi.org/
10.1007/s10562-006-0161-4.
8

R. Kuma, et al.
Applied Catalysis A, General 595 (2020) 117496
[
[
18] N. Sugishima, M. Kobayashi, Shokubai 43 (2001) 559–564.
19] S. Yamazoe, Y. Hitomi, T. Shishido, T. Tanaka, J. Phys. Chem. C 112 (2008)
92 (1988) 4961–4964, https://doi.org/10.1021/j100328a028.
[29] R. Radhakrishnan, C. Reed, S.T. Oyama, M. Seman, J.N. Kondo, K. Domen,
Y. Ohminami, K. Asakura, J. Phys. Chem. B 105 (2001) 8519–8530, https://doi.
org/10.1021/jp0117361.
[30] S. Takenaka, T. Tanaka, T. Funabiki, S. Yoshida, J. Phys. Chem. B 102 (1998)
2960–2969, https://doi.org/10.1021/jp980134n.
[31] J.A. Wharton, D.H. Ross, G.M. Treacy, G.D. Wilcox, K.R. Baldwin, J. Appl.
Electrochem. 33 (2003) 553–561, https://doi.org/10.1023/A:1024911119051.
[32] M.J. Fay, A. Proctor, D.P. Hoffmann, M. Houalla, D.M. Hercules, Mikrochim. Acta
109 (1992) 281–293, https://doi.org/10.1007/BF01242483.
[33] H.-Y. Chang, S.-P. Wang, J.-R. Chang, H.-S. Sheu, S.-G. Shyu, Appl. Catal. B Environ.
111–112 (2012) 476–484, https://doi.org/10.1016/j.apcatb.2011.10.037.
[34] Y.J. Feng, C. Wang, W.R. Cheng, J.H. Huang, T.X. Zhao, Q.H. Liu, Z. Xie, Z.Y. Pan,
S.Q. Wei, J. Phys. Conf. Ser. 430 (2013) 012090, , https://doi.org/10.1088/1742-
6596/430/1/012090.
6
869–6879, https://doi.org/10.1021/jp711250f.
[
[
[
[
[
[
[
[
[
20] B.M. Reddy, B. Chowdhury, P.G. Smirniotis, Appl. Catal. A Gen. 219 (2001) 53–60,
https://doi.org/10.1016/S0926-860X(01)00658-5.
21] F. Amano, T. Tanaka, Catal. Commun. 6 (2005) 269–273, https://doi.org/10.1016/
j.catcom.2005.01.007.
22] W.C. Vining, J. Strunk, A.T. Bell, J. Catal. 285 (2012) 160–167, https://doi.org/10.
1
016/j.jcat.2011.09.024.
23] J. Wong, F.W. Lytle, R.P. Messmer, D.H. Maylotte, Phys. Rev. B 30 (1984)
596–5610, https://doi.org/10.1103/PhysRevB.30.5596.
24] K.L. Bøyesen, K. Mathisen, Catal. Today 229 (2014) 14–22, https://doi.org/10.
016/j.cattod.2013.12.056.
25] S. Takenaka, T. Tanaka, T. Yamazaki, T. Funabiki, S. Yoshida, J. Phys. Chem. B 101
1997) 9035–9040, https://doi.org/10.1021/jp9716086.
5
1
(
26] H. Aritani, T. Tanaka, T. Funabiki, S. Yoshida, M. Kudo, S. Hasegawa, J. Phys.
Chem. 100 (1996) 5440–5446, https://doi.org/10.1021/jp9525245.
27] K. Shimura, H. Kanai, K. Utani, K. Matsuyama, S. Imamura, Appl. Catal. A Gen. 283
[35] H.W. Kang, S.B. Park, A.-H.A. Park, Int. J. Hydrogen Energy 38 (2013) 9198–9205,
https://doi.org/10.1016/j.ijhydene.2013.05.106.
[36] M.-C. Tsai, T.-T. Nguyen, N.G. Akalework, C.-J. Pan, J. Rick, Y.-F. Liao, W.-N. Su, B.-
J. Hwang, ACS Catal. 6 (2016) 6551–6559, https://doi.org/10.1021/acscatal.
6b00600.
(2005) 117–124, https://doi.org/10.1016/j.apcata.2004.12.042.
28] C.T.J. Mensch, J.A.R. Van Veen, B. Van Wingerden, M.P. Van Dijk, J. Phys. Chem.
9