Nb?V Mixed Oxide with a Random Assembly of Pentagonal Units: A Catalyst for Oxidative Dehydrogenation of Ethane and Propane

Article abstract of DOI:10.1002/cctc.202100463

High-dimensionally structured (HDS) mixed oxides of vanadium with metals (M) (e. g., Nb, Mo, and W; denoted as HDS-MVO) were constructed by {M₆O₂₁}^12? pentagonal units and {MO₆} (M=Nb, Mo, W, or V) octahedra as linkers. The materials were synthesized using a hydrothermal method and rod-shaped solids. The random assembly of the pentagonal units and octahedra in the cross-sectional plane of the rods facilitated the formation of micropore channels along the long axis of the rods. Micropore formation was directly observed in the cross-section by HAADF-STEM. These structural features are common to HDS-NbVO, HDS-MoVO, and HDS-WVO. The catalytic activity of these three HDS-MVOs with V/Mo ratios in the range 0.35–0.39 was tested for the oxidative dehydrogenation of ethane and propane. The reaction rates per surface area for ethane oxidation and propane oxidation over the HDS-MoVO and HDS-WVO catalysts were comparable, whereas the HDS-NbVO catalyst showed an appreciable difference between the two reaction rates. Both HDS-MoVO and HDS-WVO exhibited higher selectivity for olefin formation during ethane oxidation than propane oxidation. Interestingly, the olefin selectivity over the HDS-NbVO catalyst was found to be almost independent of the alkane substrate. These catalytic features were discussed on the basis of V?O?V or V?O?Mo redox coupling and pore structure effects in HDS-MoVO and HDS-WVO and also of isolated and valence stable surface V in HDS-NbVO.

Full text of DOI:10.1002/cctc.202100463

The European Society Journal for Catalysis
Supported by
Accepted Article
Title: Nb-V Mixed Oxide with Random Assembly of Pentagonal Units:
Catalyst for Oxidative Dehydrogenation of Ethane and Propane
Authors: Satoshi Ishikawa, Kosuke Shimoda, Katsuya Matsumoto, Mai
Miyasawa, Marino Takebe, Riku Matsumoto, Syutoku Lee,
and Wataru Ueda
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.202100463
Link to VoR: https://doi.org/10.1002/cctc.202100463
01/2020

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
Nb-V Mixed Oxide with Random Assembly of Pentagonal Units:
Catalyst for Oxidative Dehydrogenation of Ethane and Propane
Kosuke Shimoda, Satoshi Ishikawa*, Katsuya Matsumoto, Mai Miyasawa, Marino Takebe, Riku
Matsumoto, Syutoku Lee, Wataru Ueda*
These authors contributed equally to this work.
Department of Material and Life Chemistry, Faculty of Engineering, Kanagawa University, 3-27, Rokkakubashi, Kanagawa-ku, Yokohama, 221-8686,
Japan
E-mail: sishikawa@kanagawa-u.ac.jp
Supporting information for this article is given via a link at the end of the document.((Please delete this text if not appropriate))
Abstract: High-dimensionally structured (HDS) mixed oxides of
vanadium with metals (M) (e.g., Nb, Mo, and W; denoted as HDS-
MVO) were constructed by {M6O21}12– pentagonal units and {MO6} (M
= Nb, Mo, W, or V) octahedra as linkers. The materials were
synthesized using a hydrothermal method as rod-shaped solids. The
random assembly of the pentagonal units and octahedra in the cross-
sectional plane of the rods facilitated the formation of micropore
channels along the long axis of the rods. Micropore formation was
directly observed in the cross-section by HAADF-STEM. These
structural features are common to HDS-NbVO, HDS-MoVO, and
HDS-WVO. The catalytic activity of these three HDS-MVOs with V/Mo
ratios in the range 0.35–0.39 was tested for the oxidative
dehydrogenation of ethane and propane. The reaction rates per
surface area for ethane oxidation and propane oxidation over the
Figure 1. Structural model of crystalline Mo3VOx. Left,
HDS-MoVO and HDS-WVO catalysts were comparable, whereas the
orthorhombic Mo3VOx; right, trigonal Mo3VOx. Light blue, Mo; gray,
HDS-NbVO catalyst showed an appreciable difference between the
V; light green, mixture of Mo and V; red, O. Representative
two reaction rates. Both HDS-MoVO and HDS-WVO exhibited higher
pentagonal units are colored blue for easy understanding.
selectivity for olefin formation during ethane oxidation than propane
oxidation. Interestingly, the olefin selectivity over the HDS-NbVO
have been devoted to the development of ODH; some suitable
catalyst was found to be almost independent of the alkane substrate.
catalysts have been reported, such as Mo-V-based mixed oxides
[5-10] and Ni-Nb mixed oxides (for the oxidative dehydrogenation of
These catalytic features were discussed on the basis of V-O-V or V-
O-Mo redox coupling and pore structure effects in HDS-MoVO and
HDS-WVO and also of isolated and valence stable surface V in HDS-
NbVO.
[11-13]
[14-15]
ethane (ODHE)),
catalysts,
V-based mixed oxides,
V-supported
[16-17]
[18]
and boron nitride
for the oxidative
dehydrogenation of propane (ODHP). However, these catalysts
exhibit low selectivity, particularly for ODHP, because of the
difficulty in preventing the sequential oxidation of olefins.
For selective oxidation, Grasselli introduced the important
concept of site isolation for designing oxidation catalysts. [19] This
concept suggests that the selectivity of particular oxidation
products can be improved by limiting the number of active
oxygens available for the reaction, thereby preventing undesired
sequential oxidation. This concept emphasizes the importance of
catalyst design to isolate the catalytically active sites from each
other, in order to limit the number of active oxygen available. In
terms of this concept, crystalline Mo3VOx catalysts (C-MoVOs)
have ideal crystal structures as oxidation catalysts.[20–21] These
materials are characterized by a structure in which {Mo6O21}6–
pentagonal units and {MO6} (M = Mo, V) octahedra are arranged
to form micropores, including hexagonal and heptagonal
channels isolated from one another (Figure 1). In selective
oxidation, the local crystal structure around the heptagonal
channel functions as a catalysis site that can adsorb and activate
Introduction
Ethylene and propylene are the highest and second-highest
produced organic chemicals in the world, respectively; they are
used as precursors for the production of polymers, resins,
surfactants, and other products and their demand increases every
year.[1–3] These olefins are conventionally produced via steam
cracking of hydrocarbons and fluid catalytic cracking (FCC) of
light diesel.[1–3] However, the recent trend for using shale gas as
a feed is accelerating the gradual replacement of the steam
cracking
process.
An
alternative
process,
oxidative
dehydrogenation (ODH), has been attracting considerable
attention; because of its exothermic nature, this process can be
operated at a low reaction temperature and without the formation
of coke, which can cause catalyst poisoning.[1–2, 4] Many efforts
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
substrates, resulting in outstanding catalytic activity and
selectivity toward the oxidation of ethane, alcohol, and
unsaturated aldehydes.[20–23]
We recently discovered a series of high-dimensionally structured
(HDS) mixed oxide catalysts with structural arrangement based
on pentagonal units and octahedra.[24–28] These materials
constitute micropore channels in a fashion similar to C-MoVOs. In
this study, we have successfully synthesized an HDS Nb-V mixed
oxide catalyst (HDS-NbVO) whose structural arrangement is
based on the {Nb6O21}12– pentagonal unit and {MO6} octahedra (M
= V, Nb). The catalytic activity of HDS-NbVO for ODHE and
ODHP was investigated, and the activities and selectivities toward
olefins were compared with those of other HDS-MVO oxides
having a similar V/M ratio (M = Nb, Mo, W).
Results and Discussion
Synthesis and characterization of HDS-NbVO catalysts
Figure 2 shows the XRD patterns of Nb-V mixed oxides
synthesized by the hydrothermal method with various preparative
V/Nb ratios. When no V was used, a material showing XRD peaks
at 2θ = 22.5°, 28.4°, 36.5°, and 45.9° was formed (Figure 2 (a)).
These peaks can be attributed to the presence of
pseudohexagonal Nb2O5 (TT-Nb2O5).[24] The addition of V led to
the formation of a material showing XRD peaks at 2θ = 22.5° and
46.1° (Figure 2 (b)). The intensity of these peaks increased with
the preparative V/Nb ratio and reached a constant value when the
preparative V/Nb ratio was 0.33 (Figure 2 (c), (d)). Further
addition of V caused a decrease in their intensity without
producing impurities. The XRD patterns observed in Figure 2 (e),
(f), and (g) are similar to those of HDS Nb oxide (HDS-NbO),
which comprises the structural arrangement of {Nb6O21}12–
pentagonal units and {NbO6} octahedra (Figure 2 (h)).[24–25] In this
material, the XRD peaks at 2θ = 22.7° and 46.4° are ascribed to
the periodic stacking of octahedra along the rod direction. Based
on the similarity of the XRD patterns, we speculate that the Nb-V
mixed oxides obtained with preparative V/Nb ratios above 0.09
have a structural environment similar to that of HDS-NbO.
Therefore, these materials are hereafter abbreviated as HDS-
xNb(12–x)V, where x represents the amount (mol) of Nb used in
the preparation.
The location site of V in HDS-NbVO was then investigated. The
elemental composition of HDS-NbVO was first determined
through ICP analysis. Figure 3 (A) shows the relationship between
the preparative and bulk V/Nb ratios. With an increase in the
preparative V/Nb ratio, the V/Nb ratio in the bulk increased
continuously. However, when the bulk V/Nb ratio reached ~0.40,
the bulk V/Nb ratio became constant despite further increase in
the preparative V/Nb ratio. The observed threshold suggests the
formation of a local crystal structure determined by a V/Nb ratio of
~0.40. Figure 3 (B) shows the d values calculated from the XRD
peak at 2θ = 22°–23° for HDS-xNb(12–x)V as a function of the
V/Nb ratio in the bulk. The d value calculated from HDS-NbO is
shown at V/Nb ratio = 0 in the same figure. The d value resulting
from the periodic stacking of octahedra increased with an
increase in the V/Nb ratio and reached a constant value when the
bulk V/Nb ratio was 0.19. The observed changes in the lattice
parameter indicated the incorporation of V into the bulk HDS-NbO
structure. The saturation of the d value when the V/Nb ratio
Figure 2. XRD patterns of Nb-V mixed oxides synthesized with
preparative V/Nb ratios of (a) 0, (b) 0.09, (c) 0.20, (d) 0.33, (e)
0.50, (f) 1.00, (g) 4.00. (h) XRD pattern of HDS-NbO.
reached 0.19 might be the result of the different distribution of V
from that up to this V/Nb ratio.
We have reported a variety of high-dimensionally structured
mixed oxides. We found that the location sites of the atoms of
these materials are largely dependent on the elemental
properties.[21, 25–29] Nb can be placed in both the pentagonal unit
sites and the octahedral sites connecting the pentagonal units
(linkers), while V can occupy only the linker sites. On the basis of
these reports, we believe that the local catalyst structure in HDS-
NbVO is mainly composed of the connections between
{Nb6O21}12– pentagonal units and {VO6} octahedral linkers. If we
assume this structural environment (Figure 3 (C)), the V/Nb ratio
can be calculated to be 0.42, which is almost the same as the
threshold value of the bulk V/Nb ratio observed through ICP
analysis.
The local catalyst structure of HDS-NbVO was investigated by
HAADF-STEM measurements. Figures 4 (A) and (B) show the
HAADF-STEM images of HDS-8Nb4V (bulk V/Nb ratio = 0.35).
HDS-8Nb4V is a rod-shaped crystal showing lattice fringes with
an interstitial distance of ~0.4 nm, which matches well with the d
value calculated from the XRD peak at 2θ = 22°–23° (Figure 4
(A)). The HAADF-STEM image measured from the [001] direction
(cross-section of the rod) displayed a complex texture derived
from the random arrangement of the pentagonal units and
octahedra (Figure
4 (B), (C)). The observed structural
arrangement is similar to those of other HDS oxides—HDS-MoVO
and HDS-WVO (Figure 4 (D), (E)). Based on the elemental
analyses and HAADF-STEM observation, we conclude that HDS-
NbVO is a class of high-dimensionally structured materials
composed of {Nb6O21}12– pentagonal units and {MO6} (M = V, Nb)
octahedra.
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
Figure 5. (A) N2 adsorption isotherms of HDS-8Nb4V. (B)
Changes of external surface areas and micropore volumes as a
function of bulk V/Nb ratio measured by N2 adsorption at liquid-N2
temperature and determined by the αs-plot method.
in microporosity is considered to be a result of the development
of a local structure made up of the pentagonal units and octahedra,
as implied by the increase in the XRD peak intensities. The
addition of excess of V might prevent this structural organization,
as implied by the decrease in their XRD peak intensities, resulting
in a decrease in microporosity. Based on the results obtained
above, we concluded that HDS-NbVO comprises the structural
arrangement of {Nb6O21}12– pentagonal units and {MO6} (M = Nb,
V) octahedra, which leads to the formation of the micropore
channels. This structural environment would facilitate the isolation
of the heptagonal channel texture, providing a catalysis site in the
crystal structure, similar to that observed in C-MoVOs. In the next
section, we evaluate the catalytic properties of HDS-NbVO for
ODHE and ODHP and compare these properties with those of
HDS-MVO (M = Mo, W) with similar V/M ratios.
Figure 3. (A) Change in bulk V/Nb ratio as a function of
preparative V/Nb ratio. (B) Change of d value calculated from
XRD peak at 2θ = 23° as a function of V/Nb ratio in bulk. (C)
Linkage of {Nb6O21}12– pentagonal units and {VO6} octahedra. The
V/Nb ratio is calculated to be 0.42 for this structural model.
Catalytic properties of HDS-MVO (M = Nb, Mo, W) catalysts
for ODHE and ODHP
The ODHE and ODHP reactions were carried out over HDS-
NbVO, and its catalytic properties were compared with those of
the other HDS oxides, HDS-MoVO and HDS-WVO, synthesized
by following previously reported procedures.[27–28] The structures
of these catalysts involved the arrangement of the pentagonal
units and octahedral linkers to form micropore channels, including
hexagonal and heptagonal channels, as directly observed
through HAADF-STEM (Figure 4). SEM-mapping images
confirmed the uniform distribution of constituent elements
throughout the solid without significant elemental segregations
(Figure S1). Table 1 lists the physicochemical properties of HDS-
MVO (M = Nb, Mo, and W). The V/M ratios determined by ICP
analysis were similar, and laid in the range of 0.35–0.39. The V/M
ratios over the catalyst surface determined by XPS were also
comparable (V/M ratios: HDS-8Nb4VO, 0.16–0.17; HDS-MoVO,
0.15–0.17; HDS-WVO, 0.18–0.20) after the catalytic reactions.
The lower V/M ratio in the surface than those in the bulk are widely
observed for the materials composed of the pentagonal unit. [29-32]
The d values calculated from the [001] XRD peak were different
depending on the constituent elements. The crystallite sizes
Figure 4. HAADF-STEM images of HDS-8Nb4V in (A) side
section and (B) cross-section of the rod. The dotted circle in (B)
represents the pentagonal unit. (C) Structural model of HDS-
8Nb4V based on the HAADF-STEM image in (B). (D) HAADF-
STEM images of HDS-MoVO. (E) HAADF-STEM image of HDS-
WVO.
Owing to the structural arrangement in HDS-NbVO, micropores,
including hexagonal and heptagonal channels, are formed in the
local crystal structure. We have reported that the heptagonal
channels in the crystal structure can behave as micropores to
adsorb small molecules with sizes of <0.40 nm.[20, 28] N2 uptake
resulting from the micropores (P/P0 < 10–5) was observed in HDS-
NbVO, as shown using HDS-8Nb4V as an example (Figure 5 (A)).
The external surface areas and micropore volumes increased
with the V/Nb ratio up to 0.34, whereas these values decreased
on further increase in the V/Nb ratio (Figure 5 (B)). The increase
Table 1. Physicochemical properties of HDS-MVO
Crystallite BET surface area External surface Micropore volume h
V/M ratio
Bulk a Surface b
Catalyst
d e /nm
size f /nm
g /m2 g-1
area h /m2 g-1
/10-3 cm3 g-1
HDS-8Nb4V
HDS-MoVO
HDS-WVO
0.35 0.17 c, 0.17 d 0.3946
0.38 0.17 c, 0.15 d 0.4000
0.39 0.20 c, 0.18 d 0.3900
92
618
641
152
10
25
110
5.7
21
9.9
1.9
2.0
a Determined by ICP, b Determined by XPS, c After ODHE, d After ODHP, e d distance derived from (001) diffraction, f
Measured by XRD and determined by Scherrer equation, g Measured by N2 adsorption and determined by BET method, h
Measured by N2 adsorption and determined by αs-method.
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
alkanes increased with increasing contact time with all catalysts.
Interestingly, the order of catalytic activity was different depending
on the substrate as follows: HDS-MoVO > HDS-WVO > HDS-
8Nb4V for ODHE and HDS-8Nb4V > HDS-MoVO > HDS-WVO
for ODHP. The selectivities of these catalysts were also different
depending on the substrate and followed the order HDS-MoVO >
HDS-WVO > HDS-8Nb4V for ODHE and HDS-8Nb4V > HDS-
MoVO
> HDS-WVO for ODHP, regardless of substrate
conversion (Figure 6 (B)). It is worth noting that the selectivity to
propylene over HDS-8Nb4V was comparable with or even
superior to that of the reported V-based catalysts (Figure. S4).
The details of the catalytic activities and product selectivities are
listed in Table 2. Although catalytic activity order followed the
same trend when activities were evaluated on a surface-area
basis (Table 2), we concluded that the reaction rates and product
selectivities over these catalysts changed in a different manner
depending on the type of alkane, as can be seen in the
significantly different ratios of their reaction rates (reaction rate in
ODHP/reaction rate in ODHE: HDS-8Nb4V, 8.3; HDS-MoVO, 1.3;
HDS-WVO, 1.5). When the C-H bond dissociation energy (BDE)
of the alkane was considered, the catalytic activity of HDS-8Nb4V
followed the trend of the C–H BDE, while the activities of HDS-
MoVO and HDS-WVO did not follow this trend (Figure 7 (A)). The
selectivities of olefins obtained at the alkane conversion of
11.6%–12.6% are shown in Figure 7 (B) together with the C–H
binding energies of the olefins. The selectivities with respect to
the olefins are understandable in the cases of HDS-MoVO and
HDS-WVO because propylene, having a low C–H BDE, is
sequentially oxidized, leading to low olefin selectivity. However,
the olefin selectivity observed in HDS-8Nb4V did not follow this
trend, unlike that in HDS-MoVO and HDS-WVO. These results
are discussed in the following section.
Figure 6. (A) Changes in the conversion of the alkanes as a
function of contact time. (B) Olefin selectivity as a function of
alkane conversion. Open symbols, ODHE; closed symbols,
ODHP. Reaction conditions: catalyst amount, 0.02–0.70 g;
reaction temperature, 440 °C; reaction gas flow rate, 50 mL min–
1
(ODHE) or 100 mL min–1 (ODHP); reaction gas composition,
alkane/O2/N2 = 10/5/85.
Figure 7. (A) Reaction rate of alkanes on a surface-area basis
shown with C-H bond dissociation energy (BDE) of alkane. (B)
Olefin selectivity at the alkane conversion of 11.6~12.6% shown
with C-H BDE of olefin. Red, HDS-8Nb4V; black, HDS-MoVO;
blue, HDS-WVO.
Discussion
We have reported that catalytic oxidation occurs over the
heptagonal channel in C-MoVOs.[7] In these materials, oligomer
units such as trimer or pentamer units are facing to the heptagonal
channels. These textures are redox-active and thus produce
active oxygen species inside the heptagonal channel.[8, 20, 21, 29, 33]
Based on the similarity of the constituent structural units and
elemental composition, it is reasonable to assume that the
heptagonal channel texture acts as a highly catalytically active
structure for ODHE and ODHP over HDS-MVO. Table 3 shows
the number of heptagonal channels in the structure of HDS-MVO
counted using HAADF-STEM images. The number of heptagonal
channels in the C-MoVOs calculated from the crystal structure
was 0.66–0.72 per nm2. The numbers of heptagonal channels in
HDS-8Nb4V, HDS-MoVO, and HDS-WVO were 0.31, 0.46, and
0.32 per nm2, respectively, fewer than those in C-MoVOs owing
to the random arrangement of pentagonal units. Two types of
heptagonal channels were observed in the crystal structure of
HDS-MVO (Figure 8). One is composed of three pentagonal units,
calculated by the Scherrer equation using the [001] XRD peak
were comparable in HDS-MoVO and HDS-WVO, whereas the
crystallites of HDS-8Nb4V were much smaller than those of the
above two catalysts. The match different crystal size was also
confirmed by SEM and HAADF-STEM images (Figure S2). The
different crystallite sizes resulted in significantly different surface
areas; the BET surface area of HDS-8Nb4V (152 m2 g–1) was
much larger than those of HDS-MoVO (12 m2 g–1) and HDS-WVO
(25 m2 g–1). These materials had micropore volumes in the order
HDS-8Nb4V > HDS-WVO ≈ HDS-MoVO. The XRD patterns did
not considerably change after the catalytic reactions, indicating
that the structure was maintained during catalysis (Figure S3).
Figure 6 (A) shows the conversion of alkanes over the HDS-MVO
catalysts as a function of contact time. The catalytic activity and
product selectivities were not significantly different by changing
the pre-treatment temperature (Table S1). The conversions of the
Table 2. Results of ODHE and ODHP over HDS-MVO
Reaction
type
Conversion /%
Selectivity /%
CO
Reaction rate
rpropane
rethane
/
Catalyst
Alkane
11.6 a
8.1 b
O2
Olefin
48.2 a
60.2 b
77.3 c
36.2 d
71.4 c
30.1 c
CO2
12.6 a
15.8 b
5.5 c
mmolalkane /g·min mmolalkane /m2·min
ODHE
ODHP
ODHE
ODHP
ODHE
ODHP
43.0 a
30.1 b
26.8 c
46.6 d
32.7 e
55.8 f
39.2 a
0.06
0.48
0.15
0.18
0.09
0.12
0.4
3.1
HDS-8Nb4V
HDS-MoVO
HDS-WVO
8.3
1.3
1.5
24.0 b
12.0 c
8.4 d
17.1 c
14.6
18.3
3.4
37.1 d
26.7 d
5.1 c
11.6 e
8.6 f
23.5 c
40.0 c
29.8 c
5.0
a W/F = 10.0 mg·min /mL, b W/F = 0.75 mg·min /mL, c W/F = 3.00 mg·min /mL, d W/F = 2.00 mg·min /mL
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
pentamer units were counted on the basis of HAADF-STEM
images, and the results are summarized in Table 3. HDS-MVO
also contained these units although the trimers or pentamers in
HDS-MVO were fewer than those in C-MoVOs because of their
disordered structure. The number of these units in HDS-8Nb4V
was less than those in HDS-WVO and HDS-MoVO. This
observation implies that the fraction of B-7MR channels in the
total number of heptagonal channels is statistically higher in HDS-
8Nb4V than those in HDS-MoVO and HDS-WVO because the
oligomer units are in proximity to the heptagonal channels. In fact,
edge-sharing pentagonal units are mainly observed in HDS-
8Nb4VO (Table 3). The formation of different types of structural
textures should be related to the growth rate of the crystals, as
can be seen from the significantly different crystal sizes in HDS-
MVO (Figure S2 and Table 1).
This proposal is supported by the results of UV–vis
measurements after ODHE and ODHP (Figure 9). UV–vis
absorption was observed in HDS-MoVO and HDS-WVO at 250–
400 nm attributable to LMCT of V5+, Mo6+, and W6+, at 400–500
nm attributable to a d-d transition of W5+, at 550–600 nm
attributable to IVCT of Mo-O-V, and at 740–780 nm attributable to
d-d transitions of V4+ and Mo5+. [34-38] In the case of HDS-8Nb4V,
in contrast, almost no absorption derived from reduced V species
was observed even after the catalytic reactions. These results
support our proposal that HDS-8Nb4V is less redox active than
HDS-MoVO and HDS-WVO because of the statistically smaller
number of A-7MR channels in the structure. This observation can
also be explained from a structural aspect. HDS oxides are
composed of a structural arrangement based on pentagonal units
and octahedra in a manner similar to that in C-MoVOs. In the case
of C-MoVOs, the oxidation state of Mo is mainly 6+ and that of V
is mainly 4+.[39–40] In light of this fact, it is reasonable to speculate
that the main oxidation states of M and V are 6+ and 4+,
respectively, in HDS-MoVO and HDS-WVO because of their
similarity to C-MoVOs in terms of crystal structure and elemental
properties. However, because Nb cannot achieve an oxidation
state of 6+, the main oxidation state of V should not be 4+ but 5+
in HDS-NbVO to obtain charge neutrality in the solid. This
structural feature is one of the reasons why the number of B-7MR
channels is higher in HDS-8Nb4V.
Figure 8. Two types of heptagonal channels (7 membered rings)
mainly observed in the HAADF-STEM images. (A) A heptagonal
channel constructed from three pentagonal units, two {VO6}
linkers, and two octahedra from oligomer units (A-7MR); oligomer
units are shown as light green. (B) A heptagonal channel
constructed from four pentagonal units (two of which are
connected to each other by edge sharing) and three {VO6} linkers
(B-7MR).
Based on the above discussion, we propose that the catalytically
active oxygen species in HDS-MoVO and HDS-WVO are M6+-O-
V4+ (M = Mo, W) in the oligomer units facing the heptagonal
channels, because it has been reported that oligomer units are
redox active during oxidation catalysis.[8, 20, 21, 29, 33] In the case of
HDS-8Nb4V, we propose that trans V5+=O in {VO6} octahedra
connecting the pentagonal units is the catalytically active oxygen
species. This is based on the following observations: (1) B-7MR
is statistically high in HDS-8Nb4V; (2) almost no UV absorption
derived from IVCT was seen even after the oxidation reactions.
Regarding the reaction rates of the alkanes (Figure 7 (A)), the
reaction rate in ODHP was clearly higher than that in ODHE over
HDS-8Nb4V, whereas these rates were comparable over HDS-
MoVO and HDS-WVO despite the different C–H binding energies
Figure 9. UV–vis spectra of HDS-8Nb4V (red), HDS-MoVO
(black) and HDS-WVO (blue) measured after ODHE (solid line) or
ODHP (dotted line).
two {VO6} linkers, and two octahedra from the oligomer units (A-
7MR), and the other comprises four pentagonal units (two
connected to each other by edge sharing) and three {VO6} linkers
(B-7MR). The heptagonal channels in C-MoVOs all belong to A-
7MR, while some fractions of the heptagonal channel in the HDS-
MVO belong to B-7MR. Moreover, A-7MR should be more redox-
active than B-7MR because the oligomer units with a redox-active
texture consist of heptagonal channels. The numbers of trimer or
Table 3. Numbers of heptagonal channel, trimer unit and pentamer unit per nm2 in C-MoVOs and HDS-MVO
Heptagonal
channel /nm2
0.72
Pentamer unit
Edge sharing Pentagonal
Catalyst
Trimer unit /nm2
Pentagonal units /nm2
/nm2
0.36
0
units /nm2
Orth-MoVO a
Tri-MoVO a
0
0.72
0.66
0.70
0.60
0.68
0
0
0.25
0.01
0.05
0.66
0.31
0.47
0.32
0.22
0.08
0.12
0.15
HDS-8Nb4V b
HDS-MoVO b
HDS-WVO b
0.04
0.17
0.05
a Estimated by using the structural models and lattice parameters. b
5Counted on the basis of HAADF-STEM images.
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
The HDS Nb-V mixed oxides were synthesized by a hydrothermal
method. First, a controlled amount of Nb2O5·2.1H2O (Nb, x mmol;
CBMM) was dispersed in 20 mL of distilled water. To the Nb-
dispersed suspension, a solution of (12-x) mmol of NH4VO3
(Wako) in 20 mL of distilled water was added. After mixing for 10
min at room temperature, the mixed suspension was transferred
of the alkanes. We have reported that ethane oxidation takes
place inside the heptagonal channel micropores of C-MoVOs,
resulting in extremely high catalytic activity.[7–8, 29] In accordance
with these reports, HDS-MoVO and HDS-WVO are considered to
catalyze ethane oxidation by the lattice oxygen of M6+-O-V4+
inside the A-7MR heptagonal channels, in a manner similar to that
with C-MoVOs, resulting in extremely high catalytic activity for
ethane oxidation. This situation leads to comparable reaction
rates in ODHE and ODHP despite the different C–H binding
energies of the alkanes, because propane has some difficulty in
entering the heptagonal channel for oxidation.[41–42] In the case of
HDS-8Nb4V, although ethane might be allowed to enter the
heptagonal channel as easily as in the above-mentioned catalysts,
ethane would not be converted inside the channel, because the
heptagonal channel in HDS-8Nb4V rarely participates in the
redox reaction during ethane oxidation, because M6+-O-V4+ is
unlikely to be formed inside the channel. Therefore, the reaction
rate of alkanes over HDS-8Nb4V is followed in BDE of the alkane.
The different selectivity to olefins depending on the alkane can
also be explained on the basis of this proposal because different
catalytically active oxygen species should abstract C–H from
alkane in different ways, to afford different reaction intermediates
for the formation of olefins.[43–45] The reactivity of these active
oxygen species for the abstraction of olefin C–H and allyl C–H (for
propylene) should also be considered as a reason. These
differences are likely to result in different selectivities for olefins,
the details of which should be investigated in the future.
to an autoclave with
a 50-mL Teflon inner vessel. The
hydrothermal reaction was then performed at 175 °C for 72 h with
rotation at 1 rpm. The yellow solid formed was washed with 500
mL of distilled water and dried in air at 80 °C overnight. The solid
product was denoted as HDS-xNb(12–x)V. Prior to use in the
catalytic reactions, the catalyst was calcined at 400 °C for 2 h in
air.
Synthesis of HDS-NbO
HDS Nb oxide was prepared by a hydrothermal method, following
the reported procedure[24–25], using an aqueous solution of
ammonium
niobium
oxalate
(ANO,
NH4(NbO(C2O4)2(H2O)2·nH2O; Signa-Aldrich) as the Nb source.
First, 2.551 g of ANO (Nb, 6 mmol) was dissolved in 40 mL of
distilled water. The resulting solution was transferred to an
autoclave with a 50-mL Teflon inner vessel. The hydrothermal
reaction was performed at 175 °C for 72 h under static conditions
in an electric oven. The white solid formed was washed with 500
mL of distilled water and dried in air at 80 °C overnight. The solid
product is denoted as HDS-NbO.
Synthesis of HDS-MoVO
HDS Mo-V mixed oxide was prepared by a hydrothermal method
following the reported procedure.[28] First, 17.66
g
of
(NH4)6Mo7O24·4H2O (Mo, 100 mmol; Wako) was dissolved in 120
mL of distilled water. Separately, an aqueous solution of VOSO4
was prepared by dissolving 6.576 g of hydrated VOSO4 (V, 25.0
mmol; Mitsuwa Chemicals) in 120 mL of distilled water. The two
solutions were mixed at room temperature and stirred for 10 min
before being transferred to an autoclave with a 300-mL Teflon
inner vessel. The hydrothermal reaction was performed at 175 °C
for 48 h under static conditions in an electric oven. The dark
purple solid formed was washed with 1000 mL of distilled water
and dried in air at 80 °C overnight. The solid obtained was
denoted HDS-MoVO. Prior to the catalytic reactions, the catalyst
was calcined at 400 °C for 2 h under a flow of 50 mL min-1 of N2.
The V/Mo ratio calculated from ICP analysis was 0.38.
Conclusion
In this study, we synthesized high-dimensionally structured Nb-V
mixed oxide catalysts (HDS-NbVO) consisting of {Nb6O21}12–
pentagonal units and {MO6} (M = Nb, V) octahedra. The oxidative
dehydrogenation
of
ethane
(ODHE)
and
oxidative
dehydrogenation of propane (ODHP) was carried out over HDS-
MVO (M = Nb, Mo, W), all of which have structures based on the
pentagonal units and octahedra. Although these catalysts have
similar constituent structural units and elemental composition,
they showed significantly different catalytic activity and olefin
selectivity depending on the alkane used. Considering the crystal
structure of the HDS oxides and the results of UV–vis
spectroscopy, we concluded that the heptagonal channel texture
as the catalytically active site is constituted differently among
HDS-MVO. This difference generated different types of
catalytically active oxygen species for oxidation, resulting in
significantly different catalytic behavior for the oxidative
dehydrogenation of alkanes. The work presented here not only
provides a new class of HDS oxides active for alkane oxidation,
HDS-NbVO, but also demonstrates that catalytically active
oxygen species can be controlled by tuning the constituent
elements.
Synthesis of HDS-WVO
HDS W-V mixed oxide was prepared by a hydrothermal method
following the reported procedure.[27] First, 2.681
g
of
(NH4)6[H2W12O40]·nH2O (W, 10.5 mmol; Nippon Inorganic Colour
and Chemical Co., Ltd.) was dissolved in 20 mL of distilled water.
Separately, an aqueous solution of VOSO4 was prepared by
dissolving 1.184 g of hydrated VOSO4 (V, 4.5 mmol; Mitsuwa
Chemicals) in 20 mL of distilled water. The two solutions were
mixed at room temperature and 0.09 g of oxalic acid (1 mmol;
Wako) was added, followed by stirring at room temperature for 10
min. The resulting solution was transferred to an autoclave with a
50-mL Teflon inner vessel, and the hydrothermal reaction was
performed at 175 °C for 20 h with rotation at 1 rpm. The black
solid formed was washed with 500 mL of distilled water and dried
in air at 80 °C overnight. The solid was denoted as HDS-WVO.
Prior to the catalytic reactions, the catalyst was calcined at 400 °C
for 2 h under a flow of 50 mL min-1 of N2. The V/W ratio calculated
by ICP analysis was 0.39.
Experimental Section
Catalyst preparation
Synthesis of HDS-NbVO
Characterization of materials
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
Keywords: Hydrothermal synthesis
structured materials Nb-V mixed oxide, Oxidative
dehydrogenation of propane • Local catalyst structure
•
High-dimensionally
The materials obtained were characterized as follows. Powder
XRD patterns were recorded with a diffractometer (RINT Ultima+,
Rigaku) using Cu-Kα radiation (tube voltage, 40 kV; tube current,
20 mA). The elemental composition of the bulk was determined
using inductively coupled plasma atomic emission spectroscopy
(ICP-AES; ICPE-9820, Shimadzu). Aberration-corrected STEM
images were obtained using an ARM-200F system (JEOL Ltd.,
Japan) equipped with a cold-field emission gun at an acceleration
voltage of 200 kV. N2 adsorption isotherms at liquid-N2
temperature were obtained using an auto-adsorption system
(BELSORP MAX, MICROTRAC MRB). Prior to N2 adsorption, the
samples were calcined at 400 °C for 2 h in a muffle oven and were
further pre-treated under vacuum at 300 °C for 2 h in the
adsorption apparatus. X-ray photoelectron spectroscopy (XPS;
JPC-9010MC, JEOL) with nonmonochromatic Mg Kα radiation
was used to measure the binding energies. Powder UV
•
[1]
J. T. Grant, J. M. Venegas, W. P. McDermott, I. Hermans, Chem. Rev.
2018, 118, 2769-2815.
[2]
[3]
F. Cavani, N. Ballarini, A. Cericola, Catal. Today 2007, 127, 113-131.
J. J. H. B. Sattler, J. Ruiz-Martinez, E. Santillan-Jimenez, B. M.
Weckhuysen, Chem. Rev. 2014, 114, 10613-10653.
C. A. Carrero, R. Schlöegl, I. E. Wachs, R. Schomaecker, ACS Catal.
2014, 4, 3357-3380.
[4]
[5]
[6]
[7]
[8]
C. A. Gärtner, A. C. van Veen, J. A. Lercher, ChemCatChem 2013, 5,
3196-3217.
P. Botella, J. M. López Nieto, A. Dejoz, M. I. Vázquez, A. Martınez-Arias,
́
Catal. Today 2003, 78, 507-512.
S. Ishikawa, X. Yi, T. Murayama, W. Ueda, Appl. Catal. A: Gen. 2014,
474, 10-17.
measurements
were
carried
out
using
a
UV–vis
S. Ishikawa, D. Kobayashi, T. Konya, S. Ohmura, T. Murayama, N.
Yasuda, M. Sadakane, W. Ueda, J. Phys. Chem. C 2015, 119, 7195-
7206.
spectrophotometer (V-550, JASCO). The solid used for
measurement was prepared by mixing 2 mg of the sample with
50 mg of BaSO4 (99.998%, Aldrich).
[9]
D. Melzer, G. Mestl, K. Wanninger, Y. Zhu, N. D. Browning, M. Sanchez-
Sanchez, J. A. Lercher, Nat. Commun. 2019, 10, 4012.
Oxidative dehydrogenation of ethane and propane
[10] D. A. Svintsitskiy, T. Y. Kardash, E. V. Lazareva, A. A. Saraev, E. A.
Derevyannikova, M. Vorokhta, B. Šmíd, V. M. Bondareva, Appl. Catal. A:
Gen. 2019, 579, 141-150.
The oxidative dehydrogenation of ethane or propane was carried
out using a conventional vertical flow system with a Pyrex tubular
reactor at ambient pressure. A 0.02–0.70 g portion of the catalyst
was diluted with 4–5 g of SiC and placed in the tubular reactor.
The reactor was heated at a rate of 10 °C min–1 to 440 °C under
a flow of the reaction gas (C2H6/O2/N2 = 5/2.5/42.5 mL min–1 for
ODHE, C3H8/O2/N2 = 10/5/85 mL min–1 for ODHP). After the
temperature reached 440 °C, as measured by a thermocouple
inserted in the catalyst zone, the catalysts were maintained at that
temperature for 30 min to stabilize the state of the catalyst before
analyses of the reactant and product gases. Reactants and
products were analyzed using three online gas chromatographs.
O2, N2, and CO were detected by a TCD detector using molecular
sieve 5A. CO2, C2H6, C2H4, C3H8, C3H6, and H2O were analyzed
using a TCD detector with Gascropack54. Acetic acid, acrolein,
acetone, and acrylic acid were analyzed using an FID detector
with Porapack Q; however, their formation was negligible in the
present study. Blank runs showed that no reaction took place
without the catalysts under the experimental conditions for both
ODHE and ODHP. The carbon balance was always between 98%
and 100% in all the reactions tested.
[11] E. Heracleous, A. A. Lemonidou, J. Catal. 2006, 237, 162-174.
[12] E. Heracleous, A. A. Lemonidou, J. Catal. 2006, 237, 175-189.
[13] Z. Skoufa, E. Heracleous, A. A. Lemonidou, J. Catal. 2015, 322, 118-129.
[14] M. C. Kung, H. H. Kung, J. Catal. 1992, 134, 668-677.
[15] M. Machli, E. Heracleous, A. A. Lemonidou, Appl. Catal. A: Gen. 2002,
236, 23-34.
[16] K. Chen, A. T. Bell, E. Iglesia, J. Phys. Chem. B 2000, 104, 1292-1299.
[17] C. Carrero, M. Kauer, A. Dinse, T. Wolfram, N. Hamilton, A. Trunschke,
R. Schlögl, R. Schomäcker, Catal. Sci. Technol. 2014, 4, 786-794.
[18] J. T. Grant, C. A. Carrero, F. Goeltl, J. Venegas, P. Mueller, S. P. Burt,
S. E. Specht, W. P. McDermott, A. Chieregato, I. Hermans, Science 2016,
354, 1570.
[19] R. K. Grasselli, Catal. Today 2005, 99, 23-31.
[20] S. Ishikawa, W. Ueda, Catal. Sci. Technol. 2016, 6, 617-629.
[21] S. Ishikawa, Z. Zhang, W. Ueda, ACS Catal. 2018, 8, 2935-2943.
[22] S. Ishikawa, T. Murayama, B. Katryniok, F. Dumeignil, M. Araque, S.
Heyte, S. Paul, Y. Yamada, M. Iwazaki, N. Noda, W. Ueda, Appl. Catal.
A: Gen. 2019, 584, 117151.
[23] S. Ishikawa, N. Noda, M. Wada, S. Tsurumi, W. Ueda, ACS Catal. 2020,
10, 10535-10545.
[24] T. Murayama, J. Chen, J. Hirata, K. Matsumoto, W. Ueda, Catal. Sci.
Technol. 2014, 4, 4250-4257.
[25] S. Ishikawa, M. Shinoda, Y. Motoki, S. Tsurumi, M. Kimura, N. Hiyoshi,
A. Yoshida, W. Ueda, Inorg. Chem. 2020, 59, 9086-9094.
[26] K. Shimoda, S. Ishikawa, M. Tashiro, M. Kumaki, N. Hiyoshi, W. Ueda,
Inorg. Chem. 2020, 59, 5252-5255.
Acknowledgements
[27] T. Murayama, S. Ishikawa, N. Hiyoshi, Y. Goto, Z. Zhang, T. Toyao, K.-
I. Shimizu, S. Lee, W. Ueda, Catal. Today 2021, 363, 60-66.
[28] T. Konya, T. Katou, T. Murayama, S. Ishikawa, M. Sadakane, D. Buttrey,
W. Ueda, Catal. Sci. Technol. 2013, 3, 380-387.
1.
This work was supported by JSPS KAKENHI (Grant
Number 18K14058).
[29] S. Ishikawa, T. Murayama, M. Kumaki, M. Tashiro, Z. Zhang, A. Yoshida,
W. Ueda, Top. Catal. 2016, 59, 1477-1488.
2. This work was supported by the JSPS core-to-core program.
[30] A. C. Sanfiz, T. W. Hansen, D. Teschner, P. Schnörch, F. Girgsdies, A.
Trunschke, R. Schlögl, M. H. Looi, S. B. A. Hamid, J. Phys. Chem. C
2010, 114, 1912-1921.
3.
This work was conducted with the help of the
"Nanotechnology Platform Program" of the Ministry of Education,
Culture, Sports, Science and Technology (MEXT) at Hokkaido
University, Japan, Grant Number JPMXP09A20HK0035.
[31] M. Hävecker, S. Wrabetz, J. Kröhnert, L.-I. Csepei, R. Naumann
d’Alnoncourt, Y. V. Kolen’ko, F. Girgsdies, R. Schlögl, A. Trunschke, J
Catal. 2012, 285, 48-60.
[32] L. Masliuk, F.-P. Schmidt, W. Hetaba, M. Plodinec, G. Auffermann, K.
Hermann, D. Teschner, F. Girgsdies, A. Trunschke, R. Schlögl, T.
Lunkenbein, J.Phys. Chem. C 2020, 124, 23069-23077.
[33] S. Inukai, S. Ishikawa, T. Tanabe, Y. Jing, T. Toyao, K.-I. Shimizu, W.
Ueda, Chem. Mater. 2020, 32, 1506-1516.
4.
This study was supported by the Cooperative Research
Program of the Institute for Catalysis, Hokkaido University (Grant
18A1002).
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
[34] J. Melsheimer, J. Kröhnert, R. Ahmad, S. Klokishner, F. C. Jentoft, G.
Mestl, R. Schlögl, PhysChemChemPhys. 2002, 4, 2398-2408.
[35] L. Nakka, J. E. Molinari, I. E. Wachs, J. Am. Chem. Soc. 2009, 131,
15544-15554.
[36] D. Nitsche, C. Hess, J. Phys. Chem. C 2016, 120, 1025-1037.
[37] E. I. Ross-Medgaarden, W. V. Knowles, T. Kim, M. S. Wong, W. Zhou,
C. J. Kiely, I. E. Wachs, J. Catal. 2008, 256, 108-125.
[38] J. Albert, D. Lüders, A. Bösmann, D. M. Guldi, P. Wasserscheid, Green
Chem. 2014, 16, 226-237.
[39] H. Murayama, D. Vitry, W. Ueda, G. Fuchs, M. Anne, J. L. Dubois, Appl.
Catal. A: Gen. 2007, 318, 137-142.
[40] T. Lunkenbein, F. Girgsdies, A. Wernbacher, J. Noack, G. Auffermann,
A. Yasuhara, A. Klein-Hoffmann, W. Ueda, M. Eichelbaum, A. Trunschke,
R. Schlogl, M. G. Willinger, Angew. Chem. Int. Ed. 2015, 54, 6828-6831.
[41] S. Ishikawa, X. Yi, T. Murayama, W. Ueda, Catal. Today 2014, 238, 35-
40.
[42] Y. Liu, A. Twombly, Y. Dang, A. Mirich, S. L. Suib, P. Deshlahra,
ChemCatChem, 2020, 13, 882-899.
[43] A. Bielański, J. Haber, Catal. Rev. 1979, 19, 1-41.
[44] J. Haber, E. M. Serwicka, React. Kinet. Catal. Lett. 1987, 35, 369-379.
[45] J. Haber, Catal. Today 2009, 142, 100-113.
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.202100463
ChemCatChem
FULL PAPER
Entry for the Table of Contents
High dimensionally structured Nb-V mixed oxide catalyst was synthesized hydrothermally and the catalytic activity to oxidative
dehydrogenation of ethane and propane was carried out and the catalytic futures were compared with those of high dimensionally
structured Mo-V and W-V mixed oxide catalyst. Depending on the constituent element, the catalytic selectivity trend was much
different due to the different type of active oxygen species produced.
9
This article is protected by copyright. All rights reserved.