Mo–V–O nanocrystals synthesized in the confined space of a mesoporous carbon

Article abstract of DOI:10.1016/j.apcata.2021.118294

Ternary Mo–V oxide nanocrystals (Nano-MoVO) were hydrothermally synthesized in the confined space of a mesoporous carbon template and tested in the oxidative dehydrogenation (ODH) of ethane and propane. The synthesized nanocrystals are approximately 60 nm in length, 20 nm in diameter on average, and possess a structure resembling orthorhombic MoVO (Orth-MoVO) as indicated by spectroscopic and microscopy characterization. The Nano-MoVO catalyst has a 5-fold higher mesopore volume and a 4-fold larger external surface area than an Orth-MoVO synthesized by a conventional method (Orth-MoVO) as characterized through N₂ adsorption analysis. Nano-MoVO shows similar activation energy in the ODH of ethane compared with other conventional MoVO catalysts. However, Nano-MoVO exhibits significantly higher propane/ethane activation rate ratio and higher propene selectivity even in the absence of elements such as Te and Nb that suppress overoxidation of propane-derived species to CO_x. The results suggest the benefits of the nanocrystalline morphology to limit overoxidation.

Full text of DOI:10.1016/j.apcata.2021.118294

Applied Catalysis A, General 624 (2021) 118294
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Mo–V–O nanocrystals synthesized in the confined space of a
mesoporous carbon
,
,
Ryo Obunaia, Keisuke Tamuraa, Isao Oginoa *, Shin R. Mukaia, Wataru Uedab *
a Division of Applied Chemistry, Faculty of Engineering, Hokkaido University, N13W8, Kita-ku, Sapporo, Hokkaido, 060-8628, Japan
b Department of Material and Life Chemistry, Faculty of Engineering, Kanagawa University, Rokkakubashi, Kanagawa-ku, Yokohama, Kanagawa, 221-8686, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
Ternary Mo–V oxide nanocrystals (Nano-MoVO) were hydrothermally synthesized in the confined space of a
mesoporous carbon template and tested in the oxidative dehydrogenation (ODH) of ethane and propane. The
synthesized nanocrystals are approximately 60 nm in length, 20 nm in diameter on average, and possess a
structure resembling orthorhombic MoVO (Orth-MoVO) as indicated by spectroscopic and microscopy charac-
terization. The Nano-MoVO catalyst has a 5-fold higher mesopore volume and a 4-fold larger external surface
area than an Orth-MoVO synthesized by a conventional method (Orth-MoVO) as characterized through N2
adsorption analysis. Nano-MoVO shows similar activation energy in the ODH of ethane compared with other
conventional MoVO catalysts. However, Nano-MoVO exhibits significantly higher propane/ethane activation
rate ratio and higher propene selectivity even in the absence of elements such as Te and Nb that suppress
overoxidation of propane-derived species to COx. The results suggest the benefits of the nanocrystalline
morphology to limit overoxidation.
Alkane oxidation
Mo–V mixed oxide
Mesoporous carbon
Nano crystal
Confined space synthesis
1. Introduction
selectivity in the conversion of ethane to ethene, it remains challenging
to achieve selective conversion of propane to propene on them because
–
of overoxidation to COx. The product propene contains allylic C H
Oxidative dehydrogenation (ODH) of short-chain alkanes to olefins
[1–12], ethane into ethene, and propane into propene as examples,
provides prospective advantages in terms of energy consumption over
–
bonds whose reactivity is much higher than vinylic C H bonds,
–
allowing oxygen-insertion through facile C H activation. This reac-
– –
the existing technologies. Ternary Mo
V
O mixed oxides catalyze
tivity of C3H8 is beneficial to synthesize acrolein and acrylic acid
oxidative dehydrogenation (ODH) of ethane to ethene [11,12]. In
selectively, as demonstrated in the works using multicomponent
– –
V
– – – –
V
– –
lacks elements like Te and Nb necessary to limit overoxidation to COx.
particular, a ternary Mo
O catalyst possessing an orthorhombic
Mo
Te Nb O catalysts [13–19]. However, ternary Mo V O
structure (Orth-MoVO), known as the M1 phase, has been reported to
exhibit the highest activity (11.1 × 10–3 molC2H4 gcat–1 h–1) among the
Consequently, ternary Mo
– –
V O catalysts exhibit low selectivity to
– –
V
family of Mo
O complex oxide catalysts [2,11]. It also exhibits high
propene as the conversion is increased: ≈37 % selectivity to propene at a
ethene selectivity (82 %) at an ethane conversion of 52 % at 335 ◦C [11].
Previous works have suggested that the active sites are located at dis-
propane conversion of ≈5%. at 380 ◦C [15], ≈27 % selectivity at a
◦
conversion of ≈11 % at 400 C [16], and 19.9 % selectivity at a con-
◦
– –
O
– –
V O catalysts form
torted octahedra of V and Mo (V
Mo pairs) in empty heptagonal
version of 10.8 % at 346 C [20]. Ternary Mo
channels (0.39 nm) [11], which allow facile diffusion of ethane under
reaction conditions [12]. Beneficial roles of tight confinement of C2H6
and C2H4 in small heptagonal channels has been reported in a work on
acrylic acid in addition to COx by oxidation of propene, but the selec-
tivity to the valuable product is also low presumably because of the
limited ability of
intermediate.
α-H abstraction from propene to form a π-allylic
– – – –
V Te Nb O catalysts, by showing that it leads
multicomponent Mo
to enhanced activity through stabilized CH activation transition states
and limited undesirable oxygen-insertion in ethene [10]. Although
The substantial difference between selective ethane conversion and
unselective propane conversion on ternary MoVO catalysts may be
attributed primarily to the higher reactivity of allylic hydrogen atoms
– –
V O catalysts possessing heptagonal channels enable high
ternary Mo
* Corresponding authors.
E-mail addresses: iogino@eng.hokudai.ac.jp (I. Ogino), uedaw@kanagawa-u.ac.jp (W. Ueda).
https://doi.org/10.1016/j.apcata.2021.118294
Received 17 May 2021; Received in revised form 5 July 2021; Accepted 22 July 2021
Available online 24 July 2021
0926-860X/© 2021 Elsevier B.V. All rights reserved.

R. Obunai et al.
Applied Catalysis A, General 624 (2021) 118294
Scheme 1. Schematic drawing for the confined space synthesis of Mo–V–O nanocrystals (Nano-MoVO) in a mesoporous carbon derived from a resorcinol-
formaldehyde resin. Nano-MoVO/C represents the sample after the hydrothermal synthesis of Mo–V–O nanocrystals in the mesoporous carbon.
– –
V
relative to vinylic hydrogen atoms and the tight guest–host fit for C2H4
to limit oxygen-insertion. However, facile desorption of product olefins
from the catalyst surface may also play an important role to limit
overoxidation. C2H4 exhibits substantially weaker interactions with the
crucial for the successful synthesis of Mo
O nanocrystals because
the facile removal of carbon materials without altering the crystal
– –
structure of Mo
V O nanocrystals is a prerequisite.
The synthesized catalyst (Nano-MoVO) consists of nanocrystals
approximately 60 nm in length and 20 nm in diameter. It possesses an
approximately 5-times larger mesopore volume than conventional Orth-
MoVO. We show that the nanocrystalline morphology of Nano-MoVO
resulted in an enhanced ODH rate of propane relative to that of
ethane and higher selectivity to propene selectivity than other ternary
– – – –
V Te Nb O catalyst than C3H6 as
surface of a multicomponent Mo
indicated by in-situ Fourier transform infrared (FT-IR) spectroscopy, and
the weak interaction of C2H4 has been suggested to be responsible for the
high selectivity to C2H4 during ODH of C2H6 by this catalyst [21]. Low
selectivity from propane to acrylic acid in a zeolite-supported vanadyl
– – – –
– –
V O catalysts tested as controls.
catalyst as compared with that in a Mo
V
Te Nb O catalyst has
Mo
been attributed to factors including a long diffusion time in the micro-
pores of the zeolite-based catalyst [19]. The addition of small amounts of
additives such as K+ to suppress acid sites on VOx/Al2O3 has been re-
ported to weaken interactions between the surface and C3H6 and in-
crease the selectivity to propene during ODH of propane [22,23]. As
2. Experimental methods
2.1. Materials
– – –
propene formed by dehydrogenation of propane on Mo V–Te Nb
O
Ammonium molybdate tetrahydrate ((NH4)6Mo7O24⋅4H2O, >99.0
%) and oxalic acid dihydrate (>99.5 %) were purchased from FUJIFILM
Wako Pure Chemical Corporation. Vanadium(IV) oxide sulfate n-hy-
drate (VOSO4⋅nH2O, n = 5.21) was purchased from Mitsuwa chemicals
Co., Ltd. Deionized and distilled (DDI) water used in all experiments was
obtained from an EYELA STILL ACE SA-2100E1 system.
catalysts is released to the gas phase and travels through a catalyst bed
whose height is at least on a millimeter scale [16], promoting desorption
of propene to the bulk gas phase from the catalyst surface was antici-
pated to limit the consecutive oxidation, leading to a higher propene
– –
V O catalysts. These reports have
selectivity even on the ternary Mo
motivated us to synthesize and test nanocrystalline and ternary
– –
V O catalysts possessing mesoporosity using mesoporous carbon
Mo
2.2. Synthesis of Orth- and Amor-MoVO catalysts
as the template (Scheme 1).
– –
V O
This work aimed to synthesize nanocrystalline ternary Mo
Orthorhombic MoVO, denoted as Orth-MoVO, was synthesized by a
hydrothermal synthesis method [11]. Typically, 18.4 mL of an aqueous
solution containing (NH4)6Mo7O24⋅4H2O and VOSO4⋅5.21H2O with a
Mo concentration of 0.275 mol Lꢀ 1 and a Mo/V atomic ratio of 4 was
charged to a 23 mL Teflon-lined autoclave. After purging it with a flow
of nitrogen at a rate of 50 mL minꢀ 1 for 10 min, it was sealed and placed
in a static oven at 448 K. After 24 h of heating, the autoclave was
allowed to cool to room temperature, and the solids were collected by
filtration, washed with 50 mL of DDI water, and dried in an oven at 333
K overnight. Purification of the solids was conducted by stirring 1.0 g of
catalysts and to investigate the effects of the nanocrystalline
morphology on catalytic performance in ODH reactions of propane and
– –
V O catalysts are generally obtained as μm-sized
ethane. Ternary Mo
rod crystals with heptagonal channels running through their longest
axis. Previous attempts to retard the crystal growth by surfactants are
limited to the synthesis of sub-μm-sized rods [12]. Although carbon
templates offer advantages of inertness and easy removal by oxidative
thermal treatments [24–33], there have been no precedents for syn-
– –
V O nanocrystals using them. The latter feature is
thesizing Mo
2

R. Obunai et al.
Applied Catalysis A, General 624 (2021) 118294
the dried material in 25 mL of an aqueous solution of 0.4 mol Lꢀ 1 oxalic
acid at 333 K for 30 min, followed by washing with 500 mL of DDI water
and drying at 333 K overnight. A portion of the crystalline solids was
calcined by heating in 100 mL min–1 flowing nitrogen to 673 K at a rate
of 10 K min–1 followed by temperature holding at 673 K for 2 h.
MoVO materials having the same crystalline structure as Orth-MoVO
along the c-direction but lacking long-range order in other directions are
denoted as Amor-MoVO. Synthesis of Amor-MoVO was performed using
a method similar to that used for Orth-MoVO synthesis. The concen-
trations of Mo and V complexes were doubled and the purification step
using oxalic acid was omitted.
2.3. Synthesis of Nano-MoVO catalysts
Synthesis of MoVO nanoparticles in a mesoporous carbon was per-
formed by an impregnation method. Synthesis of the mesoporous carbon
was performed by pyrolyzing resorcinol-formaldehyde resin at 1273 K.
A detailed procedure has been reported in the literature [34]. A meso-
porous carbon (0.3 g) that had been dried in vacuum at 403 K for 2 h was
impregnated with 1.0 mL of 0.1 mol Lꢀ 1 (NH4)6Mo7O24 solution, fol-
lowed by drying in air at room temperature for 24 h. Then, the meso-
porous carbon particles were impregnated with 1.0 mL of 0.17 mol Lꢀ 1
VOSO4 solution and subsequently dried in air at room temperature for a
few hours until the water content became 0.22 g/g-carbon. The resultant
solids were transferred to a 23-mL Teflon-lined autoclave and purged
under a flow of nitrogen at 50 mL min–1 for 5 min. The autoclave was
sealed and heated at 448 K under static conditions. After 24 h of syn-
thesis, the solids were washed with 50 mL DDI water and dried at 333 K
overnight. The product solids were stirred in hot water at a concentra-
tion of 4.0 g dm–3 at 353 K for 0.5 h to remove impurities. The solids
were separated by filtration, washed with 50 mL DDI water, and dried at
333 K overnight. The resultant solids were denoted as Nano-MoVO/C.
Nano-MoVO/C was calcined by a sequence of the following steps:
heating in 100 mL minꢀ 1 flowing nitrogen to 673 K at a rate of 10 K
minꢀ 1; cooling to 298 K and switching nitrogen to air flowing at a rate of
100 mL minꢀ 1; heating to 603 K at a rate of 10 K minꢀ 1; holding tem-
perature at 603 K for 8 h; ramping temperature to 633 K at a rate of 1 K
minꢀ 1; holding temperature at 633 K for 3 h. The calcined material is
denoted as Nano-MoVO.
Fig. 1. PXRD patterns characterizing Orth-MoVO (a), Amor-MoVO (b), Nano-
MoVO/C (c), and Nano-MoVO (d).
(300 mg) was diluted with SiO2 (2.40 g, 50–80 mesh, MIYAZAKI
chemical. Co. Ltd.) and held between two end plugs of quartz wool in the
center of the reactor. The reactor was placed inside a temperature-
controlled furnace, and the temperature was measured by a K-type
thermocouple inserted into the concentric thermowell in the middle of
the catalyst zone. A feed consisting of a composition of 10.0 mol% C2H6
10.0 mol% O2 80.0 mol% N2 was passed through the catalyst bed at
30–90 mL(STP) min–1. Liquid products were collected in a cold trap
(ice/water), and gas-phase products were collected in a Tedlar bag over
a 90 min period. Liquid products were analyzed by a Shimadzu GC-17A
equipped with a WAX-HT capillary column and a flame ionization de-
tector. Gas-phase C2H6, C2H4, CO, CO2, and N2 were analyzed by a
combination of a Shimadzu GC14B equipped with a Gaskuropak54
column and a thermal conductivity detector (TCD) and a Shimadzu
GC8A equipped with an MS-5A column and a TCD. Selectivity was
calculated on a carbon basis.
2.4. Characterization of catalysts
Powder X-ray diffraction (XRD) patterns were recorded with a
diffractometer (RINT Ultima IV, Rigaku) using Cu ꢀ K
α radiation
(voltage, 40 kV; current, 20 mA). For XRD measurements, samples were
ground for 1 min and were loaded on a horizontal sample holder. Dif-
2.6. Catalytic oxidation of propane
fractions were recorded in the range of 5 ꢀ 60 at a 10 min–1 scan
speed. SEM images were taken using an electron microscope (JSM-6500
F, JEOL). HR-TEM images were taken using field emission transmission
electron microscopy (JEM-2010 F, JEOL). Fourier-transform infrared
(FT-IR) analysis was carried out using a spectrometer (FT/IR-6800,
JASCO) with an MCT detector. IR spectra were obtained by integration
of more than 256 scans with a resolution of 4 cmꢀ 1. Raman spectra
(inVia Reflex Raman spectrometer, RENISHAW) of static samples were
taken in air in the range of 0–1400 cm–1 using an Ar laser. Elemental
compositions of the bulk were determined by ICP-AES (ICPE-9000,
Shimadzu). N2 adsorption isotherms of the samples both before and after
catalytic tests were measured at liquid N2 temperature by using an auto-
adsorption system (BELSORP-max, MicrotracBEL). The samples were
heat-treated in air at 573 K for 2 h before catalytic tests. Prior to N2
adsorption, the catalysts were evacuated under vacuum at 573 K for 2 h.
Catalytic oxidation of propane was carried out using the same setup
that was used for ethane oxidation. Reactions were conducted at
558–618 K and atmospheric pressure. A feed consisting of a composition
of 7.5 mol% C3H8 10.0 mol% O2 37.0 mol% N2 45.5 mol% H2O was
passed through the catalyst bed at different total flow rates ranging from
16 to 120 mL(STP) min–1 to operate the reactor at conversions <10 %.
Liquid products were collected in a cold trap (ice/water), and gas-phase
products were collected in a Tedlar bag over a 90 min period. Products
were analyzed by the same GCs. Selectivity was calculated on a carbon
basis.
◦
◦
3. Results and discussion
3.1. Characterization of catalysts
– –
V O particles formed in the mesoporous
3.1.1. Crystal phase of Mo
2.5. Catalytic oxidation of ethane
carbon
PXRD was used to confirm the synthesis of Orth-MoVO and Amor-
MoVO and to identify the crystal phase of Nano-MoVO formed in the
carbon template. The data characterizing Orth-MoVO (Fig. 1a) show
Catalytic oxidation of ethane was carried out at 593–643 K and at-
mospheric pressure in a once-through flow reactor. An MoVO catalyst
3

R. Obunai et al.
Applied Catalysis A, General 624 (2021) 118294
Fig. 2. FE-SEM images charactering Orth-MoVO (a) and Nano-MoVO (b), and HR-TEM image characterizing Nano-MoVO (c). The inset in panel (c) represents the
power spectrum. (d) Schematic drawing of the structure of Orth-MoVO.
peaks at 2θ = 6.6, 7.9, 9.0, 22, and 45 ◦. These peaks are typically
observed for orthorhombic MoVO [11]. The first three peaks correspond
to 020, 120, and 210 reflections while the peaks at 22 ◦ (d = 0.40 nm)
hydrothermal synthesis and the weight of the sample increased by 10 wt
% (Fig. S2 in the Supplementary material). Assuming that all the metal
complexes of Mo and V used in the synthesis had been converted into
◦
– –
Mo V O crystals in the carbon, a decrease of mesopore volume by
and 45 (d = 0.20 nm) correspond to 001 and 002 reflections. In
contrast, the PXRD pattern characterizing Amor-MoVO (Fig. 1b) shows
001 and 002 reflections with a single broad peak at 2θ <10 ◦, confirming
that it possesses the same repeating units along the c-direction as
Orth-MoVO but lacks a long-range order along the a–b plane. The PXRD
data characterizing Nano-MoVO/C and Nano-MoVO (Fig. 1c and 1d)
both show a small peak at 22 ◦, indicating that Nano-MoVO consists of
regularly repeating units with a d spacing of 0.40 nm like Orth-MoVO
and Amor-MoVO.
0.03 cm3 (g-carbon)–1 is expected, which reasonably matches the
decrease in the observed mesopore volume.
Removal of the mesoporous carbon from Nano-MoVO/C requires
careful temperature control during calcination to prevent the structural
change of MoVO by exothermic processes. We found that a multi-step
calcination procedure described in the Experimental section removes
the mesoporous carbon nearly completely (less than 0.3 wt% carbon
remains as determined by thermogravimetric analysis) while retaining
– –
V O intact. The removal of the mesoporous car-
the structure of Mo
– –
3.1.2. Evidence of the formation of Mo
mesopores by FE-SEM and N2 adsorption
Both Orth- and Amor-MoVO were obtained as rods a few
V
O nanocrystals inside
bon scaffold yielded aggregates of nanoparticles 58 nm in length and 23
nm in diameter on average as characterized by FE-SEM (Fig. 2b). The
PXRD pattern characterizing the resultant Nano-MoVO (Fig. 1d) shows
μm in
◦
length as shown by FE-SEM characterization (Fig. 2, panel (a) and
Fig. S1 in the Supplementary material). In contrast, the FE-SEM image of
Nano-MoVO/C (Fig. S1 in the Supplementary material) shows only
carbon particles without any other rod-shaped crystals, suggesting that
MoVO species had formed within these carbon particles. The TEM
characterization of Nano-MoVO/C (Fig. S1 in the Supplementary ma-
terial) was unable to distinguish MoVO particles from carbon particles,
showing only aggregates of nanoparticles 20 nm in diameter. Thus, N2
adsorption experiments were conducted for the mesoporous carbon
before and after the hydrothermal synthesis to confirm the formation of
MoVO species inside mesopores of the carbon material. The data show
that the mesopore volume decreased by 0.05 cm3 (g-carbon)–1 after the
the absence of the broad feature at 2θ<10 arising from the carbon
background. The pattern for Nano-MoVO shows much broader peaks at
◦
◦
22 and 45 than those for Orth-MoVO and Amor-MoVO, consistent
with the smaller crystal sizes observed for Nano-MoVO. Analysis of the
(001) peak at 22 ◦ by the Scherrer equation indicates that the average
length of crystals is 30 nm, which is consistent with the size of particles
observed by FE-SEM (Fig. 2b). In summary, the data indicate the for-
mation of small MoVO crystals inside the mesopores of the carbon
material.
4

R. Obunai et al.
Applied Catalysis A, General 624 (2021) 118294
Thus, to obtain direct information about the structure, we charac-
terized Nano-MoVO by HR-TEM, and analyzed the powder spectrum.
The HR-TEM image characterizing Nano-MoVO (Fig. 3c) shows regular
repeating slabs with a spacing of approximately 0.4 nm along the length
of a small rod shape crystal, consistent with the PXRD data. The power
spectrum shows diffraction spots corresponding to reflections from
(001) and (200) planes. Thus, we infer that the structure of Nano-MoVO
resembles that of Orth-MoVO. The Mo/V ratios for Orth-MoVO and
Amor-MoVO are both ≈3.0 as characterized by elemental analysis. On
the other hand, the ratio for Nano-MoVO is ≈4.8, indicating that Nano-
MoVO may have a Mo-rich structure or may contain some impurities.
3.1.4. Pore structure of Nano-MoVO by N2 adsorption
N2 adsorption experiments were conducted to characterize the
porosity of synthesized catalysts after removal of carbon scaffolds. The
data characterizing Orth-MoVO and Amor-MoVO show type I isotherms
(Fig. 4), typical for these microporous materials. On the other hand, the
data characterizing Nano-MoVO show a type IV isotherm with a hys-
teresis, indicating the presence of mesopores. Analysis of the data shows
that Nano-MoVO has an approximately 5-fold higher mesopore volume
(Vmeso) and a 4-fold larger external surface area (SEXT) than those of
Orth-MoVO and Amor-MoVO (Table 1). Nano-MoVO has mesopores
about 10 nm in radius, attributable to void spaces of nanocrystals, which
had been created presumably by the removal of carbon scaffolds. In
addition, Nano-MoVO has a larger Vmicro than those for Orth-MoVO and
Amor-MoVO presumably because void spaces between nanocrystals in
addition to 7 MR micropores contribute to Vmicro. The adsorption
isotherm data show nearly the same uptake for Nano-MoVO as Orth-
MoVO at low relative pressures (P/P0 <10–4). Thus, we infer that
Nano-MoVO possesses a crystalline structure resembling Orth-MoVO
Fig. 3. FT-IR (left) and Raman (right) spectra characterizing Nano-MoVO (a),
Amor-MoVO (b), and Orth-MoVO (c).
3.1.3. Structural units of Nano-MoVO characterized by FT-IR and Raman
spectroscopies and HR-TEM
FT-IR and Raman spectroscopies were used to characterize the
structural units contained in the synthesized catalysts. FT-IR spectra
characterizing Nano-MoVO (Fig. 3) show bands characteristic for Orth-
MoVO and Amor-MoVO: 915 cm–1 for δV=O, 870 cm–1 for δMo=O, 817 and
716 cm–1 for δMo–O–Mo, 604 cm–1 for δV–O–Mo. The 870-cm–1 band rep-
–
resents the stretching mode of Mo O in the pentagonal units {Mo O
}
–
[11], which are present in both Orth-MoVO and Amor-MoVO.6 T2h1is
structural feature is also observed in the Raman spectra for all catalysts
(Fig. 3). The Raman data characterizing Nano-MoVO also show the
absence of a band ascribed to carbonaceous species at 1350 cm–1,
consistent with the TGA data. As described at the beginning of this
section, the PXRD data for Nano-MoVO (Fig. 1) show only a broad
Table 1
Textual properties of MoVO catalysts characterized by N2 adsorption
experiments.
a
b
c
Catalyst
Vmicro
/
Vmeso
/
Sext /
10ꢀ 3 cm3 g-1
cm3 gꢀ 1
m2 gꢀ 1
Nano-MoVO
Orth-MoVO
Amor-MoVO
.
10
0.14
0.03
0.03
40
10
10
◦
reflection at 2θ <10 like that for Amor-MoVO, suggesting that the
6.2
3.2
structure along the a–b plane lacks long-range order like Amor-MoVO or
that it has a crystalline order like Orth-MoVO but shows broad and
overlapping peaks because of the smallness of crystals. The data ob-
tained by FT-IR and Raman spectroscopy are inconclusive about
whether the structure of Nano-MoVO can be represented by Orth-MoVO
or Amor-MoVO.
a
Micropore volume determined by N2 uptake at P/P0 = 0.01.
b
Mesopore volume determined by subtracting Vmicro from N2 uptake at P/P0
= 0.99.
c
External surface area obtained by t-plot.
Fig. 4. (a) Nitrogen adsorption-desorption isotherms characterizing Nano-MoVO (red), Orth-MoVO (blue), Amor-MoVO (green) and (b) pore size distributions
obtained by applying the Dollimore Heal method to the adsorption isotherms (for interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article).
5

R. Obunai et al.
Applied Catalysis A, General 624 (2021) 118294
Fig. 5. Measured ratios of propane/ethane consumption rates (left bar) and
propene/ethene formation rates (right bar). ODH of ethane: T = 593 K, C2H6/
O2/N2 molar percentage = 10.0/10.0/80.0 mol%, F/W = 100 mL minꢀ 1 gꢀ 1 for
Nano-MoVO and Amor-MoVO, 300 mL minꢀ 1 gꢀ 1 for Orth-MoVO. ODH of
propane: T = 603 K, C3H8/O2/N2/H2O = 7.5/10/37.0/45.5 mol%, F/W = 200
mL minꢀ 1 gꢀ 1 for Nano-MoVO, 400 mL minꢀ 1 g-1 for Amor-MoVO, and 800 mL
minꢀ 1 gꢀ 1 for Orth-MoVO.
Fig. 6. Arrhenius plots for ethane ODH reactions catalyzed by synthesized
MoVO catalysts: Nano-MoVO ((●, red), Orth-MoVO ((◼, blue), and Amor-
MoVO ((▴, green). Feed composition: C2H6/O2/N2 = 10/10/80 mol%. F/W =
100 mL minꢀ 1 gꢀ 1 for Nano-MoVO and Amor-MoVO, 300 mL minꢀ 1 gꢀ 1 for
Orth-MoVO (for interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article).
and Amor-MoVO.
same Vmeso and Sext, Vmicro of the former is smaller than that of the latter
(Table 1). Therefore the activation of ethane proceeds less efficiently on
Amor-MoVO than Orth-MoVO whereas that of propane presumably
proceeds similarly in both catalysts, giving the higher ratio for
Amor-MoVO. The rC3H8/ rC2H6 ratio on Nano-MoVO is the largest among
the MoVO catalysts, indicating that active sites outside pores facilitate
propane activation on this catalyst. The rC3H6/rC2H4 value, as well as this
value relative to the rC3H8/ rC2H6 value, becomes larger for Nano-MoVO
than other catalysts, suggesting improved propene selectivity caused by
limited overoxidation through enhanced mass transfer. In summary, the
data indicate beneficial effects of the textural properties of Nano-MoVO
(Table 1).
3.2. Oxidative dehydrogenation catalysis of ethane and propane
The ODH reactions of ethane and propane were conducted in flow
reactors to evaluate the catalytic performance of Nano-MoVO. The
conversion-selectivity data for ODH of ethane on Orth-MoVO (Fig. S3 in
the Supplementary material) are in line with the literature data for
orthorhombic MoVO [11], showing ethene selectivity of ≈80 % at a
conversion of ≈25 %. The previous work indicates that the ODH of
– –
V O
ethane proceeds at active sites inside heptagonal channels of Mo
catalysts as well as on their external surfaces [12]. In contrast, the re-
actions of propane may preferentially occur at sites on external surfaces
because of the larger kinetic diameter of propane (4.3 Å) than heptag-
onal channels (4.0 Å). To evaluate the effects of the high external surface
area of Nano-MoVO, the ratios of the mass-specific propane/ethane
consumption rates (rC3H8/rC2H6, eq 1) and propene/ethene formation
rates (rC3H6/rC2H4, eq 2) were calculated for each catalyst (Fig. 5). ODH
Although Nano-MoVO possesses common structural units as other
MoVO catalysts, the structure of active sites may be different, which
could influence the rate ratio. Thus, the ODH of ethane was conducted at
593–643 K, and the apparent activation energies were calculated from
Arrhenius plots to investigate the nature of active sites in Nano-MoVO
and compare it with that of the other two MoVO catalysts (Fig. 6).
The apparent activation energies of all catalysts are nearly the same, and
the values are similar to those reported in the literature for complex
metal oxide catalysts (82 kJ mol–1 for the orthorhombic MoVO catalyst
[11]). All catalysts exhibit nearly the same dependence of partial pres-
sure of propane (≈0.8) on reaction rate (Fig. S4 in the Supplementary
material) whereas no explicit dependence was found for partial pressure
of O2. In addition, Nano-MoVO exhibited almost the same ethene
selectivity in ethane oxidation; ethane selectivity of Nano-MoVO was 80
%, and that of Orth-MoVO was 70 % when ethane conversion was about
10 %. These results show that the difference in the mass-specific rate
–
– –
O Mo pair
of ethane and propane are limited by C H activation at V
–
H
sites. Therefore, the rC3H8/ rC2H6 value represents the rate ratio of C
activation in C3H8 at active sites on external surfaces relative to that in
C2H6 at active sites both in the pores and on external surfaces. It avoids
general difficulties inherent to the quantification of concentrations of
active sites for three different MoVO catalysts and potential complica-
tions caused by modification of the surface structure of catalysts under
different gas reactive gas atmospheres [17,35] and impurities present in
catalysts. The variation of the rC3H6/rC2H4 ratio among the three cata-
lysts, as well as the relative value of this ratio to the rC3H8/rC2H6 ratio,
provide information about how the morphological difference influences
the olefin selectivity.
– –
V O catalysts is caused by the difference
(rC2H6) among the three Mo
in the concentrations of active sites rather than the nature of active sites.
The low mass-specific activity of Nano-MoVO is caused presumably by
blocking of some sites by Mo-rich impurities and aggregation of nano-
crystals as indicated by Fig. 2b.
/
/
r
C3H8
rate ratio of propane ethene consum ption =
(1)
r
C2H6
Selectivities in the ODH of propane were compared to investigate the
r
r
C3H6
C2H4
rate ratio of propene ethene form ation =
(2)
The rC3H8/ rC2H6 ratio on Orth-MoVO is slightly smaller than unity,
–
although propane has weaker C H bonds than ethane. This value on
Orth-MoVO is consistent with the data reported for the same type of
catalysts [36] and suggests greater accessibility of ethane to active sites
inside the pores and a more favorable contribution of pores to ethane
activation [10,37]. The rC3H8/ rC2H6 ratio on Amor-MoVO is larger than
that on Orth-MoVO. Although Amor-MoVO and Orth-MoVO possess the
Scheme 2. Reaction network of propane oxidation reaction [18].
6

R. Obunai et al.
Applied Catalysis A, General 624 (2021) 118294
Nano-MoVO, whereas they yielded more acetic acid than Orth-MoVO.
Table 2
–
Representative conversion and selectivity to major products in propane ODH on
Therefore, we infer that allylic C H activation is suppressed relative
MoVO catalystsa.
–
to C O formation on Nano-MoVO and Amor-MoVO. Nonetheless,
Nano-MoVO and Amor-MoVO show less selectivity to COx than
Orth-MoVO. In particular, COx formation is significantly suppressed on
Nano-MoVO (Table 2). Both Ortho-MoVO and Amor-MoVO show a
combined selectivity to acetic acid and COx of ≈50 % at a propane
conversion of 5%, consistent with the data reported earlier [15]. In
contrast, Nano-MoVO shows a combined selectivity to acetic acid and
COx of ≈40 % at the same conversion. These results suggest that the total
oxidation to COx is limited on Nano-MoVO relative to the other two
MoVO catalysts. Considering several reports suggesting the importance
of facile desorption of propene to limit overoxidation, we suggest that
the higher propene selectivity for Nano-MoVO is achieved by the more
facile desorption of propene into the bulk gas phase enabled by the
nanocrystalline morphology of Nano-MoVO. Nano-MoVO exhibited
stable ODH performance at 603 K, and maintained a high propene
selectivity (70 %) at a conversion of ≈3% (Fig. S5 in the Supplementary
material).
Selectivity/%
Catalyst
Conversion/%
Propene
COx
Acetic acid
Acrylic acid
Nano-MoVOb
Orth-MoVOc
Amor-MoVOd
5.54
6.88
5.26
57.7
28.2
41.5
22.9
41.5
32.3
18.2
10.9
19.2
1.2
19.4
7.0
a
T = 603 K, Feed composition: C3H8/O2/N2/H2O = 7.5/10/37.0/45.5 mol%.
b
c
F/W (Nano-MoVO) = 200 mL minꢀ 1 gꢀ 1
.
F/W (Amor-MoVO) = 400 mL minꢀ 1 gꢀ 1
.
d
F/W (Orth-MoVO) = 800 mL min–1 g–1
.
effects of nanocrystalline morphology of Nano-MoVO further. The
– –
V O complex oxide catalysts is gener-
conversion of propane on Mo
ally considered to proceed via propene as the primary product, from
which acrylic acid, acetic acid, and COx(CO/CO2) are formed (Scheme
–
2) [18]. The activation of allylic C H bonds in C3H6 and addition of one
–
oxygen atom yield acrolein. Further activation of C H bonds and the
4. Conclusion
addition of an oxygen atom at the allylic position in acrolein yield
–
–
acrylic acid. On the other hand, the activation of the C C bond in C H
3
6
In summary, we successfully synthesized Nano-MoVO that possesses
the same structural units as Orth-MoVO and Amor-MoVO as character-
ized by IR and Raman spectroscopies, XRD, and HR-TEM. Nano-MoVO
consists of nanocrystals approximately 60 nm in length and 20 nm in
diameter as characterized by FE-SEM. It possesses approximately a 5-
times larger mesopore volume and a 4-fold larger external surface
area than Orth-MoVO. In addition, it possesses micropores attributable
to 7MR like Orth-MoVO. The activation energy of Nano-MoVO in ethane
oxidation was nearly the same as that of Orth- and Amor-MoVO. Nano-
MoVO facilitated the C–H activation of C3H8 relative to C2H6 as
compared with other Mo–V–O catalysts. Moreover, it exhibited a higher
–
with C O formation produces products such as acetone, and subsequent
–
–
H
C
C scission yields acetic acid and COx. Therefore, the initial C
–
activation and C O bond formation in C3H6 influence the selectivity to
these products [37]. Orth-MoVO and Amor-MoVO exhibited nearly the
same selectivity to propene at the same conversion (Table 2 and Fig. 7).
Extrapolation of these conversion-propene selectivity data in Fig. 7(a)
agrees with the data reported by Katou et al. [38]. In contrast,
Nano-MoVO exhibited ≈20 % higher selectivity to propene and less
formation of COx than the other two catalysts (Table 2 and Fig. 7(a) and
7(d)). Nano-MoVO and Amor-MoVO formed less acrylic acid than
Fig. 7. Selectivity to (a) propene, (b) acrylic
acid, (c) acetic acid, (d) COx as a function of
propane conversion: Nano-MoVO(●, red), Orth-
MoVO(◼, blue), Amor-MoVO(▴, green). T = 603
K, Ptotal = atmospheric, vtotal = 16–120 mL(STP)/
min, W = 0.1 g, Feed gas composition: C3H8/O2/
N2/H2O = 7.5/10/37.0/45.5 mol%.★ (data re-
ported in reference [15]): T = 619 K, Ptotal
=
atmospheric, vtotal = 20 mL(STP)/min, W = 0.5 g
of Orth-MoVO, Feed gas composition:
C3H8/O2/N2/H2O
= 6.5/10/38.5/45.0 mol%
(for interpretation of the references to colour in
this figure legend, the reader is referred to the
web version of this article).
7

R. Obunai et al.
Applied Catalysis A, General 624 (2021) 118294
[9] A.M. Gaffney, O.M. Mason, Catal. Today 285 (2017) 159–165, https://doi.org/
10.1016/j.cattod.2017.01.020.
propene selectivity by limiting subsequent oxidation to acrylic acid and
COx. Considering several reports suggesting the importance of facile
desorption of propene to limit overoxidation, we infer that Nano-MoVO
allows fast transfer of propene from the catalyst surface to the bulk gas
phase, enabling a relatively higher propene selectivity than other
Mo–V–O catalysts. This work suggests an approach to improve olefin
selectivities on catalysts that overoxidation is an issue.
[10] L. Annamalai, Y. Liu, S. Ezenwa, Y. Dang, S.L. Suib, P. Deshlahra, ACS Catal. 8
(2018) 7051–7067, https://doi.org/10.1021/acscatal.8b01682.
[11] T. Konya, T. Katou, T. Murayama, S. Ishikawa, M. Sadakane, D. Buttrey, W. Ueda,
Catal. Sci. Technol. 3 (2013) 380–387, https://doi.org/10.1039/C2CY20444D.
[12] S. Ishikawa, X. Yi, T. Murayama, W. Ueda, Appl. Catal. A 474 (2014) 10–17,
https://doi.org/10.1016/j.apcata.2013.07.050.
[13] T. Ushikubo, H. Nakamura, Y. Koyasu and S.Wajiki, US Patent No. US5380933
(1995).
´
´
[14] P. Botella, J.M. Lopez Nieto, B. Solsona, A. Mifsud, F. Marquez, J. Catal. 209 (2002)
Author statement
445–455, https://doi.org/10.1006/jcat.2002.3648.
[15] W. Ueda, D. Vitry, T. Katou, Catal. Today 99 (2005) 43–49, https://doi.org/
10.1016/j.cattod.2004.09.022.
Ryo Obunai: Conceptualization, Visualization, Data curation, Formal
analysis, Writing - original draft, Writing – reviewing & editing; Keisuke
Tamura, Visualization, Data curation, Investigation, Formal analysis;
Isao Ogino, Project administration, Supervision, Conceptualization,
Investigation, Visualization, Funding acquisition, Writing - original
draft, Writing – reviewing & editing; Shin R. Mukai, Funding acquisi-
tion, Investigation, Resources, Supervision, Writing - review & editing;
Wataru Ueda, Investigation, Supervision, Writing – reviewing & editing.
[16] K. Amakawa, Y.V. Kolen’ko, A. Villa, M. Schuster, L. Csepei, G. Weinberg,
¨
S. Wrabetz, R. d’Alnoncourt, F. Girgsdies, L. Prati, R. Schlogl, A. Trunschke, ACS
Catal. 3 (2013) 1103–1113, https://doi.org/10.1021/cs400010q.
[17] A. Trunschke, J. Noack, S. Trojanov, F. Girgsdies, T. Lunkenbein, V. Pfeifer,
¨
¨
M. Havecker, P. Kube, C. Sprung, F. Rosowski, R. Schlogl, ACS Catal. 7 (2017)
3061–3071, https://doi.org/10.1021/acscatal.7b00130.
[18] N.F. Naraschewski, A. Jentys, J.A. Lercher, Top. Catal. 54 (2011) 639–649, https://
doi.org/10.1007/s11244-011-9686-x.
[19] L. Luo, J.A. Labinger, M.E. Davis, J. Catal. 200 (2001) 222–231, https://doi.org/
10.1006/jcat.2001.3214.
[20] T. Katou, D. Vitry, W. Ueda, Catal. Today 91–92 (2004) 237–240, https://doi.org/
10.1016/j.cattod.2004.03.039.
Declaration of Competing Interest
´
´
´
[21] J.M. Lopez Nieto, B. Solsona, P. Concepcion, F. Ivars, A. Dejoz, M.I. Vazquez, Catal.
Today 157 (2010) 291–296, https://doi.org/10.1016/j.cattod.2010.01.046.
[22] T. Blasco, J.M.L. Nieto, Appl. Catal. A 157 (1997) 117–142, https://doi.org/
10.1016/S0926-860X(97)00029-X.
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.
[23] Nieto, Top. Catal. 41 (2006) 3–15, https://doi.org/10.1007/s11244-006-0088-4.
[24] A.H. Liu, F. Schüth, Adv. Mater. 18 (2006) 1793–1805, https://doi.org/10.1002/
adma.200600148.
Acknowledgements
[25] D. Gu, F.S. Schüth, Chem. Soc. Rev. 43 (2014) 313–344, https://doi.org/10.1039/
C3CS60155B.
[26] W. Li, A. Lu, C. Weidenthaler, F. Schüth, Chem. Mater. 16 (2004) 5676–5681,
https://doi.org/10.1021/cm048759n.
The authors acknowledge Dr. Norihito Sakaguchi for his assistance in
the HR-TEM measurements.
[27] W. Li, M. Comotti, A. Lu, F. Schüth, Chem. Commun. 16 (2006) 1772–1774,
https://doi.org/10.1039/B601109H.
[28] C. Madsen, C.J.H. Jacobsen, Chem. Commun. 8 (1999) 673–674, https://doi.org/
10.1039/A901228A.
Appendix A. Supplementary data
[29] I. Schmidt, C. Madsen, C.J.H. Jacobsen, Inorg. Chem. 39 (2000) 2279–2283,
https://doi.org/10.1021/ic991280q.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2021.118294.
[30] S. Kim, J. Shah, T.J. Pinnavaia, Chem. Mater. 15 (2003) 1664–1668, https://doi.
org/10.1021/cm021762r.
[31] Y. Tao, H. Kanoh, K. Kaneko, J. Am. Chem. Soc. 125 (2003) 6044–6045, https://
doi.org/10.1021/ja0299405.
References
[32] W. Fan, M.A. Snyder, S. Kumar, P.S. Lee, W.C. Yoo, A.V. McCormick, R. Lee Penn,
A. Stein, M. Tsapatsis, Nat. Mater. 7 (2008), https://doi.org/10.1038/nmat2302,
984.
[1] H.H. Kung, M.C. Kung, Appl. Catal. A Gen. 157 (1997) 105–116, https://doi.org/
10.1016/S0926-860X(97)00028-8.
[33] J. Sun, K. Zhu, F. Gao, C. Wang, J. Liu, C.H.F. Peden, Y. Wang, J. Am. Chem. Soc.
133 (2011) 11096–11099, https://doi.org/10.1021/ja204235v.
[34] T. Tsuchiya, T. Mori, S. Iwamura, I. Ogino, S.R. Mukai, Carbon 76 (2014) 240–249,
https://doi.org/10.1016/j.carbon.2014.04.074.
´
´
[2] J.M. Lopez Nieto, B. Solsona, in: J.C. Vedrine (Ed.), Metal Oxides in Heterogeneous
Catalysis, Elsevier, Amsterdam, 2018, pp. 211–286.
¨
[3] C.A. Gartner, A.C. van Veen, J.A. Lercher, ChemCatChem 5 (2013) 3196–3217,
https://doi.org/10.1002/cctc.201200966.
¨
¨
[35] M. Havecker, S. Wrabetz, J. Krohnert, L. Csepei, R.N. d’Alnoncourt, Y.V. Kolen’ko,
[4] F. Cavani, N. Ballarini, A. Cericola, Catal. Today 127 (2007) 113–131, https://doi.
org/10.1016/j.cattod.2007.05.009.
¨
F. Girgsdies, R. Schlogl, A. Trunschke, J. Catal. 285 (2012) 48–60, https://doi.org/
10.1016/j.jcat.2011.09.012.
[5] J.T. Grant, J.M. Venegas, W.P. McDermott, I. Hermans, Chem. Rev. 118 (2018)
2769–2815, https://doi.org/10.1021/acs.chemrev.7b00236.
[36] S. Ishikawa, X. Yi, T. Murayama, W. Ueda, Catal. Today 238 (2014) 35–40, https://
doi.org/10.1016/j.cattod.2013.12.054.
¨
[6] C.A. Carrero, R. Schlogl, I.E. Wachs, R. Schomaecker, ACS Catal. 4 (2014)
[37] Y. Liu, A. Twombly, Y. Dang, A. Mirich, S.L. Suib, P. Deshlahra, ChemCatChem 13
(2020) 882–899, https://doi.org/10.1002/cctc.202001642.
[38] T. Katou, D. Vitry, W. Ueda, Chem. Lett. 32 (2003) 1028–1029, https://doi.org/
10.1246/cl.2003.1028.
3357–3380, https://doi.org/10.1021/cs5003417.
[7] K. Chen, A.T. Bell, E. Iglesia, J. Catal. 209 (2002) 35–42, https://doi.org/10.1006/
jcat.2002.3620.
´
´
´
[8] P. Botella, E. Garcıá -Gonzalez, A. Dejoz, J.M. Lopez Nieto, M.I. Vazquez,
´
J. Gonzalez-Calbet, J. Catal. 225 (2004) 428–438, https://doi.org/10.1016/j.
jcat.2004.04.024.
8