Oxidative dehydrogenation of ethylbenzene to styrene with CO2 over Al-MCM-41-supported vanadia catalysts

Article abstract of DOI:10.1002/aoc.5396

Catalytic performance of Al-MCM-41-supported vanadia catalysts (V/Al-MCM-41) with different V loading was investigated for oxidative dehydrogenation of ethylbenzene to styrene (ST) with CO₂ (CO₂-ODEB). For comparison, pure silica MCM

Full text of DOI:10.1002/aoc.5396

Received: 16 July 2019
DOI: 10.1002/aoc.5396
Revised: 19 October 2019
Accepted: 5 November 2019
F U L L P A P E R
Oxidative dehydrogenation of ethylbenzene to styrene with
CO2 over Al-MCM-41-supported vanadia catalysts
Shuwei Chen1
| Zheqi Xu1 | Dongchen Tan1 | Dahai Pan1 | Xingyu Cui1
|
Yan Qiao2 | Ruifeng Li1
1College of Chemistry and Chemical
Engineering, Taiyuan University of
Technology, Taiyuan, 030024, China
2State Key Laboratory of Coal Conversion,
Institute of Coal Chemistry, Chinese
Academy of Sciences, Taiyuan, 030001,
China
Catalytic performance of Al-MCM-41-supported vanadia catalysts (V/Al-
MCM-41) with different V loading was investigated for oxidative dehydrogena-
tion of ethylbenzene to styrene (ST) with CO2 (CO2-ODEB). For comparison,
pure silica MCM-41 was also used as support for vanadia catalyst. The catalysts
were characterized by N2 adsorption, X-ray diffraction (XRD) pyridine-Fou-
rier-transform infrared spectroscopy, H2-temperature-programmed reduction,
thermogravimetric analysis (TGA), UV-Raman, and diffuse reflectance (DR)
UV–vis spectroscopy. The results indicate that the catalytic behavior and the
nature of V species depend strongly on the V loading and the support proper-
ties. Compared with the MCM-41-supported catalyst, the Al-MCM-41-
supported vanadia catalyst exhibits much higher catalytic activity and stability
along with a high ST selectivity (>98%). The superior catalytic performance of
the present V/Al-MCM-41 catalyst can be attributed to the Al-MCM-41 support
being more favorable for the high dispersion of V species and the stabilization
of active V5+ species. Together with the characterization results of XRD, TGA,
and DR UV–Vis spectroscopy, the deep reduction of V5+ into V3+ during CO2-
ODEB is the main reason for the deactivation of the supported vanadia cata-
lyst, while the coke deposition has a less important impact on the catalyst
stability.
Correspondence
Shuwei Chen, College of Chemistry and
Chemical Engineering, Taiyuan
University of Technology, Taiyuan
030024, China.
Email: csw603@163.com
Ruifeng Li, College of Chemistry and
Chemical Engineering, Taiyuan
University of Technology, Taiyuan
030024, China.
Email: rfli@tyut.edu.cn
Funding information
Shanxi Provincial Key Innovative
Research Team in Science and
Technology, Grant/Award Numbers:
2014131006, No. 2014131006;
International Cooperation Project, Grant/
Award Number: 2015081048; Natural
Science Foundation of Shanxi Province,
China, Grant/Award Number:
K E Y W O R D S
Al-MCM-41, carbon dioxide, catalyst deactivation, ethylbenzene dehydrogenation, vanadium
oxide
201801D121057; Foundation of Sate Key
Laboratory of Coal Conversion, Grant/
Award Number: J18-19-906; Key Research
and Development Program of Shanxi
Province, China, Grant/Award Number:
201803D31034; National Natural Science
Foundation of China, Grant/Award
Number: 21975174
1 | INTRODUCTION
thermodynamically limited with a very high energy con-
sumption because of the loss of latent heat in the gas–liq-
Styrene (ST), one of the most important monomers in
polymer industry, is mainly produced by the dehydroge-
nation of ethylbenzene (EB) in the presence of excess
uid separator. Therefore, developing an alternative
method is of great importance. The oxidative dehydroge-
nation of EB (ODEB) with O2 is a possible alternative
solution, but suffers from the loss of selectivity for ST due
steam
at
600–650
ꢀC.[1,2]
This
process
is
Appl Organometal Chem. 2019;e5396.
wileyonlinelibrary.com/journal/aoc
© 2019 John Wiley & Sons, Ltd.
1 of 10
https://doi.org/10.1002/aoc.5396

2 of 10
CHEN ET AL.
to the deep oxidation of EB.[3,4] Utilization of CO2 as a
soft oxidant in the dehydrogenation reaction exhibits var-
ious advantages such as no loss of latent heat,
diminishing thermodynamic limitation, accelerating the
reaction rate, increasing ST selectivity, and effectively uti-
lizing the greenhouse gas CO2.[5–7] Thus, ODEB with CO2
(CO2-ODEB) has been extensively investigated as an
energy-saving, environmentally benign, and efficient
alternative process for the production of ST.[8]
In this paper, the catalytic performance of Al-MCM-
41-supported vanadia catalysts with V loading of 2–
8 wt.% for CO2-ODEB was investigated systematically.
The effect of V loading and support properties on the
structure of VOx species, the role of Al species in support
modification, and the reason for catalyst deactivation
were clarified in the light of detailed characterization of
obtained catalysts.
Several catalytic systems containing Fe,[9–13]
V,[3–7,14–20] Zr,[21,22] Ce,[23–25] Mo,[26] Mn[27] oxides have
been found to be efficient toward CO2-ODEB. Among
them, the supported vanadia catalysts have attracted spe-
cial attention because of their high activity and selectivity
for ST in the reaction. However, such catalysts still suffer
from fast deactivation, which severely hinders their prac-
tical applications. It is widely believed that the CO2-
ODEB over supported vanadia catalysts follows the
Mars–Van Krevelen redox mechanism and the highly dis-
persed V5+ species are active sites.[3–7,15] Among the fac-
tors determining the catalytic performance of supported
vanadia catalysts, the dispersion and the redox behavior
of VOx species are known to play the key roles in CO2-
ODEB. To obtain more active and stable V-based cata-
lysts for CO2-ODEB, one should achieve a high disper-
sion of the VOx species at vanadium loadings as high as
possible. Thus, it is necessary to choose a support with
high specific surface area for dispersing the VOx species.
MCM-41 material with ordered cylindrical channels, high
specific surface area, and controllable uniform pore sizes
of 2–10 nm has attracted extensive attention for its appli-
cation as support.[28–31] At present, the MCM-41 material
has been used as support for the preparation of V- and
Cr-based catalysts with excellent catalytic activities and
selectivity toward ST in the CO2-ODEB reaction.[16,32]
Unfortunately, the rapid deactivation of the resultant
MCM-41-supported catalysts remains a major problem to
be improved.
It is well-known that MCM-41 is a neutral support.
The incorporation of Al species into the MCM-41
can lead to the formation of acid sites,[33–35] which
can be preferentially anchored by the metal oxide spe-
cies. The resulting interaction of metal oxide with the
Al-MCM-41 support can induce significant changes in
the dispersion and redox behavior of active species,
which will strongly influence the catalytic performance
of catalysts. At present, it has been confirmed that Al-
MCM-41 can display excellent support properties for
the reduction of nitrophenols, the vapor-phase
hydrodeoxygenation of guaiacol, and the NO selective
reduction by hydrogen.[36–38] However, there are few
reports on Al-MCM-41-supported catalysts for CO2-
ODEB.
2 | EXPERIMENTAL SECTION
2.1 | Catalyst preparation
All solvents and materials involved in this work are pur-
chased from Tianjin Kemiou Chemical Reagent Co., Ltd.,
(Tianjin, P. R. China) and used without further purifica-
tion. Mesoporous Al-MCM-41 support was synthesized
according to the method described in the literature.[39] In
a typical synthesis, 3.0 g of cetyltrimethylammonium bro-
mide (CTAB) was dissolved in 160 mL aqueous solution
containing 73.5 g of anhydrous ethanol and 42.4 g of 25%
ammonium hydroxide. Then, 0.11 g of sodium aluminate
and 5.73 g of tetraethyl orthosilicate (TEOS) were succes-
sively added into this solution to obtain a synthesis mix-
ture with a molar composition of 1TEOS/0.05NaAlO2/
144H2O/0.ꢀ3CTAB/11NH3/58ethanol. After vigorous stir-
ring at 35 C for 3 hr, the resulting white precipitate was
ꢀ
filtrated, washed, dried at 80 C for 8 hr, and calcined at
ꢀ
550 C in air for 6 hr. Pure silica MCM-41 support was
also prepared following the same procedure except that
no sodium aluminate was added.
Al-MCM-41-supported vanadia catalysts with differ-
ent vanadium loading were prepared by the incipient
wetness impregnation method. About 1.0 g of Al-MCM-
41 was impregnated with 2 mL aqueous solution of oxalic
acid, in which the concentrations of NH4VO3 were 0.20,
0.40, 0.62, and 0.85 mol/L, respectively. After impregna-
tion, the catalysts were dried overnight at 110 ꢀC and
ꢀ
then calcined at 550 C for 4 hr. The resultant catalysts
were denoted as nV/Al-MCM-41, where n represents the
V content in wt.%. For comparison, vanadia catalyst
supported on pure silica MCM-41 with a V content of
6 wt.% (named as 6 V/MCM-41) had also been prepared
following the same procedure.
2.2 | Catalyst characterization
The textural properties of catalysts were analyzed by N2
adsorption/desorption at −196 ꢀC on a surface area and pore-
size analyzer (Quantachrome NOVA-1200e analyzer). The spe-
cific surface area was determined by the Brunauer–Emmett–
Teller (BET) equation. The BJH (Barrett–Joyner–Halenda)
method was used to calculate the pore-size distribution.

AL-MCM-41-SUPPORTED VANADIA CATALYSTS FOR CO2-ODEB
3 of 10
X-ray diffraction (XRD) patterns were recorded on a
diffractometer (Shimadzu XRD-6000) equipped with
nickel-filtered Cu-Kα (λ = 0.154 nm) radiation.
UV-Raman spectra were obtained on a spectropho-
tometer (LabRAM HR Evolution [HORIBA]) under
ambient conditions. The laser line at 325 nm of He–Cd
laser was used as the exciting source.
Pyridine adsorption experiments were conducted on a
spectrometer (Shimadzu IR Affinity-1). The samples were
pre-treated at 400 ꢀC for 4 hr under a vacuum of 10−2 Pa.
Fourier-transform infrared (FTIR) spectra were recorded
after pyridine adsorption and subsequent evacuation at
150 ꢀC.
H2-temperature-programmed reduction (H2-TPR) was
carried out on a TPR/temperature-programmed desorp-
tion apparatus (Model 5076; Micromeritics) equipped
with a thermal conductivity detector. Prior to the TPR
ꢀ
experiments, the samples were pre-treated at 300 C for
1 hr in air, and then were cooled to room temperature in
argon. H2-TPR was performed in 10 vol% H2/Ar from
room temperature to 800 ꢀC with a heating rate of 10 ꢀC/
min.
Diffuse reflectance UV–Vis spectra were recorded on
a spectrophotometer (Shimadzu UV-2501 PC) in the
range 200–800 nm under ambient conditions.
The carbon content on the used catalyst was mea-
sured by thermogravimetric analysis (TGA) using a ther-
FIGURE 1 Small-angle X-ray diffraction patterns of V/Al-
MCM-41 catalysts
mogravimetric
analyzer
(NETZSCH,
Germany;
different vanadium loadings. As can be seen in Figure 1,
the Al-MCM-41 support presents three well-resolved dif-
fraction peaks, which can be indexed as the (100), (110),
and (200) diffraction peaks of characteristic MCM-41
structure.[40] For the V/Al-MCM-41 catalysts with the
vanadium content of 2–8 wt.% and 6 V/MCM-41 catalyst,
three well-defined diffraction peaks are still observed,
suggesting that the ordered hexagonal mesostructures of
Al-MCM-41 and MCM-41 supports are well preserved
after vanadia incorporation.
The N2 adsorption–desorption isotherms and
corresponding pore-size distribution curves of the Al-
MCM-41 support and supported vanadia catalysts are dis-
played in Figures 2 and 3, respectively. From Figure 2, it
can be seen that the Al-MCM-41 support exhibits a typi-
cal type IV isotherm with a H1-type hysteresis loop along
with a very steep capillary condensation step at a relative
pressure (P/P0) ranging from 0.28 to 0.41, characteristic
of ordered mesoporous materials with large and uniform
cylindrical mesopores. For all supported vanadia cata-
lysts, isotherms similar to that of the Al-MCM-41 support
are observed, indicating the presence of uniform meso-
porous structures. As shown in Figure 3, the pore-size
distribution is narrow for all V/Al-MCM-41 catalysts with
vanadium content up to 8 wt.%. The textural properties
of V/Al-MCM-41 catalysts are summarized in Table 1. All
STA449F3) from room temperature to 800 ꢀC at a heating
rate of 10 ꢀC/min under air.
2.3 | Catalytic tests
The ꢀcatalytic activity was tested in a fixed-bed reactor at
550 C under atmospheric pressure. The catalyst (0.3 g)
was loaded into the reactor and pre-treated with CO2 for
ꢀ
15 min at 550 C to saturate its surface active sites. The
EB was introduced by a microfeeder pump at a feed rate
of 0.2 mL/hr, and the molar ratio of CO2 to EB was
maintained at 15. The condensed products (benzene, tol-
uene, EB, and ST) were collected in an ice water trap and
analyzed by gas chromatogram (7890B, Agilent) equipped
with a flame ionization detector and an HP-5 capillary
column (30 m × 0.25 mm × 0.25 μm).
3 | RESULTS AND DISCUSSION
3.1 | Catalyst characterization
Figure 1 shows the small-angle XRD patterns of Al-
MCM-41 support and V/Al-MCM-41 catalysts with

4 of 10
CHEN ET AL.
TABLE 1 Textural properties of supported vanadia catalysts
Pore
SBET
(m2/g)
Pore volume
(cm3/g)
diameter
(nm)
Catalyst
Al-MCM-41
816
0.95
0.84
0.77
0.66
0.53
0.68
2.80
2.79
2.76
2.75
2.72
2.76
2 V/Al-MCM-41 732
4 V/Al-MCM-41 638
6 V/Al-MCM-41 541
8 V/Al-MCM-41 416
6 V/MCM-41
639
The wide-angle XRD patterns of V/Al-MCM-41 cata-
lysts are shown in Figure 4. For V/Al-MCM-41 catalysts
with V content below 6 wt.%, the diffraction peaks
corresponding to vanadium oxides were not observed,
suggesting a highly homogeneous dispersion of VOx spe-
cies on the surface of Al-MCM-41 support. However, fur-
ther increasing the V loading to 8 wt.%, the crystalline
V2O5 diffraction peaks appear over the 8 V/Al-MCM-41
catalyst, pointing to the formation of “bulk-like” V2O5
crystallites at higher vanadium loading. Notably, the dif-
fraction peaks of crystalline V2O5 are also observed over
the 6 V/MCM-41 catalyst. The characterization results
from the wide-angle XRD analysis suggest that the incor-
poration of Al into MCM-41 is beneficial for the highly
homogenous dispersion of VOx species.
FIGURE 2 N2 adsorption isotherms of V/Al-MCM-41
catalysts
To further probe the structure of VOx species
supported on the Al-MCM-41 support, UV-Raman spec-
troscopy was performed, and the results are shown in
Figure 5. It is clear from Figure 5 that the Raman bands
at 341, 486, 920, and 1018 cm−1 are observed for the V/
Al-MCM-41 catalysts with V content lower than 6 wt.%.
FIGURE 3 Pore-size distribution curves of V/Al-MCM-41
catalysts
catalysts exhibit a high BET surface area (416 m2/g), a
large pore volume (0.53 cm3/g), and an average pore
diameter of 2.72 nm even with a V content of 8 wt.%.
FIGURE 4 Wide-angle X-ray diffraction patterns of V/Al-
MCM-41 catalysts

AL-MCM-41-SUPPORTED VANADIA CATALYSTS FOR CO2-ODEB
5 of 10
FIGURE 6 Pyridine-FTIR spectra of the supports and
supported vanadia catalysts
FIGURE 5 UV-Raman spectra of V/Al-MCM-41 catalysts: (a)
2 V/Al-MCM-41, (b) 4 V/Al-MCM-41, (c) 6 V/Al-MCM-41, (d)
6 V/MCM-41, (e) 8 V/Al-MCM-41
band at 1490 cm−1 is due to pyridine adsorbed onto both
B and L acid sites.[44] In comparison, the Al-MCM-41
support possesses more L acid sites than the MCM-41
support, and additionally shows a band at 1543 cm−1
which can be assigned to pyridine adsorbed onto B acid
sites.[43,44] This result indicates that Al incorporation into
MCM-41 can generate bridging hydroxyl groups (Si–OH–
Al, B acid sites) and enhance the total amount of (B + L)
acid sites on Al-MCM-41.[45] Compared with the MCM-
41 and Al-MCM-41 supports, the 6 V/MCM-41 and 6 V/
Al-MCM-41 catalysts have more L acid sites, respectively,
due to the formation of Si–O–V and Al–O–V linkages on
the surfaces of the catalysts.[46] Moreover, it can be seen
that the 6 V/Al-MCM-41 catalyst exhibits more L and B
acid sites than the 6 V/MCM-41 catalyst.
The band at 486 cm−1 is assigned to the vibration of silox-
ane rings of siliceous MCM-41.[16,41] The bands at 341
and 920 cm−1 are attributable to the bending mode of the
V–O–V bond and the symmetric stretching of O–V–O to
the polymeric VOx species, respectively.[6,42] The band at
1018 cm−1 is ascribed to the symmetric stretching mode
of the V=O bond in the isolated tetrahedral VOx spe-
cies.[43] The absence of the typical band of V2O5 at
approximately 994 cm−1 in Figure 5a–c suggests that no
crystalline V2O5 is formed over V/Al-MCM-41 catalysts
with V content lower than 6 wt.%. These results indicate
that 2D monomeric and/or polymeric VOx species are
anchored on the surface of Al-MCM-41 when V contents
are below 6 wt.%.[16,43] However, upon further increasing
the V loading content to 8 wt.%, the band at approxi-
mately 994 cm−1 is observed over the 8 V/Al-MCM-41
catalyst (Figure 5e), suggesting the existence of crystalline
V2O5 at higher vanadium contents. Interestingly, com-
pared with the 6 V/Al-MCM-41 catalyst, the 6 V/MCM-
41 catalyst exhibits an additional band at approximately
994 cm−1, which is assignable to crystalline V2O5 (Fig-
ure 5d). This indicates that at the same vanadium con-
tent, the VOx species can be better dispersed on the Al-
MCM-41 support than on the pure silica MCM-41 sup-
port. The UV-Raman characterization results are well
consistent with the results from the wide-angle XRD
analysis (Figure 4).
Figure 6 shows the FTIR spectra of the supports and
the corresponding supported vanadia catalysts after pyri-
dine adsorption. In the spectrum of the MCM-41 support,
a relatively strong band at 1454 cm−1 and a weak band at
1490 cm−1 are observed. The band at 1454 cm−1 is attrib-
uted to pyridine adsorbed onto L acid sites, whereas the
FIGURE 7 H2-temperature-programmed reduction profiles of
V/Al-MCM-41 catalysts

6 of 10
CHEN ET AL.
The H2-TPR profiles of V/Al-MCM-41 catalysts are
shown in Figure 7. One peak at about 515 C is observed
catalytic activity depends strongly on the vanadium load-
ing. The initial EB conversion first increases with increas-
ing vanadium content and then decreases. A maximum
EB conversion of 56.9% is obtained over 6 V/Al-MCM-41
(Table 2). For the V/Al-MCM-41 catalysts with vanadium
contents below 6 wt.%, the vanadium species exist mainly
in a highly dispersed form of isolated and/or low-poly-
meric VOx species (Figure 5), which are exposed at
domain surfaces in this range of vanadium content. The
accessibility of active V5+ species has hardly been affected
by vanadium content, and thus the catalytic activity
increases with the increase in vanadium content appar-
ently. However, upon further increasing vanadium con-
tents (above 6 wt.%), the crystalline V2O5 is formed due
to a further polymerization of the VOx species (Figures 4
and 5). The loss of accessibility of active V5+ sites is likely
to account for the observed decrease in EB conversion. It
is also clear from Table 2 that EB conversion over the
6 V/Al-MCM-41 catalyst in the presence of CO2 is 11%
higher than that in an inert N2 atmosphere, indicating
that CO2 markedly promoted the EB dehydrogenation.
The presence of CO2 can improve the equilibrium con-
version of EB through the reaction coupling EB dehydro-
genation with the reverse water-gas shift,[2,16,24,26] which
provides the thermodynamic driving force to achieve
higher EB conversion. In addition, the vanadium species
can be effectively kept in a high (active) oxidation state
ꢀ
over the Al-MCM-41-supported catalysts with V contents
from 2 to 6 wt.%. At higher vanadium content (8 V/Al-
ꢀ
MCM-41), the main peak shifts to 525 C along with a
ꢀ
shoulder appearing at 630 C. The peak at low tempera-
ture is attributed to the reduction of highly dispersed tet-
ꢀ
rahedral VOx species, whereas the peak at 630 C can be
assigned to the reduction of polymeric V2O5-like V5+ spe-
cies.[43,47,48] Moreover, a high-temperature shoulder at
600 ꢀC is also observed for the 6 V/MCM-41 catalyst,
suggesting the existence of V2O5-like species on this cata-
lyst. The H2-TPR results further confirm that a higher
dispersion of the VOx species can be readily achievable
on the surface of Al-MCM-41 support. By contrast, it is
also seen from Figure 7 that for the 6 V/MCM-41 catalyst,
there is a shift in the main reduction peak toward lower
temperature (495 ꢀC) as compared with the 6 V/Al-
MCM-41 catalyst. This indicates that the Al-MCM-41
support possesses a relatively stronger interaction with
VOx species than the MCM-41 support. The formation of
highly reactive Al–OH groups from the incorporation of
Al into MCM-41 can be responsible for the remarkably
enhanced interaction between the Al-MCM-41 support
and VOx species, due to the generated Al–O–V bonds
being stronger than Si–O–V bonds.[49] As a result, the
high dispersion of VOx species is readily achievable on
the surface of the Al-MCM-41 support. A similar phe-
nomenon was also observed for THE Al-SBA-15-
supported catalyst.[46]
(V5+
)
with CO2 further contributing to higher
activity.[2–6,16,19] Therefore, utilizing CO2 as an oxidant
for the dehydrogenation of EB can enrich higher conver-
sion and better ST yield.
3.2 | Catalytic activity
3.3 | Catalyst deactivation
The catalytic results of V/Al-MCM-41 catalysts with dif-
ferent V loadings for CO2-ODEB at 550 C are summa-
ꢀ
To gain further insights into the deactivation behavior of
the supported vanadia catalysts and the role of incorpo-
rating Al species in support modification, a comparative
study of the 6 V/Al-MCM-41 and 6 V/MCM-41 catalysts
for CO2-ODEB was carried out. As portrayed in Figure 8,
it is clearly noticed that the incorporation of aluminum
into the MCM-41 support has a profound effect on the
activity and stability of catalysts. The EB conversion over
the 6 V/Al-MCM-41 catalyst is higher than that on 6 V/
MCM-41, which can be mainly attributed to the higher
dispersion of VOx species achieved on the surface of the
Al-MCM-41 support, as confirmed by the wide-range
XRD (Figure 4) and UV-Raman (Figure 5) characteriza-
tion results. In addition, compared with the 6 V/MCM-41
catalyst, 6 V/Al-MCM-41 catalyst possesses more L and B
acid sites (Figure 6), which are also favorable to EB acti-
vation.[26] It is worth noting that the stability of the 6 V/
Al-MCM-41 catalyst is much better than that of the 6 V/
rized in Table 2. All of the V/Al-MCM-41 catalysts show
a high selectivity toward ST (>97%); however, their
TABLE 2 Activity of V/Al-MCM-41 catalysts for CO2-ODEB at
550 ꢀC after 1 hr on-stream
EB
ST
conversion
(%)
selectivity
(%)
ST yield
(%)
Catalyst
Al-MCM-41
2.9
71.5
97.9
97.8
98.3
98.0
98.5
2.1
2 V/Al-MCM-41
4 V/Al-MCM-41
6 V/Al-MCM-41
43.8
51.5
56.9
42.9
50.4
55.9
44.3
51.5
6 V/Al-MCM-41a 45.2
8 V/Al-MCM-41 52.3
aUnder an inert N2 atmosphere.

AL-MCM-41-SUPPORTED VANADIA CATALYSTS FOR CO2-ODEB
7 of 10
shown in Figure 9, it can be seen that both fresh catalysts
6 V/Al-MCM-41 and 6 V/MCM-41 present the absorption
bands at 260 and 380 nm, which are attributed to isolated
tetrahedral V5+ species and polymeric V–O–V spe-
cies,[3,43,47] respectively. Noticeably, for the used 6 V/Al-
MCM-41 catalyst, these absorptions are only weakened
slightly; however, for the used 6 V/MCM-41 catalyst, a
weak and broad absorption band at 600–700 nm appears,
pointing to the significant reduction of the V species dur-
ing the reaction.[3] Figure 10 shows the small-angle and
wide-angle XRD patterns of the used 6 V/Al-MCM-41
and 6 V/MCM-41 catalysts. As shown in Figure 10a, the
three
characteristic
reflections
of
hexagonal
mesostructure are still observed for both used 6 V/Al-
MCM-41 and 6 V/MCM-41 catalysts, indicating that their
ordered mesostructures well remained after 20 hr of reac-
tion. The textural properties of the used catalysts are
summarized in Table 3. The 6 V/Al-MCM-41 and 6 V/
MCM-41 catalysts still exhibit high surface area and large
pore volume after 20 hr of reaction. The wide-angle XRD
characterization (Figure 10b) demonstrates that the V2O5
diffraction lines disappear and the diffraction peaks
FIGURE 8 Ethylbenzene (EB) conversion (conv.) and styrene
(ST) selectivity (sel.) with time on-stream over 6 V/Al-MCM-41 and
6 V/MCM-41 catalysts for oxidative dehydrogenation of
ethylbenzene with CO2
MCM-41 catalyst. The relative deactivation rate (R),
defined as 100 × (X1 – X20)/X1, where X is the EB conver-
sion, was then calculated.[6] After 20 hr of reaction, EB
conversion dramatically decreases from 48.7% to 31.1%
for the 6 V/MCM-41 catalyst (R is 36%). By contrast, EB
conversion decreases from 56.9% to 47.2% for the 6 V/Al-
MCM-41 catalyst (R is only 17%). This strongly indicates
that the incorporation of aluminum into the MCM-41
support can remarkably enhance the stability of the
supported vanadia catalyst toward CO2-ODEB. To gain
more information on the stability of the catalysts, the
fresh and used catalysts 6 V/Al-MCM-41 and 6 V/MCM-
41 were characterized by different spectroscopic tech-
niques. From the diffuse reflectance UV–Vis spectra
FIGURE 9 Diffuse reflectance UV–vis spectra of fresh and
used catalysts for oxidative dehydrogenation of ethylbenzene with
CO2
FIGURE 10 (a) Small-angle X-ray diffraction (XRD) and (b)
wide-angle XRD patterns of the used catalysts for oxidative
dehydrogenation of ethylbenzene with CO2

8 of 10
CHEN ET AL.
TABLE 3 The textural properties and coke content of
6 V/Al-MCM-41 and 6 V/MCM-41 catalysts after 20 hr on-stream
for CO2-ODEB
Compared with the MCM-41 support, the Al-MCM-41
support can effectively stabilize the active V5+ species
(Figure 9) and suppress the deep reduction of V5+ into
V3+ (Figure 10b), due to the stronger interaction of VOx
species with support, which can be responsible for the
significantly enhanced stability of the 6 V/Al-MCM-41
catalyst during CO2-ODEB.
SBET
(m2/
g)
Pore
volume
(cm3/g)
Pore
diameter
(nm)
Coke
content
(wt.%)
Catalyst
6 V/Al-MCM-41 298
6 V/MCM-41 382
0.38
0.46
2.51
2.63
21.4
14.7
In general, the coke deposition is inevitable and con-
tributes to the catalyst deactivation during CO2-ODEB.
However, the effects of the coke deposition on catalytic
behavior still are not completely clear.[26] To reveal the
effect of the coke deposition on the catalyst stability, the
coke deposited on the 6 V/Al-MCM-41 and 6 V/MCM-41
catalysts was determined by TGA. As listed in Table 3,
after 20 hr of reaction, the total amount of coke deposited
on the catalysts is 21.4 wt.% for 6 V/Al-MCM-41 and
14.7 wt.% for 6 V/MCM-41. Although the coke content
over the 6 V/Al-MCM-41 catalyst is much higher than
that over the 6 V/MCM-41 catalyst, the 6 V/Al-MCM-41
catalyst exhibits higher stability during CO2-ODEB as
compared with 6 V/MCM-41 (Figure 8), suggesting that
more coke may preferentially locate on the surface of Al-
MCM-41 rather than on the active VOx species.[50] These
results indicate that the coke deposition plays a less
important role for the deactivation of the supported van-
adia catalyst. Similar results have also been reported by
Ren et al.[27] and Deng et al.[50] for the dehydrogenation
of EB.
corresponding to V2O3 are observed for 6 V/MCM-41
after 20 hr of reaction, indicating a deep reduction of V5+
into V3+ over the 6 V/MCM-41 catalyst during CO2-
ODEB. By contrast, for the used 6 V/Al-MCM-41 catalyst,
no diffraction peaks assigned to any vanadium species
are detected, which is similar to the fresh 6 V/Al-MCM-
41 catalyst (Figure 4). This suggests that the surface vana-
dium species over the 6 V/Al-MCM-41 catalyst can be
kept at a higher oxidation state against the deep reduc-
tion into V3+ during the dehydrogenation reaction.
The highly dispersed V5+ species are considered to be
active sites in CO2-ODEB and the dehydrogenation reac-
tion may proceed via the redox mechanism illustrated in
Scheme 1.[3–6,16–20] The active lattice oxygen of vanadium
oxide abstracts hydrogen from EB to produce ST and
H2O. At the same time, V5+ is reduced to lower oxidation
states (V4+ and V3+). The reduced V species are
reoxidized to V5+ with CO2 to complete the full cycle.
The reduced V3+ species are less active and hardly
reoxidized into a high oxidation state (V5+) by CO2.[3,4,6]
Thus, the redox cycle between V5+ and V4+ is crucial for
the dehydrogenation reaction (Scheme 1).[4,19] As con-
firmed in Figure 7, the reducibility of the 6 V/MCM-41
catalyst is higher than that of the 6 V/Al-MCM-41 cata-
lyst due to the weaker metal oxide–support interaction.
However, the reoxidation rate of reduced V species with
CO2 is slow, because the oxidation capability of CO2 is
relatively weak.[8,15] Thus, the reduction of the V5+ spe-
cies proceeds faster than the reoxidation of the reduced V
species. As a result, V5+ is continuously reduced to V3+
(Figure 10b), which results in the severe deactivation of
the 6 V/MCM-41 catalyst during CO2-ODEB (Figure 8).
4 | CONCLUSIONS
Highly dispersed vanadia species supported on meso-
porous Al-MCM-41 (V/Al-MCM-41) have been found to
exhibit a highly efficient catalytic performance for CO2-
ODEB. The catalytic activity as well as the nature of VOx
species on Al-MCM-41 is related to the V loading. Cata-
lytic activity increases with increasing V loading and
reaches a maximum value at V loading of 6 wt.%, because
the surface of Al-MCM-41 support is predominantly cov-
ered by isolated and/or low polymeric VOx species in
such V loading range. With further increase of the V load-
ing, the catalytic activity decreases as a result of the loss
of accessibility caused by the formation of V2O5 crystal-
lites. In addition, with the optimal V loading of 6 wt.%,
the 6 V/Al-MCM-41 catalyst shows a much higher activ-
ity and stability than the 6 V/MCM-41 catalyst. The high
dispersion of VOx species achievable on the Al-MCM-41
support and the highly dispersed V5+ active species stabi-
lized by strong interaction with the support are consid-
ered to be responsible for the superior catalytic
performance of the 6 V/Al-MCM-41 catalyst. The main
SCHEME 1 Redox mechanism of supported vanadia catalyst
for oxidative dehydrogenation of ethylbenzene (EB) with CO2. ST,
styrene

AL-MCM-41-SUPPORTED VANADIA CATALYSTS FOR CO2-ODEB
9 of 10
[18] M. A. Betiha, A. M. Rabie, A. M. Elfadly, F. Z. Yehia, Micropor.
Mesopor. Mater. 2016, 222, 44.
[19] S. W. Chen, Z. F. Qin, G. F. Wang, D. Mei, J. G. Wang, Fuel
2013, 109, 43.
[20] H. X. Zhang, S. W. Chen, X. Y. Cui, D. H. Pan, Z. F. Qin, J.
G. Wang, Acta Phys.-Chim. Sin. 2014, 30, 351.
[21] N. Jiang, A. Burri, S. E. Park, Chin. J. Catal. 2016, 37, 3.
[22] B. M. Reddy, D. S. Han, N. Jiang, S. E. Park, Catal. Surv. Jpn.
2008, 12, 56.
reason for the deactivation of the supported vanadia cata-
lysts is ascribed to the deep reduction of V5+ into V3+
,
while the coke deposition has a less important impact on
the catalyst stability. The work provides a new route for
using heteroatoms-modified MCM-41 support to enhance
the catalyst stability. The information from this work
may provide new insights for designing more efficient
catalyst systems for CO2-ODEB.
[23] H. X. Fan, J. Feng, W. Y. Li, X. H. Li, T. Wiltowski, Q. F. Ge,
Appl. Surf. Sci. 2018, 427, 973.
[24] L. Zhang, Z. Wu, N. C. Nelson, A. D. Sadow, I. I. Slowing, S.
H. Overbury, ACS Catal. 2015, 5, 6426.
[25] M. Kovacevic, S. Agarwal, B. L. Mojet, J. G. van Ommen,
L. Lefferts, Appl. Catal. A 2015, 505, 354.
[26] I. Kainthla, G. V. R. Babu, J. T. Bhanushali, K. S. R. Rao, B.
M. Nagaraja, J. CO2 Util. 2017, 18, 309.
[27] K. Ren, J. Song, Y. H. Song, H. Wang, Z. H. Liu, Z. T. Liu, J.
Q. Jiang, Z. W. Liu, J. CO2 Util. 2017, 22, 63.
[28] G. Yun, S. Ahn, A. Takagaki, R. Kikuchi, S. T. Oyama, Catal.
Today 2019, 323, 54.
[29] W. Kai, H. Qian, D. B. Liu, Z. W. Ye, Mol. Catal. 2019, 462, 85.
[30] F. J. Méndez, O. E. Franco-López, X. Bokhimi, D. A. Solís-
Casados, L. Escobar-Alarcón, T. E. Klimova, Appl Catal B
2017, 219, 479.
[31] A. P. Wight, M. E. Davis, Chem. Rev. 2002, 102, 3589.
[32] Y. Ohishi, T. Kawabata, T. Shishido, K. Takaki, Q. Zhang,
Y. Wang, K. Takehira, J. Mol. Catal. A 2005, 230, 49.
[33] C. Perego, R. Millini, Chem. Soc. Rev. 2013, 42, 3956.
[34] A. Gannouni, X. Rozanska, B. Albela, M. S. Zina, F. Delbecq,
L. Bonneviot, A. Ghorbel, J. Catal. 2012, 289, 227.
[35] N. Vardast, M. Haghighi, S. Dehghani, Renewable Energy 2019,
132, 979.
ACKNOWLEDGEMENTS
The authors are grateful for the financial support by the
National Natural Science Foundation of China
(21975174), the Key Research and Development Program
of Shanxi Province, China (201803D31034), the Founda-
tion of Sate Key Laboratory of Coal Conversion (J18-19-
906), the Natural Science Foundation of Shanxi Province,
China (201801D121057), International Cooperation Pro-
ject (2015081048), and Shanxi Provincial Key Innovative
Research Team in Science and Technology (No.
2014131006).
ORCID
Shuwei Chen https://orcid.org/0000-0003-0011-1627
REFERENCES
[1] E. H. Lee, Catal. Rev. 1973, 8, 285.
[2] F. Cavani, F. Trifiro, Appl. Catal. A 1995, 133, 219.
[3] Y. Sakurai, T. Suzaki, K. Nakagawa, N. Ikenaga, H. Aota,
T. Suzuki, J. Catal. 2002, 209, 16.
[36] Z. Derikvand, F. Rahmati, A. Azadbakht, Appl. Organomet.
Chem. 2019, e4864. https://doi.org/10.1002/aoc.4864
[37] N. T. T. Tran, Y. Uemura, S. Chowdhury, A. Ramli, Appl.
Catal. A 2016, 512, 93.
[4] J. S. Chang, V. P. Vislovskiy, M. S. Park, D. Y. Hong, J. S. Yoo,
S. E. Park, Green Chem. 2003, 5, 587.
[5] K. N. Rao, B. M. Reddy, B. Abhishek, Y. Seo, N. Jiang, S. Park,
Appl Catal B 2009, 91, 649.
[38] P. Wu, L. D. Li, Q. Yu, G. J. Wu, N. J. Guan, Catal. Today
2010, 158, 228.
[39] R. Locus, D. Verboekend, R. Zhong, K. Houthoofd,
T. Jaumann, S. Oswald, L. Giebeler, G. Baron, B. F. Sels,
Chem. Mater. 2016, 28, 7731.
[40] S. P. Naik, V. Bui, T. Ryu, J. D. Miller, W. Zmierczak, Appl.
Catal. A 2010, 381, 183.
[41] Y. M. Liu, Y. Cao, K. K. Zhu, S. R. Yan, W. L. Dai, H. Y. He, K.
N. Fan, Chem. Commun. 2002, 22, 2832.
[42] Y. T. Chua, P. C. Stair, I. E. Wachs, J. Phys. Chem. B 2001, 105,
8600.
[43] Y. M. Liu, Y. Cao, N. Yi, W. L. Feng, W. L. Dai, S. R. Yan, H.
Y. He, K. N. Fan, J. Catal. 2004, 224, 417.
[44] T. Kataoka, J. A. Dumesic, J. Catal. 1988, 112, 66.
[45] B. Dragoi, E. Dumitriu, C. Guimon, A. Auroux, Micropor.
Mesopor. Mater. 2009, 121, 7.
[6] Z. W. Liu, C. Wang, W. B. Fan, Z. T. Liu, Q. Q. Hao, X. Long,
J. Lu, J. G. Wang, Z. F. Qin, D. S. Su, ChemSusChem 2011,
4, 1.
[7] I. Kainthla, G. V. R. Babu, J. T. Bhanushali, R. S. Keri, K. S.
R. Rao, B. M. Nagaraj, New J. Chem. 2017, 41, 4173.
[8] D. Mukherjee, S. Park, B. M. Reddy, J. CO2 Util. 2016, 16, 301.
[9] V. R. B. Gurram, S. S. Enumula, K. S. Koppadi, R. R. Chada,
D. R. Burri, S. R. R. Kamaraju, Cat. Com. 2019, 118, 1.
[10] N. A. Ferreira, J. M. Filho, A. C. Oliveira, RSC Adv. 2015, 5,
20900.
[11] T. Wang, L. Qi, H. Lu, M. Ji, J. CO2 Util. 2017, 17, 162.
[12] M. D. C. Rangel, A. P. D. M. Monteiro, S. G. Marchetti, S.
B. Lima, M. D. S. Ramos, J. Mol. Catal. A 2014, 387, 147.
[13] N. Mimura, M. Saito, Appl. Organomet. Chem. 2000, 14, 773.
[14] B. S. Liu, G. Rui, R. Z. Chang, C. T. Au, Appl. Catal. A 2008,
335, 88.
[46] Y. Li, D. H. Pan, C. Z. Yu, Y. Fan, X. J. Bao, J. Catal. 2012,
286, 124.
[15] H. Fan, J. Feng, X. Li, Y. Guo, W. Li, K. Xie, Chem. Eng. Sci.
2015, 135, 403.
[47] B. Solsona, T. Blasco, J. L. Nieto, M. Pena, F. Rey, A. V. Moya,
J. Catal. 2001, 203, 443.
[16] Y. Qiao, C. Miao, Y. Yue, Z. Xie, W. Yang, W. Hua, Z. Gao,
Micropor. Mesopor. Mater. 2009, 119, 150.
[48] E. P. Reddy, R. S. Varma, J. Catal. 2004, 221, 93.
[49] L. Y. Lizama, T. E. Klimova, J. Mater. Sci. 2009, 44, 6617.
[17] C. Zhu, S. W. Chen, D. H. Pan, X. Y. Cui, Y. Qiao, R. F. Li,
Cat. Com. 2018, 115, 12.

10 of 10
CHEN ET AL.
[50] L. Deng, T. Arakawa, T. Ohkubo, H. Miura, T. Shishido,
S. Hosokawa, K. Teramura, T. Tanaka, Ind. Eng. Chem. Res.
2017, 56, 7160.
How to cite this article: Chen S, Xu Z, Tan D,
et al. Oxidative dehydrogenation of ethylbenzene
to styrene with CO2 over Al-MCM-41-supported
vanadia catalysts. Appl Organometal Chem. 2019;
e5396. https://doi.org/10.1002/aoc.5396