Molecular oxygen-assisted oxidative dehydrogenation of ethylbenzene to styrene with nanocrystalline Ti1-xVxO2

Article abstract of DOI:10.1039/c1gc15907k

Oxidative dehydrogenation of ethylbenzene to styrene has been studied with vanadium-incorporated mesoporous nanocrystalline titania (Ti_1-xV _xO₂) and molecular oxygen between 440 and 530 °C. Incorporation of V in TiO_2<

Full text of DOI:10.1039/c1gc15907k

View Article Online / Journal Homepage / Table of Contents for this issue
EvaluaGtion Ornely. eCrenated wCith Ahspoese.mPDF.iCsoptyrrighyt 2002-2021 Aspose Pty Ltd.
Dynamic Article Links
Cite this: Green Chem., 2012, 14, 461
www.rsc.org/greenchem
PAPER
Molecular oxygen-assisted oxidative dehydrogenation of ethylbenzene to
styrene with nanocrystalline Ti1-xVxO2
Kumarsrinivasan Sivaranjani,a Akrati Vermaa and Chinnakonda S. Gopinath*a,b
Received 26th July 2011, Accepted 18th November 2011
DOI: 10.1039/c1gc15907k
Oxidative dehydrogenation of ethylbenzene to styrene has been studied with
vanadium-incorporated mesoporous nanocrystalline titania (Ti1-xVxO2) and molecular oxygen
between 440 and 530 ◦C. Incorporation of V in TiO2 lattice framework has been achieved by
simple solution combustion method. Incorporation of V in TiO2 lattice has been confirmed by
X-ray diffraction, XPS and Raman spectra and other physicochemical analysis. High ethyl
benzene conversion and stable styrene yield has been observed with 10% V-containing rutile phase
titania at 500 ◦C. However, stable but relatively lower styrene yield has been observed with 2 and
5% V-containing catalysts between 440 and 500 ◦C. Highest selectivity is observed with lower
vanadium content. Comparable activity has been observed under similar experimental conditions
with four times higher air-flow than that of O2. In order to understand the structure activity
relationship, spent catalysts were analyzed by all physico-chemical methods. Although there is a
phase change from anatase to rutile Ti1-xVxO2 within 1 h of reaction, higher activity is primarily
attributed to the ionic V5+ in Ti1-xVxO2 lattice, which prevents agglomeration to V2O5. It is to be
underscored the reactivity is retained at the cost of textural properties and phase change from
anatase to rutile, which is essential for the reaction.
Exothermicity also allows the heat energy to be tapped for other
Introduction
applications. Design of new catalysts which gives reasonable
Oxidative dehydrogenation (ODH) of ethylbenzene (EB) to
conversion with maximum selectivity for this ODH reaction is
styrene is one of the commercially important reactions. Styrene
is mainly used in the production of polystyrene and several
copolymers. At present, bulk styrene monomer is produced
one of the challenges in the field of heterogeneous catalysis. Shiju
et al. achieved around 45% styrene yield with VOx/Al2O3 catalyst
using N2O as the oxidant at 500 ◦C.4 Schlo¨gl et al. synthesized
through dehydrogenation of EB on Fe–K–Cr oxide-based cata-
onion-like carbon material and explored the same for ODH
lysts with superheated steam at 700 ◦C.1 This dehydrogenation
of EB at 520 ◦C using O2 and achieved around 62% styrene
is endothermic in nature (DH = 124.85 kJ mol-1), and highly
energy intensive process.1 Although hydrogen is available as a
side product, steam-based processes utilize a large amount of
latent heat and they pose thermodynamic limitations. Also coke
formation on the catalyst leads to severe catalyst deactivation.
Moreover, EB conversion of less than 50% was achieved per
pass. Hence there are constant efforts to develop alternate
catalysts to produce styrene, especially by ODH.2,3 Unlike
dehydrogenation, ODH is an exothermic reaction and hence
the reactions can be carried out at relatively lower temperatures
than the conventional endothermic dehydrogenation reactions.
yield with 68% selectivity.5 Drago et al. also developed novel
carbon-based catalysts and◦showed high styrene yield (72%)
at lower temperatures (350 C).6 Kustrowski et al. used many
different catalytic systems for ODH of EB reaction.7 They have
also synthesised VOx supported on SBA-15 and performed EB
ODH by using N2O as an oxidant.8 Recently Liu et al. demon-
strated around 60% EB conversion and high selectivity with
V2O5/Ce0.6Zr0.4O2-Al2O3 using CO2 as an oxidant at 550 ◦C.9
Supported vanadium oxide catalysts are widely studied for
several oxidation and ODH reactions.10–16 Among all catalysts,
VOx/TiO2 is a promising catalyst which shows high activity for
oxidation of o-xylene, toluene, methanol, lactic acid as well as in
ODH of alkanes.17–25 In these reactions, activity mainly depends
on the type of vanadia species and the nature of the support
used for the reaction.26–28 Titania is an excellent heterogeneous
catalyst-support material; the major limitation of conventional
titania to act as a good support is its low surface area and
its easy conversion between anatase to rutile. If titania can be
aCatalysis Division, National Chemical Laboratory, Dr Homi Bhabha
Road, Pune, 411 008, India. E-mail: cs.gopinath@ncl.res.in;
Web: http://www.ncl.org.in/csgopinath; Fax: +91-20-2590 2633;
Tel: +91-20-2590 2043
bCentre of Excellence on Surface Science, National Chemical
Laboratory, Dr Homi Bhabha Road, Pune, 411 008, India
This journal is
The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 461–471 | 461
©

View Article Online
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
stabilised in the thermodynamically stable rutile phase, it would
be advantageous to exothermic ODH reactions.29 In general,
ordered mesoporous materials show high surface area and the
majority of active sites are present on the surface of porous
channels. Hence the reactants have to diffuse further in order
to interact with these active sites. However, in contrast to the
above, disordered wormhole mesoporous materials30,31 exhibit
smaller diffusion lengths due to the pseudo-three-dimensional
(p3D) nature of the pores compared to regular mesopores.32
Therefore the reactants and products can easily diffuse in/out
of the active sites, which increases the selectivity and yield of the
desired product and hence the rate of the reaction.
EB to styrene ODH is, generally, carried out at high tem-
perature (>500 ◦C) on supported catalysts, which leads to
segregation and agglomeration of active phase. Especially, this
is true with V-based catalysts. It is also well known that isolated
VOx species are active for partial oxidation and polymeric (VOx)n
species totally oxidises organic moieties and compounds to
CO2.4 This can be avoided by introducing V into the lattice
framework of oxide support material. Lattice incorporation of
vanadium has more advantages and it is unlikely to segregate
out from the TiO2 lattice as a result of high temperature or
exothermic nature of ODH reaction. Further, this will increase
the monomeric or isolated VOx species in the lattice, which in
turn will enhance the activity of ODH reactions. Indeed this
concept has not been proposed and evaluated for any ODH
reactions. Therefore a simple method to prepare crystalline
mesoporous supports with high surface area along with lattice
incorporation of active species is of high relevance.
In the present investigations, we follow a simple solution
combustion method (SCM) in order to meet most of the above
important criteria. SCM is an energy efficient simple process.
Short reaction time and cheap starting materials are the major
advantages of this method.32–35 For the first time we have
studied the effect of incorporation of vanadium ions in the
titania (Ti1-xVxO2) lattice, along with wormhole mesoporosity
for the ODH of EB at temperature £530 ◦C with O2 as oxidant.
Prepared catalysts were analyzed by various physico-chemical
methods and structure–activity relationship has been deduced.
The present report is a part of our ongoing investigations from
our group towards comprehensive understanding of metal oxide
catalysts for different heterogeneous catalytic reactions.36
content from 2 to 15 atom%. Here we have used urea as a fuel,
to avoid any carbon impurities. All of the prepared materials
were characterized by Energy dispersive X-ray (EDX) analysis,
X-ray diffraction (XRD), Raman, N2 adsorption-desorption
isotherm, Fourier transform infrared (FT-IR) spectroscopy,
high resolution transmission electron microscopy (HRTEM),
and X-ray photoelectron spectroscopy (XPS).
Powder X-ray diffraction (PXRD) data of Ti1-xVxO2 materials
was collected from PANalytical X’pert Pro dual goniometer
diffractometer. A proportional counter detector was used for
low angle experiments. The data were collected with a step size
of 0.02◦ and a scan rate of 0.5◦ min-1. The sample was rotated
throughout the scan for better counting statistics. The radiation
˚
used was Cu-Ka (1.5418 A) with Ni filter and the data collection
was carried out using a flat holder in Bragg–Brentango geometry
(0.20). EDX measurements were performed on an SEM system
(FEI, Model Quanta 200 3D) equipped with EDX attachment.
EDX spectra were recorded in the spot-profile mode by focusing
the electron beam onto specific regions of the sample. Nitrogen
adsorption/desorption isotherms for the materials were col-
lected from Quantachrome Autosorb Automated gas sorption
system (NOVA 1200).37 The Brunauer–Emmett–Teller (BET)
equation was used to calculate the surface area from the adsorp-
tion branch. The pore size distribution was calculated by ana-
lyzing the adsorption branch of the nitrogen sorption isotherm
using BJH method. A FEI TECNAI 3010 electron microscope
˚
operating at 300 kV (Cs = 0.6 mm, resolution 1.7 A) was used for
recording HRTEM of all materials. Samples were crushed and
dispersed in isopropanol before depositing onto a holey carbon
grid. Raman spectra were recorded on a Horiba JY LabRAM
HR 800 Raman spectrometer coupled with microscope in
reflectance mode with 633 nm excitation laser source and a
spectral resolution of 0.3 cm-1. XPS spectra were recorded on a
VG Microtech Multilab ESCA 3000 spectrometer equipped with
non-monochromatised Mg-Ka X-ray source (hn = 1253.6 eV).37
Catalytic activity
Vapour phase oxidative dehydrogenation of EB was carried out
at atmospheric pressure in a fixed bed, vertical downflow, quartz
reactor placed inside a double zone furnace (Geomechanique,
France).36,37 Fresh catalyst (1.0 g) with particle size up to 10
mesh was charged each time in the center of the reactor in such
a way that the catalyst was sandwiched between the layers of inert
glass beads. The reactant (EB) was fed using a syringe pump at a
weight hourly space velocity (WHSV) of 2. Oxygen is used as an
oxidant. 40 ml min-1 is fixed as an optimum flow rate of oxygen.
We deliberately used molecular O2 in order to increase the reac-
tion rate, minimize coke deposition as well as to inhibit catalyst
deactivation. Reaction products were collected at ice cold condi-
tions from a condenser fixed below the reactor and analyzed by
an Agilent Gas Chromatograph (19091J-413) containing an HP
5.5% phenyl methyl siloxane column equipped with a flame ion-
ization detector. Products were also analyzed by using GC-MS.
Experimental section
All the chemicals employed were of analytical grade and used
as such without any further purification. Titanyl nitrate (Spec-
trochem) as Ti precursor, ammonium meta vanadate (Sigma–
Aldrich) as vanadium precursor and urea (Merck) as a fuel were
used as such. All TVx materials were prepared with 1 : 1 molar
ratio of urea to metal ions (Ti + V). x in TVx indicating the
nominal vanadium atom percentage. Aqueous solution of the
desired amount of titanyl nitrate, ammonium meta vanadate
and urea were taken in a 250 ml beaker and introduced into
a muffle furnace maintained at 400 ◦C. Water evaporates in
the first few minutes followed by smoldering type combustion
that occurs and continues for about 1 min. Solid products were
obtained within 15 min of total preparation time. A series of
catalysts were prepared by changing the nominal vanadium
Results and discussion
Structural and textural features
PXRD measurements were carried out in order to explore the
structural features of the vanadium incorporated mesoporous
462 | Green Chem., 2012, 14, 461–471
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Table 1 Physicochemical properties of Ti1-xVxO2 catalysts
Bulk V- BET surface Pore size Pore volume Crystallite
Catalyst content area (m2 g-1) (nm)
(cc/g)
size (nm)
TiO2
TV5
TV10
TV15
0.0
4.9
6.9
10.5
160
195
210
140
3.9
4.7
4.9
5.6
0.18
0.25
0.26
0.20
15.3
12.9
9.8
10.3
titania Ti1-xVxO2 (TVx) materials. In TVx, x stands for nominal
V atom%, which is higher than the actual V-content due to
the evaporation loss of some V during solution combustion
synthesis. Fig. 1 shows XRD pattern of as prepared TVx
materials from low angle to wide angle. All TVx materials were
prepared with 1 : 1 urea to oxidizer ratio. PXRD pattern of TVx
materials could be indexed to anatase phase of titania with
small amount of rutile phase (JCPDS file 21-1272, 21-1276).
It is clearly observed that the rutile content increases from 6%
for TiO2 with increasing V-content to 11% for TV15. It is well
known that vanadium ion incorporation enhances the formation
of thermodynamically stable rutile phase.38,39 Very small amount
of brookite phase is also observed with high V-content (≥10%).
However, there is no vanadium oxide peak observed even at 15%
of vanadium loading. This observation hints that vanadium ions
are likely to occupy the lattice position of titania. Although
no VOx, particularly V2O5, has been observed in PXRD, it
cannot be ruled out a highly dispersed vanadium oxide on
titania.40 Nonetheless, it is to be noted that Feng et al.41 observed
V2O5 phase even at 7% loading itself. No V2O5 formation
in TVx indicates the potential of SCM to introduce V into
the lattice. A broad peak confirms the formation of nanosize
crystalline particles. Crystallite size was calculated by using
Debye–Scherrer equation and shown in Table 1. There is no peak
shift observed after vanadium incorporation in TiO2. However
broad XRD features, due to nanocrystallite size, does not show
any shift due to V-introduction in the TiO2 lattice.
Fig. 1 Low and wide angle XRD pattern of Ti1-xVxO2 (TVx) materials.
In Fig. 1, a single diffraction peak is observed around 0.8◦
demonstrates the mesoporous nature of the Ti1-xVxO2 materials.
Unlike ordered mesoporous materials, such as, SBA-15 and
MCM-48, all TVx materials exhibit only one peak around 0.8◦
without any extra peaks.42,43 This highlights the presence of
disordered mesoporosity30,31 for all TVx materials. Increasing
V-content affects the mesoporosity of the materials. An increase
in surface area and pore volume (Table 1) to be noted up
to V = 10% compared to TiO2 and then a marginal decrease
in surface area is observed for TV15. Nonetheless, it is clear
that all TVx materials exhibit mesoporosity along with high
crystallinity.
EDX analysis has been carried out to measure the material
composition as well as to find out the extent of homogeneity of
the material. Representative results are given in Fig. 2 for TV5
(a and b), and TV10 (c and d). Elemental mapping have been
carried out and shown with colour coding for different elements.
High intense and low intense (or diffused) color indicates the
high and low content of corresponding elements, respectively,
in a particular area/spot. Fig. 2b and d shows the vanadium
and titanium mapping of the above materials to show the
homogeneous distribution over a large area (5.2 mm2). A glance
at TV5 (Fig. 2a) and TV10 (Fig. 2c) highlights an increase in V
Fig. 2 Elemental mapping of (a and b) TV5 (c and d) TV10 (Ti – red,
V – green, O – blue) measured by EDX over an area of 2 ¥ 2.6 mm2.
content from the former to the latter. An increase in V content
on TV10 compared to TV5 is evident from Fig. 2. The above
results confirm the homogeneous distribution of, particularly,
vanadium at the microscopic level, indicating the effectiveness
of the preparation method. Bulk atom% values reported in
Table 1 are the average values obtained over large area of
>500 mm2.
N2 adsorption-desorption studies has been measured to
investigate the textural properties of the Ti1-xVxO2 materials.
Fig. 3 shows N2 adsorption-desorption results and Barret–
Joyner–Halenda (BJH) pore size distribution of TVx materials.
All materials show type IV adsorption-desorption isotherm
with H2 hysteresis loop which is typical for all mesoporous
materials.44,45 All TVx and TiO2 materials show narrow pore size
distribution. BET surface area of all materials were calculated
This journal is
The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 461–471 | 463
©

View Article Online
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
for TVx materials. This wormhole mesoporosity arises due
to intergrowth of fundamental particles and the same leads
to aggregates with significant extra framework void space.30–32
Selected-area electron diffraction (SAED) pattern confirms the
crystalline nature of the nanocrystalline TVx materials and
also it shows the anatase phase in both materials. HRTEM
image shows the majority of lattice fringes corresponding to
(101) (d(101) = 0.352 nm) crystallographic planes of anatase
phase. These observations are in excellent agreement with XRD
as well as with N2 adsorption isotherm results. Similar types
of images are already observed by Yu et al. for wormhole
type phosphated mesoporous titania prepared using co-block
polymers as a template.46 The wormhole mesoporous nature
has additional advantages like low diffusional barriers, since the
depth or length of mesopores are minimum to a few nanometres,
unlike several hundred nanometres in conventional ordered
mesoporous materials, like SBA-15. These types of mesopores
are known as p3D mesopores.32 This disordered p3D wormhole
like framework provides an easy route for the diffusion of
reactants due to less diffusion barriers.
Fig. 3 (a) N2 adsorption-desorption isotherms, and (b) BJH pore-size
distribution of TVx materials.
and shown in Table 1. Surface area increases as the V-content
increases, and TV10 shows the highest surface area (210 m2
g-1). It is interesting to note that an increase in the V-content
significantly increases the extent of mesoporosity, surface area
and the pore volume. This is attributed to a decrease in crystallite
size from 15 nm for TiO2 to 10 nm for TV10 and TV15. This
observation is in agreement with low angle XRD data. Unimodel
pore size distribution is observed with an average pore diameter
around 4.7 1 nm for all TVx materials.
Particle morphology and textural properties of TVx cata-
lysts has been studied by HRTEM carefully. Representative
HRTEM images recorded for TiO2, TV5 and TV10 materials
are shown in Fig. 4. Generally TVx exhibits spherical particles
and the size of the particles also decreases with increasing V-
content. A wormhole like mesoporous structure can be seen
Spectral analysis
Raman and FT-IR spectroscopy. Raman spectroscopy is a
versatile tool to determine the structural features of the oxide
materials. Fig. 5a shows Raman spectra of all TVx materials. In
the case of pure titania, all six Raman active fundamental modes
are observed at 145 (Eg), 198 (Eg), 398 (B1g), 516 (A1g + B1g), 640
(Eg) cm-1 for the anatase phase. Broad nature of these peaks
confirms the formation of nanosize particles. None of the four
Raman active modes (144 (B1g), 446 (Eg), 610 (A1g), 827 (B2g)
cm-1) of rutile titania was observed, possibly due to low rutile
content as well as very broad nature of the anatase features.
Raman spectra for V2O5 also included for better comparison.
None of the Raman features of V2O5 (285, 703, and 997 cm-1)
was observed in the Raman spectra underscores the absence
of bulk crystalline V2O5 phase in mesoporous TVx materials.
There is no peak that corresponds to monomeric or polymeric
vanadium species has been observed for TVx materials, unlike
Fig. 5 (a) Raman, and (b) FT-IR spectra of TVx materials. V2O5
spectral data given for comparison. None of the V2O5 features are
observed in TVx materials highlighting the absence of V2O5 and
suggesting the introduction of V into the titania framework.
Fig. 4 TEM image of (a and b) TiO2 (c) TV5 and (d) TV10. Similar
image as that of given in panel (a) is observed for all TVx materials.
464 | Green Chem., 2012, 14, 461–471
This journal is
The Royal Society of Chemistry 2012
©