Hexagonal mesoporous titanosilicates as support for vanadium oxide - Promising catalysts for the oxidative dehydrogenation of n-butane

Article abstract of DOI:10.1016/j.cattod.2012.07.028

The comparative study of structural properties and catalytic performance of V-containing high-surface mesoporous silica and mesoporous titanosilicate materials (HMS, Ti-HMS) in oxidative dehydrogenation of n-butane (C ₄-ODH) was carried out. The aim of the study was to investigate effect of different titanium amount incorporated into silica support on the texture, speciation of vanadium complexes and its impact on catalytic performance. Prepared catalysts were characterized by XRF for determination of vanadium content, DTA/TG for thermal stability of matrix, XRD, SEM and N ₂-adsorption for study of morphology and texture, FT-IR and DR UV-vis spectroscopy for verification of successful incorporation of Ti to the matrix and H₂-TPR and DR UV-vis spectroscopy for determination of vanadium complex speciation. All prepared materials were tested in n-butane ODH reaction at 460 °C. We conclude that titanium was successfully incorporated into mesoporous structure, which was preserved at least up to 600 °C. Catalytic activities of V-Ti-HMS catalysts were approximately four times higher than activity of V-HMS catalyst in spite of the fact that all samples exhibit the same amount of vanadium species with similar distribution. The selectivity to desired products was comparable for all catalysts. Enhanced catalytic activity of V-Ti-HMS materials allows activating of n-butane at significantly lower temperature (by 100 °C) compare with V-HMS materials.

Full text of DOI:10.1016/j.cattod.2012.07.028

Catalysis Today 204 (2013) 132–139
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Hexagonal mesoporous titanosilicates as support for vanadium oxide—Promising
catalysts for the oxidative dehydrogenation of n-butane
a
a
b
b
Michal Setnicˇkaa,∗, Pavel Cicˇmanec , Roman Bulánek , Arnosˇt Zukal , Jakub Pastva
ˇ
a Department of Physical Chemistry, University of Pardubice, Studentská 573, CZ532 10 Pardubice, Czech Republic
b J. Heyrovsky´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, v.v.i., Dolejsˇkova 2155/3, CZ182 23 Prague 8, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 May 2012
Received in revised form 17 July 2012
Accepted 23 July 2012
Available online 10 September 2012
The comparative study of structural properties and catalytic performance of V-containing high-surface
mesoporous silica and mesoporous titanosilicate materials (HMS, Ti-HMS) in oxidative dehydrogenation
of n-butane (C4-ODH) was carried out. The aim of the study was to investigate effect of different titanium
amount incorporated into silica support on the texture, speciation of vanadium complexes and its impact
on catalytic performance. Prepared catalysts were characterized by XRF for determination of vanadium
content, DTA/TG for thermal stability of matrix, XRD, SEM and N2-adsorption for study of morphology
and texture, FT-IR and DR UV–vis spectroscopy for verification of successful incorporation of Ti to the
matrix and H2-TPR and DR UV–vis spectroscopy for determination of vanadium complex speciation.
All prepared materials were tested in n-butane ODH reaction at 460 ◦C. We conclude that titanium was
successfully incorporated into mesoporous structure, which was preserved at least up to 600 ◦C. Catalytic
activities of V-Ti-HMS catalysts were approximately four times higher than activity of V-HMS catalyst in
spite of the fact that all samples exhibit the same amount of vanadium species with similar distribution.
The selectivity to desired products was comparable for all catalysts. Enhanced catalytic activity of V-Ti-
HMS materials allows activating of n-butane at significantly lower temperature (by 100 ◦C) compare with
V-HMS materials.
Keywords:
Mesoporous titanosilicate
Hexagonal mesoporous structure
Vanadium
Oxidative dehydrogenation
Butenes
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
their peculiar textural properties but also due to the possibility of
various modifications including different structural types and as
The great challenge of current chemical industry is function-
alization of cheap and abundant C2–C4 alkanes from crude oil to
corresponding olefins. For example, the oxidative dehydrogenation
(ODH) of light alkanes can be used for production of alkenes instead
of classically used dehydrogenation (DH) requiring very high tem-
peratures (high energy consumption) at which additional coking
and deactivation of catalyst normally occur [1–4].
Vanadium oxides are powerful redox catalysts in many indus-
trial processes and they are taken as catalysts in oxidation reactions
as well as for ODH [5,6]. However, they cannot be used in its bulk
form (it leads to nonselective reactions) [4] but have to be used as
the well dispersed VOX species anchored on the suitable support
[2,4,7–9]. Activity and selectivity of these catalysts strongly depend
on the degree of vanadium species dispersion [3,5,10,11], method
of catalyst preparation [3,10,11] and also the type of support
has a dramatic effect [4,7,11,12]. Suitable supports for anchoring
active species are mesoporous molecular sieves not only due to
well as different chemical compositions [13].
A large number of reviews [2,4,7,14–16] and papers dealing
with the ODH of light alkanes over vanadium supported materials
have been published since 1990s and most of them used different
structure of silica materials (e.g. silicalite [11] or mesoporous MCM-
41 [17,18], SBA-15 [9,19–22], SBA-16 [9] and HMS [3,10,23,24]).
The main advantages of silica mesoporous materials are: larger
surface area (a good dispersion of active particles), thermal and
hydrothermal stability and good mechanical properties [3,25,26].
Next advantage of catalysts anchored on silica support (compared
with Al2O3, TiO2, ZrO2, etc.) is that catalysts using silica supports
are more selective to desired products [27–29]. However, they
exhibit relatively low activity and C4-ODH productivity due to high
apparent activation energy of C H bonds [27,30]. The highest activ-
ities and sufficient selectivity in ODH reaction were attained using
VOX species supported on TiO2 (anatase) surface, which allows to
carry out the reaction at lower temperatures [8,11,28,31,32]. Nev-
ertheless, pure TiO2 support has also some drawbacks, such as a
relatively low specific surface area, which can be further reduced
by sintering as a consequence of thermal treatments. Low surface
area prevents dispersion of vanadium oxide species, which leads
∗
Corresponding author. Tel.: +420 46 603 73 45; fax: +420 46 603 70 68.
E-mail address: michal.setnicka@upce.cz (M. Setnicˇka).
0920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.07.028

M. Setnicˇka et al. / Catalysis Today 204 (2013) 132–139
133
to the further decrease in selectivity [3,11,31,32]. These drawbacks
prevent the use of TiO2 as conventional support for catalysts. One
possibility how to solve this problem is to prepare a mixed Si–Ti
support which combines suitable properties of both above men-
tioned SiO2 and TiO2 supports, respectively.
The vanadium oxo-species (1.5 wt.% of V) were introduced onto
the support by the wet impregnation method from EtOH solution
of vanadyl acetylacetonate. Impregnated samples were dried at
120 ◦C in air overnight and then calcined at 600 ◦C in air for 8 h
(with heating rate 5 ◦C/min).
Previous studies showed that coating of silica support with
anatase phase (impregnation or grafting by titanium alkoxide)
as one of the frequent methods for preparing TiO2/SiO2 [1,31],
TiO2/MCM-41 [28,32,33] and TiO2/SBA-15 [27,34] support. In this
case it was obtined thermostable support with high specific surface
(silica) and with good catalytic performance (TiO2). These materi-
als were doped by vanadium and obtained materials were studied
in C2-ODH [31,35], C3-ODH [1,27,31] and only in one case in C4-
ODH [11] reaction. The disadvantage of materials prepared in this
way is their complicated synthesis with multiple synthesis and
calcination steps. Moreover, these postsynthetic methods some-
times lead to the blocking of the channels by the formation of bulk
metal oxide clusters [36]. To avoid these problems, several attempts
have been made to incorporate titanium into the silica framework
directly. This problem could be overcome using direct synthesis
method as was published previously for Ti-SBA-15 [37,38], Ti-HMS
[37,39–41], Ti-MCM-41 [33,40,42] and Ti-MCM-48 [43] but accord-
ing to our knowledge these materials were not prepared in titanium
content higher than 9 wt.% [44].
In the present paper we report on one-pot synthesis of Ti-HMS
support with high content of titanium. In this case we obtain hexag-
onal mesoporous silica support with the isomorphously exchanged
titanium oxide species and they serve as an “anchor” for vana-
dium active species. The reason for this behavior is the difference
in the isoelectric point of TiO2 (IEP = 6–6.4) and SiO2 (IEP = 1–2)
supports. The acidic vanadium oxide species (IEP = 1.4) are pref-
erentially bonded to the more basic TiO2 [7]. Moreover, acid/base
properties of Ti species in the support influence the basicity of
the bridging oxygen in sup-O-V (sup = Si, Al, Ti, etc.) and there-
fore their reactivity [6,27]. The part of the surface composed of
silica positively affects selectivity to desired ODH products. We pre-
pared three HMS support with different content of titanium (0, 6
and 19 wt.%, respectively). These materials were investigated by
DTA/TG, FT-IR, N2-BET, SEM, XRD and DR UV–vis spectroscopy for
the physico-chemical characterization of the mesoporous support
structure. After impregnation of these matrices by vanadium we
verified preservation of mesoporous structure. We used DR UV–vis
spectroscopy and H2-TPR for investigation of vanadium dispersion
and for the study of catalytic activity we used ODH of n-butane
which is interesting for the industry as well as very suitable model
reaction at the same time [3,11].
The investigated samples were denoted as xTi-HMS, where x is
the titanium content in support in the weight percentage. V-xTi-
HMS is used for materials after impregnation of vanadium.
2.2. Catalysts characterization
The content of titanium was determined by means of ICP-OES
by Integra XL 2 (GBC Dandenog, Australia) for both prepared Ti-
HMS supports. The vanadium content was determined by means of
ED XRF by ElvaX (Elvatech, Ukraine) equipped with Pd anode [46].
Samples were measured against the model samples (a mechani-
cal mixture pure SiO2 and NaVO3) granulated to the same size as
catalysts.
The structure and crystallinity of catalysts were probed by scan-
ning electron microscopy (SEM) using JSM-5500LV microscope
(JEOL, Japan) and by X-ray diffraction (D8-advance diffractometer,
Bruker AXE, Germany) in the 2ꢀ range of 2–35◦ with Cu K␣ radiation
´
˚
(ꢁ = 1.5406 A).
Specific surface area and texture of investigated samples were
measured by means of nitrogen adsorption/desorption at temper-
ature of liquid nitrogen for verification of mesoporous structure by
using ASAP 2020 equipment (Micromeritics, USA). Prior to adsorp-
tion isotherm measurement, the samples were degassed at 300 ◦C
under turbomolecular pump vacuum for 8 h. The specific surface
area was calculated according to BET method. The mesopore vol-
ume was determined by DFT by using of “N2 @ 77K” model for
cylindrical pores and oxide surface.
Thermal stability of prepared matrix was studied using Jupiter
STA 449C (Netzsch, Germany) thermobalance. Around 40–50 mg of
matrix were heated in corundum TG–DTA-crucibles under a flow
of air at a heating rate of 10 ◦C/min up to 1200 ◦C and the ␣-Al2O3
was used as a reference material.
For verification of successful incorporation of titanium to the
HMS matrix the infrared spectra were collected on Nicolet 6700
FTIR spectrometer equipped with DTGS detector. Samples were
diluted using dry KBr, pressed into pellets and scanned in the range
1400–400 cm−1 with a resolution of 2 cm−1 (32 scans). The spec-
trum of blank KBr pellet was also measured to allow background
subtraction.
The UV–vis diffuse reflectance spectra of dehydrated diluted
samples were measured by using Cintra 303 spectrometer (GBC
Scientific Equipment, Australia) equipped with a Spectralon-coated
integrating sphere using a Spectralon coated discs as a stan-
dard. The spectra were recorded in the range of the wavelength
190–850 nm. The samples were diluted by the pure silica (Fumed
silica, Aldrich) in the ratio 1:100 for avoid spectra detection limits
overflow and to give better resolution of individual bands. All sam-
ples were granulated and sieved to fraction of size 0.25–0.5 mm,
dehydrated before the spectra measurement and oxidized in the
glass apparatus under static oxygen atmosphere in two steps:
120 ◦C for 30 min and 450 ◦C for 60 min and subsequently cooled
down to 250 ◦C and evacuated for 30 min. After the evacuation the
samples were transferred into the quartz optical cuvette 5 mm thick
and sealed under vacuum. Additional details about diluting and
measuring can be seen in Refs. [47,48]. The obtained reflectance
spectra were transformed into the dependencies of Kubelka–Munk
2. Experimental
2.1. Catalysts preparation
The hexagonal mesoporous silica (HMS) and hexagonal meso-
porous titanosilica (Ti-HMS), were synthesized under ambient
conditions according to the procedure reported by Tanev and
Pinnavaia [45] with their modification for Ti-HMS. In a typical
preparation, dodecylamine (DDA, Aldrich) as a neutral structure
directing agent was added to the mixture of ethanol and re-distilled
water. After 20 min of homogenization tetraethylorthosilicate
(TEOS, Aldrich) was added as silica precursor in the case of HMS
preparation or TEOS and tetraethylorthotitanate (TEOT, Aldrich)
as a titanium precursor was added simultaneously in the case of
Ti-HMS preparation. The reaction mixture was stirred at room tem-
perature for 18 h. The solid product was filtered, washed by ethanol
and calcined in air at 450 ◦C for 20 h (with heating rate 1 ◦C/min)
for the template removal.
function F(R ) on the absorption energy hꢂ using the equation:
∞
2
(1 − R
)
∞
∞
F(R ) =
(1)
∞
2R

134
M. Setnicˇka et al. / Catalysis Today 204 (2013) 132–139
Fig. 1. SEM images of mesoporous HMS and Ti-HMS supports.
where R is the measured diffuse reflectance from a semi-infinite
layer [49].
V2O5 clusters and typical orthorhombic needles were observed in
any sample (not shown here).
∞
Hydrogen temperature programmed reduction (H2-TPR) was
used for the study of redox properties and AutoChem 2920
(Micromeritics, USA) was used for the measuring. A 100 mg sam-
ple in a quartz U-tube micro reactor was oxidized in oxygen flow at
450 ◦C (for 2 h). The reduction was carried out from 35 ◦C to 850 ◦C
with a temperature gradient of 10 ◦C/min in flow of reducing gas
(5 vol.% H2 in Ar). The changes in hydrogen concentration were
monitored by online connected TCD detector.
XRD patterns of supports and catalysts show (see Fig. 2) charac-
teristic broad low-angle diffraction peak at 2ꢀ = 2–2.5◦ attributable
to a d1 0 0 diffraction typical for hexagonal lattice structure of meso-
porous materials [40,45]. The value of d1 0 0 spacing slightly changed
for individual support in the range from 3.3 to 3.9 nm in cor-
respondence with literature [42,50]. This result corresponds to
the changes in pore size distribution obtained from N2 adsorp-
tion/desorption isotherm NLDFT analysis (see inset in Fig. 3). In
addition, very broad peak with low intensity among 15–35◦ was
detected in the X-ray powder patterns of all samples. This sig-
nal is usually assigned to the presence of amorphous SiO2 wall
[51,52]. Similar patterns were reported for hexagonal mesoporous
materials in literature [10,40–42,50,53]. Absence of diffraction lines
belongs to TiO2 (anatase and/or rutile) crystallites or V2O5 crystal-
lites in the XRD patterns of all supports and catalysts confirm results
from SEM; titanium is incorporated into HMS framework and do not
form separated crystallites of TiO2 and vanadium species are finely
spread on the surface of supports without creation of oxide-like
clusters or separated V2O5 crystallites.
BET specific surface area of parent materials and catalysts are
summarized in Table 1. SBET of parent supports slightly decreases
with increasing content of titanium and ranges from 880 m2 g−1
for pure HMS to 800 m2 g−1 for 19Ti-HMS and these values are in
a good agreement with data published previously for similar type
of materials with lower content of Ti in matrix [37,40]. Such high
values of specific surface area together with type IV isotherms
with capillary condensation step and H4 hysteresis loop indicate
2.3. Catalytic tests
The n-butane ODH reaction was carried out in a glass plug-
flow fixed-bed reactor at atmospheric pressure in the kinetic
region (independently checked) and under steady state conditions
of reaction. Typically 400 mg of catalyst (grains 0.25–0.50 mm)
was diluted with 3 cm3 inert SiC to avoid the catalytic bed over-
heating. The catalysts were pre-treated in the oxygen flow at
450 ◦C for 2 h before each reaction run. The input feed composi-
tion was C4H10/O2/He = 10/10/80 vol.% – with a total flow rate of
100 cm3 min−1 STP. The catalytic activity was measured at 460 ◦C
under the steady state conditions. The analysis of reaction mixture
composition was made by on-line gas-chromatograph CHROM-5
(Laboratorní prˇístroje Praha) equipped with thermal conductivity
detector (TCD) and flame ionization detector (FID). The n-butane
and products of ODH reaction (butadiene, 1-butene, cis-2-butene,
trans-2-butene, propene and propane) were separated using a
packed column with n-octane on ResSil (Restek) at 20 ◦C. The
packed column Porapak Q (Supelco) was used for the analysis of
ethane, ethene and CO2. The molecular sieve 13 X (Supelco) was
used for the separation of permanent gases (O2, CO and traces of
N2). For details about calculation conversion, selectivity, yield and
productivity please see our previous work [3].
background
HMS (a)
V-HMS (b)
B
A
6Ti-HMS (c)
V-6Ti-HMS (d)
19Ti-HMS (e)
V-19Ti-HMS (f)
3. Results and discussion
3.1. Characterization of materials
SEM images of prepared supports (Fig. 1) show poorly defined
morphology of every studied sample. Particles exhibit uneven
shapes of sub-micrometer size. Size of particles of Ti-HMS supports
is significantly (approximately three-times) smaller than particles
of pure silica HMS support. Similar observation has been described
by Comite et al. [1] who prepared TiO2/SiO2 support by grafting
and they assign this behavior to modification in calcination step.
SEM–EDX mapping of Ti content led to the conclusion that Ti
is spread homogeneously in all parts of support and no TiO2
clusters were detected. Moreover mapping of vanadium content
in the catalysts prepared by impregnation of supports shows that
vanadium is distributed uniformly in all parts of catalysts and no
f
e
d
c
b
a
0
1
2
3
4
5 0
10
20
30
2 Θ,°
Fig. 2. (A) Low-angle diffraction patterns for parent supports and (B) X-ray diffrac-
tion patterns for parent supports and supports after impregnation by vanadium
(XRD patterns were offset for clarity).

M. Setnicˇka et al. / Catalysis Today 204 (2013) 132–139
135
Table 1
Chemical composition and results of physico-chemical characterization of investigated materials.
a
b
c
d
e
Sample
Ti(matrix) (wt.%)
Va (wt.%)
SBET (m2 g−1
)
VP (cm3 g−1
)
DME (nm)
Tonset (◦C)
Tmax (◦C)
ꢃe f
Eg g(eV)
HMS
V-HMS
6Ti-HMS
V-6Ti-HMS
19Ti-HMS
V-19Ti-HMS
0.00
0.00
6.0
6.0
12.0
12.0
0.0
1.5
0.0
1.5
0.0
1.5
880
650
890
690
800
770
0.560
–
0.532
–
0.495
–
2.4–3.1
–
2.6–3.2
–
2.4–3.0
–
–
397
–
376
–
–
562
–
586
–
–
2.1
–
2
–
–
3.73
3.78
3.66
3.72
3.65
376
609
1.6
a
Titanium resp. vanadium content determined by XRF method (error: 0.2 wt.%).
VP total pore volume determined at p/p0 = 0.97.
DME mesopore diameter determined by NLDFT.
Onset temperature of H2-TPR profile.
Position of maxima of H2-TPR profile.
b
c
d
e
f
Average change of oxidation state during H2-TPR experiment.
Energy of absorption edge determined by Tauc’s method [60].
g
mesoporous character of solids (Fig. 3). Pore size distribution for
all three matrices exhibits distribution of pore diameter in the
range from 2 to 4 nm (see inset in the Fig. 3 and Table 1). SBET of
catalyst modified by vanadium is significantly lower, as is very
often observed for impregnated catalysts [3,10,28]. Surface area
loss ranges from ca. 25% for V-HMS sample to 44% for V-6Ti-HMS
sample. This surface area loss is attributed in the literature to
partial destruction of the framework [32] or rather by blocking of
pores by oxide nanoclusters (in this case not detectable by XRD)
because our previous works showed systematic surface degree
with increasing vanadium content [3,10,32].
catalysts after reaction condition treatment have the same spectral
characteristics as the fresh catalysts; no changes in the band inten-
sity or occurrence of new band were detected. Therefore, it can be
concluded that catalysts are stable under reaction conditions.
Thermal stability of supports was investigated by DTA in the
temperature range from ambient temperature to 1200 ◦C (see
Fig. 5). DTA curves of all three supports exhibit weak endothermic
process at about 130 ◦C, which is ascribed to removal of physisorbed
molecules of water and long time drift of baseline caused by dif-
ferent values of thermal heat capacity of samples and ␣-Al2O3
standard referent respectively. No other process was detected for
pure silica HMS support. On the contrary, exothermic processes
are observed above 650 ◦C for Ti-HMS supports (peak maxima are
at 750, 900 and 1040 ◦C for 19Ti-HMS and at 1050 and 1080 ◦C for
6Ti-HMS). These signals can be assigned to destruction of meso-
porous titanosilicate framework and separation of SiO2 and TiO2
phase [55]. Similar behavior was published by Morey et al. [43]
for material Ti-MCM-48 which is stable up to 800 ◦C. On the other
hand some authors reported stability of titanium mesoporous silica
materials even up to 1000 ◦C [58].
The IR spectra of powder supports and catalysts in KBr pel-
lets before and after reaction are reported in Fig. 4 in the
1300–400 cm−1 range, where the skeletal vibrational modes
occurs. The bands at 1228, 1091, 963, 798 and 470 cm−1 per-
ceptible in all spectra are characteristic for the silica network.
The broad feature at 1228 and 1091 cm−1 is assigned to inter-
tetrahedral and intra-tetrahedral asymmetric stretch vibrations of
T
O
T, respectively. Bands at 798 and 470 cm−1 can be assigned
to symmetric stretching modes of T T vibration and T O bend-
ing modes, respectively. Finally, the feature at 950 cm−1 can be
assigned to two overlapping peaks of ꢂ(Si H) and ꢂ(Ti Si)
O
DR UV–vis spectra of both supports and vanadium catalysts
under study are presented in Fig. 6. All samples exhibit absorption
bands in the range of photon energies from 3 to 6 eV attributed to
ligand to metal charge-transfer (LMCT) transitions of the O → V+V
and/or O → Ti+IV type. HMS support exhibits only very low intensity
spectrum, whereas spectra of Ti-HMS supports exhibit very intense
absorption bands with maxima at 4.29, 4.70 and 5.74 eV (289,
264 and 216 nm, respectively), which overlap absorption bands
O
O
vibration [1,33,44,54]. In any case, no spectral feature of crystalline
TiO2 (anatase and/or rutile) with characteristic dominant broad
band at 600–650 cm−1 [55–57] was detected in the spectra of KBr
pellets of both Ti-HMS supports. This is another indication that tita-
nium is relatively homogeneously incorporated into framework. All
HMS (a)
6Ti-HMS (b)
19Ti-HMS (c)
HMS (a)
V-HMS (b)
Ti6-HMS (c)
V-6Ti-HMS (d)
Ti19-HMS (e)
V-19Ti-HMS (f)
f
e
d
c
50 cm3g-1
b
a
0.0
0.2
0.4
0.6
0.8
1.0
1200
1000
800
600
400
p / p0
Wavenumber, cm-1
Fig. 3. Nitrogen adsorption isotherms of bare supports (isotherms were offset for
clarity). Inset: pore size distribution of bare supports determined by DFT method
for cylindrical pores and metal oxide surface.
Fig. 4. FT-IR spectra of KBr pellets of supports and supports after impregnation by
vanadium (spectra were offset for clarity).

136
M. Setnicˇka et al. / Catalysis Today 204 (2013) 132–139
HMS (a)
6Ti-HMS (b)
19Ti-HMS (c)
V-HMS (a)
V-6Ti-HMS (b)
V-19Ti-HMS (c)
c
b
a
c
b
a
exo
0
200
400
600
800
1000
400
500
600
700
800
Temperature, °C
Temperature, °C
Fig. 5. Thermal analysis pattern of HMS and Ti-HMS supports (curves were offset
for clarity).
Fig. 7. H2-TPR patterns of V-HMS and V-Ti-HMS samples (curves were offset for
clarity).
belonging to vanadium species in the spectra of vanadium cata-
lysts. This is in a good agreement with spectra published previously
for mesoporous titanosilicate support [27,37,40]. LMCT transitions
are strongly influenced by the type and number of ligands sur-
rounding the central metal ion in the first coordination sphere and,
therefore, provide information on its local coordination environ-
ment. The absorption edge energy (Eg) of the spectra is usually
employed for this purpose [6,59,60]. The values of Eg of all samples
are listed in Table 1. Comparison of our Ti-HMS samples spectra
with spectra of referent materials (TS-1 representing isolated tita-
nium atoms surrounded by SiO4 tetrahedra and rutile representing
materials (lower value of Eg) is probably due to small amount of tita-
nium species coordinated in an octahedral coordination or Ti-O-Ti
clustering in the framework, which are not detected by XRD [40,62].
Presence of vanadium complexes in the sample leads to increase in
intensity of spectra and slight red-shift of energy edges to the lower
value. Energy edges of all vanadium catalysts fall to very narrow
interval from 3.73 to 3.65 eV indicating very similar distribution of
vanadium species. Based on the empirical correlation of these val-
ues with the structure of referent compounds and absorption edge
energies of their UV–vis spectra (sodium ortho-vanadate Na3VO4
with Eg = 3.83 eV and meta-vanadate NaVO3 Eg = 3.16 eV as com-
pounds containing only isolated monomeric tetrahedral units and
linearly polymerized tetrahedral units [47]) can be concluded that
all vanadium species in our investigated catalysts are in tetrahedral
coordination (no spectral signals under 3.16 eV). Unfortunately, we
cannot determine the amount of monomeric and polymeric species,
as was published previously [3,10]. Determination prevents over-
lapping absorption bands belonging to vanadium species by the
intensive bands belonging to titanium. On the other hand we can
say that most of the species are in isolated or low-polymeric form.
H2-TPR curves of VOX catalysts are depicted in Fig. 7 and values
of onset temperature and temperature of reduction peak maxima
are summarized in Table 1. It should be noted that parent supports
exhibited no reduction peaks and therefore they are not reported
here for the sake of brevity. TPR curves of all samples exhibit
only one reduction peak in the temperature range from 400 to
800 ◦C. The overall hydrogen consumption corresponds to change
of oxidation state during the reduction. The change of oxidation
state varies from 1.6 to 2.1 of electrons per vanadium atom (see
Table 1) indicating reduction from V+V to V+III or in some case par-
tially only to V+IV. Maximum of the reduction peak shifts to higher
temperature with increasing titanium content. V-HMS catalysts
exhibit reduction peak with maximum at 562 ◦C, whereas maxima
of reduction peaks of V-Ti-HMS catalysts are at 586 and 609 ◦C for V-
6Ti-HMS and V-19Ti-HMS, respectively. In addition, the reduction
peak becomes broader with increasing titanium content. However,
it is contrary to investigation of redox behavior of vanadium species
on pure titanium oxide or pure silica support reported in the lit-
erature. Most authors present that vanadium complexes on TiO2
are more reducible (reduction of vanadium on the TiO2 proceed at
a temperature about 100 ◦C lower) than vanadium on silica sup-
port [63,64]. But we can see this behavior only at V-titanosilicates
where the support is prepared by impregnation of pure silica by
bulk Ti
O
Ti network) displayed in the [F(R ) hꢂ]2 vs. hꢂ coor-
∞
dinates (see inset in Fig. 6) led to conclusion that titanium in our
samples is predominantly present in the form of isolated TiO4 tetra-
hedra, because values of energy edges of our samples is 3.78 eV and
3.72 eV for 6Ti-HMS and 19Ti-HMS, respectively. These values are
very close to the value of energy edge of TS-1 silicalite (Eg = 3.82 eV)
whereas rutile/anatase exhibits lower energy edge (Eg = 3.11 eV)
[40,61]. Small differences in energy edge value of TS-1 and our
Wavelength, 102 nm
9 8
7
6
5
4
3
2
0.6
0.5
0.4
0.3
0.2
0.1
0.0
HMS
V-HMS
6Ti-HMS
V-6Ti-HMS
19Ti-HMS
V-19Ti-HMS
2
3
4
5
6
Energy, eV
Fig. 6. Diffuse reflectance UV–vis spectra of diluted and dehydrated supports and
supports after impregnation by vanadium.

M. Setnicˇka et al. / Catalysis Today 204 (2013) 132–139
137
Table 2
Results of catalytic tests for bare supports and support with impregnated vanadium at 460 ◦C (mcat = 400 mg, C4H10/O2/He = 10/10/80 vol.%, total flow rate of 100 cm3 min−1).
Sample
Conv. (%)
C4H10
Selectivity
1-C4
Yield (%)
Productivityc
ꢀ
ꢀ
ꢀ
a
b
c-C4
t-C4
1,3-C4
C1-C3
COx
C4
C4
C4
HMS
V-HMS
6Ti-HMS
V-6Ti-HMS
19Ti-HMS
V-19Ti-HMS
1
4
5
14
6
17
10
17
8
8
10
9
6
7
7
7
9
8
2
3
10
3
13
4
18
19
4
20
5
25
8
4
2
4
39
45
67
61
59
53
36
46
28
37
37
44
0.4
2.3
1.5
5.0
2.1
7.7
0.01
0.08
0.06
0.18
0.08
0.28
23
2
a
Sum of C1–C3 hydrocarbons and acetaldehyde.
b
c
Sum of C4 alkene selectivity, yield or productivity, resp.
gprod gcat−1 h−1
.
titanium [28]. Reiche et al. [64] who investigated V-titaniumsilicate
aerogels prepared by direct synthesis and impregnated by vana-
dium showed the same results as in our work and this observation
matrices was significantly higher when compared to the pure silica
material, and ca. 5% conversion degree of n-butane was obtained
over these materials. The ability of titanosilicate matrix to acti-
vate alkanes was also reported in ODH of propane by TS-1 catalyst
[67]. Because the activity of supports cannot be neglected, it must
be taken into account in the evaluation of the activity of pre-
pared vanadium based catalyst. The catalytic activity of obtained
vanadium impregnated materials was higher for all catalysts than
activity of pure support material. The highest change of catalytic
activity was obtained for sample of V-19Ti-HMS for which the n-
butane conversion increased from 6 to 17% after VOX impregnation.
The degree of n-butane conversion obtained over our samples
did not exceed 20%, hence it can be assumed that the value of con-
version is at first approximation proportional to the reaction rate.
The summation of conversion values obtained on the Ti-HMS and
V-HMS materials can be under this assumption taken as a pro-
portional to hypothetic reaction rate occurring on the system of
two independent active sites. These sum of conversion values on
Ti-HMS and V-HMS material (in Fig. 8 presented as hollow red
circle point) are for both V-Ti-HMS materials lower than exper-
imentally obtained values of conversion of n-butane. Therefore,
some synergy effect can be observed. The V/TiO2 based cata-
lysts are reported [11,27,31,32,68] as materials with significantly
higher activity in partial oxidation or oxidative dehydrogenation
of alkanes in comparison to V/SiO2 based materials. Hence, the
observed enhancement of catalytic activity over the mesoporous
titanosilicate supported vanadium oxide catalysts is an evidence of
the interaction of vanadium atoms with titanium oxide species on
the surface of these materials.
The selectivity to sum of C4-ODH products obtained over pre-
pared samples of catalysts was below 50% and highest selectivity
to these products (46%) was obtained over the sample V-HMS.
Selectivity to C4-ODH products was in all cases lower over pure
matrices and the impregnation of vanadium increased the selec-
tivity to ODH products and decreased the selectivity to products
of n-butane cracking. The presence of titanium in catalysts slightly
decreased the selectivity to C4-ODH products, but the lowering of
this value does not exceed 10% for the sample V-6Ti-HMS. It can
be seen that V-19Ti-HMS sample exhibited comparable selectiv-
ity to C4-ODH products (44%) comparable to selectivity which was
achieved on the sample V-HMS.
proves successful synthesis of titanosilicate with Ti
O Si bonds.
However, we must take into account that redox behavior of sup-
ported vanadium species is very complex problem and can be
affected by structural changes during heating of the sample, nature
of vanadia complex (monomer/polymer/oxide) or nature of sup-
port and interaction of vanadia with support [3,65,66]. In addition,
we must take into account that only part of vanadia species is prob-
ably coordinated to surface in close vicinity of titanium. Shift of
TPR onset to lower temperature with increasing titanium content
clearly indicates interaction of vanadium complexes with titanium
(see Table 1).
3.2. Catalytic tests of n-butane ODH
Oxidative dehydrogenation of n-butane over investigated sam-
ples was studied at 460 ◦C. Main results for both prepared supports
and vanadium impregnated catalysts are presented in Fig. 8 and
Table 2. The main reaction products identified in the reaction
mixture were: 1-butene (1-C4), cis- and trans-2-butene (c-C4 and
t-C4), 1,3-butadiene (1,3-C4), methane (C1), ethane and ethene
(C2), propane and propene (C3), carbon oxides (COX) and traces of
acetaldehyde. The carbon balance was 98 3% in all catalytic tests
and no coke deposit was observed on the catalysts. The activity of
catalysts only slightly depended on the time-on-stream (TOS) and
changes in conversion degree were less than 1% during 10 h.
All prepared supports exhibited measurable activity in the acti-
vation of n-butane. The catalytic activity of both titanosilicate
To the best of our knowledge, the only reported results on
C4-ODH over VOX–titanosilicate catalysts were presented by San-
tacesaria et al. [11] who also published only small changes in
selectivity to C4-ODH products over VOX–TiO2/SiO2 based cata-
lyst in comparison to the VOX/SiO2 samples. The comparison of
our results with data from this paper also clearly shows signifi-
cant differences in distribution of individual C4-ODH products. The
1,3-butadiene was the most abundant product in our C4-ODH prod-
ucts over all vanadium containing catalysts with the selectivity ca.
20% and similar high selectivity to 1,3-butadiene was published
Fig. 8. The conversion (full red circle and line), hypothetic conversion (hollow red
circle point) and selectivity (stacked bar) of bare supports and support with impreg-
nated vanadium in ODH of n-butane at 460 ◦C. (For interpretation of the references
to color in this figure legend, the reader is referred to the web version of the article.)

138
M. Setnicˇka et al. / Catalysis Today 204 (2013) 132–139
0.30
0.25
0.20
0.15
0.10
0.05
0.00
promising supports for vanadium oxide-based catalysts allowing
their good dispersion. Advanced properties of these materials were
demonstrated in C4-ODH reaction. Productivity to desired C4-ODH
products was four times higher for titanosilicate materials com-
pared with silica-based V-HMS materials and reached value up to
280 g of C4 alkenes per 1 kg of catalyst per hour.
Acknowledgement
The authors thank to project of the Czech Science Foundation
No. P106/10/0196 for financial support.
References
[1] A. Comite, A. Sorrentino, G. Capannelli, M. Di Serio, R. Tesser, E. Santacesaria,
Journal of Molecular Catalysis A-Chemical 198 (2003) 151.
[2] L.M. Madeira, M.F. Portela, Catalysis Reviews-Science and Engineering 44
(2002) 247.
[3] M. Setnicka, R. Bulanek, L. Capek, P. Cicmanec, Journal of Molecular Catalysis
A-Chemical 344 (2011) 1.
V-HMS
V-6Ti-HMS
V-19Ti-HMS
Fig. 9. Productivity to C4 dehydrogenation products in n-butane ODH over V-HMS
and V-Ti-HMS catalysts.
[4] H.H. Kung, Advances in Catalysis, vol. 40, Academic Press Inc., San Diego, 1994,
p. 1.
previously [3] over the other set of V-HMS materials, whereas the
Santacesaria et al. [11] published nearly equimolar selectivity to 1-
C4, t-2-C4 and c-2-C4 with only a small amount of 1,3-C4 (less than
12%) for catalysts with similar vanadium concentration. Moreover
the similar distribution of C4-ODH products which was obtained in
mentioned paper was obtained also over both of our Ti-HMS sup-
ports. Differences in the selectivity to C4-ODH products between
catalysts cannot be ascribed to one simple effect, but more likely
to the superposition of two effects. The significantly lower con-
tact times were used in mentioned article, which is also indicated
by the lower reported conversion values (less than 10%) compared
to conversion values reached over our VOX catalysts. It could be
one effect which suppresses the occurrence of subsequent ODH
reactions which can be a source of 1,3-butadiene. Nevertheless,
solely this effect cannot explain the difference between the 1,3-C4
selectivity over our Ti-HMS supports and VOX–Ti-HMS catalysts.
The higher selectivity to 1,3-butadiene over our VOX containing
catalysts in comparison to the Ti-HMS support materials is hence
most likely caused by one-step subsequent ODH reaction of butenes
occurring without the desorption to the gas phase, facilitated by the
more acidic nature of VOX active sites which prolongs the time of
alkene retention on the surface as it was recently suggested [7,69]
The productivity of C4-ODH products over prepared vanadium
containing catalysts is presented in Fig. 9. It can be clearly seen that
productivity to desired C4-ODH products increases with increas-
ing amount of titanium mesoporous support. When we compare
the results obtained for vanadium catalysts impregnated on Ti-
HMS support with our recently published data [3] it is evident
that the presence of titanium in the mesoporous HMS matrix has
significant promoting effect to the catalytic activity of V-Ti-HMS
materials. The activity of V-19Ti-HMS catalyst at 460 ◦C is compa-
rable with the activity of previously reported material I-VHMS-2.1
at 540 ◦C retaining the selectivity to C4-ODH products, except for
slightly higher selectivity to 1,3-butadiene. The improvement of
activity of VOX catalyst impregnated on mesoporous titanosilicate
support hence offers the possibility to carry the reaction at lower
temperature or to use catalysts with lower loading of vanadium.
[5] B.M. Weckhuysen, D.E. Keller, Catalysis Today 78 (2003) 25.
[6] H.J. Tian, E.I. Ross, I.E. Wachs, Journal of Physical Chemistry B 110 (2006) 9593.
[7] T. Blasco, J.M.L. Nieto, Applied Catalysis A-General 157 (1997) 117.
[8] E.A. Mamedov, V.C. Corberan, Applied Catalysis A-General 127 (1995) 1.
[9] R. Bulanek, A. Kaluzova, M. Setnicka, A. Zukal, P. Cicmanec, J. Mayerova, Catal-
ysis Today 179 (2012) 149.
ˇ
ˇ
[10] R. Bulánek, P. Cicˇmanec, H. Sheng-Yang, P. Knotek, L. Capek, M. Setnicˇka, Applied
Catalysis A-General (2012).
[11] E. Santacesaria, M. Cozzolino, M. Di Serio, A.M. Venezia, R. Tesser, Applied
Catalysis A-General 270 (2004) 177.
[12] S. Albonetti, F. Cavani, F. Trifiro, Catalysis Reviews-Science and Engineering 38
(1996) 413.
[13] R.M. Martin-Aranda, J. Cejka, Topics in Catalysis 53 (2010) 141.
[14] F. Cavani, Journal of Chemical Technology and Biotechnology 85 (2010) 1175.
[15] A.A. Teixeira-Neto, L. Marchese, H.O. Pastore, Quimica Nova 32 (2009) 463.
[16] I.E. Wachs, B.M. Weckhuysen, Applied Catalysis A-General 157 (1997) 67.
[17] E.V. Kondratenko, M. Cherian, M. Baerns, D.S. Su, R. Schloegl, X. Wang, I.E.
Wachs, Journal of Catalysis 234 (2005) 131.
[18] Q.H. Zhang, Y. Wang, Y. Ohishi, T. Shishido, K. Takehira, Journal of Catalysis 202
(2001) 308.
[19] W. Liu, S.Y. Lai, H.X. Dai, S.J. Wang, H.Z. Sun, C.T. Au, Catalysis Letters 113 (2007)
147.
[20] P. Gruene, T. Wolfram, K. Pelzer, R. Schlogl, A. Trunschke, Catalysis Today 157
(2010) 137.
[21] Y.M. Liu, Y. Cao, N. Yi, W.L. Feng, W.L. Dai, S.R. Yan, H.Y. He, K.N. Fan, Journal of
Catalysis 224 (2004) 417.
[22] F. Ying, J.H. Li, C.J. Huang, W.Z. Weng, H.L. Wan, Catalysis Letters 115 (2007)
137.
[23] L. Capek, R. Bulanek, J. Adam, L. Smolakova, H. Sheng-Yang, P. Cicmanec, Catal-
ysis Today 141 (2009) 282.
[24] P. Knotek, L. Capek, R. Bulanek, J. Adam, Topics in Catalysis 45 (2007) 51.
[25] K. Cassiers, T. Linssen, M. Mathieu, M. Benjelloun, K. Schrijnemakers, P. Van Der
Voort, P. Cool, E.F. Vansant, Chemistry of Materials 14 (2002) 2317.
[26] M. Kruk, M. Jaroniec, A. Sayari, Microporous Materials 9 (1997) 173.
[27] N. Hamilton, T. Wolfram, G. Tzolova Muller, M. Havecker, J. Krohnert, C. Car-
rero, R. Schomacker, A. Trunschke, R. Schlogl, Catalysis Science and Technology
(2012).
[28] O. Ovsitser, M. Cherian, A. Bruckner, E.V. Kondratenko, Journal of Catalysis 265
(2009) 8.
[29] E. Kondratenko, M. Cherian, M. Baerns, Catalysis Today 112 (2006) 60.
[30] A. Dinse, B. Frank, C. Hess, D. Habel, R. Schomacker, Journal of Molecular Catal-
ysis A-Chemical 289 (2008) 28.
[31] D. Shee, G. Deo, Catalysis Letters 124 (2008) 340.
[32] J.H. Kwak, J.E. Herrera, J.Z. Hu, Y. Wang, C.H.F. Peden, Applied Catalysis A-
General 300 (2006) 109.
[33] A. Zhang, Z. Li, Z. Li, Y. Shen, Y. Zhu, Applied Surface Science 254 (2008) 6298.
[34] W. Zhang, B.S. Zhang, T. Wolfram, L.D. Shao, R. Schlogl, D.S. Su, Journal of Phys-
ical Chemistry C 115 (2011) 20550.
[35] M.A. Ban˜ares, X. Gao, J.L.G. Fierro, I.E. Wachs, in: S.T.O.A.M.G.R.K. Grasselli, J.E.
Lyons (Eds.), Studies in Surface Science and Catalysis, vol. 110, Elsevier, 1997,
p. 295.
4. Conclusion
[36] F. Bérubé, F. Kleitz, S. Kaliaguine, Journal of Materials Science 44 (2009) 6727.
[37] A. Tuel, Microporous and Mesoporous Materials 27 (1999) 151.
[38] Y. Chen, Y. Huang, J. Xiu, X. Han, X. Bao, Applied Catalysis A-General 273 (2004)
185.
[39] R.S. Araújo, D.C.S. Azevedo, E. Rodríguez-Castellón, A. Jiménez-López, C.L. Cav-
alcante Jr, Journal of Molecular Catalysis A-Chemical 281 (2008) 154.
[40] W. Zhang, M. Fröba, J. Wang, P.T. Tanev, J. Wong, T.J. Pinnavaia, Journal of the
American Chemical Society 118 (1996) 9164.
Here presented results indicate successful one-pot synthesis
of mesoporous titanosilicate with high titanium content (up to
19 wt.%) isomorphously incorporated into framework. No method
of characterization detected signals attributable to bulk TiO2 phase
either for neither fresh materials nor catalysts after reaction con-
ditions (460 ◦C for at least 10 h). These materials seem to be
[41] S. Gontier, A. Tuel, Zeolites 15 (1995) 601.

M. Setnicˇka et al. / Catalysis Today 204 (2013) 132–139
139
[42] T. Blasco, A. Corma, M.T. Navarro, J.P. Pariente, Journal of Catalysis 156 (1995)
65.
[57] P.M. Kumar, S. Badrinarayanan, M. Sastry, Thin Solid Films 358 (2000)
122.
[43] M. Morey, A. Davidson, G. Stucky, Microporous Materials 6 (1996) 99.
[44] M. Nandi, A. Bhaumik, Chemical Engineering Science 61 (2006) 4373.
[45] P.T. Tanev, T.J. Pinnavaia, Science 267 (1995) 865.
[46] M. Pouzar, T. Kratochvil, L. Capek, L. Smolakova, T. Cernohorsky, A. Krejcova, L.
Hromadko, Talanta 83 (2011) 1659.
[58] M. Chatterjee, H. Hayashi, N. Saito, Microporous and Mesoporous Materials 57
(2003) 143.
[59] X.T. Gao, I.E. Wachs, Journal of Physical Chemistry B 104 (2000) 1261.
[60] J. Tauc, Amorphous and Liquid Semiconductors, Plenum Press, London, 1974.
[61] C. Li, G. Xiong, Q. Xin, J.K. Liu, P.L. Ying, Z.C. Feng, J. Li, W.B. Yang, Y.Z. Wang, G.R.
Wang, X.Y. Liu, M. Lin, X.Q. Wang, E.Z. Min, Angewandte Chemie-International
Edition 38 (1999) 2220.
[47] R. Bulanek, L. Capek, M. Setnicka, P. Cicmanec, Journal of Physical Chemistry C
115 (2011) 12430.
[48] L. Capek, J. Adam, T. Grygar, R. Bulanek, L. Vradman, G. Kosova-Kucerova, P.
Cicmanec, P. Knotek, Applied Catalysis A-General 342 (2008) 99.
[49] P. Kubelka, F.Z. Munk, Technical Physics 12 (1931) 593.
[50] Y. Liu, K. Murata, M. Inaba, N. Mimura, Applied Catalysis A-General 309 (2006)
91.
[51] C. Chen, Q.H. Zhang, J. Gao, W. Zhang, J. Xu, Journal of Nanoscience and Nano-
technology 9 (2009) 1589.
[52] S.A. Karakoulia, K.S. Triantafyllidis, A.A. Lemonidou, Microporous and Meso-
porous Materials 110 (2008) 157.
[62] W.H. Zhang, J.Q. Lu, B. Han, M.J. Li, J.H. Xiu, P.L. Ying, C. Li, Chemistry of Materials
14 (2002) 3413.
[63] A. Sorrentino, S. Rega, D. Sannino, A. Magliano, P. Ciambelli, E. Santacesaria,
Applied Catalysis A-General 209 (2001) 45.
[64] M.A. Reiche, E. Ortelli, A. Baiker, Applied Catalysis B-Environmental 23 (1999)
187.
[65] B.E. Handy, A. Baiker, M. Schraml-Marth, A. Wokaun, Journal of Catalysis 133
(1992) 1.
[66] F. Arena, F. Frusteri, G. Martra, S. Coluccia, A. Parmaliana, Journal of the Chemical
Society-Faraday Transactions 93 (1997) 3849.
[67] W. Schuster, J.P.M. Niederer, W.F. Hoelderich, Applied Catalysis A-General 209
(2001) 131.
[53] J. Santamaria-Gonzalez, J. Luque-Zambrana, J. Merida-Robles, P. Maireles-
Torres, E. Rodriguez-Castellon, A. Jimenez-Lopez, Catalysis Letters 68 (2000)
67.
[54] G.W. Wallidge, R. Anderson, G. Mountjoy, D.M. Pickup, P. Gunawidjaja, R.J.
Newport, M.E. Smith, Journal of Materials Science 39 (2004) 6743.
[55] Y. Djaoued, S. Badilescu, P.V. Ashrit, D. Bersani, P.P. Lottici, J. Robichaud, Journal
of Sol–Gel Science and Technology 24 (2002) 255.
[68] I.E. Wachs, J.M. Jehng, G. Deo, B.M. Weckhuysen, V.V. Guliants, J.B. Benziger, S.
Sundaresan, Journal of Catalysis 170 (1997) 75.
[69] L.C. Marcu, L. Sandulescu, Y. Schuurman, J.M.M. Millet, Applied Catalysis A-
General 334 (2008) 207.
[56] M. Ocana, V. Fornes, J.V.G. Ramos, C.J. Serna, Journal of Solid State Chemistry
75 (1988) 364.