Electrospun vanadium oxide based submicron diameter fiber catalysts. Part I: Preparation procedure and propane ODH application

Article abstract of DOI:10.1016/j.cattod.2018.10.073

V-Zr-O submicron diameter fibers have been prepared by electrospinning of V and Zr containing solutions in only one calcination step. This is a simple and versatile synthesis method, easy to scale up, that produces porous solids, with fibrous morphology and with VO_x catalytically active species, being these solids attractive for alkane partial oxidation reactions, since the catalysts with fibrous morphology present several advantages with respect powered ones. With the new proposed method, at low vanadium loadings (V% w/w<6.4), surface dispersed VO_x were found in the zirconia fiber as shown by Raman spectroscopy analysis, whereas crystalline ZrV₂O₇ was observed when the V content was increased (V% w/w>6.4). These fibers were very active for the propane oxidative dehydrogenation (ODH) catalytic reaction, showing very promising results since the fibrous morphology of submicrometric diameter makes these solids quite suitable for fixed-bed catalytic reactors.

Full text of DOI:10.1016/j.cattod.2018.10.073

Catalysis Today xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Electrospun vanadium oxide based submicron diameter fiber catalysts. Part
I: Preparation procedure and propane ODH application
⁎
J.J. Ternero-Hidalgo, J. Torres-Liñán, M.O. Guerrero-Pérez , J. Rodríguez-Mirasol, T. Cordero
Universidad de Málaga, Departamento de Ingeniería Química, Campus de Teatinos, E29071, Málaga, Spain
A R T I C L E I N F O
A B S T R A C T
Keywords:
V-Zr-O submicron diameter fibers have been prepared by electrospinning of V and Zr containing solutions in
Propane ODH
Mixed-oxide catalysts
Submicron fibers
Electrospinning
Vanadium
only one calcination step. This is a simple and versatile synthesis method, easy to scale up, that produces porous
solids, with fibrous morphology and with VO catalytically active species, being these solids attractive for alkane
x
partial oxidation reactions, since the catalysts with fibrous morphology present several advantages with respect
powered ones. With the new proposed method, at low vanadium loadings (V% w/w < 6.4), surface dispersed
x 2 7
VO were found in the zirconia fiber as shown by Raman spectroscopy analysis, whereas crystalline ZrV O was
Zirconia
observed when the V content was increased (V% w/w > 6.4). These fibers were very active for the propane
oxidative dehydrogenation (ODH) catalytic reaction, showing very promising results since the fibrous mor-
phology of submicrometric diameter makes these solids quite suitable for fixed-bed catalytic reactors.
1. Introduction
powder catalysts can easily overcome all these intrinsic disadvantages
at laboratory scale adding a diluting inert material (glass, quartz, α-
Vanadium oxide based materials are well known active and selec-
alumina or silicon carbide [11,12]) and/or making pellets (not neces-
sary for submicrometric fibers), which may not be so operative at in-
dustrial level. Moreover, pellet size should be small enough to avoid
high resistance to internal diffusion and big enough to avoid high
pressure drop in a fixed bed, which is not possible sometimes. Conse-
quently, fibers have potentially higher fluidodynamics and transfer
properties which may make them more feasible to scale up to industrial
level. Electrospinning is a simple and straightforward technique to
obtain fibers in the submicron and nano scale [13,14]. This technique
makes use of electrohydrodynamic forces to prepare polymeric fibers
from solutions. The fibers obtained by this procedure find their use in a
wide variety of applications such as in biomedicine (tissue en-
gineering), sensor technology, energy storage (supercapacitors, elec-
trocatalysts), adsorption, heterogeneous catalysis, etc. [13–18]. The
typical set up of electrospinning mainly consists of three components: a
needle or spinneret, a grounded collecting plate or collector and a high
voltage power supply used to apply an electric field between the needle
and the collector. In the electrospinning process, an electric field is
subjected to the end of a needle that contains the polymer solution held
by its surface tension. As the intensity of the electric field is increased,
the fluid at the tip of the needle elongates to form a conical shape
known as Taylor cone. When the applied electric field reaches a critical
value, the repulsive electrical forces overcome the surface tension
tive catalysts for partial oxidation reactions [1,2]. The mechanical and
catalytic properties of Vanadium oxides can be modulated and im-
proved by the incorporation of other elements into their crystal struc-
ture and by the use of a catalytic support, such as zirconia. Several steps
x
are required to obtain a VO supported catalyst, including the pre-
paration of a suitable catalytic support, which usually requires a first
calcination process, followed by the impregnation of such support with
the active phase precursor and the subsequent calcination step [3–8]. In
general, these methods lead to a decrease of the surface area of the
catalytic support, due to pore blockage during the process of deposition
of the active phase [9].
On the other hand, the use of catalytic materials with fibrous
structure of submicrometric size may present several advantages with
respect to powdered ones. They show very small resistance to internal
diffusion and high surface area to volume ratio because of their sub-
micrometric diameter; as well as less significant temperature gradients
and lower pressure drop in a fixed bed than those in powder form due to
the high void fractions or bed porosity offered by these fibrous struc-
tures [10]. Furthermore, they are easy to handle and can be packed or
constructed in the best form to fit the particular use or application due
to their geometric flexibility. All these differences would still be more
appreciable in the use of these catalysts on an industrial scale. Since the
⁎
Corresponding author.
E-mail address: oguerrero@uma.es (M.O. Guerrero-Pérez).
https://doi.org/10.1016/j.cattod.2018.10.073
Received 17 July 2018; Received in revised form 18 October 2018; Accepted 27 October 2018
0920-5861/ © 2018 Elsevier B.V. All rights reserved.
Please cite this article as: Ternero-Hidalgo, J.J., Catalysis Today, https://doi.org/10.1016/j.cattod.2018.10.073

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Scheme 1. Schematic procedure followed for the preparation of vanadium-containing Zirconia submicron-fibers.
forces and a charged jet of the solution is ejected from the tip of the
Taylor cone to the collector. During this process, the polymer solution
jet undergoes an instability and elongation process, which allows the jet
to become very long and thin. Meanwhile, the solvent evaporates,
leaving behind a charged polymer fiber. There are several parameters
that have influence in the process such as viscosity, flow, concentration
of polymer solution, applied voltage, or distance between the needle
and the collector, controlling the diameter and length of the fibers
According to the literature, isolated and oligomerized surface dis-
persed vanadium species (VO
partial oxidation reactions [9,25,29–32]. These surface VO
x
) are the active phases during alkane
species
x
have three different oxygen atoms: forming a terminal vanadyl group
(V]O), bridging two vanadium atoms (VeOeV), and linking a vana-
dium atom and the support (VeOeS) [1,2]. The reactivity of these
catalytic systems is controlled by their redox properties, which are in-
fluenced by both redox and acid–base properties of the support
[1,33,34]. Several oxide supports have been used for preparing vana-
[
13,14,19,20].
On the other hand, the interest of petrochemical industry towards
2 2 3 2 2
dium supported catalysts, such as SiO , Al O , TiO , and ZrO [1,2].
the use of propylene is growing up because it is the raw material for a
wide variety of interest products (i.e., polypropylene, acrylonitrile, oxo-
alcohols, propylene oxide, cumene, acrylic acid, etc.) [21]. The global
demand for propylene in 2005 and in 2014 were about 67 and 89 Mt,
respectively [22,23], and this is expected to grow to about 165 Mt by
Zirconia presents quite valuable characteristics as catalyst support, such
as high thermal and chemical stability; and acidic, basic, reducing and
oxidizing properties [35].
The main objective of the present paper is the preparation of zir-
conia submicron sized fibrous catalytic materials with dispersed VO
x
2
030 [24]. In fact, propylene is the second most important starting
species active for partial oxidation of propane by electrospinning
technique. In this new procedure, the electrospinning process is per-
formed with a polymer solution that includes both the vanadium and
the zirconia precursors. These catalysts present all the advantages of
fibrous conventional catalytic materials, but with a simpler preparation
method that avoid a second process of depositing the catalytic active
phase on the zirconia fibrous support, followed by the calcination step.
In order to evaluate the effect of the synthesis method proposed on the
nature of the vanadium species in the final catalytic material of the new
methodology proposed (addition of Vanadium during the synthesis of
the fiber in only one step), a series of catalysts were prepared by con-
ventional impregnation (preparation of the support followed by im-
pregnation of the Vanadium precursor, two steps) was also prepared
and characterized.
product in the petrochemical industry after ethylene [23]. Currently,
propylene is mainly obtained as byproduct from steam cracking of
naphtha, fluid catalytic cracking in the oil refining and natural gas
processing, which are performed at high temperatures and present high
energy consumption [21,24,25]. A wide range of demanded light hy-
drocarbons can be obtained by these processes; however, the selectivity
to propylene is very low. Consequently, the propylene produced by this
way is not enough to face the current demand and growth rate in the
future [23,24]. Alternatively, the production of propylene from pro-
pane has been investigated since the early 1990s, because propane is
more abundant and significantly cheaper as feedstock [26–28]. In this
way, catalytic dehydrogenation of propane to propylene is possible, but
this is a strongly endothermic process performed at temperature above
873 K, which presents thermodynamic constraints that limit the pro-
pane conversion. Moreover, a major problem in this process is coke
formation, which causes rapid catalyst deactivation [9,25]. Oxidative
dehydrogenation (ODH) of propane is a promising alternative to pro-
duce propylene from propane. This reaction is exothermic and can be
performed at lower temperatures, it is thermodynamically unrestricted
and the coke deposition is minimized [9,25]. However, the propene
yields are still not sufficiently high to be profitable at industrial scale
2
. Materials and methods
2.1. Catalysts preparation
Scheme 1 outlines the procedure followed for the preparation of
zirconia fibers catalysts, which involves the preparation of the polymer
solution containing both vanadium catalyst and zirconia precursors to
be electrospun, followed by calcination to obtain the final submicron-
2 x
due to the undesirable combustion reactions to CO and CO (CO ) [25].
2

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 1
precursor (zirconium propoxide in this case). Thus, F-PZr is a zirconia
fiber material without vanadium, and the others (F-PZr-VX.X) are zir-
conia fibers with a V mass content of X.X%. For comparative purposes,
two samples in powder form (P-) were prepared via calcination of the
polymer solutions before being electrospun without vanadium and with
a 5.0% of V mass content, named P-PZr and P-PZr-V5.0, respectively.
These polymer solutions for the preparation of the powdered samples
were dried overnight at 60 °C in a vacuum dryer before the calcination.
This drying process was performed to try to simulate the evaporation of
the solvent that occurs during the electrospinning process for the
sample with fiber shape. In this way, the calcination was proceeded
with a dried material and in the same conditions for both cases, fibers
and powder.
Catalysts with both powder and fiber forms were also prepared by
classic incipient wetness impregnation method, for comparative pur-
poses. In this case, the zirconia samples, P-PZr and F-PZr, were im-
pregnated with aqueous solutions of vanadyl acetylacetonate obtaining
the samples PI-PZr-V5.0, PI-PZr-V1.0, and FI-PZr-V5.0. In the nomen-
clature, the letter I indicates that the samples were obtained via im-
pregnation method of the zirconia support. Table 1 summarizes all the
samples that have been prepared, their V contents, and their nomen-
clature.
Summary of the samples prepared.
Name
Shape
V content (wt
V incorporation method
%)
F-PZr
Fiber
Fiber
Fiber
Fiber
Fiber
Fiber
Fiber
Fiber
0
Electrospinning one-step
Electrospinning one-step
Electrospinning one-step
Electrospinning one-step
Electrospinning one-step
Electrospinning one-step
Electrospinning one-step
Electrospinning and Incipient wetness
impregnation
One-step
One-step
Incipient wetness impregnation
Incipient wetness impregnation
F-PZr-V1.3
F-PZr-V2.5
F-PZr-V3.7
F-PZr-V5.0
F-PZr-V6.4
F-PZr-V13.3
FI-PZr-V5.0
1.3
2.5
3.7
5.0
6.4
13.3
5.0
P-PZr
Powder
Powder
Powder
Powder
0
P-PZr-V5.0
PI-PZr-V1.0
PI-PZr-V5.0
5.0
1.0
5.0
fiber catalysts. The polymer solutions were prepared by using zirco-
nium(IV) propoxide solution (Sigma-Aldrich, CAS: 23519-77-9, 70 wt %
in 1-propanol), polyvinylpyrrolidone (PVP) (Sigma-Aldrich, CAS: 9003-
39-8, Mw ∼1,300,000), acetylacetone (Sigma-Aldrich, CAS: 123-54-6,
≥
99%), 1-propanol (Sigma-Aldrich, CAS: 71-23-8, ≥99.5%), and Va-
nadyl acetylacetonate (Sigma-Aldrich, CAS: 3153-26-2, ≥97.0%).
Zirconium(IV) propoxide solution was used as zirconia precursor
instead of zirconium oxychloride, as it has been previously published
elsewhere [36,37], because it is more environmental friendly. The use
of this precursor avoids the formation of corrosive chloride ions during
the process. Vanadyl acetylacetonate is used as vanadium precursor
since it presents an adequate solubility in the polymer solution. PVP is
used to form the polymer and to increase the viscosity of the solution in
order to be properly electrospun [13,15]. Stabilization of zirconium(IV)
propoxide is achieved by the addition of acetylacetone (chelating
agent) [19,20].
The Zr–PVP solutions have been prepared by mixing 2.1 g of zir-
conium(IV) propoxide solution, 0.317 g of PVP, 0.450 g of acet-
ylacetone and 2.984 g of 1-propanol. For preparing the V-containing
zirconia fibers, vanadyl acetylacetonate was added to the above
polymer solution, in the amounts of 0.038, 0.075, 0.110, 0.147, 0.190,
and 0.521 g, to obtain final fibers with elemental vanadium mass con-
centrations of 1.3, 2.5, 3.7, 5.0, 6.4, and 13.3%, respectively. These
amounts have been calculated in order to obtain V contents similar to
those of active and selective conventional vanadium oxide supported
catalysts that were prepared with the intention to obtain vanadium
oxide contents close to a monolayer coverage [1,2]. Then, these
polymer solutions were vigorously stirred for 24 h at room temperature
before the electrospinning process.
2
.2. Characterization
2
The porous texture of the samples was characterized by N ad-
sorption–desorption at −196 °C using a micromeritics instrument
ASAP 2020 model). The samples were previously outgassed for at least
h at 150 °C. The apparent surface area (ABET) was determined by
applying the BET equation to the N adsorption isotherm.
(
8
2
The surface morphology of the samples was studied by transmission
electron microscopy (TEM) at an accelerating voltage of 200 kV, in a
FEI Talos F200X microscope equipped with energy-dispersive X-ray
analyzer (EDXA).
Surface chemistry of the samples was analyzed by X-ray photo-
electron spectroscopy (XPS). XPS analyses of the samples were obtained
by a 5700C model Physical Electronics apparatus with MgKα radiation
(
1253.6 eV). The maximum of the C1s peak was set at 284.5 eV and
used as a reference to shift the other peaks.
X-ray diffraction patterns (XRD) were recorded in the region of
2
Ɵ = 5–80° for 30 min on an EMPYREAN diffractometer of PANalytical
using CuK 1,2 (1.5406 Å) monochromatic radiation (operational value
5 kV and 40 mA), a detector PIXcel and Soller slits (incident and dif-
α
4
fracted beam) of 0.04 rad.
In situ Raman spectra of the samples were acquired in the 200-
−1
3
200 cm region, using an Invia Reflex Raman Confocal (RENISHAW)
spectrometer equipped with a RemCam deep depletion CCD detector
Scheme 1 also shows the electrospinning set up used for the pre-
paration of the fibers. This setup consists of a syringe pump, a needle or
spinneret, a collector and two high voltage power supplies (one posi-
tively polarized connected to the needle and the other negatively po-
larized attached to the collector). In every case, the electrospinning
operation conditions were the same. The distance between the needle
and the collector was 20 cm, the electrical potential difference was
−
1
using the 514.5 nm Ar line, with a spectral resolution of ca. 1 cm . The
samples were treated at 200 °C in air 2 h for obtaining the spectra under
dehydrated conditions.
2.3. Propane oxidative dehydrogenation
The V-zirconia fibrous catalytic systems were tested for the propane
12 kV (the tip at + 6 kV and the collector was at −6 kV) and the flow
oxidative dehydrogenation (ODH) in a fixed bed microreactor (i.d.
mm), with plug-flow behavior, placed inside a vertical furnace with
rate of the polymer solution through the spinneret was 0.5 mL/h.
Then, the electrospun fibers were recovered in form of non-woven
cloth and calcined in a conventional tubular furnace. In all the cases,
the heating rate to reach the calcination temperature (500 °C) was
4
temperature control. The temperature of reaction was varied from 300
to 400 °C. The feed was prepared by a mixture of three gases: propane
(
99.95%), oxygen (99.999%) and helium (99.999%). The gases were
always mixed in a composition of 20% C , 10% O and 70% He by
mass flow controllers, yielding a molar ratio of propane to oxygen of 2.
The total volumetric gas flow (F ) and the mass of catalyst (Wcat) were
adjusted in order to vary the space-time (Wcat/FC3H8) in the range of
10 °C/min. The calcination temperature was kept for 6 h under air flow
3
H
8
2
(
150 mL STP/min) in order to eliminate the remaining solvent, the PVP
and the organic part of the V and Zr precursors, and to stabilize the
zirconia fibers. Finally, the fibers obtained were named F-PZr, F-PZr-
V1.3, F-PZr-V2.5, F-PZr-V3.7, F-PZr-V5.0, F-PZr-V6.4, and F-PZr-V13.3.
The letter F in the beginning of the names means that the sample pre-
sents a fibrillar shape whereas the letter P refers to the zirconia
T
−1
0
.1–1.0 gcat·s·mLC3H8 (Volume was always expressed in STP conditions).
In the cases of the catalysts in powder form, the particle size was quite
small, 25–75 μm, and the samples were mixed with silicon carbide
3

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
(
100–300 μm particle size) in a mass ratio of 1:1 in order to avoid both
3. Results
very high pressure drop and significant temperature gradients in the
bed [11,12]. However, in the cases of the fibrous catalysts were not
necessary to use silicon carbide. All the lines were heated above 120 °C
to pre-heat the reactant mixture before entering the reactor and to
avoid the condensation of any product leaving the reactor.
Reactants and products were analyzed in steady state conditions by
an on-line gas chromatograph (Perkin-Elmer, Clarus 500 GC), equipped
with a Hayesep D 80/100 (length: 3 m; diameter: 1/8″; internal dia-
meter; 2.1 mm) and an active carbon 80/100 (length: 2 m; diameter 1/
Fig. 1 shows TEM images of F-PZr, F-PZr-V5.0, F-PZr-V13.3 and FI-
PZr-V5.0 samples, which confirm the fibrous morphology of the sam-
ples. The TEM results for the rest of samples can be found in Fig. 2 at
the additional information section. The samples show a quite high
length to diameter ratio and they do not present fused zones or beads. V
content affects the structure of these fibers, being mechanically re-
sistant and with a smooth surface as V content is increased, up to a V
content of 13.3% (see F-PZr-V13.3 sample in Fig. 1e and 1f). At this
high V content the fiber becomes quite fragile, probably because the
maximum amount of V that can be inserted into the zirconia fiber has
been exceeded and some broken fibers are detected. F-PZr presents fiber
diameters ranging from 420 to 830 nm, while for F-PZr-V5.0 diameters
between 300 and 410 nm can be observed. These results indicate that
the incorporation of V produces more uniform fibers with smaller dia-
meters. This is probably due to the increase of the electrical con-
ductivity of the electrospinning polymer solution when the vanadium
precursor (vanadyl acetylacetonate) is added. It has been found that
with the increase of electrical conductivity of the solution, coulombic
forces further increase as result of higher amounts of charge carriers.
Therefore, jet segment experiences more stretching, what results in
smaller diameters and more uniform electrospun fibers [13]. However,
at high vanadium content (F-PZr-V13.3) the fibers are less uniform and
present a wider diameter distribution, with diameters ranging from 220
to 505 nm (Fig. 1e). Fig. 1g shows the TEM images for FI-PZr-V5.0
sample, similar to F-PZr-V5.0, but prepared by conventional incipient
wetness V impregnation of a calcined electrospun zirconia fiber. This
picture shows that some fibers have been broken during the impreg-
nation method and that some crystalline particles, tentatively assigned
8
′'; internal diameter: 2,1 mm) packed columns. Light gases (O
CO ) were detected by a thermal conductivity detector (TCD); and CH
, C , C and C were detected using a flame ionization
2
, CO and
2
4
,
C
2
H
4
2
H
6
3
H
6
3 8
H
detector (FID). The duration of each GC analysis was 15 min. In all the
cases carbon and oxygen molar balances were closed with errors lower
than 5%. The propane conversion and the selectivity to i product are
defined as XC3H8 and Si, respectively:
˙
˙
F0,C H − FC H
3
8
3
8
XC H (%) =
100
3
8
˙
F0,C H
(1)
(2)
3
8
˙
i i
n F
S
i
(%) =
100
˙
F
n
C H
X
C H
3
8
0,C H
3 8
3
8
˙
˙
C H are the molar flows of propane in the inlet and
3
Where F0,C H and
F
3
8
8
˙
F
i
is the molar flow of i product in
in the outlet streams, respectively.
the outlet stream, and n is the number of carbon atoms per molecule of
i
i product. Turn-over frequency was calculated from Eq. (3), it quantifies
the specific activity of a catalytic surface active center under defined
reaction conditions and so, describes the number of converted moles of
propane per mol of vanadium and time:
2 5
to crystalline V O , have been formed around the fibers. These crys-
˙
talline V O5 usually grows when the vanadium dispersion capacity on
the surface has been exceed. This can occur when vanadium content is
higher to the corresponding of monolayer formation, or when vana-
dium species are not well distributed along the surface.
F0,C H XC H MV
2
3
8
3 8
TOF =
Wcat WV
(3)
where M
V
is denoted as the molar mass of vanadium and W
V
mass of
vanadium per gram catalyst determined by XPS. This method of
quantification assumes that all the active sites are equal under reaction
environment and that these sites are stable during the reaction. Pro-
pylene productivity values were calculated from Eq. (4). It indicates the
propylene mass obtained per mass of catalyst and time:
The Zr and V EDXA elemental mappings of F-PZr, F-PZr-V5.0, F-PZr-
V13.3, and FI-PZr-V5.0 are presented in Fig. 2. As expected, a high
signal from Zirconium atoms is detected, and the results show that they
are quite well dispersed. The images for V containing fibers, F-PZr-V5.0
and F-PZr-V13.3 (Fig. 2f and i) indicate that vanadium oxide dispersion
along the fiber material is good, whereas only some noise is detected, as
expected, in the sample that do not contain V (Fig. 2c). The images for
the sample prepared by impregnation (Fig. 2j, 2k and 2 l) are quite
different from those in which V was incorporated during the electro-
spinning step. In this case, most of the vanadium is in the particles
instead of on the fiber surface. This support the idea exposed above
˙
F0,C H XC H SC H MC H
3
8
3
8
3
6
3 6
Propylene productivity =
Wcat
(4)
where MC H is the molar mass of propylene.
3
6
As blank reactions, catalytic tests were also performed with the
empty reactor (to check the contribution of homogeneous phase reac-
tion), and with the V-free-fiber (to see the contribution of the zirconia).
The propane conversion values were found negligible in both cases at
lower temperatures than 500 °C, where the homogeneous phase start to
have a little contribution.
The internal mass transport limitation can be considered negligible
due to the quite small diameter of the submicron-fibers and no pressure
drop has been observed through the catalyst bed under the experi-
mental conditions used, due to the fibrous morphology of the catalysts
2 5
about the formation of undesired V O in detrimental of dispersed va-
nadium species along the surface, what means that the dispersion has
not been successfully achieved, in contrast with the samples prepared
by the one-step method, whereby it is expected that the vanadium will
be very well disperse, in all the fiber, due to the preparation by one step
method.
2
Fig. 3 shows the N isotherms of the samples, whereas Table 2
presents the BET surface areas obtained from them. Figs. 3a and 3b
show the data for the samples prepared with the one-step method. The
samples with fibrous structure present a type IVa isotherm corre-
[
10]. On the other hand, diagnostic experimental tests (see additional
information Fig. 1) for analyzing the presence of external mass trans-
port limitations were performed under the most unfavorable experi-
mental conditions as possible, checking the influence of the total gas
flow rate at constant space-time in the observed conversions reaction
2
sponding to mesoporous solid [38]. Most of the N is adsorbed at high
relative pressures showing a hysteresis loop in which the desorption
occurs at lower relative pressures than the adsorption for equivalent
volumes in each case. Type H1 hysteresis loops are observed in the
fibers F-PZr, F-PZr-V1.3 and F-PZr-V6.4, associated to the presence of a
narrow range of uniform mesopores [38]. However, type H5 hysteresis
loops are observed in the fibers F-PZr-V2.5, F-PZr-V3.7, F-PZr-V5.0 and
F-PZr-V13.3, often associated to pore textures containing both open and
partially blocked mesopores [38]. The isotherm of the powder sample
[
11]. It is noteworthy that no diffusion limitations were observed in the
catalytic results shown using the catalysts with lower vanadium content
than 13.3% prepared in this study. However, in the case of F-PZr-V13.3
at 400 °C, it cannot be discarded some diffusional problems. Due to the
high activity of this catalyst the oxygen is the limiting reactant, the
selectivity to propylene is lower than those for the other catalysts and a
high production of CO
2
is observed.
P-PZr-V5.0 presents a type Ib isotherm, indicative of mainly a
4

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 1. Transmission electron micrographs of F-PZr (a, b), F-PZr-V5.0 (c, d), F-PZr-V13.3 (e, f), and FI-PZr-V5.0 (g, h).
microporous structure, including wider micropores and some narrow
mesopores [38], as all the N is adsorbed at low relative pressures. It
should be noted that samples F-PZr-V5.0 and P-PZr-V5.0, both with the
same vanadium content, show the same adsorption branch at low re-
2
than 0.3, where the adsorption of N is almost negligible in P-PZr-V5.0,
2
and is very appreciable in F-PZr-V5.0, indicating the presence of me-
soporous in the latter sample, which could suggest that the mesoporous
texture is developed during the electrospinning process and subsequent
calcination.
lative pressures in the N
consequently a very similar BET surface area. However, the N
therms of both samples are very different at relative pressures higher
2
adsorption-desorption isotherms (Fig. 3b) and
2
iso-
N
2
isotherms of the submicron-fibers (Fig. 3a and b) with (F-PZr-
V1.3, F-PZr-V2.5, F-PZr-V3.7, F-PZr-V5.0, F-PZr-V6.4, and F-PZr-V13.3)
5

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 2. HAADF-STEM images and EDXA elemental mappings of Zr and V for the samples F-PZr (a, b, c), F-PZr-V5.0 (d, e, f), F-PZr-V13.3 (g, h, i), and FI-PZr-V5.0 (j, k,
l). Note that the scale of each figure has been adapted to the size of the sample.
and without (F-PZr) vanadium, indicate that the porous texture is in-
fluenced by the presence of the vanadium precursor (vanadyl acet-
ylacetonate) during the electrospinning procedure. The isotherm profile
found in the sample with the lowest vanadium content (F-PZr-V1.3) is
very similar to that of the sample without vanadium (F-PZr), and only a
closure is shifted to lower relative pressures (0.4 - 0.5), indicating the
presence of narrower mesopores. These changes in the structure of V
containing fibers results in a wider size distribution of mesoporous.
Fig. 3c shows the isotherms of the samples prepared by the incipient
wetness impregnation method. The samples with fibrous structure (F-
PZr and FI-PZr-V5.0) present type IVa isotherms with a type H1 hys-
teresis loops. The powder samples (P-PZr, PI-PZr-V1.0, and PI-PZr-
V5.0) show type II isotherms with hysteresis loops type H3 character-
istic of nonporous or macroporous adsorbents [38]. These results show
that the electrospinning process promote the development of the porous
decrease in the N
2
adsorption at high relative pressures is observed
ad-
(
Fig. 3a). When V content is increased, also an increment of N
2
sorption is found to occur at low and medium relative pressures,
achieving a maximum for F-PZr-V5.0 sample (Fig. 3a). It is also de-
tected a change in the hysteresis loop from type H1 to H5, whose
6

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 2
Physicochemical characteristics of the submicron-fibers.
aABET
2
V EDXA
(wt %)
XPS (wt %)
(
m /g)
V2p
Zr3d
O1s
C1s
F-PZr
26
29
59
52
66
59
38
24
10
64
10
6
0
0
67.7
65.3
66.1
65.4
64.3
59.8
48.0
66.5
69.5
64.5
62.7
29.5
28.0
26.3
24.2
28.2
27.9
24.8
31.4
28.1
26.3
27.7
28.9
40.6
4.3
7.0
7.0
3.1
2.9
9.4
7.7
3.2
4.2
3.1
3.1
4.2
F-PZr-V1.3
F-PZr-V2.5
F-PZr-V3.7
F-PZr-V5.0
F-PZr-V6.4
F-PZr-V13.3
FI-PZr-V5.0
P-PZr
1.2
2.4
3.6
4.7
6.2
11.4
–
–
–
–
–
1.4
2.7
3.3
4.9
6.0
12.9
2.2
0
P-PZr-V5.0
PI-PZr-V1.0
PI-PZr-V5.0
4.7
5.3
25.7
a
2
From N adsorption-desorption isotherms.
samples and the atomic compositions calculated from XPS spectra.
Vanadium-containing samples prepared with the one-step method
present the highest BET area values, achieving a maximum value
2
around 66 m /g for the samples with 5.0% of vanadium (P-PZr-V5.0
and F-PZr-V5.0), significantly higher than the samples without vana-
2
2
dium, 26 m /g for F-PZr, and 10 m /g for P-PZr. Regarding the samples
prepared by the incipient wetness impregnation method, as expected,
vanadium incorporation produces a slight decrease in the BET specific
2
surface value, from 26 to 24 m /g for the samples with fibrous structure
2
(
F-PZr and FI-PZr-V5.0, respectively), and from 10 to 6 m /g for the
powdered samples (P-PZr and PI-PZr-V5.0, respectively).
Fig. 4 shows the X-ray diffractograms of the samples. The pattern of
zirconia tetragonal phase is detected in all the samples, as expected,
although the monoclinic phase is the thermodynamically stable one for
pure zirconia up to 1170 °C. This is because amorphous zirconia pre-
cursors tend to first crystallize forming the metastable tetragonal phase
of zirconia and then, at temperature higher than 600 °C, they start to be
transformed into the monoclinic structure [15,39]. For fibrous shaped
2 5
samples prepared by electrospinning, no crystalline V O pattern is
detected, or at least they are not so big to be detected by this technique,
indicative of a probable well dispersion of this oxide in the fibrous
material, in line with the EDXA mapping results (Fig. 2). At high V
loadings (F-PZr-V13.3 sample), it is detected the pattern of ZrV
2 7
O
(
Fig. 4a). The formation of this crystalline mixed oxide usually occurs at
high temperatures (> 600 °C) [40] that should not have been reached
in this case. However, the high V content could be responsible of this
fact, given that in the presence of high concentration of V (as in F-PZr-
2 7
V13.3), ZrV O could be form. Nevertheless, it is also possible that
combustion of PVP and organic part of V and Zr precursor during cal-
cination gives rise to hot spots in the fiber matrix, increasing the local
temperatures of part of the fibers. On the other hand, PI-PZr-V5.0 and
FI-PZr-V5.0, the reference materials prepared by impregnation, show
V
2
O
5
pattern (Fig. 4b).
Fig. 5 shows the V2p3/2 XPS spectra of the samples and the typical
binding energies of the powdered oxide standards of V (517.2 eV),
VO (516.2 eV) and V (515.4 eV) as a reference, where vanadium is
present in oxidation states of
18,41–45]. The data indicate that as the vanadium content increases,
2 5
O
2
2 3
O
+5
4+
3+
V
,
V
and
V
, respectively
[
the maximum value of the V2p3/2 spectrum shifts to higher binding
energies corresponding to more oxidized species (Fig. 5a). By this
manner, the spectra of F-PZr-V1.3 and F-PZr-V2.5, are centered at
516 eV, close to the typical binding energy of V+4 present in the stoi-
2
Fig. 3. N adsorption-desorption isotherms at -196 °C of the (a, b) catalysts
prepared with the one-step method, and (c) the ones prepared by using the
incipient wetness impregnation method.
chiometric oxide VO
17.2 eV, corresponding to the typical binding energy of V present in
the stoichiometric oxide V , but in this case vanadium is mainly
forming ZrV crystals as previously observed by the XRD results
(Fig. 4a). Fig. 5b shows the V2p3/2 XPS spectra of the samples prepared
2
; and the spectrum of F-PZr-V13.3 is centered at
+5
5
texture of the final catalyst during the one-step method used in this
study.
2 5
O
2 7
O
Table 2 summarizes some physicochemical characteristics of all the
7

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 4. XRD patterns of (a) the zirconia fibers prepared by the one-step method, and (b) the samples FI-PZr-V5.0, P-PZr-V5.0, PI-PZr-V1.0, and PI-PZr-V5.0.
Tetragonal ZrO (JCPDS 01-079-1769), cubic ZrV (JCPDS 98-005-9396) and orthorhombic V (JCPDS 01-085-0601) are found in the X-ray diffractograms.
2
2
O
7
2 5
O
by V-impregnation, FI-PZr-V5.0, PI-PZr-V1.0 and PI-PZr-V5.0, with a
maximum value around 517.2 eV corresponding to V+5 forming V
Moreover, these samples present a broad band between 700 and
−1
2
O
5
950 cm
associated to the presence of monovanadates and poly-
crystals, as also mentioned in the XRD results (Fig. 4b). The surface
vanadium oxide species of the catalysts prepared in this study should
vanadates, due to VeO modes along VeOeV and/or VeOeZr chains
[5,49,50]. It can also be identified a broad band centered near
+
5
−1
practically be present as V with a very moderate extent of reduction,
unless they are under reaction/reduction conditions [5,46]. However,
most of the prepared material present a V2p3/2 spectrum with binding
1020 cm , due to the V]O terminal stretching of isolated mono-
−
1
−1
vanadates (1019 cm ) and polyvanadates (1030 cm ) [5,51,52]. It
−1
can be also observed that the broad band at 700–950 cm
grows in
+
4
+5
energies between V (in V
of catalysts where vanadium is dispersed in zirconia [41], which in-
dicates electron transfer from ZrO to V.
2 5 2
O ) and V (in VO ), typical in this kind
intensity by rising the vanadium content, at the same time that the V]
−1
O mode is blue-shifted close to 1030 cm . Therefore, it can be derived
that increasing the vanadium content in the sample results in a higher
proportion of polyvanadates, as reported in other studies in the litera-
ture [5,40,53]. As mentioned, the shape of spectra is different for
samples with higher vanadium contents. F-PZr-V6.4 presents bands
assigned to both monoclinic (637, 616, 559, 533, 502, 476, 381, 330,
2
Fig. 6 shows the Raman spectra obtained under dehydrated condi-
tions. Fig. 6a shows the spectra for the catalysts in fiber form prepared
by electrospinning in one step; it can be observed a different shape of
the spectra for samples with higher V contents (F-PZr-V13.3 and F-PZr-
−1
−1
6.4) with respect to samples with lower V concentrations. The catalysts
305, 224, 190, 175 and 115 cm
[48]) and tetragonal (146 cm
)
with lower V contents (F-PZr-V1.3, F-PZr-V2.5, F-PZr-V3.7 and F-PZr-
V5.0) present bands associated to tetragonal phase of zirconia (642,
phases (Fig. 6a). Since only tetragonal phase was detected in the XRD
pattern (Fig. 4), this indicates that the monoclinic aggregates present a
very small size. This could be also related to the hot spots (reaching
temperatures > 600 °C) occurring by combustion of PVP and organic
part of the V and Zr precursors during calcination of the fibers. With
respect to the bands associated to vanadium species, it can be detected
−1
604 (shoulder), 459, 314, 268 and 146 cm
[47]) in agreement with
the XRD results. For F-PZr-V3.7, it can be also observed some low in-
−1
tensity bands assigned to monoclinic phase (381, 189 and 177 cm
48]), which are not detected by XRD, due to the reduced size of these
crystalline structures that are not able to give an XRD pattern.
[
−1
the presence of two peaks, one at 747 cm and other one, smaller, at
8

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 5. V2p3/2 XPS spectra of the (a) zirconia fibers prepared with the one-step
method, and (b) the samples prepared by using the incipient wetness impreg-
nation method.
−
1
9
25 cm , which can be related to a mixture of monovanadates/poly-
−1
2 7
vanadates species and crystalline ZrV O (988 and 777 cm [5,50,54])
indicating somehow that at this vanadium content the vanadium dis-
persion capacity in zirconia has been exceeded. In fact, as the vanadium
content increases there is a turning point in the type of hysteresis loop
observed in the N
which could be related to the fact that vanadium species start to appear
in a more condensed form as crystalline ZrV having a repercussion
in the final porous texture. In addition, a band at 1020 cm due to V]
O mode is detected, indicating the presence of VO . With respect to the
sample with highest vanadium content (F-PZr-V13.3), only bands as-
sociated to crystalline ZrV are found. Therefore, it can be concluded
2
isotherms from H5 to H1 in this sample (F-PZr-V6.4),
2 7
O
−
1
x
2 7
O
that the V/Zr ratio in the initial polymer solution determine the vana-
dium species formed in the final sample after calcination, which in turn
influences the porous texture.
Fig. 6. In situ Raman spectra of (a) the dehydrated vanadium-containing zir-
conia fibers prepared by the one-step method, and (b) the samples P-PZr-V5.0,
PI-PZr-V1.0, and PI-PZr-V5.0.
Fig. 6b shows the spectra for the reference samples in powder form,
those prepared by incipient wetness impregnation method (PI-PZr-V1.0
and PI-PZr-V5.0) and the one prepared in one step (P-PZr-5.0). It can be
observed that the Raman spectrum of the sample P-PZr-V5.0 is very
similar to that of F-PZr-V5.0, indicative that the surface chemistry of
the catalysts is not influenced by the electrospinning process, as sug-
gested by the XRD and XPS results. On the other hand, the samples PI-
−
1
crystals (994, 700, 526, 479, 404, 302, 283, and 143 cm
whereas ZrV is not detected. Thus, by impregnation of the fibers
even with low amount of V, as in the case of PI-PZr-V1.0, V crystals
are detected, whereas the new proposed methods allow the in-
corporation of disperses VO species on the surface of the zirconia fi-
bers, as shown by the Raman spectra results (Fig. 6a).
[55]),
2 7
O
2 5
O
x
2 5
PZr-V1.0 and PI-PZr-V5.0 show Raman bands associated to V O
9

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 3
Propane conversions and selectivities to propylene, CO and CO
2
, in the propane ODH reaction for the different vanadium-containing catalysts, as a function of
−
1
temperature with a space-time of 0.1 g·s·mLC3H8
.
Sample
300 °C
C3 (%)
350 °C
400 °C
X
S
C3= (%)
S
CO (%)
SCO2 (%)
X
C3
S
C3=
S
CO
S
CO2
X
C3
S
C3=
S
CO (%)
SCO2 (%)
(%)
(%)
(%)
(%)
(%)
(%)
F-PZr-V13.3
F-PZr-V6.4
F-PZr-V5.0
F-PZr-V3.7
F-PZr-V2.5
F-PZr-V1.3
P-PZr-V5.0
PI-PZr-V5.0
PI-PZr-V1.0
0.3
0.3
0.4
–
–
–
0.3
–
–
100
100
100
–
–
–
100
–
–
0
0
0
–
–
–
0
–
–
0
0
0
–
–
–
0
–
–
17.4
1.0
1.7
0.5
0.1
–
1.6
0.3
0.2
27.4
71.6
87.7
100
100
–
86.2
100
100
58.2
28.4
12.3
0
0
–
13.8
0
0
14.4
0
0
0
0
–
0
0
0
17.4
4.2
4.0
2.7
0.7
0.2
3.8
1.0
0.8
27.0
30.5
64.6
66.7
85.7
100
65.3
88.8
88.9
58.4
56.6
30.6
27.3
14.3
0
28.4
11.2
11.1
14.6
12.9
4.8
6.0
0
0
6.3
0
0
Table 3 shows steady state propane conversions and selectivity to
propylene, CO and CO , as a function of reaction temperature at a
space-time of 0.1 g·s·mLC3H8. At very low propane conversions the se-
lectivity to propylene is 100%, as can be observed at 300 °C for all the
catalysts studied. At 350 °C, when the propane conversions are higher,
the selectivity to CO starts also to be appreciable for some catalysts, as
V6.4 and F-PZr-V13.3 catalysts are quite different, with lower values for
the entire range of temperature studied, with respect to those for the
other catalysts with lower V contents, which could indicate the pre-
sence of different active sites, in line with the characterization results
2
−
1
2 7
that demonstrated the presence of the crystalline phase of ZrV O in F-
PZr-V6.4 and F-PZr-V13.3 samples (Figs. 4a and Fig. 6a). It is inter-
esting to note that the oxygen conversion is 100% during reaction of F-
PZr-V13.3 at 350 and 400 °C in the studied conditions. This is because
the propane conversions are very similar at both different temperatures
(Fig. 7b and Table 3), since the reaction is truncated when all the
oxygen is consumed.
2
well as that to propylene, but not to CO . Finally, at 400 °C the presence
of CO is also observed for most of the catalysts, together with propy-
2
lene and CO. These results suggest that the reaction mechanism could
occur in a consecutive pathway, where propylene is firstly formed,
followed by CO and finally CO , although the direct combustion of
2
propane cannot be discarded, especially at high temperatures.
Fig. 7 shows steady state propane conversions and selectivity to
propylene as a function of reaction temperature and space time. The
values for the samples prepared by the one-step method are shown in
Figs. 7a and b. The propylene selectivity profiles obtained for F-PZr-
The conversion-selectivity profiles for the vanadium-containing
catalysts F-PZr-V5.0, P-PZr-V5.0, PI-PZr-V1.0, and PI-PZr-V5.0 are
shown in Figs. 7c and d. These values are quite similar for the samples
F-PZr-V5.0 and P-PZr-V5.0, in agreement with the results obtained by
Raman spectroscopy (Fig. 6), which suggests that both samples present
Fig. 7. Reaction profiles in the propane ODH reaction of the different vanadium-containing catalysts. (●) Propane conversion and (▲) selectivity to propylene (a, c)
−
1
as a function of space-time (Wcat/FC3H8) at 300 °C; and (b, d) as a function of temperature with a space-time of 0.1 g·s·mLC3H8
.
10

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
the one-step method with the only difference of using the electrospin-
ning in F-PZr-V5.0. With respect to the porous texture, although the N
2
isotherms of F-PZr-V5.0 and P-PZr-V5.0 are different (Fig. 3b), both
samples present similar BET surface areas (Table 2), since they have the
same adsorption branch at low relative pressures and this method only
considers this range of low relative pressures to estimate the surface
area. In terms of catalytic activity, both samples are very similar
(
Figs. 7c and d), although the sample F-PZr-V5.0 present a more de-
veloped mesoporous texture. Therefore, taking into account that the
only differences among the two samples are the shape of the catalyst
and the porous texture in the mesopore range, it can be concluded that
the mesoporosity and the shape of the catalyst (fiber or powder) have
not influence in the catalytic results of this reaction in the studied
conditions. So, the type of isotherms at high relative pressures including
the type of hysteresis loops are not relevant in the catalytic results, the
only important points are the surface chemistry and the surface area of
the catalyst, which it will be determined by the isotherm at low relative
pressures corresponding to the micropores range. It is noteworthy that
although the fiber shape does not have influence in the activity results,
they present some important advantages with respect to the powder
shape due to its superior fluidodynamics and transfer properties, which
allow the use of this catalysts without adding a diluting inert material
as silicon carbide, that it is necessary in the case of the powdered cat-
alysts.
In order to compare the best propane ODH results of the catalysts
prepared in this study with those reported in the literature, propylene
productivity (Fig. 9a) and propane TOF (Fig. 9b) were used. The pro-
ductivity and TOF results for other catalysts have been collected from
two reviews articles published by Cavani et al. [25] and by Carrero
et al. [9], respectively. Fig. 9 shows that both parameters increase with
temperature and with vanadium loading for the catalysts prepared in
this work, in general. The results clearly indicate that the best values of
propane TOF and propylene productivities obtained in this study cor-
Fig. 8. Time on Stream study of the activity in the propane ODH reaction of the
vanadium-containing zirconia fibers (F-PZr-V1.3, F-PZr-V2.5, F-PZr-V3.7, F-
PZr-V5.0, and F-PZr-V6.4) at different conditions. (●) Propane conversion and
(
▲) selectivity to propylene.
a similar surface chemistry. Nevertheless, it has to be pointed out that
F-PZr-V5.0 catalyst works properly in its fibrous conformation under
the operation conditions studied. However, P-PZr-V5.0 catalyst needs to
be mixed with silicon carbide particles of 100–300 microns in size to
avoid very high-pressure drop and temperature gradients through the
catalytic bed or to prevent catalyst from severe reduction in surface
area (and thus available active sites) that happen when agglomeration
of catalyst particles takes place to increase the catalyst particle size, by
addition of a binder. Figs. 7c and 7d also show that PI-PZr-V1.0 and PI-
PZr-V5.0 present relatively low conversion values, in agreement with
its low BET area values (Table 2).
Propane conversions and selectivities to propylene as a function of
time on stream for F-PZr-V1.3, F-PZr-V2.5, F-PZr-V3.7, F-PZr-V5.0, and
F-PZr-V6.4 samples are shown in Fig. 8. The results demonstrate that
the fibrous catalysts are stable under time of stream up to 24 h at the
operation conditions studied, without noticeable deactivation.
x
respond to F-PZr-V5.0, suggesting that VO species are the most active
and selective for propane ODH. Note that the results of propane TOF
and propylene productivity obtained with P-PZr-V5.0 are not shown for
the sake of brevity, but these results are very similar to those obtained
with F-PZr-V5.0, since they present the same catalytic results as it was
mentioned before (Fig. 7 and Table 3).
4. Discussion
Table 3 results show that the highest propylene yields are obtained
with the samples prepared with the one step method, which enhances
The method described for the incorporation of V active sites
throughout the synthesis of the fibrous catalysts, instead of by the
synthesis of the catalyst support followed by the impregnation of the
active phase, presents several advantages, generating fibers more uni-
form and mechanically resistant, as underlined the TEM results (Fig. 1),
and with a larger porosity and higher specific surface area values (Fig. 3
and Table 2). In addition, by comparison with the samples prepared by
impregnation, it has been showed how the one-step method allows the
the formation of VO
crystalline structures. If the surface V loading exceeds a certain
value, corresponding probably to the monolayer, 5% in this case, the
formation of ZrV crystalline structures increases and that of VO
x
decreases, reducing the TOF and propylene productivity, as in the case
of F-PZr-V6.4 catalysts, shown in Fig. 9.
On the other hand, it can be also observed in Fig. 9 that the catalysts
prepared in this study present similar results, in terms of TOF and
productivity with respect to those reported in the literature. In fact, F-
PZr-V5.0 presents higher or very similar TOF values in the temperature
range of 300–350 °C than those of the catalysts reported in the litera-
ture and the highest propylene productivity values in the range of
350–400 °C. In this sense, it has to be pointed out that the propylene
productivity values obtained for this catalyst between 300 and 400 °C
correspond to typical values obtained at higher temperatures for the
catalysts reported in the literature [25].
x
dispersed species and avoids the formation of
2 5
V O
2
O
7
incorporation of a high amount of VO
which seems to be active and selective for the propane ODH reaction to
propylene, avoiding the formation of non-selective V crystals. Pre-
x
species dispersed on the surface,
2 5
O
vious studies with powdered materials showed that carrying out the
formation of the zirconia support and the dispersion of metal oxides in
the same calcination step [56] presents also some advantages. Ac-
cording to Zhao et al. [56], the growth and sintering of crystallites takes
place during the calcination process of the support, in addition to phase
transformation via a mechanism of surface diffusion, which lead to a
less developed porous texture. However, if in this process the active
component is dispersed on the surface, it segregates the particles of the
support from each other before the particles have grown into bigger
ones. In this way, the surface diffusion is suppressed, and consequently
the porous texture is improved.
5. Conclusions
A new, versatile, simple and straightforward method to obtain
highly dispersed vanadium oxide species on zirconia submicronfibers
by electrospinning has been described in present work. This new
The best catalytic activity results were obtained with the samples F-
PZr-V5.0 and P-PZr-V5.0 (Fig. 7 and Table 3), both with the same va-
nadium content, similar surface chemistry as showed in the XRD, XPS
and Raman results (Figs. 4, 5 and 6, respectively), and prepared using
method develops quite porous solids and enhances the formation of VO
dispersed species on the surface of the ZrO fiber without an additional
step for impregnation. Since these VO species are well-known as active
x
2
x
11

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
Competitiveness and FEDER [CTQ-2012-36408 and CTQ-2015-68654-
R]. J.J.T.H. and J.T.L. acknowledge the assistance of the Ministry of
Economy and Competitiveness of Spain for the award of FPI Grants
[
BES-2013-064425 and BES-2016-079237, respectively].
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.cattod.2018.10.073.
References
[
[
1] I.E. Wachs, Molecular engineering of supported metal oxide catalysts, Chem. Eng.
Sci. 45 (1990) 2561–2565, https://doi.org/10.1016/0009-2509(90)80142-2.
2] M.O. Guerrero-Pérez, Supported, bulk and bulk-supported vanadium oxide cata-
lysts: A short review with an historical perspective, Catal. Today 285 (2017)
2
26–233, https://doi.org/10.1016/j.cattod.2017.01.037.
[
3] A. Dinse, B. Frank, C. Hess, D. Habel, R. Schomäcker, Oxidative dehydrogenation of
propane over low-loaded vanadia catalysts: impact of the support material on ki-
netics and selectivity, J. Mol. Catal. Chem. 289 (2008) 28–37, https://doi.org/10.
1
016/j.molcata.2008.04.007.
[
4] A. Khodakov, B. Olthof, A.T. Bell, E. Iglesia, Structure and catalytic properties of
supported vanadium oxides: support effects on oxidative dehydrogenation reac-
tions, J. Catal. 181 (1999) 205–216.
[
5] A. Christodoulakis, M. Machli, A.A. Lemonidou, S. Boghosian, Molecular structure
and reactivity of vanadia-based catalysts for propane oxidative dehydrogenation
studied by in situ Raman spectroscopy and catalytic activity measurements, J.
Catal. 222 (2004) 293–306, https://doi.org/10.1016/j.jcat.2003.10.007.
6] J. Bedia, J.M. Rosas, J. Rodríguez-Mirasol, T. Cordero, Pd supported on mesoporous
activated carbons with high oxidation resistance as catalysts for toluene oxidation,
Appl. Catal. B Environ. 94 (2010) 8–18, https://doi.org/10.1016/j.apcatb.2009.10.
[
[
[
[
0
15.
7] I. Hita, T. Cordero-Lanzac, A. Gallardo, J.M. Arandes, J. Rodríguez-Mirasol,
J. Bilbao, T. Cordero, P. Castaño, Phosphorus-containing activated carbon as acid
support in a bifunctional Pt–Pd catalyst for tire oil hydrocracking, Catal. Commun.
7
8 (2016) 48–51, https://doi.org/10.1016/j.catcom.2016.01.035.
8] M.O. Guerrero-Pérez, J.L.G. Fierro, M.A. Vicente, M.A. Bañares, Support effect on
the structure and reactivity of VSbO4 catalysts for propane ammoxidation to ac-
rylonitrile, Chem. Mater. 19 (2007) 6621–6628, https://doi.org/10.1021/
cm702022d.
9] C.A. Carrero, R. Schloegl, I.E. Wachs, R. Schomaecker, Critical literature review of
the kinetics for the oxidative dehydrogenation of propane over well-defined sup-
ported vanadium oxide catalysts, ACS Catal. 4 (2014) 3357–3380, https://doi.org/
1
0.1021/cs5003417.
[
[
[
[
10] E. Reichelt, M.P. Heddrich, M. Jahn, A. Michaelis, Fiber based structured materials
for catalytic applications, Appl. Catal. Gen. 476 (2014) 78–90, https://doi.org/10.
1
016/j.apcata.2014.02.021.
11] J. Pérez-Ramírez, R.J. Berger, G. Mul, F. Kapteijn, J.A. Moulijn, The six-flow reactor
technology a review on fast catalyst screening and kinetic studies, Catal. Today 60
(2000) 93–109, https://doi.org/10.1016/S0920-5861(00)00321-7.
12] R.J. Berger, J. Pérez-Ramírez, F. Kapteijn, J.A. Moulijn, Catalyst performance
testing: The influence of catalyst bed dilution on the conversion observed, Chem.
Eng. J. 90 (2002) 173–183, https://doi.org/10.1016/S1385-8947(02)00078-5.
13] N. Bhardwaj, S.C. Kundu, Electrospinning: A fascinating fiber fabrication technique,
Biotechnol. Adv. 28 (2010) 325–347, https://doi.org/10.1016/j.biotechadv.2010.
01.004.
Fig. 9. Comparison of the propane ODH results between the best catalysts
prepared in this study and others reported in the literature. (a) Propylene
productivity and (b) Propane consumption TOF rates as a function of the
temperature at a space-time (Wcat/FC3H8) of 0.1 g·s·mLC3H8. The literature re-
sults may have been obtained with different space-times and the numbers in (a)
and in (b) indicate the cited reference in the reported reviews [25] and [9],
respectively.
[14] Z.-M. Huang, Y.-Z. Zhang, M. Kotaki, S. Ramakrishna, A review on polymer nano-
fibers by electrospinning and their applications in nanocomposites, Compos. Sci.
Technol. 63 (2003) 2223–2253, https://doi.org/10.1016/S0266-3538(03)00178-7.
−
1
[
15] R. Ruiz-Rosas, J. Bedia, J.M. Rosas, M. Lallave, I.G. Loscertales, J. Rodríguez-
Mirasol, T. Cordero, Methanol decomposition on electrospun zirconia nanofibers,
Catal. Today 187 (2012) 77–87, https://doi.org/10.1016/j.cattod.2011.10.031.
16] R. Berenguer, F.J. García-Mateos, R. Ruiz-Rosas, D. Cazorla-Amorós, E. Morallón,
J. Rodríguez-Mirasol, T. Cordero, Biomass-derived binderless fibrous carbon elec-
trodes for ultrafast energy storage, Green Chem. 18 (2016) 1506–1515, https://doi.
org/10.1039/c5gc02409a.
17] M. Lallave, J. Bedia, R. Ruiz-Rosas, J. Rodríguez-Mirasol, T. Cordero, J.C. Otero,
M. Marquez, A. Barrero, I.G. Loscertales, Filled and hollow carbon nanofibers by
coaxial electrospinning of alcell lignin without binder polymers, Adv. Mater. 19
(2007) 4292–4296, https://doi.org/10.1002/adma.200700963.
18] R. Berenguer, J. Fornells, F.J. García-Mateos, M.O. Guerrero-Pérez, J. Rodríguez-
Mirasol, T. Cordero, Novel synthesis method of porous VPO catalysts with fibrous
structure by electrospinning, Catal. Today. 277 (Part 2) (2016) 266–273, https://
doi.org/10.1016/j.cattod.2016.03.002.
19] S. Singh, V. Singh, M. Vijayakumar, V.V. Bhanu Prasad, Dual fiber behavior of
polyvinyl alcohol/zirconium n-propoxide composite fibrous mats prepared via
electrospinning, Ceram. Int. 39 (2013) 5031–5037, https://doi.org/10.1016/j.
ceramint.2012.11.101.
[
species for alkane activation, the prepared solids are quite attractive as
partial oxidation catalysts. The surface dispersion of vanadium oxide
[
[
[
[
(
VO
crystallization of amorphous ZrO
tastable tetragonal to monoclinic phase in zirconia, while favoring the
2 5
development of the porous texture and avoiding the formation of V O
x
) during the zirconia oxide sub microfiber formation, delay the
2
and inhibit the conversion of me-
crystals. It has been also shown that these materials do not present
deactivation processes under time on stream experiments. The propane
ODH results in terms of conversion and selectivity are quite promising
when they are compared with those already reported elsewhere.
20] S. Singh, V. Singh, M. Vijayakumar, V.V. Bhanu Prasad, Electrospun ZrO2 fibers
obtained from polyvinyl alcohol/zirconium n-propoxide composite fibers processed
through halide free sol–gel route using acetic acid as a stabilizer, Mater. Lett. 115
Acknowledgements
(2014) 64–67, https://doi.org/10.1016/j.matlet.2013.10.026.
This work was supported by the Spanish Ministry of Economy and
12

J.J. Ternero-Hidalgo et al.
Catalysis Today xxx (xxxx) xxx–xxx
[
[
21] Propene (Propylene), (n.d.). http://www.essentialchemicalindustry.org/chemicals/
propene.html (Accessed March, 2018).
22] Ethylene, propylene demand will experience increased growth in 2005-10, (n.d.).
http://www.ogj.com/articles/print/volume-103/issue-45/special-report/ethylene-
propylene-demand-will-experience-increased-growth-in-2005-10.html (Accessed
March 20, 2018).
zirconia, Appl. Phys. Mater. Sci. Process. 84 (2006) 431–437, https://doi.org/10.
1007/s00339-006-3631-z.
[40] A. Khodakov, J. Yang, S. Su, E. Iglesia, A.T. Bell, Structure and properties of va-
nadium oxide-zirconia catalysts for propane oxidative dehydrogenation, J. Catal.
177 (1998) 343–351.
[41] C.L. Pieck, M.A. Bañares, J.L.G. Fierro, Propane oxidative dehydrogenation on
VOx/ZrO2 catalysts, J. Catal. 224 (2004) 1–7, https://doi.org/10.1016/j.jcat.2004.
02.024.
[42] B. Horvath, J. Strutz, J. Geyer‐Lippmann, E.G. Horvath, Preparation, properties,
and ESCA characterization of vanadium surface compounds on silicagel, I, ZAAC ‐ J.
Inorg. Gen. Chem. 483 (1981) 181–192, https://doi.org/10.1002/zaac.
19814831223.
[43] M.J. Valero-Romero, A. Cabrera-Molina, M.O. Guerrero-Pérez, J. Rodríguez-
Mirasol, T. Cordero, Carbon materials as template for the preparation of mixed
oxides with controlled morphology and porous structure, Catal. Today 227 (2014)
233–241, https://doi.org/10.1016/j.cattod.2013.10.093.
[44] R. López-Medina, J.L.G. Fierro, M.O. Guerrero-Pérez, M.A. Bañares, Nanoscaled
rutile active phase in Mo-V-Nb-O supported catalysts for the oxidation of propane to
acrylic acid, Appl. Catal. Gen. 375 (2010) 55–62, https://doi.org/10.1016/j.apcata.
2009.12.017.
[
[
[
23] The Propylene Gap: How Can It Be Filled?, Am. Chem. Soc. (n.d.). https://www.acs.
org/content/acs/en/pressroom/cutting-edge-chemistry/the-propylene-gap-how-
can-it-be-filled.html (Accessed April, 2018).
24] A. Agarwal, D. Sengupta, M. El-Halwagi, Sustainable process design approach for
on-purpose propylene production and intensification, ACS Sustain. Chem. Eng. 6
(2018) 2407–2421, https://doi.org/10.1021/acssuschemeng.7b03854.
25] F. Cavani, N. Ballarini, A. Cericola, Oxidative dehydrogenation of ethane and
propane: how far from commercial implementation? Catal. Today 127 (2007)
1
13–131, https://doi.org/10.1016/j.cattod.2007.05.009.
26] J.F. Brazdil, Strategies for the selective catalytic oxidation of alkanes, Top. Catal. 38
2006) 289–294, https://doi.org/10.1007/s11244-006-0027-4.
[
[
[
[
(
27] R.K. Grasselli, Advances and future trends in selective oxidation and ammoxidation
catalysis, Catal. Today 49 (1999) 141–153.
28] I. Amghizar, L.A. Vandewalle, G. Van, G.B. Marin, New trends in olefin production,
Engineering 3 (2017) 171–178, https://doi.org/10.1016/J.ENG.2017.02.006.
29] A. Corma, J.M. López-Nieto, N. Paredes, M. Pérez, Y. Shen, H. Cao, S.L. Suib,
Oxidative dehydrogenation of propane Over supported-vanadium oxide catalysts,
Stud. Surf. Sci. Catal. 72 (1992) 213–220, https://doi.org/10.1016/S0167-
[45] E. Hryha, E. Rutqvist, L. Nyborg, Stoichiometric vanadium oxides studied by XPS,
Surf. Interface Anal. 44 (2012) 1022–1025, https://doi.org/10.1002/sia.3844.
[46] M.A. Bañares, M.V. Martınez-Huerta, X. Gao, J.L.G. Fierro, I.E. Wachs, Dynamic
́
behavior of supported vanadia catalysts in the selective oxidation of ethane: in situ
Raman, UV–Vis DRS and reactivity studies, Catal. Today 61 (2000) 295–301,
https://doi.org/10.1016/S0920-5861(00)00388-6.
2
991(08)61673-0.
[
30] B. Frank, A. Dinse, O. Ovsitser, E.V. Kondratenko, R. Schomäcker, Mass and heat
transfer effects on the oxidative dehydrogenation of propane (ODP) over a low
loaded VOx/Al2O3 catalyst, Appl. Catal. Gen. 323 (2007) 66–76, https://doi.org/
[47] L. Fernandez, E. Sanchez, M. Panizza, M.M. Carnasciali, G. Busca, Vibrational and
electronic spectroscopic properties of zirconia powders, J. Mater. Chem. 11 (2001)
1891–1897, https://doi.org/10.1039/b100909p.
1
0.1016/j.apcata.2007.02.006.
[
31] Y.-M. Liu, Y. Cao, K.-K. Zhu, S.-R. Yan, W.-L. Dai, H.-Y. He, K.-N. Fan, Highly ef-
ficient VOχ/SBA-15 mesoporous catalysts for oxidative dehydrogenation of pro-
pane, Chem. Commun. (2002) 2832–2833.
[48] X. Zhao, D. Vanderbilt, Phonons and lattice dielectric properties of zirconia, Phys.
Rev. B - Condens. Matter Mater. Phys. 65 (2002) 0751051–07510510.
[49] S.C. Su, A.T. Bell, A study of the structure of vanadium oxide dispersed on zirconia,
J. Phys. Chem. B 102 (1998) 7000–7007.
[50] D. Gazzoli, R. De, G. Ferraris, M. Valigi, L. Ferrari, S. Selci, Morphological and
textural characterization of vanadium oxide supported on zirconia by ionic ex-
change, Appl. Surf. Sci. 255 (2008) 2012–2019, https://doi.org/10.1016/j.apsusc.
2008.06.139.
[
32] Y.-M. Liu, Y. Cao, N. Yi, W.-L. Feng, W.-L. Dai, S.-R. Yan, H.-Y. He, K.-N. Fan,
Vanadium oxide supported on mesoporous SBA-15 as highly selective catalysts in
the oxidative dehydrogenation of propane, J. Catal. 224 (2004) 417–428, https://
doi.org/10.1016/j.jcat.2004.03.010.
[
33] T. Blasco, J.M.L. Nieto, Oxidative dyhydrogenation of short chain alkanes on sup-
ported vanadium oxide catalysts, Appl. Catal. Gen. 157 (1997) 117–142, https://
doi.org/10.1016/S0926-860X(97)00029-X.
[51] I.E. Wachs, Raman and IR studies of surface metal oxide species on oxide supports:
supported metal oxide catalysts, Catal. Today 27 (1996) 437–455, https://doi.org/
10.1016/0920-5861(95)00203-0.
[34] J. Haber, Fifty years of my romance with vanadium oxide catalysts, Catal. Today
1
42 (2009) 100–113, https://doi.org/10.1016/j.cattod.2008.11.007.
[52] M.A. Bañares, I.E. Wachs, Molecular structures of supported metal oxide catalysts
under different environments, J. Raman Spectrosc. 33 (2002) 359–380, https://doi.
org/10.1002/jrs.866.
[53] E. Heracleous, M. Machli, A.A. Lemonidou, I.A. Vasalos, Oxidative dehydrogenation
of ethane and propane over vanadia and molybdena supported catalysts, J. Mol.
Catal. Chem. 232 (2005) 29–39, https://doi.org/10.1016/j.molcata.2005.01.027.
[54] J.R. Sohn, J.S. Han, J.S. Lim, Spectroscopic study of V2O5 supported on zirconia
and modified with MoO3, Mater. Chem. Phys. 91 (2005) 558–566, https://doi.org/
10.1016/j.matchemphys.2004.12.024.
[55] M.O. Guerrero-Pérez, J.L.G. Fierro, M.A. Vicente, M.A. Bañares, Effect of Sb/V ratio
and of Sb + V coverage on the molecular structure and activity of alumina-sup-
ported Sb-V-O catalysts for the ammoxidation of propane to acrylonitrile, J. Catal.
206 (2002) 339–348, https://doi.org/10.1006/jcat.2001.3494.
[56] B.-Y. Zhao, X.-P. Xu, H.-R. Ma, D.-H. Sun, J.-M. Gao, Monolayer dispersion of oxides
and salts on surface of ZrO2 and its application in preparation of ZrO2-supported
catalysts with high surface areas, Catal. Lett. 45 (1997) 237–244.
[
35] K.V.R. Chary, C.P. Kumar, T. Rajiah, C.S. Srikanth, Dispersion and reactivity of
monolayer vanadium oxide catalysts supported on zirconia: The influence of mo-
lybdena addition, J. Mol. Catal. Chem. 258 (2006) 313–319, https://doi.org/10.
1
016/j.molcata.2006.05.049.
[
36] C. Shao, H. Guan, Y. Liu, J. Gong, N. Yu, X. Yang, A novel method for making ZrO2
nanofibres via an electrospinning technique, J. Cryst. Growth 267 (2004) 380–384,
https://doi.org/10.1016/j.jcrysgro.2004.03.065.
37] J.Y. Li, Y. Sun, Y. Tan, F.M. Xu, X.L. Shi, N. Ren, Zirconium nitride (ZrN) fibers
prepared by carbothermal reduction and nitridation of electrospun PVP/zirconium
oxychloride composite fibers, Chem. Eng. J. 144 (2008) 149–152, https://doi.org/
[
1
0.1016/j.cej.2008.04.037.
[
38] M. Thommes, K. Kaneko, A.V. Neimark, J.P. Olivier, F. Rodriguez-Reinoso,
J. Rouquerol, K.S.W. Sing, Physisorption of gases, with special reference to the
evaluation of surface area and pore size distribution (IUPAC technical report), Pure
Appl. Chem. 87 (2015) 1051–1069, https://doi.org/10.1515/pac-2014-1117.
39] G.-Y. Guo, Y.-L. Chen, Unusual structural phase transition in nanocrystalline
[
13