J.-L. Sánchez-García et al. / Catalysis Communications 27 (2012) 154–158
155
2
. Experimental
to 673 K. The feedstock composition was kept at 10 vol % of n-butane
and 90 vol % of O /He (molar ratio of 5/95) and a total flow rate of
100 cm NTP/min using mass flow controllers. Butane conversion was
2
3
2
.1. Catalyst synthesis
determined according to:
2 2
Catalysts 10 VZrS and 20 VZrS: ZrO –SiO mixed oxide supports
were prepared by contacting an appropriate amount of Zirconium(IV)
n-butoxide (Aldrich, 80%) with 473 K dried, pure silica (Degussa,
Aerosil 380 m /g) to give surface Zr densities of 4 and 8 atom Zr/nm .
During this procedure the solid powder was maintained 12 h continu-
ously immersed and stirred under toluene reflux at room temperature.
The solids were then dried and calcined in flowing oxygen at
%XnC4 ¼ ðmoles nC4 consumed=moles nC4 in feedÞ⋅100
2
2
while the selectivity to reaction products was calculated as follows:
%Si ¼ ðvnC4=viÞ:ðmoles product I=moles nC4 consumedÞ⋅100
6
73 K for 8 h, before contacting with vanadyl (V) triisopropoxide
where vnC4 and v
i
are, respectively, the stoichiometric factors of nC
4
and
(
Alfa Aesar, 98%) to achieve 4 wt % V loading. This procedure was anal-
the product I, in the reaction nC
4
→I.
ogous to that used with the Zr grafting procedure described, but using
isopropanol as the solvent, maintaining reflux at 333 K for 12 h, and fi-
nally drying and calcining the solid at 823 K for 8 h.
With mass spectroscopy, traces of other compounds were
detected, which included acrolein, benzene, 2-propenoic acid, formic
acid, and acetic acid.
2 2
Catalysts VCeGS and VCeIS: In both preparations, CeO -SiO sup-
ports were synthesized such that the Ce surface density corresponded
to 5 atom/nm . In the first preparation, an appropriate amount of
3. Results and discussion
2
Cerium (IV) isopropoxide (Alfa Aesar) was contacted with the dried
Aerosil silica immersed in isopropanol under continuous stirring
and reflux at 343 K for 24 h, left to dry overnight, and calcined at
3.1. Catalyst and support characterization
Table 1 summarizes the composition of each catalyst and the
823 K under flowing oxygen for 8 h. In the VCeIS preparation, the silica
2
BET surface areas calculated from the N physisorption data. After
powder was wet to incipient wetness with an appropriate concentration
of aqueous cerium ammonium nitrate solution (Mallincrodt Baker),
dried overnight at 373 K, and calcined at 823 K. In a second step, the
vanadium grafting procedure described above with the Zr catalysts
was applied to obtain the VCeGS and VCeIS catalysts.
grafting/impregnation, all the surface areas experienced an area re-
duction (up to 55%) from the original Aerosil (380 m /g). For the Zr
2
materials, the X-ray diffraction pattern shows only the amorphous
nature of pure silica in the 10 VZrS material, suggesting that with
the grafting method, high dispersions of zirconium and vanadium
species on the surface were achieved (Fig. 1). In contrast, the 20
VZrS diffraction pattern shows peaks at 30°, 50° and 60° that are
characteristic of tetragonal zirconia crystals, however there is no
evidence of a vanadium oxide crystalline phase present, implying
that vanadium is well dispersed over the modified support. The
literature shows that at the calcination temperature used for this
work, the formation of a stable vanadium oxide monolayer is expected
2
.2. Catalyst characterization techniques
Specific surface area and micropore volumes were determined
with nitrogen isotherms at 77 K using a homemade volumetric ad-
sorption apparatus equipped with precision capacitance manometers
−
6
and capable of achieving 10
obtaining adsorption isotherms, samples were degassed (10
at 473 K for 2 h. BET and t-plot analysis were conducted on at least
ten points obtained in the 0.01–0.40 p/p range. X-ray diffraction
XRD) patterns were obtained using a X'pert PRO (PANalytical) auto-
mated diffractometer system, using Cu Kα radiation (λ=0.15405 nm,
5 kV, 30 mA) and collecting data in the 5°b2θb80° range. Ammonia
Torr under dynamic pumping. Prior to
−
3
torr)
over the ZrO
of the vanadium oxide must be confined to the ZrO
[12]. For the Ce materials, VCeGS and VCeIS, X-ray diffraction shows the
presence of CeO crystals. Additionally, in the VCeGS it shows also the
presence of CeVO . This compound is the product of a solid state reac-
tion that can occur between highly dispersed vanadium and cerium at
temperatures above 500 °C [13]. The CeVO formed during calcination
2
, and since the affinity of V for Zr is higher than for Si, most
2
in the mixed oxide
o
(
2
4
3
adsorption microcalorimetry was conducted at 473 K on a homemade
Tian–Calvet calorimeter whose design and operation have been de-
scribed elsewhere [11]. In brief, 0.2 g sample sizes were previously
4
(550 °C) and it was favored by the grafting method. As in the zirconium
catalysts, vanadium oxide crystals were not detected. It is expected that
−
5
degassed under vacuum (10
greater than 10
torr) at 623 K to achieve a leak rate no
4
the vanadium that is not part of CeVO must be interacting mainly with
−
4
torr/min at that temperature, before inserting the
the CeO , since the vanadia–silica interactions are very weak [13,14].
2
cells in the calorimeter and stabilizing the heat flow baseline. Ammonia
gas doses of 10 μmol/g were admitted in succession onto the sample,
allowing sufficient time (ca. 30 min) for each dose to achieve thermal
equilibrium before proceeding with the next dose. This procedure was
The microcalorimetry data at 200 °C show differential heat values
≤100 kJ/mol for all the materials (Fig. 2). These values are lower than
those reported by other researchers for ammonia adsorption over
pure ZrO
values for V/SiO
loading, with the highest values (100 kJ/mol) determined for mate-
rials with crystalline V . For highly dispersed vanadium, as in the
materials studied here, the differential heat of adsorption is below
0 kJ/mol [15]. Since the values measured in this work are between
those of pure ZrO and those for highly dispersed V/SiO , the V
2
, which typically are as high as 180 kJ/mol [15]. The heat
continued up to a saturation pressure of at least 3 torr NH
both isotherm and differential heat of adsorption data. For acidity com-
parison purposes, important NH adsorption site densities are those
that correspond to adsorption heats greater than 60 kJ/mol.
3
, obtaining
2
are usually lower and dependent on the vanadium
3
2 5
O
6
2
.3. Catalytic activity
2
2
must be moderating the acidity of Zr, resulting in a less acidic material
The catalytic activity for the conversion of n-butane to ODH, par-
tial oxidation and cycloaddition products was evaluated in a continu-
ous flow quartz reactor at ambient pressure. Reactant and product
concentrations were measured on-line with gas chromatography
Table 1
Synthesized catalysts and BET surface area of zirconium and cerium catalysts.
Catalyst
Zr (nm−2)
Ce (nm−2
)
BET area (m /g)
2
V micro (cm /g)
3
(
Chromepak MicroGC, TCD detector, CPSil-5 and Hayesep-A columns).
Product gases were also collected via a sampling bulb and analyzed
with a Selective Mass Detector 5973 Network coupled with a 6890 N
GC system (Agilent Technologies). The catalyst charge of 100 mg was
pretreated at 823 K in flowing oxygen for 60 min, followed by cooling
10 VZrS
20 VZrS
VCeGS
VCeIS
4
8
0
0
0
0
5
5
239
266
168
230
0.011
0.014
0
0
Sánchez-García, José-Luis
