2
Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 394 (2014) 1–9
sol–gel catalysts of Si–Al–Mo mixed oxides [20], flame spray
pyrolysis (FSP) synthesis and thermal spreading [21,22]. These
investigations have demonstrated interesting and important
results.
In the present work, a high performance catalyst with molyb-
dena as active species and a kind of well-shaped one-dimensional
and the 1-butene (99.5%) was introduced to the reaction system.
−
1
The reaction conditions were as following: 393 K, 0.1 MPa, 2.4 h
of WHSV (1-C H ). The reaction products were analyzed online by
4
8
gas chromatography equipped with a 50-m fused silica chrompack
PLOT Al O /Na SO capillary column and a flame ionization detec-
2
3
2
4
tor. Regeneration of the used catalysts was performed at 823 K for
6 h in the atmosphere of oxygen.
␥
-Al O as the support is prepared for 1-butene self-metathesis.
2 3
The catalyst shows ∼40 mol% yield of propene and ethene at 393 K
A series of reaction pathways have been observed in the
self-metathesis of 1-butene on catalysts: isomerization, self-
metathesis, cross-metathesis, secondary cross-metathesis and
under 0.1 MPa with a high weight hourly space velocity (WHSV,
−
.4 h ). The one-dimensional ␥-Al O , uniform in size and mor-
2 3
1
2
=
phology (10–20 nm of diameters and ∼100 nm of lengths), supports
most of the active molybdena species in suitable aggregation
sizes. It offers quite active sites and is also stable during multi-
regeneration for 1-butene metathesis.
oligomerization [24]. We classify alkene products as Cn , where
=
n represents the carbon numbers. The conversion X(Cn ) is deter-
mined by subtracting the mass percentage of 1-butene feed in the
blank
outlet
W
−W
1−butene
blank effluent gas, as Eq. (1).(1)X(C=1−butene) =
1
−butene
blank
W
1
−butene
blank
2
2
2
. Experimental
W
W
: 1-butene feed from the blank reaction
1
−butene
outlet
: 1-butene feed from the outlet after reaction
1−butene
.1. Catalyst preparation
The
lated
molar
yield
of
the
product
Eq.
is
calcu-
(2):(2)Y(C ) =
=
according
to
following
n
=
=
.1.1. One-dimensional ꢀ-Al O support
X(C )×((W(C ))/n)
2
3
n
n
(
(W(C ))/2)+(W(C=))/3)+(W(C4 ))/4)+(W(C ))/5)+(W(C=))/6)+(W(C=))/7)
=
=
=
All chemical reagents were of analytical grade and used with-
out further purification. One-dimensional ␥-Al O was prepared
2
3
5
6
7
=
where W(Cn ) is the mass percent of the product with the given
2
3
=
=
carbon number, and Y(Cn ) is the molar yield. C7 is used to denote
all the polymerized alkenes with carbon numbers over six. The spe-
cific activity is defined as the number of moles of propene produced
during the reaction per m2 of the catalyst and per second.
by hydrothermal synthesis method. Typically, oleylamine (80–90%,
Aladdin Industrial Corporation) and NH ·H O (25%) was dissolved
3
2
in distilled water at 353 K. The alumina sol (Zhejiang Yuda Chemical
Industry Co., Ltd.) was then gradually added into the above solution
understirringfor2 hat353 K. Thenthemixturewas transferredinto
the autoclave, which was sealed and maintained at 453 K for 72 h.
After the reaction, the solids were collected, washed with distilled
water and alcohol, then dried in the air at 373 K for 12 h and cal-
2.3. Characterization
XRD patterns of the samples were recorded in the 2ꢁ region of
◦
cined at 823 K for 24 h. One-dimensional ␥-Al O3 was designated
10–80 using a diffractometer (Shimadzu XRD-6000) with Cu K␣
2
◦ −1
A˚ ). The scan speed was set at 10 min with
as Al(n).
radiation (ꢂ = 1.5418
a step size of 0.02 .
◦
2
.1.2. Silica modification
The specific surface areas and pore volumes of catalysts were
obtained by means of the nitrogen adsorption measurement at 77 K
on a Micrometrics ASAP 2020 instrument. Samples were outgassed
at 573 K (<10 Torr) for 4 h prior to analysis. Adsorption data were
analyzed by the Barrett–Joyner–Halenda (BJH) method. Pore-size
distribution curves were calculated using the Dolimore–Heal (D–H)
method.
Commercial ␥-Al O powder was obtained from Fu Shun
2
3
Research Institute of Petroleum and Petrochemicals, China, which
was designated as Al(c). The silica modified ␥-Al O was prepared
according to the procedure reported elsewhere [23]. Typically,
the samples of silica modified Al(c) and Al(n) were prepared by
immersing 1 g of commercial ␥-Al O3 or one-dimensional ␥-Al O3
−
3
2
3
2
2
into an ethanol solution (20 mL) containing a certain amount of
TEM measurements were conducted with JEOL JEM-100S, using
an accelerating voltage of 80 kV. Samples for TEM measurements
were suspended in ethanol and ultrasonically dispersed. Drops
of the suspensions were applied to a carbon-coated copper grid.
FTIR spectra of catalysts were obtained on a FT-IR spectrometer
3
2
-aminopropyl-triethoxylsilane (Aldrich, 99.9%) under stirring for
h at the room temperature. The obtained solid was dried and then
calcined at 823 K for 12 h in air. They were designated as ySi/Al(c)
and ySi/Al(n), respectively, where y is the SiO2 loading in weight
percent (wt.%).
−
1
(Bruker Vector 22) with a resolution of 4 cm at room tempera-
ture. Raman spectra were recorded with a Renishaw invia system
equipped with a confocal microscope. A 514.5 nm exciting line was
focused using a 50× objective lens. The laser power at the sam-
ple was 20 mW. Diffuse reflectance UV–vis spectra were recorded
with a UV-2401PC spectrometer under air-exposed conditions in
2.1.3. MoO3 loaded on ySi/Al(c) and ySi/Al (n)
The catalysts of supported molybdena were prepared by
impregnating the supports, i.e. Al(c), Al(n), ySi/Al(c) and ySi/Al(n),
with an appropriate amount of ammonium molybdate (AR) solu-
tion under stirring for 6 h, then dried at 373 K, and finally
calcined at 823 K for 8 h in air. The prepared catalysts were
denoted as xMo/ySi/Al(c) and xMo/ySi/Al(n), respectively, where
x is the MoO3 loading in weight percent (wt.%). The cata-
lysts were pressed and crushed to particles of 20–40 mesh
for 1-butene metathesis reaction and performed corresponding
characterization.
−
1
the range 200–800 nm. The scan speed was 120 nm min and the
BaSO4 powder was used as the reference.
NH3-TPD profiles were obtained by using a micromeritics
AutoChem TP-5080 apparatus with a thermal conductivity detec-
tor (Tianjin Golden Eagle Technology Limited Corporation). Prior to
measurement, 0.1 g of the sample was pretreated in a helium flow
at 873 K for 2 h and then cooled down to 373 K. Subsequently a
flow of ammonia/helium mixed gas (5 vol.% ammonia) was passed
and maintained for 30 min. Then the excess ammonia was removed
by a helium flow at 373 K for 1 h. The ammonia desorption was
performed in the temperature range of 373–873 K at a rate of
2.2. Catalytic test
The 1-butene metathesis reaction was carried out in
a
−
1
continuous-flow, fixed-bed system. 0.2 g of catalyst was placed
onto a quartz wool plug located in a quartz U-tube. After the catalyst
10 K min . The desorbed NH3 was detected using a thermal con-
ductivity detector (TCD).
−
1
was pretreated at 873 K for 2 h under Ar (99.9%, 60 mL min ), then
cooled down to the desired reaction temperatures in the Ar flow
Temperature-programmed reduction of H2 (H -TPR) measure-
2
ments were carried out in a conventional setup (Tianjin Xianquan