The role of alumina in the supported Mo/HBeta-Al2O3 catalyst for olefin metathesis: A high-resolution solid-state NMR and electron microscopy study

Article abstract of DOI:10.1016/j.jcat.2007.05.019

High-resolution solid-state MAS NMR; hyperpolarized ¹²⁹Xe NMR, SEM, and TEM; XRD; and N₂ adsorption were used to investigate the role of alumina in the Mo/HBeta-Al₂O₃ catalyst. XRD and N₂ adsorption showed that introduction of alumina into the support may protect the HBeta framework from destruction. Quantitative ²⁹Si MAS, ²⁷Al MAS, and MQ MAS NMR spectra demonstrated that dealumination occurs preferentially at specific T-sites in the Mo/HBeta catalyst; however, the framework aluminum in HBeta zeolites remains at the same crystallographic sites after the addition of alumina. Hyperpolarized ¹²⁹Xe NMR, HRTEM, and SEM images indicate that Mo species are readily dispersed in alumina rather than in HBeta zeolites. The preferential migration of Mo species into the alumina pores may preserve the integrity of the HBeta zeolite framework and result in the moderate distribution of Mo species and acidity in the composite support. These may be correlated with the high performance of Mo/HBeta-Al₂O₃ catalysts in the metathesis of ethene and butene-2 to propene.

Full text of DOI:10.1016/j.jcat.2007.05.019

Journal of Catalysis 250 (2007) 55–66
www.elsevier.com/locate/jcat
The role of alumina in the supported Mo/HBeta–Al2O3 catalyst
for olefin metathesis: A high-resolution solid-state NMR
and electron microscopy study
Xiujie Li, Weiping Zhang ∗, Shenglin Liu, Longya Xu, Xiuwen Han, Xinhe Bao ∗
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, China
Received 5 February 2007; revised 25 May 2007; accepted 28 May 2007
Available online 28 June 2007
Abstract
High-resolution solid-state MAS NMR; hyperpolarized
role of alumina in the Mo/HBeta–Al O catalyst. XRD and N adsorption showed that introduction of alumina into the support may protect the
129
Xe NMR, SEM, and TEM; XRD; and N adsorption were used to investigate the
2
2
3
2
29
27
HBeta framework from destruction. Quantitative Si MAS, Al MAS, and MQ MAS NMR spectra demonstrated that dealumination occurs
preferentially at specific T-sites in the Mo/HBeta catalyst; however, the framework aluminum in HBeta zeolites remains at the same crystallo-
129
graphic sites after the addition of alumina. Hyperpolarized
Xe NMR, HRTEM, and SEM images indicate that Mo species are readily dispersed
in alumina rather than in HBeta zeolites. The preferential migration of Mo species into the alumina pores may preserve the integrity of the HBeta
zeolite framework and result in the moderate distribution of Mo species and acidity in the composite support. These may be correlated with the
high performance of Mo/HBeta–Al O catalysts in the metathesis of ethene and butene-2 to propene.
2
3
© 2007 Elsevier Inc. All rights reserved.
27
129
Keywords: Olefin metathesis; Mo/HBeta–Al O catalyst; Solid-state NMR; Al MQ MAS NMR; Hyperpolarized
Xe NMR; Electron microscopy
2
3
1. Introduction
calcined [13], which may restrict its range of practical appli-
cations. The supported tungsten oxide catalysts are less active
and require much higher reaction temperatures (523–773 K)
to achieve considerable metathesis activity [14,15]. Recently,
Olefin metathesis has received increasing attention and is
now considered one of the most important C–C bond forma-
tion reactions [1–3]. Chauvin, Grubbs, and Schrock won the
Nobel Prize in chemistry in 2005 for their development of the
olefin metathesis in homogeneous systems. Heterogeneous cat-
alysts for metathesis reactions are of further interest due to
their ease of separation, good persistence, and recyclability
[4,5]. A number of solid catalysts have been reported for the
olefin metathesis, the most feasible of which are those based
on Re, W, and Mo [6–16]. Of these, supported Re2O7/Al2O3
is highly active and selective even at room temperature, and
its catalytic performance can be improved by using composite
supports, such as Al2O3–SiO2 [9], Al2O3–B2O3 [10,11], and
Al2O3–P2O5 [12]. However, rhenium oxide will sublime when
our laboratory reported that Mo supported on HBeta–Al2O3
composites exhibited high activity in the metathesis of ethene
and butene-2 to propene at ca. 393 K, which offers an alter-
native pathway for the production of propene [16]. In contrast,
Mo supported on HBeta zeolites performed poorly due to the
strong interaction between Mo species and HBeta zeolites [17].
Thus, the composition and structure of the support play an im-
portant role in the heterogeneous olefin metathesis reaction. In
this work, we demonstrate the importance of choosing a suit-
able support for optimizing the propene production. The role
of alumina in the composite support have been revealed by
29Si MAS NMR, 27Al MAS NMR, two-dimensional 27Al 3Q
MAS NMR, N2 adsorption, hyperpolarized 129Xe NMR, high-
resolution SEM, and TEM. The varying distributions of Mo
species in the HBeta and alumina composite support have been
demonstrated. These may be correlated with the high catalytic
*
Corresponding authors. Fax: +86 411 8469 4447.
E-mail addresses: wpzhang@dicp.ac.cn (W. Zhang), xhbao@dicp.ac.cn
(X. Bao).
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.05.019

56
X. Li et al. / Journal of Catalysis 250 (2007) 55–66
performance of Mo/HBeta–Al2O3 in the metathesis of ethene
and butene-2 to propene.
used for the last conversion step (0Q → 1Q), which was the
central transition selective soft 90◦ pulse. A two-dimensional
(2D) Fourier transformation followed by a shearing transforma-
tion gave a pure absorption mode 2D contour plot [21–23]. The
second-order quadrupolar effect (SOQE) and isotropic chemi-
cal shift (δiso) values were calculated according to the proce-
dures in Ref. [21]. 29Si MAS NMR spectra with high power
proton decoupling were recorded at 79.4 kHz using a 7.5-mm
MAS probe with a spinning rate of 4 kHz. 4,4-Dimethyl-4-
silapentane sulfonate sodium (DSS) was used as the chemical
shift reference for 29Si MAS NMR spectroscopy. A total of
1024 scans were accumulated with a π/4 pulse width of 2 µs
2. Experimental
2.1. Catalyst preparation
The HBeta and γ -Al2O3 composite support was prepared by
extruding a mixture of γ -Al2O3 and HBeta zeolite (Si/Al = 15
provided by Fushun Petroleum Company, China) powder into
strips of ca. 2 mm diameter. The supports, designated HB-nAl,
were then calcined at 773 K for 2 h and ground to 16–32 mesh.
Catalysts containing ca. 4.0 wt% Mo were prepared by wet im-
pregnation of HBeta–Al2O3 strips with an aqueous solution of
(NH4)6Mo7O24·4H2O, then dried at 393 K, and finally calcined
at 953 K for 2 h. The catalysts are designated 4Mo/HB-nAl,
where n (%) represents the weight percent of γ -Al2O3 in the
support.
1
and a 4-s recycle delay. Before the H MAS NMR measure-
ments, samples were dehydrated at 400 ◦C at a pressure below
1
10−2 Pa for 20 h. H MAS NMR spectra were collected at
399.9 MHz using a single-pulse sequence with a π/4 pulse,
a 4-s recycle delay, and a spinning speed of 10 kHz. Chem-
ical shifts were referenced to DSS. For the determination of
quantitative results, all samples were weighed, and the spectra
were calibrated by measuring a known amount of 1,1,1,3,3,3-
hexafluoro-2-propanol performed in the same conditions [19].
The Dmfit software was used for deconvolution using fitted
Gaussian–Lorentzian line shapes [24].
2.2. Catalyst characterization
2.2.1. XRD, N2 adsorption, SEM, and TEM measurements
XRD patterns were obtained at room temperature on a
Rigaku D/Max-RB diffractometer using CuKα radiation. Pow-
der diffractograms of samples were recorded over a range of
2θ values from 5◦ to 50◦ at 40 kV and 100 mA with a scan-
ning rate of 8 deg/min. The integrated intensity of the signal at
2θ = 22.4◦ was used to evaluate the crystallinity of the Beta ze-
olites. Nitrogen sorption experiments were performed at 77 K
on an ASAP 2000 system in the static measurement mode.
Samples were degassed at 623 K for 10 h before the mea-
surements. Specific surface areas were calculated by the BET
method, and the pore volume was determined by N2 adsorption
at a relative pressure of 0.98. SEM images were obtained using
a Hitachi S4800 field-emission microscope. TEM images were
obtained with a Philips CM 200 microscope equipped with a
CCD camera at an acceleration voltage of 200 kV. Energy-
dispersive spectroscopy (EDS) analysis also was done using
this microscope equipped with an X-ray energy analyzer.
2.2.3. Hyperpolarized 129Xe NMR measurements
Continuous-flow hyperpolarized 129Xe NMR spectra were
collected at 110 MHz under static conditions. Optical polariza-
tion of xenon was achieved with a home-built pumping cell in
the fringe field of the spectrometer magnet and a 60 W dual
diode laser array (Coherent Inc.). A flow of 1% Xe–1% N2–
98% He gas mixture was delivered to the sample region via
plastic tubing. Variable temperature measurements also were
performed from room temperature to 153 K. All of the spec-
tra were acquired with a 3.0-µs π/2 pulse, 200–400 scans, and
a 2-s recycle delay. Chemical shifts were referenced to the sig-
nal of xenon gas.
2.3. Catalyst evaluation
The catalysts (∼2.0 g) were tested in a 10-mm-diameter,
fixed-bed flow microreactor [17]. After activation for 1 h at
823 K under nitrogen to remove the moisture, the catalysts
were cooled to the reaction temperature at 393 K. The reaction
products were analyzed using a Shimadzu GC-8A gas chro-
matograph with a flame ionization detector. Propene was the
main product; trace amounts of C5–C10 oligomers also were
detected. The conversion of ethene and butene-2 was calculated
on the basis of carbon number using methane and butane as the
internal standards.
2.2.2. MAS NMR measurements
All NMR spectra were recorded on Varian Infinityplus-
400 spectrometer. 27Al MAS NMR experiments were carried
out at 104.2 MHz using a 4-mm MAS NMR probe with
a spinning rate of 15 kHz. Chemical shifts were referenced
to (NH4)Al(SO4)2·12H2O at −0.4 ppm as a secondary ref-
erence. The spectra were accumulated for 1024 scans with
a π/12 flip angle and 2-s pulse delay. For a quantitative
comparison, all samples were weighted and hydrated com-
pletely in a desiccator with saturated NH4NO3 solution, and
the spectra were calibrated by measuring a known amount of
(NH4)Al(SO4)2·12H2O under the same NMR acquisition con-
ditions [18,19]. 27Al 3Q MAS NMR experiments were per-
formed using a three-pulse sequence incorporating a z-filter at a
spinning speed of 25 kHz with a 2.5-mm probe [20]. An rf field
of 200 kHz was used for the creation (0Q → 3Q) and the
first conversion ( 3Q → 0Q) pulses. An rf field of 18 kHz was
3. Results and discussion
3.1. XRD patterns and N2 adsorption data
XRD patterns of Mo species loaded on HBeta zeolite with
different alumina contents are shown in Fig. 1. It can be seen
that the resultant HBeta zeolites have typical BEA structures

X. Li et al. / Journal of Catalysis 250 (2007) 55–66
57
Fig. 1. XRD patterns of HBeta zeolites and 4Mo/HB-nAl catalysts.
Table 1
Framework Si/Al ratios, Al contents and relative crystallinity of the HBeta and 4Mo/HB-nAl samples
Sample
HBeta
4Mo/HB
4Mo/HB-10Al
4Mo/HB-30Al
4Mo/HB-70Al
a
Relative crystallinity (%)
Framework Si/Al ratio
100
24
592
60
34
400
72
31
455
70
28
486
72
28
482
b
−1
c
Framework Al content (µmol g
a
(
5%))
◦
Determined by XRD patterns and the intensity of the signal at 2θ = 22.4 was used to evaluate the crystallinity of the Beta zeolites.
b
c
29
Framework Si/Al ratio of HBeta zeolites in the composite support determined by Si MAS NMR.
27
Framework Al content of HBeta zeolites in the composite support determined by Al MAS NMR using (NH )Al(SO ) ·12H O as the external standard.
4
4 2
2
Table 2
Surface areas and pore volumes of HB-nAl supports and 4Mo/HB-nAl catalysts
Sample
HBeta 4Mo/HB HB-10Al 4Mo/HB-10Al HB-30Al 4Mo/HB-30Al HB-70Al 4Mo/HB-70Al Al O 4Mo/Al O
2
3
2
3
2
BET surface area (m /g support) 632
Total pore volume (ml/g support) 0.47
334
0.24
589
0.46
380
0.43
483
0.45
384
0.40
334
0.38
296
0.35
223
0.39
210
0.35
and good crystallinity [25]. On loading Mo species, the struc-
ture of HBeta zeolites is retained, but their relative crystallinity
decreases significantly compared with the parent zeolites. As
shown in Table 1, introduction of 4% Mo leads to a crystallinity
loss of 40% for HBeta zeolites. Interestingly, 4% Mo loading
results in only a 28% decrease in crystallinity after addition of
10% alumina into the support, indicating that alumina in the
composite support may protect the HBeta zeolites framework
from destruction. In addition, no γ -Al2O3 phase is detected in
the XRD spectra until its content reaches 30%, and no diffrac-
tions of bulk MoO3 can be observed in any sample, indicating
that Mo species are highly dispersed on the support.
port. This may be due to the strong interaction between Mo
species and HBeta support, which leads to partial dealumina-
tion of the framework [17]. It should be noted here that the loss
of BET surface area and pore volume on Mo species loaded
on HB-nAl composite support is not as obvious as that occur-
ring on HBeta support. For example, introduction of 4% Mo
species into HBeta support leads to nearly a 50% reduction in
the BET surface area. For the 4Mo/HB-30Al catalyst, only a
20% decrease is observed, and the surface area of γ -Al2O3 re-
mains relatively unchanged after addition of Mo species. This
indicates that addition of alumina may preserve the framework
structure of HBeta zeolites.
Table 2 summarizes the BET surface areas and pore vol-
umes of 4Mo/HB-nAl samples measured by N2 adsorption. It
is found that the parent HBeta zeolites have the largest surface
area and pore volume. Introduction of Mo species leads to a
significant decrease in the surface area and total pore volume
for all supports, with the effect most obvious on the HBeta sup-
3.2. 29Si MAS NMR
Fig. 2 shows the 29Si MAS NMR spectra of the 4Mo/HB-
nAl catalysts. The lines at −117 and −113 ppm are attributed
to the crystallographically inequivalent sites of the Si(OSi)4

58
X. Li et al. / Journal of Catalysis 250 (2007) 55–66
29
Fig. 2. Si MAS NMR spectra of HBeta zeolites and 4Mo/HB-nAl catalysts, which can be deconvoluted using four Gaussian–Lorentzian lines.
27
Fig. 3. Al MAS NMR spectra of HBeta zeolites and 4Mo/HB-nAl catalysts recorded at a resonance frequency of 104.2 MHz with the sample spinning rate of 15
kHz and 1024 scans. (Prior to measurements, samples were completely hydrated in a desiccator with saturated NH NO solution.)
4
3
groupings in HBeta zeolites. The peak at −107 ppm is supposed
to come from the contribution of Si(OAl)(OSi)3 groupings [26].
The peak at −103 ppm is assigned to the silanols with the struc-
ture of Si(OSi)3OH because it would be highly enhanced after
1H → 29Si cross-polarization, as shown in our previous study
[17]. The framework Si/Al ratios of HBeta zeolites in the com-
posite support can be calculated by deconvolution of the 29Si
MAS NMR spectra. As shown in Table 1, at a fixed Mo loading,
the framework Si/Al ratio decreases with increasing alumina
content in the support, indicating that addition of alumina into
the support may prevent dealumination of the HBeta zeolitic
framework. This is in accordance with the above results from
N2 adsorption and XRD.
3.3. 27Al MAS and two-dimensional MQ MAS NMR
Fig. 3 illustrates the 27Al MAS NMR spectra of 4Mo/HB-
nAl samples. The peaks at about 54 and 0 ppm in spectrum
of HBeta zeolites are assigned to the tetrahedrally coordinated
framework and octahedrally coordinated extra-framework Al,
respectively [27]. In addition, the introduction of Mo species
on the support produces two new peaks at −14 and 14 ppm,

X. Li et al. / Journal of Catalysis 250 (2007) 55–66
59
(A)
27
27
Fig. 4. Two-dimensional Al MQ MAS NMR spectra of samples: (A) HBeta, (B) 4Mo/HB, (C) 4Mo/HB-30Al. The corresponding Al MAS NMR spectrum is
given on the top of the MQ MAS plot. The F projection is the pure isotropic spectrum. Asterisk denotes the sidebands.
1
corresponding to the nonhydrated and hydrated forms of the
Al2(MoO4)3 phase [17,28]. Two broad lines gradually emerge
at around 6 and 67 ppm with increasing alumina content in the
support; these peaks are assigned to the octahedral and tetra-
hedral Al in γ -alumina, respectively [29]. Table 1 shows the
quantitative analysis of the framework Al content from 27Al
MAS NMR measurements. The framework Al content of HBeta
zeolites decreases from 592 to 400 µmol/g after introduction of
Mo species, which means the dealumination of HBeta frame-
work. However, the concentration of framework Al in HBeta
zeolites increases to 486 µmol/g after addition of 30% alumina
into the support. This indicates that addition of alumina into the
support may protect the HBeta framework from destruction.
Two-dimensional 27Al MQ MAS NMR spectra are used to
remove the anisotropic line broadening, allowing identification
of species with similar isotropic chemical shifts but different
quadrupolar coupling constants. As shown in Fig. 4, the over-
lapped peaks observed in the 27Al MAS spectra at 54 ppm
are clearly resolved in the MQ MAS spectra for HBeta zeo-
lites. Two distinct framework Al species, designated Al(IV)a
and Al(IV)b, can be seen in the tetrahedral region and a third
one, Al(VI)a, is also seen in the octahedral environment at
ca. 0 ppm. When HBeta was impregnated with Mo species,
the 27Al MQ MAS spectra of Mo/HBeta reveals an addi-
tional species, Al(IV)c, with an isotropic chemical shift (δiso) at
58.8 ppm along with the line at 14 ppm assigned to a hydrated
species of aluminum molybdate, Al2(MoO4)3. The signal of
Al(IV)c appears in the resonance region of the tetrahedral alu-
minum framework and exhibits a large quadrupolar interaction
with the SOQE of 4.0 MHz. This large anisotropic quadrupolar
broadening makes it difficult to identify in the usual 27Al MAS
NMR spectra, however, it is clearly noticeable in the isotropic
F1 projection. At the same time, we note that the intensity of
the Al(IV)a signal is readily decreased from the HBeta frame-
work after impregnation of Mo species. The Al(IV)a/Al(IV)b
ratio in the framework aluminum content decreases from 0.6
to 0.06. This implies that introduction of Mo species preferen-
tially removes aluminum atoms from crystallographic positions
represented by Al(IV)a. In the meantime two new peaks of
Al(IV)c and Al(VI)b can be seen in the spectra. Consequently,
it is the introduction of Mo species that leads to formation
of a third framework Al(IV)c species and the subsequent ap-
pearance of nonframework Al2(MoO4)3. This suggests that the
Al atoms corresponding to peak Al(IV)c can be assigned to

60
X. Li et al. / Journal of Catalysis 250 (2007) 55–66
(B)
Fig. 4. (continued)
the distorted extra-framework tetrahedral aluminum, similar to
the findings reported by van Bokhoven et al. in the hydrother-
mal treatment of HBeta zeolites [30,31]. For the 4Mo/HB-30Al
catalyst, along with the tetrahedral Al(IV)d in alumina, some
Al(IV)a can be seen in the HBeta framework, with an increase
in Al(IV)a/Al(IV)b ratio from 0.06 to 0.23. At the same time,
the peak belonging to the distorted tetrahedral Al(IV)c is not
so evident as that in the 4Mo/HB catalyst. It exhibits a smaller
quadrupolar interaction (δiso = 58.3 ppm, SOQE = 3.2 MHz).
Therefore, alumina can protect the specific T-sites in the frame-
work of HBeta zeolites when impregnating Mo species. The
framework aluminum in HBeta zeolites can remain at the same
crystallographic sites after addition of alumina into the sup-
port.
and long relaxation time, especially for the mesoporous mate-
rials [33]. An increase in sensitivity of several orders of mag-
nitude can be achieved using laser polarization techniques to
produce hyperpolarized xenon [34–36]. Fig. 5 shows the hyper-
polarized 129Xe NMR spectra of Mo/HB-nAl samples acquired
at different temperatures. The hyperpolarized 129Xe NMR spec-
trum acquired for HBeta zeolites at room temperature shows a
peak at 80 ppm in addition to the xenon gas signal at 0 ppm (see
Fig. 5A). The former peak is characteristic of xenon adsorbed
in the 12-membered ring channels of HBeta [37,38]. There is
no signal for xenon adsorbed in pure alumina at room tem-
perature; however, when the measurement temperature drops to
173 K, a signal at about 103 ppm can be seen, which can be as-
signed to xenon adsorbed in the pores of alumina [39]. For the
HBeta–Al2O3 composite support, only the peak at 80 ppm cor-
responding to xenon in the HBeta pores is detectable. This may
be due to the preferential adsorption of xenon in the microp-
ores rather than in the mesopores [32]. After introduction of Mo
species, a small downfield shift of xenon signal in HBeta zeo-
lites is observed at lower temperatures (see Figs. 5B and 5C).
This indicates that a very small portion of Mo species is de-
posited into the pores of HBeta, slightly narrowing the pore
size of HBeta. At the same time, a small peak at ca. 103 ppm
3.4. Hyperpolarized 129Xe NMR
129Xe NMR spectroscopy is a powerful tool for characteriz-
ing the porous structure and location of guest species in ma-
terials [32]. Its main advantage is the high sensitivity of the
chemical shift of 129Xe to its local environments. However,
the application of thermally polarized 129Xe NMR to materi-
als is often hampered by a relatively weak signal-to-noise ratio

X. Li et al. / Journal of Catalysis 250 (2007) 55–66
61
(C)
Fig. 4. (continued)
emerges on Mo loading, which is consistent with the signal
3.5. SEM and TEM images
from the empty alumina. This peak becomes more evident and
shifts to the lower field when the temperature decreases down
to 153 K for the HBeta–Al2O3 composite supported with 4 or
9 wt% Mo. However, the signal from the empty alumina is
absent in the 129Xe NMR spectra of HBeta–Al2O3 composite
support. This demonstrates that addition of Mo species into the
composite support may inhibit the exchange of xenon gas in the
HBeta and alumina domains, and thus Mo species may locate
at the boundary of the HBeta and alumina. At a Mo loading as
high as 9 wt%, an additional peak is centered at ca. 145 ppm
along with the peak of xenon in the empty alumina at 131 ppm.
This peak can be ascribed to the xenon adsorbed in the alu-
mina with Mo species, because the deposition of Mo species
into the pores of alumina may decrease the average free path of
xenon in alumina, leading to a low-field shift of the xenon sig-
nal [32]. Thus, the hyperpolarized 129Xe NMR measurements
clearly show that the distribution of Mo species is heteroge-
neous in the HBeta zeolites and alumina composite support.
These species are deposited more at the boundary and in the
pores of alumina than in the pores of HBeta zeolites at higher
Mo loadings.
The morphology and distribution of Mo species in HBeta
and alumina composite support were further investigated by
high-resolution SEM and TEM. Fig. 6A shows the SEM image
of the 4Mo/HB-30Al sample, showing the HBeta particles sur-
rounded by irregular alumina floccules. Table 3 gives the quan-
titative analysis of the components by X-ray EDS, demonstrat-
ing the concentrations of Al and Mo elements are higher in site
I than in site II. Therefore, Mo species are distributed more in
the alumina than in the HBeta zeolites. This finding is verified
by the TEM images. Fig. 6B presents a cross-sectional TEM
image of the 4Mo/HBeta-30Al sample, showing a clear bound-
ary between site aa and site bb. Quantitative EDS measurement
shows that site bb has a much higher Al content than site aa
(see Table 3); thus, site bb should be associated with alumina,
whereas site aa corresponds to HBeta zeolites. The concentra-
tion of Mo elements is higher in alumina than in HBeta zeolites,
indicating that Mo species are preferentially distributed in alu-
mina rather than in HBeta zeolites, which is consistent with the
above results from hyperpolarized 129Xe NMR. This phenom-
ena was also observed by Li et al. in Ni–Mo sulfide catalysts

62
X. Li et al. / Journal of Catalysis 250 (2007) 55–66
(A)
(B)
129
Fig. 5. Variable-temperature hyperpolarized
(B) 173 K, (C) 153 K.
Xe NMR spectra of HBeta zeolites, alumina, and Mo/HB-nAl catalysts acquired at different temperatures: (A) 293 K,
supported on mixed USY zeolites and alumina for hydrotreat-
ment [40].
the nonacidic unperturbed extra-framework aluminum hydrox-
yls, silanol groups, and Al–OH experiencing hydrogen bonds
with neighboring oxygen atoms. The peak at 3.9 ppm should be
assigned to the bridging hydroxyl groups, i.e. Brønsted acidic
sites [41]. The broad line centered at 5.1 ppm also can be seen in
the deconvoluted spectra, which may be attributed to a second
Brønsted acidic site interacting electrostatically with the zeolite
framework because it could be significantly suppressed after
Al irradiation [17]. Quantitative analysis of the spectra demon-
strates that introduction of 4% Mo leads to the consumption
of Brønsted acid sites from 228 to 127 µmol/g of the sup-
port, likely due to the strong interaction between the Mo species
and the HBeta zeolites [17]. When alumina is incorporated into
the support, more Brønsted acid sites (i.e., 151 µmol/g) are
preserved on the 4Mo/HB-30Al catalyst. This is in agreement
3.6. 1H MAS NMR
High-resolution 1H MAS NMR is a powerful and direct
method for characterizing the acidic sites in solid catalysts.
Compared with IR, it can provide quantitative information on
the hydroxyl species without the problem of extinction co-
efficients [41]. Fig. 7 shows the 1H MAS NMR spectra of
4Mo/HB-nAl catalysts. Two broad peaks can be seen on the
4Mo/Al2O3 sample. The line at about −0.2 ppm is assigned
to the basic hydroxyls on alumina, and the line centered at
ca. 2.2 ppm belongs to the acidic hydroxyls [42]. In addition,
three peaks at about 0.8, 1.7, and 2.3 ppm can be differenti-
ated on HBeta zeolites, which are attributed, respectively, to

X. Li et al. / Journal of Catalysis 250 (2007) 55–66
63
(C)
Fig. 5. (continued)
Table 3
Chemical compositions of different parts in 4Mo/HB-30Al catalyst detected by energy dispersive spectroscopy (EDS)
Element
I (molar fraction%)
II (molar fraction%)
bb (molar fraction%)
aa (molar fraction%)
O
Al
Si
63.1
20.2
16.1
0.6
63.5
10.4
25.7
0.4
65.9
24.8
8.2
67.2
3.2
29.2
0.4
Mo
1.1
with the above 27Al MQ MAS NMR results in which more
tetrahedral aluminum remained in the framework of 4Mo/HB-
30Al compared with 4Mo/HBeta. Further increasing the alu-
mina content in the support to 70% decreases the concentration
of Brønsted acid sites to ca. 90 µmol/g, due mainly to the lower
content of HBeta in the support.
alumina support. Clearly, the changes in catalytic activity are
related to the structure and property of the support.
Comparing the results from solid-state NMR and SEM/TEM
studies with the above catalytic performance of Mo/HBeta–
Al2O3 seems to show that addition of γ -alumina into the sup-
port protects the framework of HBeta zeolites from destruction.
In addition, the preferential migration of Mo species into the
alumina pores may preserve the aluminum atoms at specific
T-positions in HBeta zeolites. This may result in the moder-
ate distribution of Mo species and also a proper acidity of the
3.7. Catalytic performance of cross-metathesis of ethene and
butene-2 to propene
1
support, as evidenced by H MAS NMR spectra. Combined
Fig. 8 illustrates the catalytic performance of olefin metathe-
sis on 4Mo/HBeta-nAl catalysts as a function of alumina con-
tent. Mo supported on HBeta zeolites clearly shows the worst
performance, with butene-2 conversion of <20% and propene
selectivity of ca. 15%. Addition of alumina in the support
(e.g., 4Mo/HBeta-10Al) produces significantly improved cat-
alytic performance. As the alumina content is increased, the
catalytic activity increases accordingly. At an alumina content
of 30–50 wt%, the 4Mo/HB-30Al catalyst shows the highest
metathesis activity. Under these conditions, the butene-2 con-
version is up to 80%, and the propene selectivity is about 95%.
These catalysts also have high stability, with no obvious de-
activation after 7 h. Further increases in alumina content (e.g.,
4Mo/HBeta-70Al) lead to decreased catalyst activity, however.
The conversion of butene-2 is only 35% for Mo loaded on pure
with catalytic performance data, this may indicate that Brøn-
sted acidic sites are involved in the metathesis reaction. This is
consistent with the results from Mol et al. [43–45], who also
observed a correlation between metathesis activity and Brøn-
sted acidic sites on Re2O7/Al2O3 catalysts. We suspect that the
dispersion of Mo species and addition of alumina into the sup-
port, as well as the proper acidity, may be advantageous for the
metathesis of ethene and butene-2 to propene.
4. Conclusion
4Mo/HBeta–30%Al2O3 catalyst demonstrates much better
performance for the metathesis of ethene and butene-2 to
propene compared with the catalyst without alumina in the sup-
port. XRD, N2 adsorption, and solid-state MAS NMR results

64
X. Li et al. / Journal of Catalysis 250 (2007) 55–66
(A)
(B)
Fig. 6. SEM (A), cross-sectional TEM (B) images, and EDS analysis of 4Mo/HB-30Al catalyst.
demonstrate that the addition of Al2O3 into the support protects
the framework of HBeta zeolites from destruction. Hyperpo-
larized 129Xe NMR, SEM, and TEM images indicate that Mo
species are dispersed more in the alumina than in the HBeta ze-
olites. Such a distribution may preserve the aluminum atoms
at specific T-positions in HBeta zeolites, as evidenced by 27Al
MQ MAS NMR, and result in the moderate distribution of Mo
species and acidity in the composite support. These effects may
contribute to the good performance of Mo/HBeta–Al2O3 cata-
lysts in olefin metathesis.

X. Li et al. / Journal of Catalysis 250 (2007) 55–66
65
1
Fig. 7. H MAS NMR spectra of 4Mo/HB-nAl catalysts. The spinning rate was 10 kHz, and 200 single-pulse scans were accumulated.
(A)
(B)
Fig. 8. Catalytic conversions (A) and selectivities (B) of 4Mo/HB-nAl catalysts with different alumina contents in the metathesis of ethene and butene-2 to propene
−1
(reaction temperature: 393 K, pressure: 1.0 MPa, ethene/2-butene = 3:1, WHSV of ethene: 1.2 h ).
Acknowledgments
Dr. D. Su (Fritz-Haber Institute, Germany) for help with the
SEM and TEM measurements and the anonymous reviewers
for their helpful discussions and suggestions.
Financial support was provided by the National Natural Sci-
ence Foundation of China (grants 20403017 and 20303019)
and the Ministry of Science and Technology of China through
the National Key Project of Fundamental Research (grant
2003CB615806). The authors thank Dr. J.M. Sun (DICP) and
References
[1] Y. Chauvin, Angew. Chem. Int. Ed. 45 (2006) 3741.
[2] R.R. Schrock, Angew. Chem. Int. Ed. 45 (2006) 3748.

66
X. Li et al. / Journal of Catalysis 250 (2007) 55–66
[3] R.H. Grubbs, S.J. Miller, G.C. Fu, Acc. Chem. Res. 28 (1995) 446.
[4] K.J. Ivin, J.C. Mol, Olefin Metathesis and Metathesis Polymerization,
Academic Press, San Diego, 1997, p. 397.
[25] M.J. Eapen, K.S.N. Reddy, V.P. Shiralkar, Zeolites 14 (1994) 295.
[26] J. Perez-Pariente, J. Sanz, V. Fornes, A. Corma, J. Catal. 124 (1990)
217.
[5] J.C. Mol, J. Mol. Catal. A 213 (2004) 39.
[27] J.A. van Bokhoven, D.C. Koningsberger, P. Kunkeler, A.P.M. Kentgens, J.
Am. Chem. Soc. 122 (2000) 12842.
[28] W.P. Zhang, D. Ma, X.W. Han, X.M. Liu, X.H. Bao, X.W. Guo, X.S.
Wang, J. Catal. 188 (1999) 393.
[29] H. Kraus, M. Müller, R. Prins, J. Phys. Chem. B 102 (1998) 3862.
[30] A. Omegna, M. Vasic, J.A. van Bokhoven, G. Pringruber, R. Prins, Phys.
Chem. Chem. Phys. 6 (2004) 447.
[31] A. Abraham, S.H. Lee, C.H. Shin, S.B. Hong, R. Prins, J.A.
van Bokhoven, Phys. Chem. Chem. Phys. 6 (2004) 3031.
[32] J.L. Bonardet, J. Fraissard, A. Gedeon, M.A. Springuel-Huet, Catal. Rev.
Sci. Eng. 41 (1999) 115.
[33] T. Pietrass, J.M. Kneller, R.A. Assink, M.T. Anderson, J. Phys. Chem.
B 103 (1999) 8837.
[34] M. Haake, A. Pines, J.A. Reimer, R. Seydoux, J. Am. Chem. Soc. 119
(1997) 11711.
[35] I.L. Moudrakovski, A. Nossov, S. Lang, S.R. Breeze, C.I. Ratcliffe, B.
Simard, G. Santyr, J.A. Ripmeester, Chem. Mater. 12 (2000) 1181.
[36] W.P. Zhang, C.I. Ratcliffe, I.L. Moudrakovski, J.A. Ripmeester, Anal.
Chem. 77 (2005) 3379.
[37] S.B. Liu, J.F. Wu, L.J. Ma, T.C. Tsai, I. Wang, J. Catal. 132 (1991)
432.
[6] J.C. Mol, Catal. Today 51 (1999) 289.
ˇ
[7] R. Hamtil, N. Žilková, H. Balcar, J. Cejka, Appl. Catal. A 302 (2006) 193.
[8] A.M.J. Rost, H. Schneider, J.P. Zoller, W.A. Herrmann, F.E. Kühn, J.
Organomet. Chem. 690 (2005) 4712.
[9] M. Sibeijn, J.C. Mol, Appl. Catal. 67 (1991) 279.
[10] X. Xu, C. Boelhouwer, J.I. Benecke, D. Vonk, J.C. Mol, J. Chem. Soc.
Faraday Trans. 82 (1986) 1945.
[11] F.C. Sheu, C.T. Hong, W.L. Hwang, C.J. Shih, J.C. Wu, C.T. Yeh, Catal.
Lett. 14 (1992) 297.
[12] M. Sibeijn, R. Spronk, J.A.R. van Veen, J.C. Mol, Catal. Lett. 8 (1991)
201.
[13] R. Spronk, J.A.R. van Veen, J.C. Mol, J. Catal. 144 (1994) 72.
[14] W. Grunert, R. Feldhaus, K. Anders, E.S. Shpiro, K.H.M. Minachev, J.
Catal. 120 (1989) 444.
[15] C.V. Schalkwyk, A. Spamer, D.J. Moodley, T. Dube, J. Reynhardt, J.M.
Botha, Appl. Catal. A 255 (2003) 121.
[16] S.L. Liu, S.J. Huang, W.J. Xin, J. Bai, S.J. Xie, L.Y. Xu, Catal. Today 93–
95 (2004) 471.
[17] X.J. Li, W.P. Zhang, S.L. Liu, X.W. Han, L.Y. Xu, X.H. Bao, J. Mol. Catal.
A 250 (2006) 94.
[18] H. Kraus, R. Prins, J. Catal. 164 (1996) 260.
[19] M. Müller, G. Harvey, R. Prins, Microporous Mesoporous Mater. 34
(2000) 281.
[20] J.P. Amoureux, C. Fernandez, S. Steuernagel, J. Magn. Reson. 123 (1996)
116.
[38] Q.J. Chen, J.L. Guth, A. Seive, P. Caullet, J. Fraissard, Zeolites 11 (1991)
798.
[39] K. Hagiwara, T. Ebihara, N. Urasato, T. Fujikawa, Appl. Catal. A 285
(2005) 132.
[40] D. Li, A. Nishijima, D.E. Morris, G.D. Guthrie, J. Catal. 188 (1999)
111.
[21] M.E. Simth, E.R.H. van Eck, Prog. Nucl. Magn. Reson. Spectrosc. 34
(1999) 159.
[41] M. Hunger, Catal. Rev. Sci. Eng. 39 (1997) 345.
[42] W.P. Zhang, M.Y. Sun, R. Prins, J. Phys. Chem. B 106 (2002) 11805.
[43] X. Xu, J.C. Mol, C. Boelhouwer, J. Chem. Soc. Faraday Trans. 82 (1986)
2707.
[44] A. Andreini, X. Xu, J.C. Mol, Appl. Catal. 27 (1986) 31.
[45] A. Andreini, J. Mol. Catal. 65 (1991) 359.
[22] J. Rocha, C.M. Morais, C. Fernandez, Top. Curr. Chem. 246 (2004) 141.
[23] A. Medek, J.S. Harwood, L. Frydman, J. Am. Chem. Soc. 117 (1995)
12779.
[24] D. Massiot, F. Fayon, M. Capron, I. King, S. LeCalvé, B. Alonso, J.O.
Durand, B. Bujoli, Z. Gan, G. Hoatson, Magn. Reson. Chem. 40 (2002)
70.