Olefin metathesis with mesoporous rhenium-silicium-aluminum mixed oxides obtained via a one-step non-hydrolytic sol-gel route

Article abstract of DOI:10.1016/j.jcat.2013.02.016

Mesoporous Re-Si-Al, Re-Al, and Re-Si mixed oxides were obtained via a one-step non-hydrolytic sol-gel route followed by calcination. The texture and surface properties of the resulting catalysts were characterized by a combination of EDX, XRD, N₂-physisorption, XPS, ToF-SIMS, and NH ₃-TPD. The loss of rhenium during calcination, the texture and the acidity of the catalysts depended on their composition. Migration of rhenium toward the surface occurred during the calcination treatment, as evidenced by XPS and ToF-SIMS. After calcination, ToF-SIMS showed the presence of well-dispersed ReO_x surface species. The influence of the composition of Re-Si-Al catalysts on their performances in the cross-metathesis of ethene and butene to propene was investigated. The specific activity of the Re-Si-Al catalysts was much higher than that of Re-Al catalysts, whereas the Re-Si sample was not active. The best Re-Si-Al catalysts displayed excellent specific activities (up to 45 mmol g^-1 h^-1) and apparent TOF values (98 × 10^-3 s^-1).

Full text of DOI:10.1016/j.jcat.2013.02.016

Journal of Catalysis 301 (2013) 233–241
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Olefin metathesis with mesoporous rhenium–silicium–aluminum mixed oxides
obtained via a one-step non-hydrolytic sol–gel route
a
a,
⇑
b
c
c
Karim Bouchmella , P. Hubert Mutin , Mariana Stoyanova , Claude Poleunis , Pierre Eloy ,
b
c
c,
⇑
Uwe Rodemerck , Eric M. Gaigneaux , Damien P. Debecker
a
Institut Charles Gerhardt, UMR 5253, CNRS-UM2-ENSCM-UM1, Université Monpellier 2, cc 1701, Montpellier 34095, France
Leibniz-Institut für Katalyse e.V. an der Universität Rostock, Albert-Einstein-Str. 29a, 18059 Rostock, Germany
Institute of Condensed Matter and Nanoscience – MOlecules, Solids and reactiviTy (IMCN/MOST), Université catholique de Louvain, Croix du Sud 2 box L7.05.17, 1348
b
c
Louvain-La-Neuve, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
Mesoporous Re–Si–Al, Re–Al, and Re–Si mixed oxides were obtained via a one-step non-hydrolytic
sol–gel route followed by calcination. The texture and surface properties of the resulting catalysts were
characterized by a combination of EDX, XRD, N -physisorption, XPS, ToF-SIMS, and NH -TPD. The loss of
2 3
rhenium during calcination, the texture and the acidity of the catalysts depended on their composition.
Migration of rhenium toward the surface occurred during the calcination treatment, as evidenced by XPS
Received 13 October 2012
Revised 14 January 2013
Accepted 18 February 2013
x
and ToF-SIMS. After calcination, ToF-SIMS showed the presence of well-dispersed ReO surface species.
Keywords:
Alkene metathesis
XPS
The influence of the composition of Re–Si–Al catalysts on their performances in the cross-metathesis
of ethene and butene to propene was investigated. The specific activity of the Re–Si–Al catalysts was
much higher than that of Re–Al catalysts, whereas the Re–Si sample was not active. The best Re–Si–Al
ToF-SIMS
ꢀ
1
ꢀ1
catalysts displayed excellent specific activities (up to 45 mmol g
h ) and apparent TOF values
Rhenium catalysts
Non-aqueous
Mixed oxides
(98 ꢁ 10ꢀ
3 ꢀ1
s
).
Ó 2013 Elsevier Inc. All rights reserved.
2 7
Re O
Silica
Alumina
Aluminosilicate
1
. Introduction
Olefin metathesis is one of the very few fundamentally novel
advantage of rhenium catalysts is that they are more tolerant to
functional groups such as esters than tungsten and molybdenum
catalysts [2,3,11]. Rhenium catalysts are generally based on
reactions discovered in the last 50 years. Olefin metathesis over
heterogeneous catalysts has various industrial applications in the
fields of petrochemicals, oleochemicals, polymers, and specialty
chemicals [1–3]. The first industrial process (the Phillips Triolefin
Process) was used to produce ethene and 2-butene from propene
with a heterogeneous catalyst. Nowadays, the reverse reaction is
of interest because the demand for propene is high. Propene is
mainly used in the production of polypropylene. The metathesis
reaction is catalyzed by supported transition metal oxides [4],
and the most successful metathesis catalysts are based on rhenium
Re
mixed oxide (SiO
catalysts in olefin metathesis is influenced by the nature of the
support [14]: for example, Re /SiO –Al catalysts are known
to be more active than Re /Al [15], while Re /SiO catalysts
are virtually inactive in olefin metathesis [16,17].
Impregnation is classically used to prepare supported rhenium
oxide catalysts [13,18,19]. Well-dispersed surface ReO species
can also be obtained by thermal spreading of bulk rhenium oxide
physically mixed with a support [20,21]. Spreading of rhenium
oxide species has also been observed during alcohol oxidation
reactions [22]. Supported rhenium oxide catalysts can be prepared
by grafting rhenium alkoxo complexes onto chlorinated alumina or
2
O
7
[12,13] supported on different oxide (TiO
2 2 3 2
, Al O , SiO ) or
2
–Al ) supports. The activity of heterogeneous
2 3
O
O
2 7
2
2 3
O
2
O
7
2
O
3
2
O
7
2
x
[
5], molybdenum [6–8], or tungsten oxides [9]. Among them, the
rhenium-based catalysts are highly interesting because they are
very active and selective even at room temperature [10]. Another
2 7
silica–alumina [23] or by reaction of gaseous Re O with the
hydroxyl groups of a zeolite to form Si–OReO –Al species [24].
3
⇑
Corresponding authors. Fax: +33 4 6714 3852 (P.H. Mutin), fax: +32 10 47 3649
D.P. Debecker).
E-mail addresses: hubert.mutin@univ-montp2.fr (P. H. Mutin), damien.debecker@
All of these methods are limited either by the number of com-
mercially available supports or by the need to prepare tailored sup-
ports in separate steps. In addition, controlling the deposition of
(
uclouvain.be (D.P. Debecker).
0
021-9517/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2013.02.016

2
34
K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
the active phase onto the surface of the support is not straightfor-
ward [25]. The active species must be dispersed in their most ac-
tive form on the available surface avoiding the precipitation of
rhenium oxide and the plugging of the pores. A specific problem
encountered with rhenium-based catalysts is linked to the volatil-
ity of rhenium heptoxide [26]. Rhenium heptoxide sublimes at
2. Experimental
2.1. Preparation of the catalysts
The non-hydrolytic sol–gel syntheses [47] were performed un-
der an argon atmosphere using a glove box. SiCl4 (Alfa Aesar,
99.9%), AlCl (Alfa Aesar, 99.9%), and ReCl (Alfa Aesar, 99.5%) were
5
35 K, which may lead to the loss of rhenium during the calcina-
3
5
i
tion step. Several authors point out this difficulty, independently
of the support and the preparation method [21,27–31].
used as received. Diisopropyl ether ( Pr O) was purchased from
Aldrich with 99% purity and was further dried by distillation over
2
The sol–gel process appears attractive since it allows in princi-
ple the preparation of mixed oxide catalysts in one step. However,
there are very few examples of rhenium-based catalysts obtained
sodium wire. The catalysts were prepared in 2-g quantities in
80 ml autoclaves. The chloride precursors were introduced first
i
in the autoclave, then a stoichiometric amount of Pr O was added.
2
i
by a sol–gel method. Scheckler-Nahama et al. reported the prepa-
More precisely, the number of moles of Pr O is calculated, so that
the number of Pr groups in Pr O is equal to the total number of Cl
2
i
i
i
ration from HReO
4
and Al(O Pr)
3
precursors of a Re–Al metathesis
2
catalyst, which proved to be far less active than a catalyst of similar
composition obtained by impregnation [32]. This low activity was
ascribed to the low Lewis acidity of the sol–gel derived catalyst and
to the fact that a large proportion of Re atoms was occluded in the
mass of the material and not accessible to propylene. Fabris et al.
groups in the precursors. Finally, the solvent (10 ml of CH Cl ) was
2
2
introduced. The solution obtained was heated at 110 °C for 4 days
under autogenous pressure (ca. 0.7 MPa). After cooling down to
room temperature, the gel was washed with CH Cl and dried at
2
2
20 °C under vacuum (10 Pa) for 1 h and then for 4 h at 120 °C.
The xerogel was then crushed in a mortar and calcined in a muffle
oven for 5 h at 500 °C (heating rate 10 K/min). The color of the
resulting powders ranged from white to brown depending on their
Al content. The Re–Al and Re–Si materials are labeled as ReX–Al
and ReX–Si, respectively, where X represents the experimental
Re/(Re + Si + Al) atomic ratio (in %). The Re–Si–Al materials are la-
beled ReX–SiAlY, where X is the Re/(Re + Si + Al) atomic ratio (in %)
and Y the nominal Si/Al atomic ratio (see Table 1).
reported the synthesis of a Re–Si catalyst from (NH
4 4
)ReO and
Si(OEt) , but this sample showed a low activity in octene self-
4
metathesis [33], ascribed again to a lower accessibility of Re spe-
cies and to the general poorer activity of silica-supported rhenium
oxide catalysts in the olefin metathesis. Yoboue et al. prepared a
Re–Si catalyst from HReO
ity in the direct conversion of methanol to dimethoxymethane
34]. The authors used this one-pot sol–gel method to lessen the
4 4
and Si(OEt) which displayed high activ-
[
loss of rhenium and to tune the interactions between the metal
and the support.
A rhenium catalyst supported on alumina with a Re/(Re + Si + Al)
atomic ratio of 3% was prepared by the incipient wetness (IW)
impregnation method and used as a reference catalyst; this sample
Although rhenium oxide catalysts supported on silica–alumina
are more active in olefin metathesis than Re catalysts supported
either on alumina or on silica, one-step sol–gel methods have not
been used for the synthesis of Re–Si–Al ternary mixed oxides.
In recent years, the non-hydrolytic sol–gel (NHSG) method has
attracted growing interest for the preparation of various oxide and
mixed oxide materials [35]. The application of NHSG chemistry to
the preparation of heterogeneous catalysts has been reviewed re-
cently [36]. The NHSG route based on the reaction of chloride pre-
cursors with an ether (‘‘ether route’’) is recognized as a powerful
method for the single-step synthesis of mesoporous mixed oxides
with controlled composition, homogeneity, and texture, avoiding
the use of any reactivity modifier, templating agent, or further
supercritical drying step [35,37]. As reviewed recently [38], this
route has been successfully used to prepare various efficient binary
and ternary mixed oxide catalysts, including Nb–V(–Si) [39], Si–Ti
is labeled IW-Re3/Al. Perrhenic acid, HReO (Merck), was used as a
4
precursor.
2 3
c-Al O (Evonik) with a specific surface area of
2
ꢀ1
3
ꢀ1
240 m g and a pore volume of 0.6 cm g was used as support
material. After impregnation, the catalyst was dried at 150 °C for
3 h then calcined at 550 °C for 8 h.
2.2. Characterization of the catalysts
The atomic percentages of Re, Si, Al, and Cl were measured by
energy dispersive X-ray spectroscopy (EDX). Measurements were
carried out using an X-Max Silicon Drift Detector mounted on a
FEI Quanta FEG 200 scanning electron microscope. To obtain reli-
able statistics in the analysis, the data for each sample were taken
as the average of three separate measurements.
[
40], Si–Zr [41], Ti–V [42,43], Ag(–Nb)–Al [44], or Si–Al–Mo [45].
The ether route is based on the in situ formation of alkoxide
Powder X-ray diffraction (XRD) diffractograms of the fresh cat-
alysts were obtained with a Philips X-Pert Pro II diffractometer
groups by reaction of halide groups with the ether (Eq. (1)), fol-
lowed by non-hydrolytic condensation between these alkoxide
groups and the remaining chloride groups (Eq. (2)).
using the K
corded between 20° and 80° with rate of 0.02° s
physisorption experiments were performed at 77 K on a
a radiation of Cu (k = 1.5418 Å). The 2h range was re-
ꢀ1
.
N
2
Micromeritics Tristar sorptometer. The samples were outgassed
M-Cl þ ROR ! M-OR þ RCl
ð1Þ
ð2Þ
at 150 °C under vacuum (2 Pa) overnight. The specific surface area
was determined via the BET method in the 0.05–0.30 P/P range.
0
The pore size distribution was derived from the desorption branch
using the BJH method. The average pore diameter was calculated
as (4 ꢁ Pore Volume/BET specific surface area). The total micropore
volume of the fresh samples was estimated using t-plot analysis.
X-ray photoelectron spectroscopy (XPS) was performed on a
SSX 100/206 photoelectron spectrometer from Surface Science
Instruments (USA) equipped with a monochromatized microfocus
Al X-ray source (powered at 20 mA and 10 kV). The sample pow-
ders pressed in small stainless steel troughs of 4 mm diameter
were placed on a ceramic carousel. The pressure in the analysis
M-Cl þ M-OR ! M—O—M þ RCl
Recently, we have reported the NHSG synthesis of Mo–Si–Al mixed
oxide catalysts [45] and on their excellent performances in propene
self-metathesis [46].
In the present work, we describe the one-step NHSG synthesis
of Re–Si–Al, Re–Al, and Re–Si catalysts by the ether route. The
resulting materials were thoroughly characterized, and their cata-
lytic performances in the metathesis of ethene and butene were
investigated and compared to that of reference metathesis cata-
lysts, including the molybdenum–silica–alumina catalysts pro-
posed recently [46].
ꢀ6
chamber was around 10 Pa. The angle between the surface nor-
mal and the axis of the analyzer lens was 55°. The analyzed area
2
was approximately 1.4 mm , and the pass energy was set at

K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
235
density was lower than 1011 Bi /cm . The powders were supported
þ
2
Table 1
3
Nominal atomic ratios (N) of the (mixed) oxide samples compared to the experi-
mental atomic ratios of the catalysts and supports before calcination (xerogels labeled
X), after calcination (labeled C), and after the catalytic test (labeled P) derived from
EDX. Because in some cases calcination resulted in the loss of Re, the catalysts have
been named according to their actual composition after calcination and not after their
initially targeted composition.
Ò
by pressing them onto the adhesive side of ‘‘Post-it ’’ pieces.
The acidity was evaluated by temperature-programmed
3 3
desorption of NH (NH -TPD) on a Micromeritics AutoChem 2910
apparatus with a thermal conductivity detector. A 100 mg sample
was preheated in helium at 500 °C for 60 min (ramp 10 °C min ).
ꢀ1
ꢀ1
Adsorption of NH
3
(5 vol.% in helium; flow rate 30 ml min ) was
Sample names
Atomic ratios
done at 100 °C for 45 min. Physisorbed NH
ing with helium at 100 °C for 1 h (flow rate 30 ml min ). The TPD
3
was removed by purg-
Re/(Re + Si + Al)
Si/Al
Cl/(Re + Si + Al)
ꢀ1
Re3–Al
N
X
C
P
0.024
0.031
0.029
0.031
0
0
0
0
0
measurement was conducted by heating the sample from 100 to
0.460
0.012
0.013
ꢀ1
5
00 °C at a 10 °C min rate.
Re0–Al
N
X
C
0
na
0.00
0
na
0
0
na
0.026
2
.3. Metathesis reaction
Although industrial feeds comprise both cis- and trans-2-butene,
Re3–SiAl0.3
N
X
C
P
0.025
0.024
0.025
0.028
0.24
0.30
0.28
0.26
0
the cross-metathesis of ethene and pure trans-2-butene to propene
was selected as a model reaction to allow better comparison
between catalysts with different cis–trans isomerization activity.
Indeed, trans-2-butene usually reacts faster than cis-2-butene on
Re catalysts. The activity of the catalysts was measured in a mul-
ti-channel apparatus, which allows fully automated control of gas
flows and of three temperature zones (gas pre-heating, reactor,
and post-reactor lines with 16-port valve) along with reactor
switching and product sampling [48]. All catalysts were sieved
0.410
0.001
0.003
Re0–SiAl0.3
Re1–SiAl14
N
X
C
0
0.27
0.23
0.22
0
0.000
0.000
0.350
0.031
N
X
C
P
0.028
na
0.007
0.010
14.3
na
14.2
15.1
0
na
0.000
0.000
and selected within the 200–315
catalysts (200 mg) were introduced in quartz straight reactors
4 mm i.d.). Prior to reaction, the catalysts were activated for 2 h
lm granulometric fraction. The
Re0–SiAl14
Re2–Si
N
X
C
0
13.4
11.3
13.3
0
0.000
0.000
0.007
0.000
(
ꢀ1
ꢀ1
at 550 °C (temperature ramp of 5 K min ) in dry N
flow in each reactor). Afterward, the system was cooled down to
the reaction temperature (40 °C) under the same N flow. The
reaction was carried out at 40 °C in butene:ethene:N
45:45:10 vol.%) total flow of 8 ml min . trans-2-Butene (99.00%),
ethene (99.95%), and N (99.999%) were further purified over
Molsieve 3A (Roth) filters. N was also purified by an oxygen filter
2
(8 ml min
N
X
C
P
0.029
0.036
0.015
0.020
NA
NA
NA
NA
0
0.025
0.001
0.002
2
a
2
ꢀ1
(
Re0–Si
N
X
C
0
na
0.000
NA
na
NA
0.0
na
0.000
2
2
IW-Re3/Al
N
C
P
0.03
0.03
0.03
0
0
(Oxysorb-glass, Linde). The composition of the reaction gas was
analyzed online by an Agilent 6890 GC. The separation of hydrocar-
bons was performed on a HP-AL/M column (30 m length, 0.53 mm
0.000
0.000
0.000
0.000
na = not analyzed, NA: not applicable.
i.d., 0.15
9
6
l
m film thickness) applying a temperature ramp between
5 and 140 °C and FID detection. Product analysis took about
.5 min for each injection. N , used as internal standard, was ana-
2
lyzed on a HP Plot-Q column with TCD detection. The experiments
were carried out at atmospheric pressure. The specific activity is
defined as the number of moles of propene produced per gram of
catalyst and per hour. The apparent turnover frequency (TOF) is
defined as the number of moles of propene produced per mole of
Re (considering the total Re content, even if only part of the Re is
actually active) and per second.
1
50 eV. Under these conditions, the resolution determined by the
full width at half maximum (FWHM) of the Au 4f7/2 peak was
around 1.6 eV. A flood gun set at 8 eV and a Ni grid placed 3 mm
above the sample surface were used for charge stabilization. The
following sequence of spectra was recorded: survey spectrum, C
1
s, O 1s, Si 2p, Al 2p, Re 4f, and C 1s again to check the stability
of charge compensation in function of time and the absence of deg-
radation of the sample during the analysis. The binding energies
were calculated with respect to the Al 2p peak fixed at 74.5 eV
3. Results
(
or for the sample without alumina, the binding energies were cal-
culated with respect to the Si 2p peak fixed at 103.0 eV). Data eval-
uation was performed with CasaXPS (Casa Software Ltd., UK), and
some spectra were decomposed with the least squares fitting rou-
tine provided by the software with a Gaussian/Lorentzian (85/15)
product function after subtraction of a non-linear baseline. Molar
fractions were calculated using peak areas normalized on the basis
of acquisition parameters and sensitivity factors provided by the
manufacturer.
3.1. Synthesis, composition, and structure
A series of Re–Al, Re–Si, and Re–Si–Al mixed oxides with a
nominal Re/(Re + Si + Al) atomic ratio of approximately 0.03
(corresponding to ꢂ10 wt% Re
2 7
O ) were prepared for this study.
The corresponding silica, alumina, and silica–alumina materials
(without Re) were synthesized under the same conditions to inves-
tigate the influence of rhenium on texture and acidity. These Re-
free oxides are referred to as ‘‘supports’’ in this manuscript, by
analogy with more classical preparation studies, even though they
were not actually used as supports here. All the samples were ob-
ToF-SIMS measurements were performed with an IONTOF V
spectrometer (IONTOF GmbH, Münster, Germany). The samples
þ
were bombarded with a pulsed Bi3 ion beam (30 keV). The ana-
lyzed area used in this work was a square of 500 ꢁ 500
l
m, and
tained by a non-hydrolytic sol–gel route based on the reaction of
i
the data acquisition time was 60 s. Charge effects were
the chlorides with Pr
2
O at 110 °C for 4 days [49], except for IW-
compensated by means of an interlaced pulsed electron flood
Re3/Al which was prepared by incipient wetness impregnation of
k
gun (E = 20 eV). With these parameters, the primary ion dose
a commercial alumina. The nominal atomic ratios (labeled N) of

2
36
K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
the samples (based on the amounts of reactants) are given in
Table 1 and compared to the experimental atomic ratios deter-
mined by EDX analysis before calcination (labeled X, for ‘‘xerogel’’),
after calcination (C, for ‘‘calcined’’), and after catalytic testing (P,
for ‘‘post-test’’).
samples. This behavior might be related to a catalytic effect of
Lewis acidic Re species on the condensation reactions during the
synthesis, leading to a higher degree of condensation for the
Re-containing samples.
In all cases, the experimental Si/Al ratio calculated from EDX
data was close to the nominal values, indicating that there was
no loss of Si or Al during the synthesis steps. The chlorine content
of the xerogels (uncalcined samples) was rather high, particularly
for the Al-rich samples. This high chlorine content has to be related
to the presence of uncondensed residual Cl groups (instead of hy-
3.3. Surface analysis
The surface composition of selected xerogels and catalysts was
studied by XPS in order to evidence a possible migration of
rhenium toward the surface during the calcination. As calcination
resulted in important Re losses in the silica-rich samples, only
the alumina-rich samples for which no loss of Re occurred were
analyzed (Table 3).
droxyl groups in conventional sol–gel). In addition, adsorbed sol-
i
2
vent (CH Cl
2
) or reaction by-product ( PrCl) could also participate
to the chlorine content. Nevertheless, most of this chlorine was
removed upon calcination, and the final chlorine contamination
was quite low in the silica-rich samples (0.1%) and higher in the
Al-rich samples (1.2–3%).
The high amount of carbon and chlorine found for the xerogels
has to be related to the presence of O Pr and Cl residual groups
i
inherent to the non-hydrolytic route used. After calcination, the
surface carbon content decreased to about 12 at.%, corresponding
to the usual contamination by atmospheric carbon. The surface
chlorine content decreased to 1.1 at.% for the alumina–rhenium
catalyst and to 0.1 at.% for the silica–alumina–rhenium catalysts.
The surface Re/(Re + Si + Al) atomic ratios found for the xerogels
by XPS were lower than the bulk Re/(Re + Si + Al) atomic ratio
found by EDX, suggesting that Re species are located more in the
bulk of the materials. This might be caused by a higher condensa-
tion rate around the Re precursors. On the other hand, after calci-
nation, the surface Re/(Re + Si + Al) atomic ratio increased
significantly. Since the bulk composition did not change for these
materials, this implies that Re species do migrate toward the
surface upon thermal treatment. A similar phenomenon was previ-
ously observed for vanadium, molybdenum, and tungsten in Ti–V,
Ti–V–Mo, Ti–V–W, and Si–Al–Mo mixed oxide catalysts prepared
by NHSG [43,45,51]. However, the surface Re/(Re + Si + Al) atomic
ratio found for the IW-Re3/Al sample prepared by impregnation
remains higher than the ratio found for the calcined xerogels,
suggesting that part of the Re species remains in the bulk after cal-
cination. It has to be recalled that XPS explores a depth of up to
No loss of rhenium was observed after calcination for the two
samples with high Al content, Re3–SiAl0.3 and Re3–Al. Conversely,
the experimental Re/(Re + Si + Al) ratio measured after calcination
for the samples with low Al contents (Re2–Si and Re1–SiAl14) was
significantly lower than the nominal one, indicating a loss of about
7
5% of the rhenium for Re1–SiAl14 and of about 50% for Re2–Si.
Analysis of the Re2–Si xerogel (before calcination) showed that
the loss of rhenium occurred during the calcination step (500 °C
for 5 h) and not during the synthesis, washing, or drying steps.
The catalysts were analyzed using X-ray diffraction (diffracto-
grams available as Supplementary material, Fig. S1). Regardless
of the composition of the mixed oxides, all the samples appeared
amorphous, indicating the absence of crystalline rhenium oxide
domains larger than about 5 nm. For the IW-Re3/Al catalyst, broad
peaks were observed at 2h = 32°, 37°, 46°, 61°, and 67° characteris-
tic of the
c-alumina support (according to the PCPDFWIN
database).
3.2. Texture
1
0 nm, with the very large proportion of the signal arising from
The texture of the samples strongly depended on their compo-
the first few nanometers. As we are considering high surface area
solids, the analysis depth is not negligible compared to the thick-
ness of the pore walls; thus, the effect seen in Fig. 2 is presumably
underestimated. ToF-SIMS, probing the outermost surface of the
samples, confirms the effect (vide infra).
XPS also indicated that at the surface of the xerogels, rhenium
was mainly present at the +4 oxidation state. Calcination resulted
in the oxidation of these surface species to mainly the +7 and +6
oxidation states (see Fig. S2 in Supplementary material).
ToF-SIMS measurements were performed to probe the composi-
tion of the outermost surface of the catalysts (before and after cal-
cination) and the nature of the ReOx surface species. The total
count can vary from one analysis to another, and the sensibility
factor is different and unknown for each ion. Therefore, the relative
intensities do not reflect the actual surface concentration of each
2
ꢀ1
sition: the specific surface area varied between 150 and 800 m g
3
ꢀ1
and the pore volume between 0.3 and 1.5 cm g (Table 2). The N
2
adsorption–desorption isotherms were all of type IV (according to
the BDDT classification), typical of mesoporous materials (Fig. 1)
50]. t-Plot analysis indicated that most of the specific surface area
>88%) was located in mesopores.
[
(
The texture of the supports strongly depended on the Si/Al ratio,
with a minimum (in terms of specific surface area and pore vol-
ume) for Si/Al = 0.3. In all cases, the addition of Re led to a signifi-
cant decrease in the pore volume. For the samples with high Al
content (Re3–Al, Re3–SiAl0.3), the specific surface area did not de-
pend on the Re content. On the other hand for the samples with
low Al content (Re1–SiAl14 and Re2–Si), the presence of Re
increased markedly the specific area as compared to the Re-free
ꢀ
species (for example, the Cl detected from the Re3–Al xerogel ac-
counts for ꢂ43% of the total intensity, whereas XPS indicated a
much lower surface concentration). However, different catalysts
can be compared on the basis of the relative intensities of ions of
interest. We focused on the ions containing rhenium, on chloride,
and on the organic ions (Table 4). ToF-SIMS results confirmed
XPS experiments by showing that both Cl and C contamination de-
creased significantly after calcination and that chlorine contamina-
tion remained more important in alumina-rich samples than in
silica-rich samples. The carbon contamination on NHSG-made cat-
alysts was slightly higher than on the impregnated catalyst. The
proportion of Re-containing ions in silica-rich xerogels was signif-
icantly lower than in alumina-rich xerogels.
Table 2
Textural properties of the calcined samples obtained by N
2
-physisorption at 77 K.
Sample
S
BET (m2
g
ꢀ1
)
V
p
(cm3
g
ꢀ1
)
D
p
(nm)
S
l
(m2
g
ꢀ1
)
Re3–Al
Re0–Al
360
360
190
160
630
290
810
550
230
0.6
1.1
0.3
0.7
0.9
1.5
0.8
1.5
0.5
6
12
5
17
6
21
4
11
9
13
37
20
18
24
4
73
63
17
Re3–SiAl0.3
Re0–SiAl0.3
Re1–SiAl14
Re0–SiAl14
Re2–Si
Re0–Si
IW-Re3/Al

K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
237
Re0-SiAl0.3
Re0-SiAl14
Re0-Si
Re3-SiAl0,3
Re1-SiAl14
Re2-Si
1
000
6
4
2
00
00
00
0
Re0-Al
Re3-Al
8
6
4
2
00
00
00
00
0
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P0)
Fig. 1. N
2
adsorption–desorption isotherms at 77 K obtained for Re0–Al, Re0–SiAl0.3, Re0–SiAl14, Re0–Si, Re3–Al, Re3–SiAl0.3, Re1–SiAl14, and Re2–Si.
Table 3
Table 4
Surface characterization by XPS on the xerogels (X) and calcined (C) alumina-rich
catalysts.
Relative contribution of peaks of interest in the ToF-SIMS spectra. Each data point is
an average on three measurements.
Sample
C (at.%) O (at.%) Cl (at.%) Al (at.%) Si (at.%) Re (at.%)
Re/(Re + Si + Al)a Cl/(Re + Si + Al)b Organic/(Re + Si + Al)c
Re3–Al X
Re3–Al C
Re3–SiAl0.3 X 29.7
Re3–SiAl0.3 C 12.3
23.8
12.1
46.8
56.7
43.7
56.9
59.1
7.3
1.1
4.8
0.1
0
21.7
29.1
15.9
22.7
31.1
0
0
5.5
7.1
0
0.3
1.0
0.3
0.8
1.4
Re2–Si X
Re2–Si C
Re1–SiAl14 X
Re1–SiAl14 C
Re3–SiAl0.3 X 0.39
Re3–SiAl0.3 C 0.52
Re3–Al X
Re3–Al C
IW-Re3/Al C
0.11
0.10
0.16
0.08
0.17
0.01
0.10
0.03
1.78
0.46
3.15
0.68
0.02
0.33
0.15
0.29
0.28
0.31
0.15
0.45
0.13
0.08
IW-Re3/Al C
8.4
0.55
0.78
0.82
a
ꢀ
ꢀ
ꢀ
Intensity of all Re-containing ions (Re
x
O
, ReAlOx , Re
x
Cl ) summed and divided
y
y
4
3
2
1
0
ꢀ
ꢀ
ꢀ
ꢀ
by the sum of the intensity of all inorganic peaks (Al
x
O
, Al
x
O
y
H
, Si
x
O
, Si
x
O
y
; H
,
y
z
y
z
ꢀ
ꢀ
ꢀ
ꢀ
AlSiO , Re
x
O
, ReAlOx , Re
x
Cl ).
4
y
y
b
ꢀ
Intensity of the Cl peak divided by the sum of the intensity of all inorganic
peaks.
c
ꢀ
ꢀ
ꢀ
Intensity of all organic ions (C , C
intensity of all inorganic peaks.
x
H
, C
x
H
y
O
) divided by the sum of the
y
z
similar to the proportion found in the IW-Re3/Al sample (0.82),
where the Re is deposited onto the surface.
It must be noted that in the alumina-rich xerogels (Re3–SiAl0.3,
Re3–Al), about 40–60% of the Re-containing anions were chlori-
ꢀ
nated anions Re
x
Cl (Table 5). Surprisingly, no oxychloride
anions were detected. The proportion of chlorinated Re
y
ꢀ
Re
x
Cl
y
O
z
anions drops to ꢃ2–3% for the silica-rich xerogels (Re1–SiAl14
and Re2–Si). In the catalysts (after calcination), the percentage of
chlorinated Re anions is very low (<2%).
ToF-SIMS brings further information on the nuclearity of the Re
species present at the outermost surface of the catalysts. Before
calcination, di- and trinuclear Re anions are detected, particularly
for the xerogels with high alumina content. After calcination, prac-
tically all of the Re anions detected in the mixed oxide catalysts
contained only one Re atom, as in the sample prepared by impreg-
nation (also calcined), suggesting a similar degree of dispersion for
all the catalysts (Table 5).
Fig. 2. Surface Re/(Si + Al + Re) atomic ratios (XPS, gray bars) and bulk Re/
(
Si + Al + Re) atomic ratios (EDX, black squares) for xerogels and calcined catalysts.
After calcination, the proportion of Re-containing ions de-
creased in silica-rich samples, in agreement with the loss of rhe-
nium observed by EDX. In the case of alumina-rich samples (no
loss of Re observed by EDX), the proportion of Re-containing ions
in the calcined samples was significantly higher than in the xero-
gels. This is consistent with the migration of Re species toward
the surface already evidenced by XPS. Remarkably, the proportion
of Re-containing ions in the calcined Re3–Al sample (0.78) is
3.4. Acidity
The metathesis activity of rhenium oxide supported on silica–
alumina catalysts is known to be strongly influenced by the acidity

2
38
K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
Table 5
a silica–alumina was used as a ‘‘support’’ instead of pure alumina
or silica (Fig. 4). As expected, the Re–Si catalyst was virtually inac-
tive. The Re3–SiAl0.3 sample produced propene at a specific activ-
Composition of the rhenium-containing negative ions detected by ToF-SIMS on fresh
xerogels (X) and calcined samples (C). All Re-containing ions detected in the negative
ToF-SIMS spectra have been split in four categories whose contributions are
expressed in% of the total intensities of Re-containing ions: ‘‘Mono-ReO’’ refers to
mononuclear oxygenated Re anions (ReO , ReAlO ); ‘‘Poly-ReO’’ refers to polynuclear
oxygenated Re anions (Re
ꢀ1
ꢀ1
ity of 45 mmol g
h , which corresponds to a yield in propene of
ꢀ
ꢀ
y
ꢃ51%, thus lower than the maximum yield predicted by the ther-
modynamic equilibrium (63%). Fig. 4 also shows that deactivation
was relatively slow compared to Mo-based catalysts tested under
the same experimental conditions [46]. After 2 h on stream, the
activity of the Re-based samples was about 90% of their initial
activity. Importantly, EDX analysis of the catalysts after the cata-
lytic test showed that no loss of rhenium occurred during the test
y
ꢀ
x
O , x > 1); ‘‘Mono-ReCl’’ refers to mononuclear chlorinated
y
ꢀ
ꢀ
Re anions (ReCl ), ‘‘Poly-ReCl’’ refers to polynuclear chlorinated anions (Re
x
Cl , x > 1).
y
y
‘
‘Mono-ReO’’
‘‘Poly-ReO’’
‘‘Mono-ReCl’’
‘‘Poly-ReCl’’
Re2–Si X
95.0
95.9
51.3
37.4
97.9
97.2
97.9
98.4
99.7
2.3
2.3
0.8
0.6
0.7
0.6
0.2
0.2
0.2
1.0
0.4
33.2
47.4
0.5
1.2
1.8
1.3
0.1
1.7
1.4
14.7
14.6
0.9
1.0
0.1
0.1
0.0
Re1–SiAl14 X
Re3–SiAl0.3 X
Re3–Al X
(Table 1).
Re2–Si C
Under our experimental conditions, ethene and trans-2-butenes
were mainly converted to propene and cis-2-butene (Fig. 5). The
formation of other by-products by secondary processes (1-butene,
Re1–SiAl14 C
Re3–SiAl0.3 C
Re3–Al C
IW-Re3/Al
1
-pentene, trans-2-pentene, cis-2-pentene, and hexenes) remained
very low in all cases. Propene selectivity increased with conver-
sion. It must be recalled that cis-2-butene, which is formed by
isomerization of trans-2-butene over acid sites, can still undergo
the metathesis reaction with ethene, to produce the desired pro-
pene (see Scheme in Supplementary information). As the cis/trans
isomerization is an equilibrated reaction, cis-2-butene can also
convert back to trans-2-butene. Thus, cis-2-butene is not ‘‘lost’’
for the reaction of interest and can to some extent be considered
as a reactant, in which case the selectivity for propene is always
very high (above 90%) for all the catalysts displaying significant
activity. It is noteworthy that the cis/trans ratio (taken after
14 min on stream) was similar (ꢃ0.44) in all cases (excepted over
Re2Si which exhibited negligible conversion), suggesting that the
cis/trans isomerization reaction was fast compared to the metath-
esis reaction.
of the support. We used temperature-programmed desorption of
ammonia (NH -TPD) to probe the acidity of the different catalysts
3
and of the corresponding supports. The thermograms (Fig. 3) of the
different samples showed a broad desorption peak centered at
about 200 °C, indicating the presence of weak acidic sites, and
ammonia desorbed up to 450 °C, indicating the presence of stron-
ger acidic sites. The desorption profiles of Re0–Si and Re2–Si sam-
ples were nearly flat, indicating very low acidity.
The amount of NH
the end of the TPD experiment (mmol NH
To take into account the difference in specific surface area, the re-
3
molecules desorbed per gram of sample at
ꢀ
1
3
g ) is given in Table 6.
sults are also expressed as the number of molecules of NH
3
des-
2
ꢀ2
orbed per nm of sample surface (‘‘acid site density,’’ NH
3
nm ).
When alumina is not present in the formulation (Re0–Si and
Re2–Si), the samples are virtually not acidic. The acid site density
4
. Discussion
ꢀ2
was similar (ꢂ0.8 NH
3
nm ) for silica–alumina and alumina sup-
ports (Re0–Al, Re0–SiAl0.3, and Re0–SiA14). According to the data
in Table 6, the acid site density decreased in the following order:
Re3–SiAl0.3 > Re3–Al > Re1–SiAl14 > Re2–Si. The acidity of the alu-
mina-rich catalysts (Re3–Al and Re3–SiAl0.3) was much higher
than that of the corresponding supports.
4.1. Effect of the composition on the properties of the catalysts
prepared by NHSG
The non-hydrolytic sol–gel method offers a very simple route to
mesoporous Re(–Si)–Al catalysts. The composition of the silica–
alumina matrix can be easily adjusted, and this parameter plays
an important role on the loss of Re during calcination, as well as
on the texture, the acidity and the activity of the catalysts. Most
Re-based metathesis catalysts reported in the literature are sup-
ported catalysts prepared by impregnation of alumina or
silica–alumina supports [13,18,19]. In this work, we prepared by
3.5. Catalytic activity
The calcined samples were tested in the cross-metathesis of bu-
tene and ethene to propene at 40 °C. The activity was dependent on
the composition of the catalyst and was markedly improved when
Re0-Si
Re2-Si
Re0-SiAl14
Re1-SiAl14
Re0-SiAl0.3
Re0-Al
Re3-SiAl0.3
Re3-Al
γ-Al O
IW-Re3/Al
2
3
100
200
300
400
500 100
Temperature / °C
200
300
400
500
3
Fig. 3. NH -TPD thermograms of the ‘‘supports’’ (left) and the catalysts (right), normalized to sample weight.

K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
239
Table 6
3
Amount of NH desorbed, determined by TPD in the 100–500 °C temperature range
for the rhenium catalysts and the ‘‘supports.’’
ꢀ1) Acid site density (NH
nm
ꢀ2
)
Sample
NH
3
desorbed (mmol NH
3
g
3
Re3–Al
Re0–Al
Re3–SiAl0.3 1.5
Re0–SiAl0.3 0.2
Re1–SiAl14
Re0–SiAl14
Re2–Si
1.1
0.4
1.8
0.7
4.8
0.8
0.3
0.8
0.01
0.02
0.5
0.3
0.4
0.02
0.02
0.2
Re0–Si
IW-Re3/Al
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
Re3-SiAl0.3
Fig. 5. Selectivity for propene ( ), cis-2-butene ( ), and by-products ( ) which
includes traces of 1-butene, isobutene, butadiene, pentenes, and hexenes. The full
symbols correspond to NHSG samples, empty symbols correspond to the IW-Re3Al
sample (the empty red triangle is almost superposed with the full red triangle).
Data taken at 14 min time on stream.
Mo13-SiAl14
IW-Re3/Al
Re3-Al
Re1-SiAl14
Re2-Si
during calcination. Furthermore, ToF-SIMS results suggest that this
migration can be very effective since the relative contribution of Re
ions from Re3–Al and IW-Re3/Al is similar. After calcination, prac-
tically, only monomeric rhenium oxide ions are detected from the
surface of the NHSG catalysts, as in the case of the catalyst pre-
pared by incipient wetness impregnation. Calcination also results
in the ‘‘cleaning’’ of the catalyst surface, as Cl and carbon contam-
inations are drastically reduced.
The highest specific surface areas are obtained for the Re–Si
sample and the Re–Si–Al sample with high silica content. Addition
of alumina in the composition of the mixed oxides significantly de-
creases the specific surface area (see Table 2). The specific activity
0
20
40
60
80
100 120 140 160 180
Time / min
Fig. 4. Specific metathesis activity versus time on stream. The effect of the support
and the Re contents can be visualized within the series: Re3–Al, Re3–SiAl0.3, Re1–
SiAl14, Re2–Si compared to the best molybdenum catalyst reported Mo13–SiAl14
[
45] and to IW-Re3/Al.
(Table 7), however, does not follow the same trend, as the Re–Si
incipient wetness impregnation of a
alyst (IW-Re3/Al) that was used as a ‘‘reference’’ for comparing to
the samples prepared by NHSG.
c
-alumina support a Re/Al cat-
sample with the highest specific surface area is practically inactive,
while Re3–SiAl0.3, the sample with the lowest specific surface
area, clearly outperforms all other Re-based catalysts, including
the catalyst IW-Re3/Al prepared by impregnation.
To take into account the active phase loading, Table 7 gives the
apparent turnover frequency (TOF). One should keep in mind that
(i) only a very small fraction (typically around 1% [52]) of the active
element in metathesis catalysts actually leads to active sites, and
(ii) this fraction probably varies from one formulation to another.
Additionally, part of the Re may remain in the bulk of the mixed
oxide. To also take into account the effect of the texture, the appar-
ent TOF values were divided by the specific surface area (‘‘areal
TOF’’ in Table 7).
The acidity of the mixed oxide samples strongly depends on
their composition. While the silica and the Re–Si mixed oxide sam-
ples are virtually not acidic, silica–alumina and alumina-based
samples exhibited a significant acidity. As classically observed with
Re-based catalysts, we were confronted to the loss of rhenium dur-
ing the calcination of the catalysts. Rhenium volatilization is, how-
ever, suppressed in the alumina-rich samples. This is consistent
with previous observations made for impregnated catalysts [18].
Losses of Re have been ascribed to the sublimation of rhenium
heptoxide, which can be significant when the interactions between
ReO
x
species and the support are poor, as in the case of silica which
In terms of apparent TOF, the most active catalysts are again
the Re–Si–Al samples. The very high TOF value observed for
Re1–SiAl14 is difficult to interpret in comparison with Re3–SiAl0.3
because several parameters vary at the same time. The Si/Al ratio is
higher, the loss of rhenium during calcination leads to a low Re
loading, and the Re surface concentration determined by XPS
(0.24 at.%) is also significantly lower than the values found for
the other calcined catalysts (Table 3). At the same time, ToF-SIMS
experiments indicate the presence of similar Re surface species
in Re1–SiAl14 and Re3–SiAl0.3. Thus, the high TOF value found
for the Re3–SiAl0.3 is likely related to its particularly high specific
surface area, as suggested by the ‘‘areal TOF’’ values reported in
Table 7.
is not acidic. Conversely, when rhenium species interact strongly
with more acidic supports such as alumina or silica–alumina, the
loss of rhenium during calcination is much lower.
According to XRD, all the samples are amorphous, suggesting a
good dispersion of Re species. One important drawback of catalysts
prepared by one-step sol–gel synthesis is that a large part of the
active phase is located in the pore walls and thus not accessible
to the reactants. This is also true in the case of the xerogels in
which the Re species are most probably dispersed in the bulk of
the material. However, as anticipated from the particularly low
Tammann temperature of rhenium oxide, migration of Re species
toward the surface occurs during the calcination step. Both XPS
and ToF-SIMS data show that the Re surface concentration in-
creases after calcination for the samples that do not lose rhenium
Both the specific activities and the areal TOF values vary in the
following order: Re3–SiAl0.3 > Re1–SiAl14 > Re3–Al ꢄ Re2–Si. This

2
40
K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
Table 7
Comparison of the activity of the rhenium- and the molybdenum-based metathesis catalysts. Activity expressed as specific activity (in mmol of metathesis products per gram of
catalyst and per hour), apparent turnover frequency TOF (in mol of reactants converted per mole of active metal and per second), and as areal TOF (mol of reactants converted per
2
mole of active metal per second and per m of catalyst).
SSA (m2
g
ꢀ1
)
Specific activity (mmol g
ꢀ1 ꢀ1
h
)
TOF (10
ꢀ3 ꢀ1
s
)
Areal TOF (10
ꢀ6 ꢀ1
s
m )
ꢀ2
Sample
M/(M + Si + Al) (%)
Re2–Si
Re3–Al
2
3
3
0.7
3
13
7
1
810
360
190
630
230
490
480
167
0.2
0.5
14.6
57.4
98.5
32.3
9.1
0.6
11.3
45.2
20.7
24
44.5
32
40.6
302
156
140
1.9
Re3–SiAl0.3
Re1–SiAl14
IW-Re3/Al
Mo13–SiAl14a
AER-Mo7/SiAl13
FSP-Mo1/SiAl1c
b
12.8d
71.3
d
26.7
17.8
430d
a
Reported in [45]. The catalyst (10 wt% MoO
Re-based catalysts.
3 2 3 2
, 5.7 wt% Al O , 84 wt% SiO ) was re-tested here in the metathesis of ethene and 2-butene, under the same conditions as the
b
Reported in [52]. The catalyst (10 wt% MoO
3
, 5.7 wt% Al
2
O
3
, 84 wt% SiO
2
, 0.4 wt%) was prepared by a one-pot aerosol synthesis and tested under identical conditions.
c
Reported in [8]. The catalyst (1 wt% MoO
Calculated from the published data.
3
, 69 wt% Al
2
O
3
, 30 wt% SiO ) was prepared by a one-pot flame spray pyrolysis and tested in the propene metathesis at 40 °C.
2
d
points out to a clear effect of the composition of the matrix on the
metathesis activity. A similar support effect was already reported
for supported rhenium- or molybdenum-based metathesis cata-
lysts [14–17]. This support effect is most likely related to the pres-
ence of acidic sites, which are considered to play a major role in the
generation of active species for the metathesis reaction [13]. Based
on studies on impregnated catalysts, it was put forward that the
activity of the Re active centers was determined by their interac-
[46,53], which itself outclassed in terms of specific activity other
Mo-based metathesis catalysts, including the ones prepared by
the aerosol route [53] and flame spray pyrolysis [8]. The latter
method was shown to yield highly active catalysts, but only at very
low loadings, resulting in relatively low specific activities.
5. Conclusions
x
tions with the support. ReO species deposited on an acidic support
such as silica–alumina are themselves more acidic (or electron
deficient), which is beneficial for their activation upon contact with
the olefin. It is noteworthy that the sample with the highest spe-
cific activity (Re3–SiAl0.3) is also the sample with the highest
This study presents a simple, one-step NHSG method to prepare
highly active Re–Si–Al metathesis catalysts with well-controlled
Si/Al ratios and good textures, starting from commercially avail-
able chloride precursors.
amount of acidic sites according to NH
or Brönsted) and the strength of the acidic sites are most likely
important parameters which deserve further study.
The specific activity and apparent TOF values of the Re3–SiAl0.3
catalyst are significantly higher than that of the catalyst IW-Re3/Al
prepared by impregnation, showing the importance of the compo-
sition of the matrix. On the other hand, the activity of the Re3–Al
sample is lower than that of IW-Re3/Al. This could be due in part
to a lower surface concentration of ReO
XPS and also to the fact that Re–3Al is amorphous, whereas the
support used for the impregnation is a crystalline
support.
3
-TPD. The nature (Lewis
The loss of rhenium during calcination could be avoided by
playing on the Al content: no loss of rhenium occurred in alu-
mina-rich formulations. Moreover, no loss of rhenium occurred
during catalytic tests. During the calcination, Re species migrate
toward the surface, leading to highly dispersed Re surface species,
comparable to the species found at the surface of a calcined Re/Al
catalyst prepared by impregnation.
The activity of these NHSG catalysts in the cross-metathesis of
ethene and trans-2-butene directly depends on their composition.
The Re–Si–Al catalysts are significantly more active than the Re–Al
catalyst, while the Re–Si catalyst was inactive. The most active
NHSG catalyst (Re3–SiAl0.3) showed a very high specific activity,
as high as that of the best Mo–Si–Al catalyst reported and signifi-
cantly higher than a reference Re/Al catalyst with a similar Re con-
tent prepared by impregnation. A large number of acidic sites, a
x
species as suggested by
c-alumina
4.2. Comparison of Si–Al–Re and Si–Al–Mo metathesis catalysts
prepared by different methods
x
high surface area, and well-dispersed ReO species appear to be
the key parameters to reach a high activity.
In Table 7, we compare the metathesis activity of the rhenium-
based catalysts prepared by NHSG and incipient wetness impreg-
nation to that of molybdenum-based catalysts synthesized by dif-
ferent methods. All of these catalysts were tested under exactly the
same conditions in the cross-metathesis of butene and ethene or in
the self-metathesis of propene. The Mo–Si–Al catalyst prepared by
NHSG (Mo13–SiAl14) [46] was initially tested in the self-metathe-
sis of propene but was re-tested here in the cross-metathesis of
ethene and 2-butene, leading to a similar activity. The Mo–Si–Al
catalyst prepared via a one-pot aerosol route (AER-Mo7/SiAl13)
was tested under the same conditions [53], while the catalyst ob-
tained by flame spray pyrolysis (FSP-Mo1/SiAl1) was tested under
the same conditions but in the metathesis of propene [8].
Acknowledgments
The authors from ICGM acknowledge the Ministère de l’Ens-
eignement Supérieur et de la Recherche and the Centre National
de la Recherche Scientifique (CNRS) in France for financial support.
D.P. Debecker thanks the FNRS for his Postdoctoral Researcher po-
sition. The authors from UCL are involved in the ‘‘Inanomat’’ IUAP
network sustained by the Service public fédéral de programmation
politique scientifique (Belgium), in the Cost Action D41 sustained
by the European Science Foundation.
The molar content of the active phase used in rhenium catalysts
is much lower than in molybdenum catalysts. This is important, gi-
ven the price of rhenium starting materials. Nevertheless, the spe-
cific activity of our most active catalyst, Re3–SiAl0.3, is as high as
that of the Mo-based catalyst prepared by NHSG, Mo13–SiAl114
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2013.02.016.

K. Bouchmella et al. / Journal of Catalysis 301 (2013) 233–241
241
References
[29] R. Spronk, J.A.R. van Veen, J.C. Mol, J. Catal. 144 (1993) 472–483.
[
[
30] F.D. Hardcastle, I.E. Wachs, J.A. Horsley, G.H. Via, J. Mol. Catal. 46 (1988) 15–36.
31] M. Valigi, D. Cordischi, D. Gazzoli, C.P. Keijzers, A.A.K. Klaassen, J. Chem. Soc.,
Faraday Trans. 1 (77) (1981) 1871–1882.
[
[
[
[
1] J.C. Mol, J. Mol. Catal. A: Chem. 213 (2004) 39–45.
2] J.C. Mol, Green Chem. 4 (2002) 5–13.
3] B.B. Marvey, Int. J. Mol. Sci. 10 (2009) 213–220.
4] K.J. Ivin, I.C. Mol (Eds.), Olefin Metathesis and Metathesis Polymerization,
second ed., 1996.
[
[
[
32] F. Schekler-Nahama, O. Clause, D. Commereuc, J. Saussey, Appl. Catal., A 167
(
1998) 237–245.
33] M. Fabris, C. Aquino, A.J. Ward, A. Perosa, T. Maschmeyer, M. Selva, Top. Catal.
2 (2009) 315–321.
5
[
[
5] J. Aguado, J.M. Escola, M.C. Castro, B. Paredes, Appl. Catal., A 284 (2005) 47–57.
6] D.P. Debecker, M. Stoyanova, U. Rodemerck, A. Leonard, B.-L. Su, E.M.
Gaigneaux, Catal. Today 169 (2011) 60–68.
34] A. Yoboue, A. Susset, A. Tougerti, D. Gallego, S.V. Ramani, M. Kalyanikar, D.S.
Dolzhnikov, S.G. Wubshet, Y. Wang, S. Cristol, V. Briois, C. La Fontaine, R.M.
Gauvin, J.-F. Paul, E. Berrier, Chem. Commun. 47 (2011) 4285–4287.
35] P.H. Mutin, A. Vioux, Chem. Mater. 21 (2009) 582–596.
36] D.P. Debecker, P.H. Mutin, Chem. Soc. Rev. 41 (2012) 3624–3650.
37] V. Lafond, P.H. Mutin, A. Vioux, Chem. Mater. 16 (2004) 5380–5386.
38] D.P. Debecker, V. Hulea, P.H. Mutin, Appl. Catal., A 451 (2013) 192–206.
39] F. Barbieri, D. Cauzzi, F. De Smet, M. Devillers, P. Moggi, G. Predieri, P. Ruiz,
Catal. Today 61 (2000) 353–360.
[
[
[
7] D.P. Debecker, M. Stoyanova, U. Rodemerck, E.M. Gaigneaux, J. Mol. Catal. A:
Chem. 340 (2011) 65–76.
8] D.P. Debecker, B. Schimmoeller, M. Stoyanova, C. Poleunis, P. Bertrand, U.
Rodemerck, E.M. Gaigneaux, J. Catal. 277 (2011) 154–163.
9] H. Liu, S. Huang, L. Zhang, S. Liu, W. Xin, L. Xu, Catal. Commun. 10 (2009) 544–
[
[
[
[
[
5
48.
[
[
[
10] J.C. Mol, J.A. Moulijn, Adv. Catal. 24 (1975) 131–171.
11] C. Coperet, New J. Chem. 28 (2004) 1–10.
12] H. Balcar, R. Hamtil, N. Zilkova, Z. Zhang, T.J. Pinnavaia, J. Cejka, Appl. Catal., A
[
[
[
[
40] A.M. Cojocariu, P.H. Mutin, E. Dumitriu, F. Fajula, A. Vioux, V. Hulea, Chem.
Commun. (2008) 5357–5359.
41] H. Kaper, K. Bouchmella, P.H. Mutin, F. Goettmann, ChemCatChem 4 (2012)
3
20 (2007) 56–63.
1
813–1818.
42] A.F. Popa, P.H. Mutin, A. Vioux, G. Delahay, B. Coq, Chem. Commun. (2004)
214–2215.
[
[
13] J.C. Mol, Catal. Today 51 (1999) 289–299.
14] D.P. Debecker, D. Hauwaert, M. Stoyanova, A. Barkschat, U. Rodemerck, E.M.
Gaigneaux, Appl. Catal., A 391 (2011) 78–85.
2
43] D.P. Debecker, K. Bouchmella, R. Delaigle, P. Eloy, C. Poleunis, P. Bertrand, E.M.
Gaigneaux, P.H. Mutin, Appl. Catal., B 94 (2010) 38–45.
[
[
[
[
15] A. Andreini, X. Xu, J.C. Mol, Appl. Catal. 27 (1986) 31–40.
16] C.J. Lin, A.W. Aldag, A. Clark, J. Catal. 45 (1976) 287–296.
17] J.C. Mol, A. Andreini, J. Mol. Catal. 46 (1988) 151–156.
18] B. Mitra, X. Gao, I.E. Wachs, A.M. Hirt, G. Deo, Phys. Chem. Chem. Phys. 3 (2001)
[
[
44] C. Petitto, H.P. Mutin, G. Delahay, Chem. Commun. 47 (2011) 10728–10730.
45] D.P. Debecker, K. Bouchmella, C. Poleunis, P. Eloy, P. Bertrand, E.M. Gaigneaux,
P.H. Mutin, Chem. Mater. 21 (2009) 2817–2824.
1
144–1152.
[
46] D.P. Debecker, K. Bouchmella, M. Stoyanova, U. Rodemerck, E.M. Gaigneaux,
P.H. Mutin, Catal. Sci. Technol. 2 (2012) 1157–1164.
[
19] M. Stoyanova, U. Rodemerck, U. Bentrup, U. Dingerdissen, D. Linke, R.W.
Mayer, H.G.J. Lansink Rotgerink, T. Tacke, Appl. Catal., A 340 (2008) 242–249.
20] H. Balcar, R. Hamtil, N. Zilkova, J. Cejka, Catal. Lett. 97 (2004) 25–29.
21] X. Secordel, A. Yoboue, S. Cristol, C. Lancelot, M. Capron, J.-F. Paul, E. Berrier, J.
Solid State Chem. 184 (2011) 2806–2811.
[
[
[
[
47] A. Vioux, Chem. Mater. 9 (1997) 2292–2299.
48] M. Sibeijn, J.C. Mol, Appl. Catal. 67 (1991) 279–295.
49] W. Gerrard, A.H. Woodhead, J. Chem. Soc. (1951) 519–522.
50] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T.
Siemieniewska, Pure Appl. Chem. 57 (1985) 603–619.
[
[
[
[
22] C.-B. Wang, Y. Cai, I.E. Wachs, Langmuir 15 (1999) 1223–1235.
23] P.C. Bakala, E. Briot, Y. Millot, J.-Y. Piquemal, J.-M. Bregeault, J. Catal. 258
[
[
[
51] D.P. Debecker, R. Delaigle, K. Bouchmella, P. Eloy, E.M. Gaigneaux, P.H. Mutin,
Catal. Today 157 (2010) 125–130.
52] K. Amakawa, S. Wrabetz, J. Kroehnert, G. Tzolova-Mueller, R. Schloegl, A.
Trunschke, J. Am. Chem. Soc. 134 (2012) 11462–11473.
53] D.P. Debecker, M. Stoyanova, F. Colbeau-Justin, U. Rodemerck, C. Boissiere,
E.M. Gaigneaux, C. Sanchez, Angew. Chem., Int. Ed. 51 (2012) 2129–2131.
(
2008) 61–70.
[
[
[
[
24] H.S. Lacheen, P.J. Cordeiro, E. Iglesia, Chem. Eur. J. 13 (2007) 3048–3057.
25] R. Spronk, J.C. Mol, Stud. Surf. Sci. Catal. 75 (1993) 2071–2074.
26] R.H. Blom, C.H. Kline, Hydrocarbon Process. Pet. Refin. 42 (1963) 132–134.
27] M.A. Banares, L.J. Alemany, M. Lopez Granados, M. Faraldos, J.L.G. Fierro, Catal.
Today 33 (1997) 73–83.
[
28] M.A. Vuurman, D.J. Stufkens, A. Oskam, J. Mol. Catal. 76 (1992) 263–285.