Moving from surfactant-stabilized aqueous rhodium (0) colloidal suspension to heterogeneous magnetite-supported rhodium nanocatalysts: Synthesis, characterization and catalytic performance in hydrogenation reactions

Article abstract of DOI:10.1016/j.cattod.2011.08.046

Wet impregnation of pre-synthesized surfactant-stabilized aqueous rhodium (0) colloidal suspension on silica was employed in order to prepare supported Rh⁰ nanoparticles of well-defined composition, morphology and size. A magnetic core-shell support of silica (Fe₃O₄@SiO ₂) was used to increase the handling properties of the obtained nanoheterogeneous catalyst. The nanocomposite catalyst Fe₃O ₄@SiO₂-Rh⁰ NPs was highly active in the solventless hydrogenation of model olefins and aromatic substrates under mild conditions with turnover frequencies up to 143,000 h^-1. The catalyst was characterized by various transmission electron microscopy techniques showing well-dispersed rhodium nanoparticles (～3 nm) mainly located at the periphery of the silica coating. The heterogeneous magnetite-supported nanocatalyst was investigated in the hydrogenation of cyclohexene and compared to the previous surfactant-stabilized aqueous Rh⁰ colloidal suspension and various silica-supported Rh⁰ nanoparticles. Finally, the composite catalyst could be reused in several runs after magnetic separation.

Full text of DOI:10.1016/j.cattod.2011.08.046

Catalysis Today 183 (2012) 124–129
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Moving from surfactant-stabilized aqueous rhodium (0) colloidal suspension to
heterogeneous magnetite-supported rhodium nanocatalysts: Synthesis,
characterization and catalytic performance in hydrogenation reactions
Carl-Hugo Pélissona,b, Lucas L.R. Vonoc, Claudie Huberta,b, Audrey Denicourt-Nowickia,b
,
Liane M. Rossic, Alain Roucouxa,b,∗
a Ecole Nationale Supérieure de Chimie de Rennes, CNRS, UMR 6226, Avenue du Général Leclerc, CS 50837, 35 708 Rennes Cedex 7, France
b Université Européenne de Bretagne, France
c Instituto de Química, Universidade de São Paulo, Av. Prof. Lineu Prestes 748, São Paulo 05508-000, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 July 2011
Received in revised form 31 August 2011
Accepted 31 August 2011
Available online 5 October 2011
Wet impregnation of pre-synthesized surfactant-stabilized aqueous rhodium (0) colloidal suspension on
silica was employed in order to prepare supported Rh0 nanoparticles of well-defined composition, mor-
phology and size. A magnetic core–shell support of silica (Fe3O4@SiO2) was used to increase the handling
properties of the obtained nanoheterogeneous catalyst. The nanocomposite catalyst Fe3O4@SiO2-Rh0 NPs
was highly active in the solventless hydrogenation of model olefins and aromatic substrates under mild
conditions with turnover frequencies up to 143,000 h−1. The catalyst was characterized by various trans-
mission electron microscopy techniques showing well-dispersed rhodium nanoparticles (∼3 nm) mainly
located at the periphery of the silica coating. The heterogeneous magnetite-supported nanocatalyst was
investigated in the hydrogenation of cyclohexene and compared to the previous surfactant-stabilized
aqueous Rh0 colloidal suspension and various silica-supported Rh0 nanoparticles. Finally, the composite
catalyst could be reused in several runs after magnetic separation.
Keywords:
Metallic nanoparticles
Magnetic support
Impregnation
Hydrogenation
Recycling
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
Another very elegant strategy consists of the immobilization of pre-
synthesized metal nanoparticles [11]. In that case, the control on
A great but still largely unmet goal of modern catalysis is
to transfer to heterogeneous catalysis the synthetic control of
nanoparticle synthesis in solution by the bottom-up approach.
Despite recent exciting progress in nanocatalysis [1,2], the prepa-
ration of supported nanoparticles with precise control over factors
such as composition, size, size-distribution and shape remains a
challenge. In that context, can we bring what has been learned in
the modern “soluble nanoparticle catalysts” [3–5] to the develop-
ment of supported-nanoparticle heterogeneous catalysts [6], which
will be more suitable for industrial applications [7–9].
In general, supported nanoparticles on solids such as oxides,
carbon or zeolites are prepared by various deposition–precipitation
or co-precipitation procedures, oxidation–reduction steps and the
nanoparticle size is tentatively adjusted by varying experimental
parameters such as the pH, the reducing agent, the concentration
of precursors in solution and the temperature of calcinations [10].
particles size and shape obtained in solution will be better than any
method to prepare supported metal particles by reduction and/or
deposition of metal precursors over solid supports.
Among other attempts to improve the reusability of “solu-
ble metal nanocatalysts”, the development of biphasic catalytic
conditions, such as biphasic aqueous/organic systems [12,13] or
ionic-liquid/organic systems [14,15] have largely been developed.
In those systems, the catalysts are stabilized in the aqueous or
ionic liquid phase and the product/organic phase can be collected
by phase decantation. The catalyst-containing phase is easily
recovered and recycled, although the catalytic rates are compro-
mised because of mass-transfer limitations. Metal nanoparticles
deposited on solid supports, which typically increase the reaction
rates by few orders of magnitude, have emerged as sustainable
alternatives to conventional materials [1]. Magnetically driven
separation of catalysts immobilized on magnetic supports by just
applying an external magnetic field to the reaction mixture has
recently received great attention as a non-laborious, cheap and
often highly scalable technique [16,17] although the early research
in the field of magnetic nanoparticles can be dated back several
decades. A variety of methodologies have been reported in the
literature for the immobilization of homogeneous and
∗
Corresponding author at: Ecole Nationale Supérieure de Chimie de Rennes,
CNRS, UMR 6226, Avenue du Général Leclerc, CS 50837, 35 708 Rennes Cedex 7,
France. Fax: +33 223 23 80 37.
E-mail address: alain.roucoux@ensc-rennes.fr (A. Roucoux).
0920-5861/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2011.08.046

C.-H. Pélisson et al. / Catalysis Today 183 (2012) 124–129
125
heterogeneous catalysts on such kind of supports possessing
superparamagnetic properties. The developed approaches include
the direct deposition of metals on magnetic nanoparticles, the
functionalization of magnetic particle surfaces or the coating of
the magnetic cores with inorganic or organic materials before
metal loading. The use of inorganic oxides, such as SiO2, is a strat-
egy to give an extra protection to naked iron oxides (especially
magnetite) against oxidation and to help the immobilization of
catalysts through a covalent approach [18–23]. This methodology
was efficiently used to prepare Fe3O4@SiO2-NH2-Rh0 NPs and
relies on the uptake of Rh3+ by amino-functionalized silica-coated
magnetic nanoparticles followed by metal reduction under con-
trolled H2 conditions [24]. Tremendous efforts are still ongoing
to improve the preparation of supported metal NP catalysts with
good control of composition, morphology, and size distribution.
The examples of magnetically recoverable metal NP catalysts
found in the literature are based on reduction of metal precursors
over a magnetic solid.
Recently, we have described a simple methodology based on
the wet impregnation onto an inorganic matrix of a pre-stabilized
aqueous colloidal rhodium suspension. SiO2-Rh0 NPs [25,26] and
TiO2-Rh0 NPs [27,28] were prepared by this method and well
characterized in terms of organization and size. In this paper,
we described the immobilization of Rh0 NPs on the silica coated
magnetite core–shell support. Fe3O4@SiO2-Rh0 NPs material was
prepared by impregnation of the inorganic silica coating of the
magnetite core by pre-synthesized aqueous Rh0coll suspension. The
composite catalyst was characterized by transmission electron
microscopy (TEM, HTREM) and evaluated in the hydrogenation of
cyclohexene at 75 ◦C and 6 bar of H2 in comparison with SiO2-Rh0
NPs, the aqueous Rh0coll suspension and the analogous Fe3O4@SiO2-
NH2-Rh0 NPs. Finally, the heterogeneous magnetite-supported
rhodium nanocatalyst was investigated in the solventless hydro-
genation of model olefins and aromatic substrates under mild
conditions as well as its stability and reusability.
and centrifugation, the silica-coated magnetic particles were dried
under vacuum.
2.2.2. Preparation of the surfactant-stabilized aqueous Rh0
colloidal suspension
To an aqueous solution of the ammonium surfactant HEA16X
(with X = Cl or BF4) or THEA16Cl (2 eq., 7.6 × 10−4 mol, 95 mL H2O),
was added 36 mg of sodium borohydride (9.5 × 10−4 mol, 2.5 eq.).
Then this solution was quickly added under vigorous stirring to
an aqueous solution (5 mL) of the precursor RhCl3·3H2O (100 mg,
3.8 × 10−4 mol) to obtain a stable aqueous Rh0 colloidal suspension
(100 mL). The reduction occurs instantaneously and is observed by
a colour change from red to black. The Rh0coll suspension was stirred
overnight before use.
2.2.3. Metal immobilization on silica-coated magnetic particles
This preparation follows the preparation of SiO2-supported
metallic nanoparticles reported elsewhere [26]. The magnetic
support (50 mg) was dispersed in distilled water (2.5 mL). After
addition of the ammonium surfactant (2 eq.), the mixture was
stirred at room temperature overnight. The adequate amount of the
pre-stabilized Rh0 nanoparticles (0.1 wt%) was added and the mix-
ture was stirred for 2 h. The particles were magnetically separated
and dried at 100 ◦C.
2.3. General procedure for the solventless hydrogenation of the
cyclohexene under 6 bar of H2
Catalytic reactions were carried out in a modified Fischer–Porter
glass reactor connected to a pressurized dihydrogen gas supply
tank. The Fischer–Porter bottle was set at a constant pressure for
the entire course of the reaction by leaving the reactor connected
to the dihydrogen gas supply through a gas regulator set to the
desired pressure. In a typical experiment, the catalyst (35 mg of
magnetic solid, 3 × 10−4 mmol Rh), cyclohexene (1.2 g, 14.6 mmol)
and a magnetic stir bar were added to the reactor under inert atmo-
sphere. The reactor was evacuated and connected to the dihydrogen
gas supply apparatus for gas admission. The reaction was conducted
under magnetic stirring (700 rpm) at 6 bar and 75 ◦C. The temper-
ature was maintained constant by a hot-stirring plate connected
to a digital temperature controller. The reactions were monitored
by the fall in the pressure of the dihydrogen gas supply tank until
completion. After the dihydrogen gas consumption stopped, the
catalyst was recovered magnetically by placing a magnet on the
reactor wall and the products were collected and analyzed by gas
chromatography (GC).
2. Experimental
2.1. General
Rhodium trichloride was obtained from Strem Products. Sodium
borohydride, all the organic compounds investigated as substrates
were purchased from Acros Organics or Sigma–Aldrich–Fluka
and used without further purification. Water was distilled twice
before use by conventional method. The N,N-dimethyl-N-cetyl-N-
(2-hydroxyethyl) ammonium chloride and tetrafluoroborate salts
(HEA16X, X = Cl or BF4) were synthesised as previously described
[29] and the N-cetyl-N-tris-(2-hydroxyethyl) ammonium chloride
(THEA16Cl) was prepared by quaternarization of hexadecylamine
with chloroethanol [30]. The silica was obtained from Merck and
presents the following characteristics: mean diameter: 80 ␮m,
pore diameter 5.4 nm, specific area 490 m2 g−1, pore volume
0.77 cm3 g−1 and internal porosity 58%.
2.4. General procedure for the solventless hydrogenation of
model olefins and aromatic substrates under 10 bar of H2
Catalytic reactions were carried out in a stainless steel auto-
clave under magnetic stirring. In a typical experiment, the reactor
was charged with the catalyst Fe3O4@SiO2-Rh0 NPs (37.6 mg,
3.6 × 10−4 mmol of Rh0) and a magnetic stir bar. Then, the appro-
priate substrate (14.6 mmol) was added to the autoclave and the
dihydrogen was admitted to the system at the desired pressure
(10 bar). The system was purged with H2 three times after that the
reaction was magnetically stirred until completion. The complete
conversion was followed by GC analysis.
2.2. Synthesis of magnetic particles supported Rh0 nanoparticles
2.2.1. Synthesis of the silica-coated magnetic nanoparticles
The silica-coated magnetic particles were prepared by a reverse
microemulsion method reported elsewhere [24]. To 44.6 g of
polyoxyethylene (5) isooctylphenyl ether surfactant dispersed in
600 mL of cyclohexane was added oleic acid coated magnetite
(200 mg). After the addition of 9.44 mL NH4OH (29% aqueous
solution), the microemulsion was formed and 7.7 mL of TEOS
were added. The reaction mixture was slowly stirred during 16 h.
Then, 200 mL of methanol was added and the alcoholic phase
was centrifuged (7000 rpm, 10 min). After washing with ethanol
2.5. Transmission electron microscopy
Samples for TEM observations were prepared by placing a drop
containing the nanoparticles in carbon coated TEM grids. The trans-
mission electron microscopy (TEM) work was partially performed
at the C2NANO – Center for Nanoscience and Nanotechnology

126
C.-H. Pélisson et al. / Catalysis Today 183 (2012) 124–129
Rh0coll - HEA16Cl
SiO2
Magnetic core
Fe3O4
Fe3O4@SiO2
Fe3O4@SiO2 - Rh0 NPs
Fig. 1. Preparation of Fe3O4@SiO2-Rh0 NPs.
(LNLS, Campinas, SP, Brazil) using a JEOL-3010 ARP microscope and
at the Institut des Matériaux de Paris-Centre – Université Pierre et
Marie Curie using a JEOL JEM 2010 UHR microscope.
the substrate and based on the introduced amount of metal and not
on the exposed surface metal and thus may be underestimated. As
previously reported by Janiak et al. [31], the catalytic activity results
not only from the exposed surface metal atoms since the surface
can restructure, atoms can shift positions during heterogeneous
processes and partial aggregation could occur during catalysis,
modifying the fraction of surface atoms. These changes in the
surface render the determination of the number of surface atoms
difficult.
3. Results and discussion
We here describe the immobilization of pre-synthesized
surfactant-stabilized aqueous Rh0 colloidal NPs on a silica coated
magnetite core–shell support, as depicted in Fig. 1. The aque-
ous Rh0 colloidal (Rhc0oll) suspension was prepared by chemical
reduction of an aqueous solution of rhodium chloride salt in the
presence of water-soluble ammonium salts as stabilizing agents
and fully characterized as previously reported elsewhere [29,30].
The core–shell magnetic support comprised of magnetite nanopar-
ticles spherically coated with silica (Fe3O4@SiO2) was synthesized
and characterized as also previously reported by some of us
[24]. Based on the method we have previously developed for the
deposition of metallic nanospecies on an inorganic support such
as SiO2 or TiO2 [26,28], the wet impregnation of the silica coated
magnetic particles Fe3O4@SiO2 was investigated at room temper-
ature using pre-synthesized aqueous Rh0coll suspensions. Various
protective agents (HEA16Cl, Br or THEA16Cl salts) were tested both
to stabilize the Rh0 NPs and to facilitate the impregnation process.
The results are gathered in Table 1 and the metal loading was
checked by ICP AES analysis and compared to the nominal value.
First, concerning the choice of the surfactant, the metal loading
is much better using HEA16Cl or HEA16BF4 as surfactants (Table 1,
entries 1 and 3) and very close to the nominal value (0.1 wt%),
compared to THEA16Cl (Table 1, entry 4). After several experi-
ments we have seen that metal loadings up to 0.5 wt% could also
be obtained with the same Rh0coll-HEA16Cl suspension containing
a molar ratio Rh/HEA16Cl = 2 (Table 1, entry 2). However, attempts
to get higher metal loadings close to 1 wt% were unfruitful.
The Fe3O4@SiO2-Rh0 NP catalyst with 0.09 wt% Rh loading
determined by ICP AES (0.1 wt% nominal value, Table 1, entry 1)
was evaluated with a turnover number of 43,000 mol of substrate
per mol of metal (total metal) and showed very high catalytic
activities up to 143,000 h−1 with total conversion of the substrate
after 0.3 h (Table 2, entry 1). As a reference, the use of the colloidal
suspension Rh0coll-HEA16Cl leads to the conversion of cyclohexene
into cyclohexane with a TOF of 33,000 h−1 in the same catalytic
conditions (Table 2, entry 3). Consequently, the TOF observed with
the heterogeneous magnetite-supported rhodium nanocatalysts is
4.3 times the catalytic activity of Rh0coll-HEA16Cl used under bipha-
sic conditions (Table 2, entries 1 and 3). Moreover, the performance
of the Fe3O4@SiO2-Rh0 NPs was also superior to the Rh0 NPs pre-
pared by impregnation of Rh3+ ions followed by in situ H2 reduction
over the same magnetic solid (Fe3O4@SiO2 with 0.1 wt% of Rh) and
thus showed the advantage of the use of pre-formed nanoparticles
(Table 2, entries 1 and 4). In that case, the turnover frequency is
enhanced by a factor of 70 when using pre-synthesized Rh0 NPs.
The reduction of Rh3+ ions by NaBH4 leads to a significant increase
in catalytic activity (Table 2, entry 6). This result could be explained
by a better reduction of the metal salts using this strong reductive
agent, providing zerovalent rhodium nanoparticles with a suitable
size for catalytic applications. Indeed, the same catalyst prepared by
in situ H2 reduction with 1.5 wt% Rh loading decreases the catalytic
reaction time from 21 h to 3 h in the same conditions probably due
to a better growth processes giving a suitable nanoparticle size for
catalytic applications (Table 2, entry 5). Finally, a weak influence
of the silica support was also observed when comparing the
immobilization process on commercially available silica instead of
In a first approach, we have compared the catalytic perfor-
mance of the optimized system with previously described Rh0 NP
catalysts in the hydrogenation of cyclohexene at 75 ◦C and 6 bar
H2 (Table 2). The turnover frequencies (TOFs) were calculated
considering an optimized reaction time for complete conversion of
Table 1
Optimization of the synthesis of Fe3O4@SiO2-Rh0 NPs: choice of the surfactant.
Entry
Surfactant
Equivalents of surfactant
Rhtheorical (wt%)
RhE.A. (wt%)a
1
2
3
4
HEA16Cl
HEA16Cl
HEA16BF4
THEA16Cl
2
2
2
2
0.1
0.5
0.1
0.1
0.09
0.5
0.09
0.02
a
Determined by atomic absorption analysis (ICP AES).

C.-H. Pélisson et al. / Catalysis Today 183 (2012) 124–129
127
Table 2
Comparison of various catalysts in the hydrogenation of cyclohexene at 75 ◦C and 6 bar H2.
h
Entry
Catalyst
Rh loading (%)
TONg
Time (h)
TOF (h−1
)
1
2
3
4
5
6
Fe3O4@SiO2-Rh0 NPsa
SiO2-Rh0 NPsb
0.09e
0.08e
–
43,000
35,600
43,000
43,000
35,600
35,600
0.3
0.3
1.3
21
3
143,000
120,000
33,000
2,000
12,000
62,500
Rh0coll-HEA16Cla
Fe3O4@SiO2-NH2-Rh0 NPsc
Fe3O4@SiO2-NH2-Rh0 NPsc
Fe3O4@SiO2-NH2-Rh0 NPsd
0.1f
1.5f
0.1f
0.57
a
This work.
b
c
d
e
f
Catalyst prepared according to Ref. [26].
Supported Rh0 NPs prepared by impregnation of Rh3+ ions followed by in situ H2 reduction prepared according to Ref. [24].
Supported Rh0 NPs obtained by the reduction of Rh3+ ions with NaBH4.
Determined by ICP AES in this set of experiments.
Nominal value.
g
h
TON = mol substrate converted per mol catalyst (total metal).
TOF = mol substrate converted per mol catalyst (total metal) per hour at 100% conversion.
the silica coated magnetite core–shell support (Table 2, entries 1
and 2). In fact, the catalytic activity of the SiO2-Rh0 NPs with a TOF of
120,000 h−1 was very close to the value observed for Fe3O4@SiO2-
Rh0 NPs. This result could be explained by heterogeneous surface
properties of silica. In fact, different preparation method and
post-treatments have been used providing different character-
istics of silica based supports in terms of surface area, pore and
grain sizes.
160 000
120 000
80 000
40 000
In
a
second step, the Fe3O4@SiO2-Rh0 NPs stability and
reusability were studied in the solventless hydrogenation of cyclo-
hexene with turnover number as high as 40,000 mol of substrate
per mol of metal (total metal) at 75 ◦C and 6 bar H2. The cata-
lyst was easily recovered by placing a magnet in the reactor wall
and reused for several runs with >99% conversion. The catalytic
activity remains constant on the first three runs (accumulated TON
of 129,000 mol/mol) with TOF values ∼143,000 h−1. However, a
slight decrease in the catalytic activity was observed starting the
fourth run, although the catalyst remains active for up to 7 cycles
with accumulated TON of 300,000 mol/mol. The atomic absorption
analysis (ICP AES) of the recovered catalyst confirmed the pres-
ence of 93% of the initial metal loading after 7th run (0.084 in the
spent catalyst vs. 0.09 wt%). Consequently, the decrease in kinetic
properties of the catalyst is not only explained by metal leaching but
also by other deactivation processes such as particles aggregation
or poisoning effect.
In a third set of experiments, we investigated the hydrogena-
tion of several olefins and various benzene derivatives with a
turnover number of 40,000 mol of substrate per mol of metal
(total metal) under standard conditions (10 bar H2, room tempera-
ture). These experimental conditions have been used to activate
the reaction and limit the time in the case of less reducible
substrates as aromatic compounds. Indeed, no reduction was
observed under atmospheric hydrogen pressure in opposition with
the use of the aqueous suspension of Rh0coll-HEA16Cl in biphasic
conditions [13]. The conversion was followed by GC analysis or
0
Run 1
Run 2
Run 3
Run 4
Fig. 2. Cyclohexene hydrogenation at 75 ◦C and 6 bar H2: recycling experiments
with Fe3O4@SiO2-Rh0 NPs (TON each run = 43,000).
determined for a defined time (6 h). The results are gathered in
Table 3.
As model olefinic substrates, cyclooctene and tetradecene
(Table 3, entries 1 and 2) have been investigated. An efficient
turnover frequency was observed in the hydrogenation of the
lipophilic tetradecene (TOF = 32,000 h−1) but a significant decrease
was obtained with the cyclic compound. This result could be
explained by the rigid ring leading to a more steric hindrance
around the particle and thus limiting the accessibility of the
reducible C–C double bond to the active sites. We have also
studied the hydrogenation of substituted aromatic compounds.
The best activities were observed for toluene hydrogenation with a
complete conversion in 2.5 h providing a TOF = 16,000 h−1 (Table 3,
entry 5). In the same time, styrene was transformed in a mixture
of ethylbenzene and ethylcyclohexane (ratio 70/30, Table 3, entry
3). After 4 h, 70% of ethylcyclohexane was obtained (Table 3, entry
Table 3
Catalytic investigation of olefins and aromatic derivatives with Fe3O4@SiO2-Rh0 NPs.a
b
Entry
Substrate
Product (%)
Time (h)
TOF (h−1
)
1
2
3
4
5
6
7
8
9
Cyclooctene
Tetradecene
Styrene
Styrene
Toluene
p-Xylene
iPr-Benzene
Anisole
Cyclooctane (100)
Tetradecane (100)
2.5
1.25
2
4
2.5
6
6
6
6
16,000
32,000
20,000
10,000
16,000
6,667
n.c.c
Ethylbenzene/Ethylcyclohexane (70/30)
Ethylbenzene/Ethylcyclohexane (30/70)
Methylcyclohexane (100)
p-dimethylcyclohexane (cis/trans:75/25)
iPr-cyclohexane (15)
Methoxycyclohexane (10)
m-Methylcyclohexanol (1.5)
n.c.c
m-Cresol
n.c.c
a
Fe3O4@SiO2-Rh0 NPs (37.6 mg, 3.6 × 10−4 mmol of Rh0), substrate (14.6 mmol), 10 bar H2, room temperature.
TOF = mol substrate converted per mol catalyst (total metal) per hour at 100% conversion.
Not calculated.
b
c

128
C.-H. Pélisson et al. / Catalysis Today 183 (2012) 124–129
Fig. 3. TEM images of (a) Fe3O4@SiO2 catalyst support, (b) Fe3O4@SiO2-Rh0 NPs catalyst, (c) detail of Fe3O4@SiO2-Rh0 NPs catalyst showing Rh0 at the surface of the silica
and (d) high resolution (HRTEM) of a single Rh0 NP over the support with an average diameter of about 3 nm (scale bar = 1 nm).
4), showing no deactivation of the catalyst during the process and
a complete conversion could be observed after a more important
reaction time. When we compared alkylated derivatives such
as toluene, p-xylene and iPr-benzene, the presence of bulkier
substituents on the aromatic ring tends to decrease the kinetic
properties. Indeed, a total conversion was obtained in a double time
(6 h) for the hydrogenation of p-xylene as a model disubstituted
compound (TOF = 6667 h−1, Table 3, entry 6) in opposition with the
4. Conclusion
We have described a synthetic methodology based on the
use of pre-synthesized metal nanoparticles to develop supported
Rh0 NPs over silica-based supports. This approach takes advan-
tage of the recent advances in the synthesis of size, shape and
composition-controlled metal and magnetic oxide nanomaterials,
which were efficiently combined to formulate pertinent nanocom-
posite catalysts. An important feature is that the developed strategy
allows metal nanoparticles to be well-dispersed on support sur-
faces, selection of the desired support properties, as for example,
the magnetic support used in this study, and by choosing the
metal nanoparticles synthesis one can select the desired particle
size and shape with few restrictions on their surface proper-
ties. The magnetically recoverable Rh0 nanocatalyst prepared by
using surfactant-stabilized Rh0 NPs proved to be highly active in
the hydrogenation of cyclohexene (TOF = 143,000 h−1) and usual
benzene derivatives with TOF up to 20,000 h−1. The overall per-
formance of the supported NPs is improved if compared to its
activity as a soluble NP catalyst and the magnetic properties allow
the catalyst to be isolated and recycled by the assistance of an
external magnet, which greatly simplifies the work-up procedure
and purification of products minimizing the use of solvents, costly
consumables, energy and time. Also, the pre-formed Rh0 NPs are
prepared through reproducible and well-controlled syntheses in
opposition to traditional colloids, which is a major advantage of
this methodology.
toluene as a monoalkyl-substituted derivative (TOF = 16,000 h−1
,
Table 3, entry 5). Similarly, a lower activity was also observed
with iPr-benzene (Table 3, entry 7). This phenomenon has already
been reported in various media [32] and could be explained by
an increased steric hindrance on the aromatic ring that limits the
access to the Rh0 species on the silica coated magnetite core–shell
support. Finally, the hydrogenation of functionalized oxygenated
aromatic substrates such as anisole or cresol derivatives, which
correspond to model lignin compounds, was also investigated.
These reactions could be an advanced step for the production of
fuels from biomass [33]. Unfortunately, very poor results were
obtained, proving the difficulties to reduce functionalized aromatic
derivatives with Fe3O4@SiO2-Rh0 NPs in mild conditions in terms
of hydrogen pressure and short reaction time Fig. 2.
The morphology and the location of the Fe3O4@SiO2-Rh0 NPs
catalyst were assessed from analysis of transmission electron
microscopy (TEM) micrographs such as presented in Fig. 3. The
support is comprised of magnetic cores (Fe3O4 ∼ 8–10 nm) well-
coated with silica (Fig. 3a). The rhodium nanoparticles (∼3 nm) are
mainly located at the periphery of the silica coating with good dis-
persion (Fig. 3b and c). High resolution TEM was also carried out
(Fig. 3d). Analyses of the fast Fourier transformation (FFT) of the
HRTEM images reveal the presence of inter-reticular distances of
nearly 0.2196 nm and 0.1902 nm, which are characteristic of the
(1 1 1) and (2 0 0) lattice planes of Rh0 and different from the Fe
nanoparticles (2.53 and 2.96 nm).
Acknowledgements
The authors are grateful to the Région Bretagne (PhD fellow-
ships – MAGFOCAT and PARTICAT programs) for financial supports
and the Université Européenne de Bretagne for the international
mobility grant to C. Hubert. They also thank the support from

C.-H. Pélisson et al. / Catalysis Today 183 (2012) 124–129
129
the international cooperation programs PICS Brazil-France (grant
nos. 490678/2008-4 and 3637/2007-2010) and CAPES-COFECUB
(grant nos. 965/09 and Ph695/10). Support from the Brazilian
agencies CNPq and FAPESP is also acknowledged. The authors are
indebted to Patricia Beaunier from Université Pierre et Marie Curie,
and C2NANO – Center for Nanoscience and Nanotechnology/MCT
(Brazil) for TEM images.
[15] J. Dupont, G.S. Fonseca, A.P. Umpierre, P.F.P. Fichtner, S.R. Teixeira, J. Am. Chem.
Soc. 124 (2002) 4228–4229.
[16] V. Polshettiwar, R. Luque, A. Fihri, H.B. Zhu, M. Bouhrara, J.M. Bassett, Chem.
Rev. 111 (2011) 3036–3075.
[17] S. Shylesh, V. Schunemann, W.R. Thiel, Angew. Chem. Int. Ed. 49 (2010)
3428–3459.
[18] L.M. Rossi, F.P. Silva, L.L.R. Vono, P.K. Kiyohara, E.L. Duarte, R. Itri, R. Landers, G.
Machado, Green Chem. 9 (2007) 379–385.
[19] L.M. Rossi, L.L.R. Vono, F.P. Silva, P.K. Kiyohara, E.L. Duarte, J.R. Matos, Appl.
Catal. A: Gen. 330 (2007) 139–144.
[20] L.M. Rossi, I.M. Nangoi, N.J.S. Costa, Inorg. Chem. 48 (2009) 4640–4642.
[21] F.W. Zhang, J. Jin, X. Zhong, S.W. Li, J.R. Niu, R. Li, J.T. Ma, Green Chem. 13 (2011)
1238–1243.
[22] D. Guin, B. Baruwati, S.V. Manorama, Org. Lett. 9 (2007) 1419–1421.
[23] A.J. Amali, R.K. Rana, Green Chem. 11 (2009) 1781–1786.
[24] M.J. Jacinto, P.K. Kiyohara, S.H. Masunaga, R.F. Jardim, L.M. Rossi, Appl. Catal. A:
Gen. 338 (2008) 52–57.
[25] V. Mévellec, A. Nowicki, A. Roucoux, C. Dujardin, P. Granger, E. Payen, K. Philip-
pot, New J. Chem. 30 (2006) 1214–1219.
[26] L. Barthe, A. Denicourt-Nowicki, A. Roucoux, K. Philippot, B. Chaudret, M.
Hemati, Catal. Commun. 10 (2009) 1235–1239.
[27] C. Hubert, E. Guyonnet Bile, A. Denicourt-Nowicki, A. Roucoux, Green Chem. 13
(2011) 1766–1771.
[28] C. Hubert, A. Denicourt-Nowicki, P. Beaunier, A. Roucoux, Green Chem. 12
(2010) 1167–1170.
[29] E. Guyonnet Bile, R. Sassine, A. Denicourt-Nowicki, F. Launay, A. Roucoux, Dal-
ton Trans. 40 (2011) 6524–6531.
[30] C. Hubert, A. Denicourt-Nowicki, J.P. Guegan, A. Roucoux, Dalton Trans. (2009)
7356–7358.
References
[1] V. Polshettiwar, R.S. Varma, Green Chem. 112 (2010) 743–754.
[2] N. Yan, C.X. Xiao, Y. Kou, Coord. Chem. Rev. 254 (2010) 1179–1218.
[3] D. Astruc, F. Lu, J.R. Aranzaes, Angew. Chem. Int. Ed. 44 (2005) 7852–7872.
[4] J.A. Widegren, R.G. Finke, J. Mol. Catal. A: Chem. 198 (2003) 317–341.
[5] A. Roucoux, J. Schulz, H. Patin, Chem. Rev. 102 (2002) 3757–3778.
[6] K. Mori, H. Yamashita, Phys. Chem. Chem. Phys. 12 (2010) 14420–14432.
[7] J. Grunes, J. Zhu, A. Somorjai, Chem. Commun. (2003) 2257–2260.
[8] V. Ponec, G.C. Bond, Catalysis by metals and alloys, in: B. Delmon, Y.T. Yates
(Eds.), Studies in Surface Science and Catalysis, Elsevier, Amsterdam, 2005.
[9] G. Ertl, H. Knoezinger, J. Weitkamp, Catalytic Process. Part B. Handbook of
Heterogeneous Catalysis, Wiley-VCH, New-York, 1997.
[10] J.M. Campelo, D. Luna, R. Luque, J.M. Marinas, A.A. Romero, ChemSusChem 2
(2009) 18–45.
[11] N. Zheng, G.D. Stucky, J. Am. Chem. Soc. 128 (2006) 14278–14280.
[12] B. Cornils, Org. Process Res. Dev. 2 (1998) 121–127.
[13] J. Schulz, S. Levigne, A. Roucoux, H. Patin, Adv. Synth. Catal. 344 (2002)
266–269.
[14] J. Dupont, D. de Oliveira Silva, Transition-metal nanoparticle catalysis in imida-
zolium ionic liquids, in: D. Astruc (Ed.), Nanoparticles and Catalysis, Wiley-VCH,
Weinheim, 2008, pp. 195–218.
[31] E. Redel, J. Krämer, R. Thomann, C. Janiak, J. Organomet. Chem. 694 (2009)
1069–1075.
[32] A. Gual, C. Godard, S. Castillon, C. Claver, Dalton Trans. 39 (2010) 11499– 11512.
[33] L. Petrus, M.A. Noordermeer, Green Chem. 8 (2006) 861–867.