Magnetically recoverable palladium(0) nanocomposite catalyst for hydrogenation reactions in water

Article abstract of DOI:10.1002/cctc.201402761

An easy and straightforward strategy has been used for loading palladium(0) nanoparticles onto the magnetic surface of maghemite (?3-Fe₂O₃), without the use of organic modifiers. The nanocomposite was fully characterized by TEM, XRD, ⁵⁷Fe M??ssbauer spectroscopy, X-ray photoelectron spectroscopy, and superconducting quantum interference device measurements. The Pd⁰@?3-Fe₂O₃ nanocatalyst has been investigated in various catalytic reactions, under mild conditions (100 kPa H₂, RT), and in neat water. Relevant catalytic activities were achieved in the hydrogenation of olefinic substrates, as well as in hydrodehalogenation reactions of halogenoarenes, that constitutes a promising process for wastewater treatment. These nanocatalysts proved also pertinent for the reduction of nitroarene derivatives into the corresponding anilines, which are promising substrates for fine chemistry. Finally, these catalysts proved to be easily recoverable through the use of an external magnet, without significant loss of activity.

Full text of DOI:10.1002/cctc.201402761

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201402761
Magnetically Recoverable Palladium(0) Nanocomposite
Catalyst for Hydrogenation Reactions in Water
Carl-Hugo Pꢀlisson,[a, b] Audrey Denicourt-Nowicki,*[a, b] Cristelle Meriadec,[c]
Jean-Marc Greneche,[d] and Alain Roucoux*[a, b]
An easy and straightforward strategy has been used for load-
ing palladium(0) nanoparticles onto the magnetic surface of
maghemite (g-Fe2O3), without the use of organic modifiers.
The nanocomposite was fully characterized by TEM, XRD,
57Fe Mçssbauer spectroscopy, X-ray photoelectron spectrosco-
py, and superconducting quantum interference device meas-
urements. The Pd0@g-Fe2O3 nanocatalyst has been investigated
in various catalytic reactions, under mild conditions (100 kPa
H2, RT), and in neat water. Relevant catalytic activities were
achieved in the hydrogenation of olefinic substrates, as well as
in hydrodehalogenation reactions of halogenoarenes, that con-
stitutes a promising process for wastewater treatment. These
nanocatalysts proved also pertinent for the reduction of nitro-
arene derivatives into the corresponding anilines, which are
promising substrates for fine chemistry. Finally, these catalysts
proved to be easily recoverable through the use of an external
magnet, without significant loss of activity.
Introduction
In the quest for a sustainable chemistry,[1] the development of
novel cost-effective cleaner catalytic technologies, as well as
the use of environmentally safer solvents such as water or non-
aqueous neoteric solvents (ionic liquids, supercritical CO2, fluo-
rous media), remains of great interest.[2–3] Besides the proper-
ties of the catalysts, such as catalytic activity and selectivity,
their recyclability remains essential for their applicability on an
industrial scale.[4] Among the various catalytic systems, nano-
meter-sized metal particles proved to be relevant and sustaina-
ble catalyst candidates for synthetic transformations,[5–6] com-
bining pertinent activities and selectivities with improved re-
covery potentialities.[7–9] Various strategies have been devel-
oped for the recoverability of these nanocatalysts, including
decantation by using biphasic water/organic solvent or two-
phase systems with ionic liquids,[10–11] or filtration through their
immobilization on inorganic supports.[12–13] In the last decade,
superparamagnetic nanoparticles have emerged as novel sup-
porting materials for catalysts immobilization to further facili-
tate their recovery[14–18] through the use of an external magnet
or by magnetically assisted cross-flow filtration and centrifuga-
tion.[19] This magnetically driven separation approach could be
considered as a clean, cheap, and highly scalable technology,
avoiding filtration steps.[20] Bare magnetic nanoparticles (NPs)
were successfully used as catalysts in some catalytic process-
es,[21] but could also be protected with inorganic supports[22–23]
or directly “decorated” with transition-metal NPs. In the case of
palladium nanospecies, many works have been reported re-
garding their immobilization on the magnetic surface, which
was modified by external organic reagents.[24–25] However, the
functionalization of the magnetic surface remains time-con-
suming and needs the use of organic reagents. Therefore, the
direct immobilization of nanospecies on the unmodified sur-
face of iron oxide, without the use of external modifiers, has
been recently reported.[26–28] Herein we report the straightfor-
ward formation of palladium NPs on magnetic iron oxide sup-
ports, following an impregnation methodology and without
the use of surface modifiers. After characterization, this novel
magnetically recoverable nanocomposite (Pd0@FexOy) was first
evaluated in a model reaction such as the hydrogenation of
olefinic compounds in neat water and under mild conditions.
The catalytic hydrodehalogenation of halogenated arenes was
then investigated as an environmentally friendly and cost-
saving alternative for wastewater treatment, competing with
oxidation processes.[29] Finally, the reduction of nitroarenes[30]
was explored owing to the great potential of the resulting ani-
lines as building blocks in the chemical industry.
[a] Dr. C.-H. Pꢀlisson, Dr. A. Denicourt-Nowicki, Prof. A. Roucoux
Ecole Nationale Supꢀrieure de Chimie de Rennes
CNRS, UMR 6226
11 Allꢀe de Beaulieu, CS 50837, 35708 Rennes Cedex 7 (France)
E-mail: audrey.denicourt@ensc-rennes.fr
alain.roucoux@ensc-rennes.fr
[b] Dr. C.-H. Pꢀlisson, Dr. A. Denicourt-Nowicki, Prof. A. Roucoux
Universitꢀ Europꢀenne de Bretagne
[c] C. Meriadec
UMR Institut de Physique de Rennes
UR1-CNRS 6251, Universitꢀ de Rennes1
Campus de Beaulieu, 35042 Rennes Cedex (France)
Results and Discussion
[d] Dr. J.-M. Greneche
IMMM UMR CNRS 6283, Institut des Molꢀcules et Matꢀriaux du Mans
LUNAM, Universitꢀ du Maine
72085 Le Mans Cedex 9 (France)
Magnetic iron oxide NPs were synthesized through a usual co-
precipitation method using aqueous Fe2+/Fe3+ salt solu-
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
1
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
tions.[31–32] The magnetic materials were characterized with
a wide range of physicochemical techniques to determine the
particle size and morphology, as well as their chemical compo-
sition, their crystalline structure, and their magnetic behavior.
In the TEM images, relatively uniform iron oxide NPs with an
average diameter of 11 nm were observed (Figure 1a). The size
Figure 2. XRD pattern of iron oxide particles.
Figure 1. TEM analyses of iron oxide nanoparticles prepared by the copreci-
pitation method. a) TEM picture (scale bar=100 nm). b) Size distribution.
c) HRTEM picture of an isolated particle. d) Fast Fourier transformation spec-
trum (Ø=diameter).
Figure 3. Magnetization curve of the bare magnetic NPs measured at 300 K.
sat =Saturation magnetization, Hc=coercive force.
M
distribution is relatively narrow, with 80% of the nanoobjects
having a size between 6 and 15 nm (Figure 1b). Structural
characterization of the magnetic particles was performed by
means of high resolution (HR) TEM (Figure 1c) and selected-
area electron diffraction (SAED). The fast Fourier transformation
diffraction patterns (Figure 1d) reveal the presence of interre-
ticular distances of 0.252 and 0.161 nm, which are in good
agreement with those of magnetite iron oxide (Fe3O4, JCPDS
data 19-0629) and/or maghemite iron oxide (g-Fe2O3, JCPDS
data 39-1346). However, as repeatedly described in the litera-
ture, diffraction does not allow distinguishing magnetite from
maghemite.[33]
The XRD patterns demonstrate that the particles display the
typical spinel structure of iron oxide, which could be either
magnetite or maghemite (Figure 2). The average particle size,
calculated by using the Debye–Scherrer formula from the main
reflection peak (311) at 35.78 2q, is approximately 9.6 nm, con-
sistent with the sizes obtained from the TEM pictures.
Figure 4. Fe2p XPS spectrum of the bare iron oxide particles.
The magnetization curve of bare NPs at room temperature
(Figure 3) is characteristic of a superparamagnetic sample and
the saturation magnetization is approximately 66.7 emugÀ1, in
fair agreement with that recently reported for FeNPs exhibiting
similar morphological characteristics.[14]
As previously reported,[34–35] XPS analyses constitute a promis-
ing tool for the discrimination of magnetite (Fe3O4) from ma-
ghemite (g-Fe2O3). The XPS spectrum in the Fe2p region for
the magnetic particles obtained by the coprecipitation ap-
proach is shown in Figure 4. The peak positions of Fe2p3/2 and
Fe2p1/2 are observed at 710.4 and 723.6 eV, respectively, which
agree well with literature values.[36] Moreover, the presence of
an associated satellite peak for Fe2p3/2, located at 719 eV, at
a binding energy approximately 8.5 eV higher than that of the
main peak, constitutes an evidence of the g-Fe2O3 form.[37
To corroborate the XPS results, 57Fe Mçssbauer spectrometry
was performed as a useful technique for determining the oxi-
dation state of Fe species.[38] The Mçssbauer spectra recorded
at 300 and 77 K on the bare magnetic NPs are illustrated in
Figure 5. The spectrum at 300 K exhibits a hyperfine complex
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
2
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Figure 7. X-ray photoelectron spectrum of Pd3d of Pd0@g-Fe2O3.
particles were deposited by chemical reduction of palladium
chloride salts with sodium borohydride under basic conditions.
The palladium loading of the nanomaterials was determined
by ICP analysis as 1.08 wt%, corresponding to the expected
value. In the TEM images (Figure 6a), palladium(0) nanoparti-
cles of 3 nm size are well dispersed on the magnetic surface.
The analyses of the fast Fourier transformation of the HRTEM
picture of an isolated Pd0@g-Fe2O3 nanocomposite (Figure 6b)
mainly reveal interreticular distances of 0.223 nm and 0.250 nm
corresponding to the (111) plane lattice of palladium and
(311) one of maghemite, respectively.
Figure 5. Mçssbauer spectra on NPs samples recorded at 300 K (bottom)
and 77 K (top), V=velocity.
structure consisting of a magnetic sextet with broadened lines
and a central quadrupolar doublet: these two components are
attributed to the larger and the smaller nanoparticles, which
exhibit slower and faster superparamagnetic relaxation phe-
nomena, respectively. In the spectrum at 77 K, only a sextet
with fine and asymmetric lines was observed in agreement
with rather blocked magnetic states. The values of the mean
The X-ray photoelectron spectrum of the Pd3d core level in
the Pd0@g-Fe2O3 material is shown in Figure 7 The Pd3d5/2 and
Pd3d3/2 peaks, which were fitted by constraining the spin-orbit
separation of 5.25 eV, could be resolved into two sets of spin–
orbit doublets.[41–42] Accordingly, the two peaks observed at
335.2 eV (Pd3d5/2) and 340.45 eV (Pd3d3/2) could be attributed
to metallic Pd,[43] whereas those at 337.3 and 342.5 eV in the
deconvoluted spectrum are assigned to the Pd3d5/2 and
Pd3d3/2 peaks of PdO2, respectively.[44] This result indicates that
the surface of palladium particles was partially oxidized in air
to PdO2. Additionally, the peaks corresponding to oxygen
(O1s, 530 eV) and carbon (C1s, 285 eV) used as references
were observed, contrary to those of chlorine (Cl2p3/2, 198.2 eV)
and boron (B1s, 187.3 eV) species issued from the Na2PdCl4
precursor and the NaBH4 reducing agent. The absence of the
peaks corresponding to these species could be explained by
several aqueous washings during the catalyst preparation.
Firstly, the magnetically separable Pd0@g-Fe2O3 catalyst was
investigated in hydrogenation reactions of linear or cyclic ole-
fins under atmospheric hydrogen pressure and at room tem-
perature. The reactions were performed in two different sol-
vents in terms of polarity, water and hexane. The turnover fre-
quencies (TOFs) were calculated, considering an optimized re-
action time for complete conversion of the substrate and
based on the total introduced amount of metal and not on
the exposed surface metal. These TOFs were clearly underesti-
mated but were suitable from an economic point of view.[45] As
previously reported by Janiak et al.,[46] the catalytic activity re-
¯
isomer, which give information on the valency state (d=0.32
and 0.42 mmsÀ1 at 300 and 77 K, respectively), suggest the
presence of rather exclusively Fe3+ species and thus of maghe-
mite form, according to previously reported works.[39] As al-
ready reported,[39] the obtained magnetic particles could not
be labelled as magnetite, rather mainly as maghemite (g-Fe2O3)
as a result of the oxidation process of Fe3O4, which is facilitated
by the very small particles that are characterized by a high sur-
face/volume ratio.
The direct immobilization of palladium(0) NPs on the mag-
netic surface (Pd0@g-Fe2O3) was performed by the wet impreg-
nation method, according to the modified Mizuno’s procedure
used for ruthenium hydroxide on magnetite.[40] After redisper-
sion of the magnetic particles in neat water, the palladium(0)
Figure 6. Pd0@g-Fe2O3 nanocomposite. a) TEM image (scale bar=50 nm).
b) HRTEM image (scale bar=2 nm).
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
3
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
dard mild conditions, no reduction of the aromatic ring was
observed (entry 3).
Table 1. Hydrogenation of olefinic compounds with Pd0@g-Fe2O3 nano-
catalyst in n-hexane or water.[a]
Secondly, the Pd0@g-Fe2O3 nanocatalyst was also studied in
the dehalogenation reactions of halogenoarenes under the
same conditions (100 kPa H2, RT). The results are reported in
Table 2. In a first set of experiments (Table 2, entries 1–4), the
influence of the solvent was studied. Whatever the solvent
was, good activities were achieved with the Pd0@g-Fe2O3 cata-
lyst for the transformation of chlorobenzene into benzene (en-
tries 1, 2), particularly with a TOF in n-hexane of approximately
400 hÀ1. However, the dehalogenation of 2-chloroanisole in n-
hexane (entry 3) led to poor conversion (13%) and destabiliza-
tion of the Pd0@g-Fe2O3 suspension. This result was attributed
to the production of chlorine species in the reaction mixture,
which acts as a poison towards the catalytic system, as already
reported.[48–49] The addition of chlorohydric acid hydrochloric
acid in the hydrogenation of 1-tetradecene in n-hexane after
5 min led to deactivation of the catalyst, with a 40% conver-
sion after 2 h, instead of a complete transformation under clas-
sical conditions (Table 1, entry 2). Moreover, the maghemite (g-
Fe2O3) support was totally inactive towards the dechlorination
reaction of 2-chloroanisole in water (Table 2, entry 5), with no
production of anisole after 24 h. In a second set of experi-
ments, the scope of the reaction was extended to other halo-
genoarenes in neat water (Table 2, entries 7–9). The reactivity
of bromobenzene is lower than of the chloroarenes (Table 2,
entry 7 vs. 2), as already reported.[49–51] This result could be ex-
plained by the lower electron affinity of Br (3.364 eV) than that
of Cl (3.615 eV), which results in a less effective activation of
the bromo reactant through surface s-complex formation. The
additional presence of a second chloro group (Cl or Br) on the
ring leads to a deactivating effect, lowering the halogenoarene
reactivity because the halogen substitution reduces the elec-
tron density associated with the ring carbons (entries 8–9).[51]
Thirdly, the reduction of nitroarenes into aniline was investi-
gated by using the Pd0@g-Fe2O3 nanocomposite under stan-
dard conditions (100 kPa H2, RT) in neat water (Table 3). As al-
ready observed in previous experiments, no reduction of the
nitro groups was observed with the g-Fe2O3 support (Table 3,
entry 2). However, the Pd0@g-Fe2O3 nanocomposite was active
towards the hydrogenation of nitrobenzene in neat water with
a TOF up to 150 hÀ1 (entry 1). Moreover, the tandem hydroge-
nation–dehalogenation reaction of 4-chloronitrobenzene was
Entry
Substrate
Product
TOF[b] [hÀ1
]
in n-hexane
in water
1
2
3
cyclohexene
1-tetradecene
styrene
cyclohexane
tetradecane
ethylbenzene
200
50
600
150
n.c.[c] (82%)
100
[a] Reaction conditions: catalyst Pd0@g-Fe2O3 (50 mg, 1% wt Pd, 4.6ꢂ
10À6 mol Pd), substrate (0.47 mmol), 100 kPa H2, 10 mL solvent, RT. [b] De-
termined by GC analysis and defined as the number of consumed H2 per
mole of introduced Pd per hour. [c] Not calculated; conversion deter-
mined by GC analysis shown in parentheses.
sults not only from the exposed surface metal atoms because
the surface can restructure, atoms can shift positions during
the 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. The results are gathered in
Table 1.
Notably, the bare g-Fe2O3 particles are not active towards
the reduction of unsaturated compounds. In contrast, the
Pd0@g-Fe2O3 nanocomposite presents remarkable activities to-
wards the hydrogenation of linear and cyclic olefins (Table 1,
entries 1–3) in both solvents. However, the catalytic system
was more active in n-hexane, with TOFs up to 600 hÀ1 for sty-
rene (entry 3). These results could be attributed to a better sol-
ubilization of the substrate and product in organic solvents,
and also to a better diffusion of the substrate towards the met-
allic surface, as already reported.[47] Concerning the 1-tetrade-
cene (entry 2), the kinetics are slower owing to the hydropho-
bic nature of the substrate and to the isomerization of the ter-
minal olefin, leading to the formation of more hindered olefins,
which are more difficult to reduce. Finally, under these stan-
Table 2. Dehalogenation of halogenoarenes with Pd0@g-Fe2O3 nano-
catalyst in n-hexane or water.[a]
Entry Substrate
Solvent
Product[b] [%] t [h] TOF[c] [hÀ1
]
1
2
3
4
5[e]
6
7
8
chlorobenzene
n-hexane benzene (98)
water
n-hexane anisole (13)
0.25 392
benzene (100) 1.5 67
Table 3. Hydrogenation of nitroarene derivatives into aniline with Pd0@g-
chlorobenzene
2-chloroanisole
2-chloroanisole
2-chloroanisole
4-chloroaniline
bromobenzene
Fe2O3 in water.[a]
0.5
2.2
24
n.c.[d]
water
water
water
water
anisole (100)
anisole (0)
aniline (100)
benzene (100)
45
–
Entry Substrate
Time [h] TOF[b] [hÀ1
]
3.1
2
32
50
18
22
1
nitrobenzene
nitrobenzene
4-chloronitrobenzene
nitrobenzene/4-chloronitrobenzene (1:1)
2
–
5
5
150
–
80
70
2[c]
3
1,2-dichlorobenzene water
2,4-dichloroanisole water
benzene (100) 5.7
anisole (100) 4.5
9
4
[a] Reaction conditions: catalyst Pd0@g-Fe2O3 (50 mg, 1% wt Pd, 4.6ꢂ
10À6 mol Pd), substrate (0.47 mmol), 100 kPa H2, 10 mL solvent, RT.
[b] Percentage of conversion determined by GC analysis. [c] Number of
consumed H2 per mole of introduced Pd per hour. [d] Not calculated.
[e] g-Fe2O3.
[a] Reaction conditions: catalyst Pd0@g-Fe2O3 (50 mg, 1% wt Pd, 4.6ꢂ
10À6 mol Pd), substrate (0.47 mmol), 100 kPa H2, 10 mL solvent, RT.
[b] TOF determined by GC analysis and defined as the number of con-
sumed H2 per mole of introduced Pd per hour. [c] Reaction performed
with g-Fe2O3 support.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
4
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
was performed (Table 3, entry 4). The kinetic profile is shown in
Figure 9.
The kinetics study (Figure 9) reveals a rapid conversion of ni-
trobenzene into aniline after 1 h, whereas the 4-chloronitro-
benzene ratio remains relatively constant. These results con-
firm the hypothesis that the reduction of nitro groups is fa-
vored over the dechlorination step and that the chlorinated
compound is the preferential intermediate (Scheme 1).
Figure 8. Kinetics of the tandem hydrogenation–dechlorination reaction of
4-chloronitrobenzene in neat water with Pd0@g-Fe2O3 nanocomposite.
performed, with a complete conversion into aniline in 5 h
(entry 3). To understand the preferential reaction pathway, the
kinetics of the reaction was investigated (Figure 8).
Scheme 1. Proposed reaction pathway for the tandem hydrogenation–dech-
lorination reaction of 4-chloronitrobenzene. Dashed arrows: disfavored path-
ways.
The kinetics profile (Figure 8) indicates a reaction pathway
involving the reduction of the nitro group in a first step, pro-
ducing 4-chloroaniline, followed by the dechlorination step.
This hypothesis is reinforced by the fact that the nitro reduc-
tion (Table 3, entry 1) seems to be faster than the dehalogena-
tion of 4-chloroaniline (Table 2, entry 6), thus explaining the
low ratio of nitrobenzene observed during the reaction. More-
over, the conversion of 4-chloronitrobenzene seems to be di-
rectly correlated with the formation of 4-chloroaniline, which
reaches a maximum with the complete conversion of the sub-
strate. To confirm this preferential pathway, the hydrogenation
of a 1:1 mixture of nitrobenzene and 4-chloronitrobenzene
Finally, the catalytic lifetime of Pd0@g-Fe2O3 catalyst was in-
vestigated in water, through successive hydrogenations of
three reference substrates, such as styrene, chlorobenzene,
and nitrobenzene. The catalyst was easily separated from the
reaction mixture, by using an external neodymium magnet,
and washed with water before reuse under the same condi-
tions. This method avoids loss of catalyst during the filtration
step and a potential oxidation of active species. The results
over five runs are reported in Figure 10. For hydrogenation re-
actions on styrene and nitrobenzene, the catalyst could be effi-
ciently reused over five successive runs without loss of catalyt-
ic activity. In the case of styrene, the reuse has been performed
up to 20 runs, showing the excellent catalytic lifetime of
Figure 9. Kinetics of the reduction of a nitrobenzene/4-chloronitrobenzene
(1:1) mixture with the Pd0@g-Fe2O3 catalyst.
Figure 10. Activities of the recycled Pd0@g-Fe2O3 nanocatalyst over five runs
in the hydrogenation of nitrobenzene, styrene, and chlorobenzene.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
5
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Pd0@g-Fe2O3 nanocatalyst. However, in the case of the dehalo-
genation reaction, the production of a co-product, hydrochlo-
ric acid, led to a progressive destabilization of the Pd0@g-Fe2O3
suspension and a deactivation of the catalytic system in the
fourth run. This phenomenon could be attributed to the disso-
lution of the magnetic support under acidic conditions, as al-
ready reported.[52] Nevertheless, the results prove the efficient
adsorption of metallic nanospecies on the magnetic surface,
with no metal leaching in the sequential reuse reactions in re-
duction reactions.
a-Fe foil was used as the calibration sample. The values of isomer
shift are quoted to that of a-Fe at 300 K.
Synthesis of Fe2O3 magnetic NPs
Ferrous chloride hydrate (2.6 g, 13 mmol) was added to a hydro-
chloric acid solution (20 mL, 2m). Ferric chloride hydrate (7.1 g,
26 mmol) was dissolved in nitrogen-purged water (200 mL), fol-
lowed by the addition of the ferrous chloride solution. After
15 min of mechanic stirring, A 20 mL volume of nitrogen-purged
aqueous NH3 solution (20 wt%) was slowly introduced. The color
of the solution turned from orange to dark brown and the stirring
was maintained for 1 h. The nanoparticles were magnetically deca-
nted and washed with deionized water.
Conclusion
We have developed a highly stable magnetically recoverable
nanocomposite based on nanometer-sized palladium particles
that are well dispersed on the magnetic support. XPS and
Mçssbauer analyses enabled us to conclude that the iron
oxide particles were composed of maghemite (g-Fe2O3). This
magnetic support was directly functionalized by zerovalent
palladium nanoparticles having sizes of 3 nm, through a wet
impregnation method and without the use of surface modifi-
ers. The obtained nanocomposite (Pd0@g-Fe2O3) exhibited ex-
cellent activities towards the reduction of olefinic compounds
and nitro derivatives, as well as in dehalogenation processes.
Moreover, this nanocatalyst could be easily recycled through
magnetic separation from the reaction products, avoiding the
use of organic solvents. The simple and straightforward meth-
odology for immobilizing metallic nanospecies on magnetic
iron oxide leads to the design of an attractive catalyst in terms
of activity, selectivity, and recyclability in the drive towards
sustainable processes.
Synthesis of maghemite-supported Pd0 NPs (Pd0@g-Fe2O3)
Maghemite particles (1 g) were redispersed in deionized water
(50 mL) by sonication. The pH was adjusted at over 11 through ad-
dition of a NaOH solution (1m). An aqueous solution of Na2PdCl4
(27 mg, 9.4ꢂ10À6 mol, 10 mL) was slowly added and the stirring
was maintained for 1 h. After cooling at 08C, an aqueous solution
of NaBH4 (1 mg, 2.6ꢂ10À5 mol, 10 mL) was added dropwise. After
30 min at 08C, the solution was stirred at RT for 2 h. The nanoparti-
cles were magnetically decanted, washed with deionized water,
and dried at 808C over the night.
Catalytic tests and recycling
Catalysts (50 mg) were introduced in a glass recipient with a stirring
bar and redispersed in deionized water (10 mL) by sonication. The
adequate quantity of substrate was added. The solution was put
under vacuum then hydrogen. The operation was renewed three
times and the final pressure of hydrogen was adjusted to 1 atm. To
recycle the catalyst, a neodymium magnet was applied under the
recipient to precipitate the catalyst magnetically and remove the
liquid phase. The catalyst was washed with water and precipitated
three times before a new run.
Experimental Section
General
FeCl2·4H2O, FeCl3·6H2O, and Na2PdCl4 were purchased from Strem
Chemicals. Sodium borohydride and all organic substrates were
obtained from Aldrich, Acros, or Alfa Aesar and were used as re-
ceived. Water was distilled twice by conventional methods before
use. TEM images were performed at the Universitꢀ Pierre et Marie
Curie. They were recorded with a JEOL TEM 100CXII electron micro-
scope operated at an acceleration voltage of 100 kV. The nanoma-
terial was dispersed in ethanol or water with ultrasounds and
a drop of this solution was deposited on a carbon-coated copper
grid and dried in air. HRTEM and energy-dispersive X-ray analysis
were performed by using a JEOL JEM 2010 UHR microscope com-
bined with a PGT-IMIX PC microanalyzer. Hysteresis cycles of the
magnetic NPs were recorded at RT by using a MPMS XL5 magneto-
meter. XPS measurements were performed with an MgKa X-ray
source (hn=1254 eV) by using a VSW HA100 photoelectron spec-
trometer with a hemispherical photoelectron analyzer, working at
an energy pass of 22 eV. The experimental resolution was 1.0 eV.
The binding energy for the main CÀC peak was taken at 285.0 eV
as an internal reference level for all measurements. Spectral analy-
sis included a Shirley background subtraction and peak separation
using mixed Gaussian–Lorentzian functions. 57Fe Mçssbauer spectra
were recorded at 300 and 77 K by using a conventional constant
acceleration transmission spectrometer with a 57Co (Rh) source.
The spectra were fitted by means of the MOSFIT program and an
Acknowledgements
The Rꢀgion Bretagne is acknowledged for financial supports (PhD
fellowship MAGFOCAT). The authors are indebted to Patricia
Beaunier from Universitꢀ Pierre et Marie Curie (UMPC) for trans-
mission electron microscopy analysis. They also thank Isabelle
Marlart for XRD analyses at ENSCR and Thierry Guizouarn for su-
perconducting quantum interference device measurements at
Universitꢀ de Rennes I.
Keywords: magnetic properties
nanoparticles · palladium
· hydrogenation · iron ·
[1] R. A. Sheldon, Chem. Commun. 2008, 3352–3365.
[2] R. S. Varma, Green Chem. 2014, 16, 2027–2041.
[3] P. T. Anastas, M. M. Kirchhoff, Acc. Chem. Res. 2002, 35, 686–694.
[4] D. J. Cole-Hamilton, R. P. Tooze, Catalyst Separation, Recovery and Recy-
cling: Chemistry and Porcess Design, Springer, 2006.
[5] H. Cong, J. A. Porco, ACS Catal. 2012, 2, 65–70.
[6] L. L. Chng, N. Erathodiyil, J. Y. Ying, Acc. Chem. Res. 2013, 46, 1825–
1837.
[7] K. An, G. A. Somorjai, ChemCatChem 2012, 4, 1512–1524.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
6
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[8] V. Polshettiwar, R. S. Varma, Green Chem. 2010, 12, 743–754.
[9] D. Astruc, Nanoparticles and Catalysis, Wiley-VCH, Weinheim, 2008.
[10] A. Denicourt-Nowicki, A. Roucoux in Nanomaterials in Catalysis (Eds.: P.
Serp, K. Philippot), Wiley-VCH, Weinheim, 2013, pp. 55–95.
[11] J. D. Scholten, B. C. Leal, J. Dupont, ACS Catal. 2012, 2, 184–200.
[12] J. M. Campelo, D. Luna, R. Luque, J. M. Marinas, A. A. Romero, ChemSu-
sChem 2009, 2, 18–45.
[13] R. J. White, R. Luque, V. L. Budarin, J. H. Clark, D. J. Macquarrie, Chem.
Soc. Rev. 2009, 38, 481–494.
[14] L. M. Rossi, N. J. S. Costa, F. P. Silva, R. Wojcieszak, Green Chem. 2014, 16,
2906–2933.
[15] R. B. Nasir Baig, R. S. Varma, Green Chem. 2013, 15, 398–417.
[16] M. B. Gawande, P. S. Branco, R. S. Varma, Chem. Soc. Rev. 2013, 42,
3371–3393.
[17] R. B. N. Baig, R. S. Varma, Chem. Commun. 2013, 49, 752–770.
[18] V. Polshettiwar, R. Luque, A. Fihri, H. Zhu, M. Bouhrara, J.-M. Basset,
Chem. Rev. 2011, 111, 3036–3075.
[33] A. Shavel, B. Rodrꢅguez-Gonzꢆlez, M. Spasova, M. Farle, L. M. Liz-Marzꢆn,
Adv. Funct. Mater. 2007, 17, 3870–3876.
[34] C. Ruby, B. Humbert, J. Fusy, Surf. Interface Anal. 2000, 29, 377–380.
[35] S. Asuha, B. Suyala, X. Siqintana, S. Zhao, J. Alloys Compd. 2011, 509,
2870–2873.
[36] T. Yamashita, P. Hayes, Appl. Surf. Sci. 2008, 254, 2441–2449.
[37] M. Muhler, R. Schlçgl, G. Ertl, J. Catal. 1992, 138, 413–444.
[38] J. Fouineau, K. Brymora, L. Ourry, F. Mammeri, N. Yaacoub, F. Calvayrac,
S. Ammar-Merah, J. M. Greneche, J. Phys. Chem. C 2013, 117, 14295–
14302.
[39] J. Santoyo Salazar, L. Perez, O. de Abril, L. Truong Phuoc, D. Ihiawakrim,
M. Vazquez, J.-M. Greneche, S. Begin-Colin, G. Pourroy, Chem. Mater.
2011, 23, 1379–1386.
[40] M. Kotani, T. Koike, K. Yamaguchi, N. Mizuno, Green Chem. 2006, 8, 735–
741.
[41] S. Santra, P. Ranjan, P. Bera, P. Ghosh, S. K. Mandal, RSC Adv. 2012, 2,
7523–7533.
[19] A.-H. Lu, E. L. Salabas, F. Schꢃth, Angew. Chem. Int. Ed. 2007, 46, 1222–
1244; Angew. Chem. 2007, 119, 1242–1266.
[42] R. Wojcieszak, M. J. Genet, P. Eloy, P. Ruiz, E. M. Gaigneaux, J. Phys.
Chem. C 2010, 114, 16677–16684.
[20] S. Shylesh, V. Schꢃnemann, W. R. Thiel, Angew. Chem. Int. Ed. 2010, 49,
3428–3459; Angew. Chem. 2010, 122, 3504–3537.
[21] R. Hudson, Y. Feng, R. S. Varma, A. Moores, Green Chem. 2014, 16,
4493–4505.
[43] M. C. Militello, S. J. Simko, Surf. Sci. Spectra 1994, 3, 387–394.
[44] P. H. Wang, C. Y. Pan, Colloid Polym. Sci. 2001, 279, 171–177.
[45] A. P. Umpierre, E. de Jesffls, J. Dupont, ChemCatChem 2011, 3, 1413–
1418.
[22] Z. Dong, X. Le, C. Dong, W. Zhang, X. Li, J. Ma, Appl. Catal. B 2015, 162,
372–380.
[46] E. Redel, J. Krꢄmer, R. Thomann, C. Janiak, J. Organomet. Chem. 2009,
694, 1069–1075.
[23] Z. Dong, X. Tian, Y. Chen, Y. Guo, J. Ma, RSC Adv. 2013, 3, 1082–1088.
[24] L. M. Rossi, M. A. S. Garcia, L. L. R. Vono, J. Braz. Chem. Soc. 2012, 23,
1959–1971.
[47] L. Barthe, A. Denicourt-Nowicki, A. Roucoux, K. Philippot, B. Chaudret,
M. Hemati, Catal. Commun. 2009, 10, 1235–1239.
[48] J. Andersin, P. Parkkinen, K. Honkala, J. Catal. 2012, 290, 118–125.
[49] C. Hubert, E. Guyonnet Bilꢀ, A. Denicourt-Nowicki, A. Roucoux, Appl.
Catal. A 2011, 394, 215–219.
[50] B. Lꢀger, A. Nowicki, A. Roucoux, J.-P. Rolland, J. Mol. Catal. A 2007, 266,
221–225.
[51] M. A. Keane, Appl. Catal. A 2004, 271, 109–118.
[52] W. Wu, Q. G. He, C. Jiang, Nanoscale Res. Lett. 2008, 3, 397–415.
[25] Q. M. Kainz, O. Reiser, Acc. Chem. Res. 2014, 47, 667–677.
[26] Y. Kim, M. J. Kim, Bull. Korean Chem. Soc. 2010, 31, 1368–1370.
[27] J. Liu, X. Peng, W. Sun, Y. Zhao, C. Xia, Org. Lett. 2008, 10, 3933–3936.
[28] K. V. S. Ranganath, J. Kloesges, A. H. Schꢄfer, F. Glorius, Angew. Chem.
Int. Ed. 2010, 49, 7786–7789; Angew. Chem. 2010, 122, 7952–7956.
[29] D. Richard, L. D. Nunez, C. de Bellefon, D. Schweich in Environmental
Chemistry (Eds.: E. Lichtfouse, J. Schwarzbauer, D. Robert), Springer,
Berlin, 2005.
[30] P. Lara, K. Philippot, Catal. Sci. Technol. 2014, 4, 2445–2465.
[31] L. Bartolome, M. Imran, K. G. Lee, A. Sangalang, J. K. Ahn, D. H. Kim,
Green Chem. 2014, 16, 279–286.
[32] Y. S. Kang, S. Risbud, J. F. Rabolt, P. Stroeve, Chem. Mater. 1996, 8, 2209–
2211.
Received: September 23, 2014
Revised: October 22, 2014
Published online on && &&, 0000
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
7
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
C.-H. Pꢀlisson, A. Denicourt-Nowicki,*
Active, selective, and reusable! A mag-
netically Pd0 nanocomposite was easily
prepared and efficiently used in hydro-
genation reactions of various substrates
(nitro- and/or halogenoarenes) in neat
water.
C. Meriadec, J.-M. Greneche, A. Roucoux*
&& – &&
Magnetically Recoverable Palladium(0)
Nanocomposite Catalyst for
Hydrogenation Reactions in Water
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
8
&
ÞÞ
These are not the final page numbers!