How to modulate catalytic properties in nanosystems: The case of iron-ruthenium nanoparticles

Article abstract of DOI:10.1002/cctc.201300907

Ultrasmall FeRu bimetallic nanoparticles were prepared by co-decomposition of two organometallic precursors, {Fe[N(Si(CH₃)₃) ₂]₂}₂ and (η⁴-1,5- cyclooctadiene)(η⁶-1,3,5-cyclooctatriene)ruthenium(0) (Ru(COD)(COT)), under dihydrogen at 150 °C in mesitylene. A series of FeRu nanoparticles of sizes of approximately 1.8 nm and incorporating different ratios of iron to ruthenium were synthesized by varying the quantity of the ruthenium complex introduced (Fe/Ru=1:1, 1:0.5, 1:0.2, and 1:0.1). FeRu nanoparticles were characterized by TEM, high-resolution TEM, and wide-angle X-ray scattering analyses. Their surface was studied by hydride titration and IR spectroscopy after CO adsorption and their magnetic properties were analyzed by using a superconducting quantum interference device (SQUID). The FeRu nanoparticles were used as catalysts in the hydrogenation of styrene and 2-butanone. The results indicate that the selectivity of the nanoparticle catalysts can be modulated according to their composition and therefore represent a case study on fine-tuning the reactivity of nanocatalysts and adjusting their selectivity in a given reaction. Singing a bimetallic tune: The selectivity of FeRu nanocatalysts in hydrogenation reactions can be tuned by adjusting the Ru content in bimetallic FeRu ultrasmall nanoparticles.

Full text of DOI:10.1002/cctc.201300907

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201300907
How to Modulate Catalytic Properties in Nanosystems:
The Case of Iron–Ruthenium Nanoparticles
Vinciane Kelsen,[a] Anca Meffre,[a] Pier-Francesco Fazzini,[a] Pierre Lecante,[b] and
Bruno Chaudret*[a]
Ultrasmall FeRu bimetallic nanoparticles were prepared by co-
decomposition of two organometallic precursors, {Fe[N(Si-
(CH3)3)2]2}2 and (h4-1,5-cyclooctadiene)(h6-1,3,5-cyclooctatrie-
ne)ruthenium(0) (Ru(COD)(COT)), under dihydrogen at 1508C
in mesitylene. A series of FeRu nanoparticles of sizes of approx-
imately 1.8 nm and incorporating different ratios of iron to
ruthenium were synthesized by varying the quantity of the
ruthenium complex introduced (Fe/Ru=1:1, 1:0.5, 1:0.2, and
1:0.1). FeRu nanoparticles were characterized by TEM, high-res-
olution TEM, and wide-angle X-ray scattering analyses. Their
surface was studied by hydride titration and IR spectroscopy
after CO adsorption and their magnetic properties were ana-
lyzed by using
a superconducting quantum interference
device (SQUID). The FeRu nanoparticles were used as catalysts
in the hydrogenation of styrene and 2-butanone. The results
indicate that the selectivity of the nanoparticle catalysts can be
modulated according to their composition and therefore repre-
sent a case study on fine-tuning the reactivity of nanocatalysts
and adjusting their selectivity in a given reaction.
Introduction
Iron-based catalysts are of great interest because of their low
toxicity and the abundance of iron on Earth. Iron nanoparticles
(NPs) have been known to be good catalysts for Fischer–
Tropsch synthesis for approximately 100 years.[1] Recently, the
use of iron NPs as catalysts for hydrogenation reactions has
emerged. For example, iron NPs supported on graphene were
reported to display activity in alkene hydrogenation,[2] and
Morris and co-workers described the application of iron NPs in
the asymmetric transfer hydrogenation of ketones.[3] An inter-
esting general method for the synthesis of soluble iron NPs
active in alkene and alkyne hydrogenation has been reported
that allows the preparation of Fe NPs supported on MgO.[4] In-
dependently, our group has developed the synthesis of Fe NPs
of various sizes and shapes[5] and recently published the use of
ultrasmall iron(0) NPs as catalysts for the hydrogenation of un-
saturated CÀC bonds.[6] These NPs could hydrogenate alkenes
and alkynes under mild conditions (10 bar H2 and room tem-
perature) and also displayed weak activity in the direct hydro-
genation of C=O bonds. Our group has a long-standing inter-
est in ruthenium NPs,[7] which display a rich catalytic chemistry,
in particular, for arene[8] and C=O bonds[9] hydrogenation.
A challenge in this field is the design of nanocatalysts with
precise selectivity. Tuning the selectivity is very important be-
cause NPs are expected to play a more extended role in the
transformation of multifunctional molecules such as molecules
of biological interest. For example, we have recently reported
that Ru NPs can be selective for reducing aromatic ketones. In
some cases, only the arene ring is hydrogenated and not the
ketone function.[10] However, molecules of biological impor-
tance often associate aromatic rings that need to be preserved
for their biological activity with functions relatively difficult to
reduce such as C=O or C=N bonds. The question is, therefore,
the following: can we reverse this selectivity by manipulating
the NPs? To answer this question, we decided to look at a com-
bination of two different metals displaying very different cata-
lytic properties, namely iron and ruthenium.
FeRu NPs have proven to be interesting catalysts for various
reactions, such as Fischer–Tropsch synthesis,[11] water gas
shift,[12] and hydrogenation of unsaturated aldehydes and ke-
tones.[13] In most cases, the catalysts are immobilized on a sup-
port and are synthesized by co-impregnation on the solid sup-
port or in ionic liquids.[3,11c,d,14] Other syntheses of FeRu NPs in-
volve microwave irradiation,[15] hydrazine reduction,[11a,b] or
metal organic chemical vapor deposition.[16] A synthesis of
FeRu bimetallic NPs by a modified polyol method was also re-
cently reported.[12] By using poly-N-vinyl-2-pyrrolidone (PVP) as
a stabilizing agent, spherical NPs of approximately 8 nm were
obtained. To our knowledge, the synthesis of FeRu bimetallic
NPs through an organometallic route is unprecedented.
[a] Dr. V. Kelsen, Dr. A. Meffre, Dr. P.-F. Fazzini, Dr. B. Chaudret
LPCNO–INSA, Universitꢀ de Toulouse
135 avenue de Rangueil 31077 Toulouse Cedex 4 (France)
Fax: (+33)05-62-17-28-16
E-mail: chaudret@insa-toulouse.fr
[b] Dr. P. Lecante
Herein we report the synthesis and full characterization of
FeRu bimetallic NPs with different ratios of iron to ruthenium
by co-decomposition of two organometallic precursors,
{Fe[N(Si(CH3)3)2]2}2 and Ru(COD)(COT), under H2 pressure. We
Centre d’Elaboration des Matꢀriaux et d’Etudes Structurales
29 rue Jeanne Marvig, 31055 Toulouse Cedex 4 (France)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201300907.
ꢀ 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
also investigated their catalytic properties for the hydrogena-
tion of styrene and 2-butanone.
iron to ruthenium did not influence the morphology of the re-
sulting NPs.
High-resolution transmission electron microscopy (HRTEM)
as well as scanning transmission electron microscopy (STEM)
imaging combined with energy-dispersive X-ray analysis (EDX)
were used to analyze the structure and composition of the re-
sulting NPs. In Figure 2 the HRTEM images of NPs with differ-
Results and Discussion
Structural properties
FeRu NPs with different molar Ru/Fe ratios were synthesized to
study the effect of ruthenium concentration on their catalytic
properties. These properties were compared to those of both
Ru NPs and ultrasmall iron(0) NPs described in a previous
paper.[6] For this purpose, a mesitylene solution containing vari-
ous relative concentrations of {Fe[N(Si(CH3)3)2]2}2 and (h4-1,5-cy-
clooctadiene)(h6-1,3,5-cyclooctatriene) ruthenium(0) [Ru(COD)-
Figure 2. HRTEM images of FeRu NPs with a) 0.5 equiv. Ru and b) 0.2 equiv.
Ru.
ent molar Ru/Fe ratios are displayed. It is clear from this analy-
sis that the FeRu NPs do not have a well-defined crystalline
structure.
STEM/EDX analyses allow the cartography of the particles
composition. The resulting signal-to-noise ratios were, howev-
er, too low to obtain reliable STEM/EDX data on the particles,
owing to their small size and the presence of free ligands on
the TEM grid. To determine whether bimetallic NPs can be
formed in these conditions, we decided to increase the size of
NPs by adding a ligand, hexadecylamine (HDA). For these NPs,
one equivalent of ruthenium and three equivalents of HDA
were added to the iron precursor. In Figure 3 the TEM, HRTEM,
Figure 1. TEM images of FeRu bimetallic NPs with a) 1, b) 0.5, c) 0.2, and
d) 0.1 equiv. of ruthenium (scale bar=100 nm).
(COT)] (1Ru:1Fe, 0.5Ru:1Fe,
0.2Ru:1Fe,
0.1Ru:1Fe;
Fig-
ure 1a–d, respectively) was treat-
ed under 3 bar of dihydrogen at
1508C overnight. In all cases, the
samples were analyzed by ele-
mental analysis after precipita-
tion and washing with toluene
to determine the composition of
the NPs (see the Supporting In-
formation). The values found
(Fe/Ru=1:0.09, 1:0.16, 1:0.49,
1:1) matched the initial composi-
tion of the solutions (Fe/Ru=
1:0.1, 1:0.2, 1:0.5, and 1:1).
TEM images obtained after
evaporation of the solution only
revealed one population of
spherical NPs displaying a size
between
1.6
and
1.9 nm
Figure 3. a) TEM, b) HRTEM, and c) STEM/EDX images of FeRu (1:1) NPs synthesized in the presence of HDA with
(Figure 1). The molar ratio of the cartographies of d) iron and e) ruthenium. Scale bars (c,d,e)=40 nm.
ꢀ 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
and STEM/EDX images are shown characterizing the FeRu NPs
synthesized in the presence of HDA. Interestingly, these NPs
presented a different morphology: a wormlike irregular shape.
We have previously demonstrated that HDA promotes the coa-
lescence of Ru NPs to lead to a wormlike morphology as
a result of weak coordination of HDA on the metal surface.[17]
The mean diameter of the worms is approximately 2–3 nm,
which allows characterization by STEM/EDX. The cartography
of elements revealed that Fe and Ru are both present in the
same particles and confirmed the bimetallic nature of the FeRu
NPs.
atomic number has been shown to be a direct proof of segre-
gation in the particles. In the present case, such scaling strong-
ly indicates that FeRu/HDA NPs, although bimetallic as ob-
served by STEM/EDX, are actually not alloyed but organized as
Ru@Fe core–shell particles.
For the other samples, a much smaller and nearly constant
amplitude could be observed. In addition, RDF patterns are
very similar for all compositions with no significant shift of the
first interatomic distance, indicating an average bond length
nearly independent from the Ru/Fe ratio. This pattern is also
very close to the one previously observed for pure Fe particles
prepared by using similar conditions.[18] All these results
strongly suggest that most of the order in the FeRu PVP sam-
ples, as observed by using X-ray scattering, is derived from the
same small particles in all four cases, namely Fe NPs adopting
the manganese beta structure, all with a coherence length of
approximately 1.5 nm. This structure does not exclude a bimet-
allic character of the particles, which would explain the dis-
crepancy between the nearly constant coherence length and
the different sizes observed by TEM. Considering the large
atomic number of Ru compared to that of Fe and the lack of
distances clearly related to the bigger Ru atoms, these results
exclude any kind of extended structure involving Ru, for exam-
ple, Janus-like organization.
Owing to the small size of these NPs, X-ray diffraction meas-
urements only provided broad peaks, which did not allow
a clear understanding of the structure of our NPs. Therefore,
wide-angle X-ray scattering (WAXS) analyses were performed,
which provided more information (see Figure 4).
In summary, the WAXS studies evidence the presence of
a well-ordered core–shell structure Ru@Fe in the presence of
HDA and a more disordered but also core–shell structure of
the type Fe@Ru in the absence of HDA, with a core structure
similar to the one observed for pure iron NPs. The origin of
this discrepancy is not clear but is probably related to the ac-
cessibility of Ru surface to HDA during the initial steps of the
growth of Ru NPs, which then coalesce as in the case of pure
Ru NPs.[17] In the case of the particles prepared in the absence
of HDA, the core is clearly essentially composed of Fe NPs on
which Ru is adsorbed in a disordered way. One explanation for
this surprising reversal of reactivity could be the fact that Ru-
(COD)(COT) reacts in the absence of added amine with aromat-
ic solvents at elevated temperature under H2 to give stable
(arene)(COD) ruthenium(0) species much more stable than Ru-
(COD)(COT).[20] If an excess of HDA is present in the solution, it
coordinates on ruthenium, hence preventing the formation of
the ruthenium–arene complex. This demonstrates once more
the importance of the organometallic chemistry present in so-
lution prior to the formation of NPs.[21] For the purpose of our
catalytic studies, we did not consider the Fe/Ru/HDA system,
which is too different, and concentrated on the NPs obtained
in the absence of additional ligand. It is, however, necessary to
gain more evidence for the presence of increasing quantities
of Ru on the surface of the Fe NPs. For this purpose, surface
and magnetic studies were performed.
Figure 4. WAXS analyses of FeRu NPs with b) 0.2, c) 0.5, and d) 1 equiv. of
Ru, e) in the presence of HDA, and their comparison with a) the experimen-
tal function obtained for pure Fe NPs[18,19] and f) the function computed
from the Fe1Ru1 structure (expanded by 3% for comparison purposes).
r=Coherence length.
The observed coherence length, which corresponds to the
size of crystalline domains, is close to 1.5 nm for FeRu NPs pre-
pared in the absence of HDA. It exceeds 2 nm for the NPs pre-
pared in the presence of HDA. These values are in good agree-
ment with the observations made by TEM/HRTEM analyses.
The structure of the particles prepared in the presence of HDA
(FeRu/HDA) appears very similar to that of both bulk Fe1Ru1
alloy and pure hexagonal close-packed (hcp) ruthenium. Thus,
the obtained radial distribution function (RDF) of FeRu/HDA is
close to the one computed from a model based on the struc-
ture of Fe1Ru1. The coherence length is above 2 nm and a sig-
nificant deviation from the perfect hexagonal structure can be
observed at increasing distances. The average bond length es-
timated on the RDF is, however, larger than the one in the
Fe1Ru1 alloy (0.2624 nm) by a 3% expansion factor, which
makes the average bond length in FeRu/HDA almost identical
to the value in pure ruthenium (0.2705 nm). In previous WAXS
studies on different bimetallic NPs,[19] scaling of all the intera-
tomic distances on the parameter of the species of higher
Surface state studies
We performed first the titration of hydrides present at the sur-
face of the FeRu NPs. The general procedure for the quantifica-
tion of hydrogen atoms adsorbed onto the surface of metallic
NPs through hydrogenation of 2-norbornene has been previ-
ꢀ 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
ously described.[22] This procedure allows not only the determi-
nation of the number of hydrides present on the surface of
a given nanoparticle but also the demonstration that the sur-
face of the particle is not oxidized, because, especially for iron,
any trace of oxidation would immediately lead to the disap-
pearance of all hydrides. The procedure is the following: three
freeze–pump cycles were performed on each fresh colloidal so-
lution of bimetallic FeRu NPs to eliminate the dihydrogen pres-
ent in the solvent. Then, one equivalent of olefin (2-norbor-
nene) was added. After 24 h of magnetic stirring at room tem-
perature, samples were analyzed by gas phase chromatogra-
phy. With the conversion of 2-norbornene into norbornane
and the estimated percentage of surface atoms of Ru and Fe
of the particle (we considered 70% for a mean size of 1.7 nm),
the number of hydrides per surface metal atom can be esti-
mated. Results are reported in Table 1.
Magnetization measurements were performed on a Quantum
Design model MPMS5.5 superconducting quantum interfer-
ence device (SQUID) magnetometer. They were performed on
dried powders and the samples were sealed under argon at-
mosphere to preserve the NPs from any uncontrolled oxida-
tion. In Figure 5 the hysteresis curves are shown relative to the
Table 1. Number of hydrides per surface metal atom of some FeRu bi-
metallic NPs.
NPs
Number of surface hydrides per surface metal atom
FeRu (1:1)
FeRu (1:0.5)
FeRu (1:0.1)
0.40
0.40
0.46
Figure 5. Hysteresis curves measured at 2 K of bimetallic FeRu NPs with dif-
ferent molar ratios of iron to ruthenium. m0H(T)=Applied magnetic field
intensity.
We found approximately 0.4 hydrides per surface metal
atom for all the FeRu NPs studied. This is a lower limit because
the freeze–pump cycles lead to the elimination of a number of
hydrides from the NPs surface. These results demonstrate, nev-
ertheless, the nonoxidized character of these FeRu NPs and,
hence, their potential in catalytic hydrogenation reactions.
To have a better understanding of the surface state of our
bimetallic FeRu NPs, experiments of CO absorption were also
performed, because the CO stretch can give information on
the nature of the metal present on the surface (see the Sup-
porting Information).[23]
series of FeRu NPs and measured at 2 K. The absolute magneti-
zation was deduced from the iron(0) total content determined
by microanalysis using the inductively coupled plasma mass
spectrometry technique (ICP). Bimetallic FeRu NPs exhibit a fer-
romagnetic behavior at 2 K. The measured saturation magneti-
zations (Ms) are lower than that of bulk iron (210 Am2 kgÀ1 at
2 K).[26] FeRu NPs saturation magnetizations vary between 187
and 18 Am2 kgÀ1 at 2 K following the increase of the ruthenium
molar ratio as expected for a dilution effect of nonmagnetic Ru
atoms.[27]
For this study, solid samples of FeRu NPs were placed in
a Fischer–Porter bottle under CO (3 bar) for 2 h. After this time,
the powder was analyzed by IR spectroscopy. Very broad CO
stretches were detected at 1927 cmÀ1 and 1946 cmÀ1 for FeRu
NPs of a molar ratio of 0.1Ru:1Fe and 0.5Ru:1Fe, respectively.
These values are again in agreement with the absence of oxi-
dation of the NPs surfaces and are intermediate between
Catalytic properties
Having proven the bimetallic nature of the FeRu particles, their
catalytic performance was investigated in the hydrogenation
of styrene and 2-butanone (Scheme 1) and compared with the
those resulting from CO adsorption on pure Fe (1899 cmÀ1
)
and pure Ru/PVP NPs (ꢀ1945 cmÀ1) or close pure Ru/PVP.[24]
Furthermore, the CO band frequency varies as a function of
the Ru content as expected. This is in excellent agreement
with the proposal of the bimetallic nature of the NPs and the
presence of Ru on their surfaces.
Scheme 1. Hydrogenation of styrene catalyzed by FeRu bimetallic NPs.
catalytic performance of monometallic iron and ruthenium
NPs. We chose these two unsaturated substrates to evaluate
the potential of our NPs for the hydrogenation of a terminal
alkene and an arene (with styrene) and of a keto function
(with 2-butanone). Reference Ru NPs were synthesized by
Magnetic properties
Magnetic studies are a very sensitive tool to monitor the for-
mation of bimetallic NPs, previously used, for example, to
follow the formation of RuCo NPs.[25]
ꢀ 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
adding two equivalents of bis(trimethylsilyl)amine (HMDS, the
ligand of the bimetallic NPs) on Ru(COD)(COT) in mesitylene
and by stirring the mixture under 3 bar of H2 at room tempera-
ture overnight.
As discussed in a previous work,[6] ultrasmall Fe0 NPs selec-
tively hydrogenated styrene to give ethylbenzene without
touching the phenyl ring in the conditions used (3 bar H2,
room temperature). Moreover, Fe0 NPs were inactive for the
hydrogenation of 2-butanone in the conditions used herein (in
our previous work, a 7% conversion was observed at an H2
pressure of 10 bar).
However, these bimetallic NPs could hydrogenate 2-buta-
none into 2-butanol with a conversion increasing as a function
of the increased number of Ru equivalents in the mild condi-
tions used (3 bar H2, room temperature, 24 h, 5 mol% NPs). In
principle, only one surface ruthenium atom is necessary to hy-
drogenate the C=O bond. NPs at very low ruthenium loading
(0.1Ru:1Fe) selectively hydrogenated both styrene into ethyl-
benzene and butanone into butanol (Table 2, entry 6; Table 3,
entry 6). Although the reaction conditions were far from opti-
mum, these results are interesting in that the selectivity of the
hydrogenation reaction could be easily tuned by doping iron
NPs with ruthenium metal.
In the case of the use of pure ruthenium NPs as the cata-
lysts, styrene was completely converted into ethylcyclohexane
and 2-butanone was completely converted into 2-butanol with
high yield, in both cases after 20 h under an H2 pressure of
3 bar at room temperature and by using 5 mol% of Ru
(Tables 2 and 3).
Conclusions
FeRu nanoparticles were synthesized by a simple chemical
route consisting of the co-decomposition of {Fe[N(Si(CH3)3)2]2}2
and Ru(cyclooctadiene)(cyclooctatriene) under an H2 pressure
of 3 bar at 1508C overnight. Thus, ultrasmall FeRu nanoparti-
cles of approximately 1.8 nm were obtained. Structural charac-
terization methods (TEM, high-resolution TEM, scanning trans-
mission electron microscopy combined with energy-dispersive
X-ray analysis, and wide-angle X-ray scattering analysis) proved
the bimetallic composition of the nanoparticles. A magnetic
study on a superconducting quantum interference device
demonstrated the ferromagnetic behavior of the FeRu nano-
particles and showed that their magnetization decreased with
the ruthenium content of the particles. Surface state studies
(hydride titration and IR spectroscopy on CO-exposed nanopar-
ticles) revealed that the FeRu nanoparticles were not oxidized
and, therefore, suitable for hydrogenation reaction.
Table 2. Styrene hydrogenation under 3 bar H2 at room temperature
during 24 h using 5 mol% of NPs as the catalyst and mesitylene as the
solvent: influence of the NPs metallic composition on the reaction
selectivity.
Entry NPs
Styrene
conv.[a] [%]
Ethylbenzene[a]
yield [%]
Ethylcyclohexane
yield[a] [%]
1
2
3
4
5
6
Fe0
Ru0
>99
>99
>99
0
75
90
95
91
0
>99
FeRu (1:1) >99
FeRu (1:0.5) >99
FeRu (1:0.2) >99
FeRu (1:0.1) >99
1
0
0
0
[a] Determined by using GC with n-dodecane as an internal standard.
The catalytic properties of bimetallic FeRu nanoparticles in
the hydrogenation of styrene and 2-butanone were investigat-
ed to elucidate the influence of the presence of the two
metals on the hydrogenation reactions. FeRu nanoparticles
could selectively hydrogenate styrene to ethylbenzene and
catalyzed the hydrogenation of 2-butanone to 2-butanol with
good yields. Therefore, the present work is a case study on
fine-tuning the reactivity of nanocatalysts and adjusting their
selectivity for a given reaction. The next step will be to apply
this methodology to the reduction of more complex
molecules.
Table 3. 2-Butanone hydrogenation under 3 bar H2 at room temperature
during 24 h using 5 mol% of NPs as the catalyst and mesitylene as the
solvent: influence of the molar ratio of iron to ruthenium on the reaction
activity.
Entry
NPs
2-Butanone conv.[a] [%]
2-Butanol yield[a] [%]
1
2
3
4
5
6
Fe0
0
>99
>99
>99
78
0
>99
70
92
68
Ru0
FeRu (1:1)
FeRu (1:0.5)
FeRu (1:0.2)
FeRu (1:0.1)
67
44
Experimental Section
[a] Determined by using GC with n-dodecane as an internal standard.
Preparation of FeRu bimetallic NPs
The FeRu NPs were prepared in the same conditions as those used
for the synthesis of ultrasmall Fe0 NPs.[6] The method consisted of
the simultaneous decomposition of {Fe[N(Si(CH3)3)2]2}2 (1 mmol,
376.5 mg) and (h4-1,5-cyclooctadiene)(h6-1,3,5-cyclooctatriene)ru-
thenium(0) [Ru(COD)(COT)] in mesitylene (20 mL) under a H2 pres-
sure of 3 bar at 1508C overnight. The quantity of the ruthenium
complex was adjusted to the desired molar ratio of iron to rutheni-
um (1, 0.5, 0.2, and 0.1 equiv. of Ru corresponding to 1 mmol,
315.4 mg; 0.5 mmol, 157.7 mg; 0.2 mmol, 63.1 mg; and 0.1 mmol,
31.5 mg of Ru(COD)(COT)). After the reaction time, the solution
turned from dark green to black in each case. The solution was
The bimetallic FeRu NPs displayed basically the same reactiv-
ity as the Fe0 NPs, that is, hydrogenation of the vinyl group
but not of the arene moiety. Only in the case of high rutheni-
um loading, we observed some hydrogenation to ethylcyclo-
hexane. This is consistent with the fact that hydrogenation of
arenes occurs on a nanocrystal face, which, in the present
case, needed to be exclusively of ruthenium. Statistically, this
situation is not met for the present RuFe NPs.
ꢀ 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
[2] M. Stein, J. Wieland, P. Steurer, F. Tçlle, R. Mꢂlhaupt, B. Breit, Adv. Synth.
Catal. 2011, 353, 523–527.
then dried under vacuum to remove the solvent and to give
a black powder in all cases.
[3] J. F. Sonnenberg, N. Coombs, P. A. Dube, R. H. Morris, J. Am. Chem. Soc.
2012, 134, 5893–5899.
[4] a) R. B. Bedford, M. Betham, D. W. Bruce, S. A. Davis, R. M. Frost, M. Hird,
Chem. Commun. 2006, 1398–1400; b) P.-H. Phua, L. Lefort, J. A. F. Boo-
gers, M. Tristany, J. G. de Vries, Chem. Commun. 2009, 3747–3749; c) C.
Rangheard, C. de J. Fernandez, P.-H. Phua, J. Hoorn, L. Lefort, J. G. de V-
ries, Dalton Trans. 2010, 39, 8464–8471; d) A. Welther, M. Bauer, M.
Mayer, A. J. von Wangelin, ChemCatChem 2012, 4, 1088–1093.
[5] a) F. Dumestre, B. Chaudret, C. Amiens, P. Renaud, P. Fejes, Science 2004,
303, 821–823; b) L.-M. Lacroix, S. Lachaize, A. Falqui, M. Respaud, B.
Chaudret, J. Am. Chem. Soc. 2009, 131, 549–557; c) A. Meffre, S.
Lachaize, C. Gatel, M. Respaud, B. Chaudret, J. Mater. Chem. 2011, 21,
13464–13469.
Preparation of FeRu bimetallic NPs in the presence of
hexadecylamine
FeRu NPs with a molar ratio of iron to ruthenium equal to 1 were
synthesized in the presence of HDA. {Fe[N(Si(CH3)3)2]2}2 (0.5 mmol,
188.3 mg), [Ru(COD)(COT)] (0.5 mmol, 157.7 mg) and HDA
(1.5 mmol, 362.2 mg) were mixed together in mesitylene (10 mL)
under an inert atmosphere. The mixture was then pressurized to
three bars of dihydrogen and heated at 1508C overnight. At the
end of the reaction, a black precipitate and a green supernatant
were observed. After removing of the supernatant, the black pre-
cipitate was washed with toluene and then dried under vacuum to
give a black powder.
[6] V. Kelsen, B. Wendt, S. Werkmeister, K. Junge, M. Beller, B. Chaudret,
Chem. Commun. 2013, 49, 3416–3418.
[7] P. Lara, K. Philippot, B. Chaudret, ChemCatChem 2013, 5, 28–45.
[8] a) M. Guerrero, A. Roucoux, A. Denicourt-Nowicki, H. Bricout, E. Monflier,
V. Colliꢃre, K. Fajerwerg, K. Philippot, Catal. Today 2012, 183, 34–41;
b) M. H. G. Prechtl, M. Scariot, J. D. Scholten, G. Machado, S. R. Teixeira,
J. Dupont, Inorg. Chem. 2008, 47, 8995–9001; c) F. Schwab, M. Lucas, P.
Claus, Angew. Chem. Int. Ed. 2011, 50, 10453–10456; Angew. Chem.
2011, 123, 10637–10640; d) D. Gonzꢄlez-Gꢄlvez, P. Nolis, K. Philippot, B.
Chaudret, P. W. N. M. van Leeuwen, ACS Catal. 2012, 2, 317–321.
Characterization
FeRu bimetallic NPs were characterized by TEM. Conventional
images were obtained by using JEOL1011 and JEOL1400 micro-
scopes operated at 100 kV and 120 kV, respectively. HRTEM and
STEM images were obtained by using a JEOL2100F microscope op-
erated at 200 kV and equipped with a high-angle annular dark-field
detector and an EDX spectrometer. Crystallographic characteriza-
tions were conducted by WAXS performed at CEMES–CNRS. Sam-
ples were sealed in 1 mm diameter Lindemann glass capillaries.
Data collection was performed by using a dedicated two-axis dif-
fractometer with graphite-monochromated MoKa radiation (l=
0.7107 ꢁ). Magnetic studies were conducted on a SQUID (MPMS
Quantum Design5.5). IR spectra were recorded on a Perkin–Elmer
´
´
´
[9] D. Duraczynska, A. Drelinkiewicz, E. Bielanska, E. M. Serwicka, L. Lityn-
´
ska-Dobrzynska, Catal. Lett. 2011, 141, 83–94.
[10] D. Gonzalez-Galvez, P. Lara, O. Rivada-Wheelaghan, S. Conejero, B. Chau-
dret, K. Philippot, P. W. N. M. Van Leeuwen, Catal. Sci. Technol. 2013, 3,
99–105.
[11] a) G. L. Ott, T. Fleisch, W. N. Delgass, J. Catal. 1979, 60, 394–403; b) G. L.
Ott, T. Fleisch, W. N. Delgass, J. Catal. 1980, 65, 253–262; c) F. J. Berry, L.
Liwu, W. Chengyu, T. Renyuan, Z. Su, L. Dongbai, J. Chem. Soc. Faraday
Trans. 1 1985, 81, 2293–2305; d) M. C. Bahome, L. L. Jewell, K. Padaya-
chy, D. Hildebrandt, D. Glasser, A. K. Datye, N. J. Coville, Appl. Catal. A
2007, 328, 243–251.
GX2000 spectrometer in the wavelengths range of 4000–400 cmÀ1
.
[12] J.-Q. Du, Y. Zhang, T. Tian, S.-C. Yan, H.-T. Wang, Mater. Res. Bull. 2009,
44, 1347–1351.
[13] a) B. Bachiller-Baeza, A. Guerrero-Ruiz, P. Wang, I. Rodrꢅguez-Ramos, J.
Catal. 2001, 204, 450–459; b) J.-M. Andanson, S. Marx, A. Baiker, Catal.
Sci. Technol. 2012, 2, 1403–1409.
[14] a) K. Lꢄzꢄr, W. M. Reiff, W. Mçrke, L. Guczi, J. Catal. 1986, 100, 118–129;
b) K. R. Kannan, G. U. Kulkarni, C. N. R. Rao, Catal. Lett. 1992, 14, 149–
163.
[15] X. Wang, W. Yue, M. He, M. Liu, J. Zhang, Z. Liu, Chem. Mater. 2004, 16,
799–805.
[16] N. Popovska, K. Danova, I. Jipa, U. Zenneck, Powder Technol. 2011, 207,
17–25.
Catalytic hydrogenation reactions
The as-prepared solutions of NPs in mesitylene were used directly
for catalysis. The concentration of the catalyst solution was
5 mol% and the amount of the substrates used (styrene and 2-bu-
tanone) was 2 mmol. The reactions were conducted in a Fischer–
Porter vessel under H2 (3 bar) at RT during 24 h. At the end of the
reaction, the Fischer–Porter vessel was depressurized and the inter-
nal standard (n-dodecane) was added to the reaction mixture for
analysis by gas chromatography (GC Perkin–Elmer Clarus500).
[17] C. Pan, K. Pelzer, K. Philippot, B. Chaudret, F. Dassenoy, P. Lecante, M.-J.
Casanove, J. Am. Chem. Soc. 2001, 123, 7584–7593.
[18] L. M. Lacroix, S. Lachaize, A. Falqui, T. Blon, J. Carrey, M. Respaud, F. Du-
mestre, C. Amiens, O. Margeat, B. Chaudret, P. Lecante, E. Snoeck, J.
Appl. Phys. 2008, 103, 07D521.
Acknowledgements
[19] a) D. Ciuculescu, C. Amiens, M. Respaud, A. Falqui, P. Lecante, R. E. Ben-
field, L. Jiang, K. Fauth, B. Chaudret, Chem. Mater. 2007, 19, 4624–4626;
b) C. Bergounhou, C. Blandy, R. Choukroun, P. Lecante, C. Lorber, J. L.
Pellegatta, New J. Chem. 2007, 31, 218–223.
[20] P. Pertici, G. Vitulli, R. Lazzaroni, P. Salvadori, P. L. Barili, J. Chem. Soc.
Dalton Trans. 1982, 1019–1022.
The authors thank UPS-TEMSCAN for electron microscopy facili-
ties. The authors gratefully acknowledge financial support from
CNRS and the European Union for ERC Advanced Grant (NANO-
SONWINGS 2009-246763).
[21] N. Liakakos, B. Cormary, X.-J. Li, P. Lecante, M. Respaud, L. Maron, A.
Falqui, A. Genovese, L. Vendier, S. Koꢆnis, B. Chaudret, K. Soulantica, J.
Am. Chem. Soc. 2012, 134, 17922–17931.
[22] J. Garcꢅa-Antꢇn, M. R. Axet, S. Jansat, K. Philippot, B. Chaudret, T. Pery,
G. Buntkowsky, H.-H. Limbach, Angew. Chem. Int. Ed. 2008, 47, 2074–
2078; Angew. Chem. 2008, 120, 2104–2108.
[23] J. S. Bradley, E. W. Hill, C. Klein, B. Chaudret, A. Duteil, Chem. Mater.
1993, 5, 254–256.
[24] F. Novio, K. Philippot, B. Chaudret, Catal. Lett. 2010, 140, 1–7.
[25] D. Zitoun, C. Amiens, B. Chaudret, M.-C. Fromen, P. Lecante, M.-J. Casa-
nove, M. Respaud, J. Phys. Chem. B 2003, 107, 6997–7005.
Keywords: electron microscopy
nanoparticles · ruthenium
· iron · hydrogenation ·
[1] a) H. M. Torres Galvis, J. H. Bitter, C. B. Khare, M. Ruitenbeek, A. Lulian
Dugulan, K. P. de Jong, Science 2012, 335, 835–838; b) H. M. Torres Gal-
vis, J. H. Bitter, T. Davidian, M. Ruitenbeek, A. Iulian Dugulan, K. P.
de Jong, J. Am. Chem. Soc. 2012, 134, 16207–16215; c) A. Gual, C.
Godard, S. Castillon, D. Curulla-Ferre, C. Claver, Catal. Today 2012, 183,
154–171.
ꢀ 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
[26] A. Meffre, B. Mehdaoui, V. Kelsen, P. F. Fazzini, J. Carrey, S. Lachaize, M.
[27] a) M. Kitada, K. Yamamoto, K. Momata, J. Magn. Magn. Mater. 1996, 161,
397–403; b) M. Takahashi, H. Shoji, S. Kadowaki, D. D. Djayaprawira, Y.
Komori, H. Domon, Appl. Phys. Lett. 2000, 76, 3457–3459.
Respaud, B. Chaudret, Nano Lett. 2012, 12, 4722–4728.
Received: October 23, 2013
Revised: January 8, 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
V. Kelsen, A. Meffre, P.-F. Fazzini,
P. Lecante, B. Chaudret*
Singing a bimetallic tune: The selectivi-
ty of FeRu nanocatalysts in hydrogena-
tion reactions can be tuned by adjust-
ing the Ru content in bimetallic FeRu
ultrasmall nanoparticles.
&& – &&
How to Modulate Catalytic Properties
in Nanosystems: The Case of Iron–
Ruthenium Nanoparticles
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
8
&
ÞÞ
These are not the final page numbers!