Rhodium nanoparticles stabilized by ferrocenyl-phosphine ligands: Synthesis and catalytic styrene hydrogenation

Article abstract of DOI:10.1039/c9dt01006h

A series of ferrocenylphosphine-stabilized rhodium nanoparticles has been prepared in one pot from the organometallic [Rh(η³-C₃H₅)₃] precursor. This complex has been decomposed by hydrogen treatment (3 bar) in dichloromethane in the presence of five different ferrocene-based phosphine ligands. Very small rhodium nanoparticles in the size range of 1.1-1.7 nm have been obtained. These nanoparticles have shown activity in a model catalytic reaction, namely the hydrogenation of styrene. These results evidence that the metal surface is not blocked despite the steric bulk of the stabilizing ligands. Moreover, certain selectivity has been observed depending on the ligand employed. To the best of our knowledge, such a type of compound has not yet been used for stabilizing metal nanoparticles and our findings highlight the interest to do so.

Full text of DOI:10.1039/c9dt01006h

Dalton
Transactions
View Article Online
View Journal
PAPER
Rhodium nanoparticles stabilized by ferrocenyl-
phosphine ligands: synthesis and catalytic styrene
Cite this: DOI: 10.1039/c9dt01006h
hydrogenation
a,b
M. Ibrahim, a,b M. M. Wei,a,b E. Deydier,
*
a,b,c E. Manoury,
*
a,b R. Poli,
a,b
P. Lecante d and K. Philippot
*
A series of ferrocenylphosphine-stabilized rhodium nanoparticles has been prepared in one pot from the
organometallic [Rh(η3-C3H5)3] precursor. This complex has been decomposed by hydrogen treatment
(3 bar) in dichloromethane in the presence of five different ferrocene-based phosphine ligands. Very small
rhodium nanoparticles in the size range of 1.1–1.7 nm have been obtained. These nanoparticles have
shown activity in a model catalytic reaction, namely the hydrogenation of styrene. These results evidence
that the metal surface is not blocked despite the steric bulk of the stabilizing ligands. Moreover, certain
selectivity has been observed depending on the ligand employed. To the best of our knowledge, such a
type of compound has not yet been used for stabilizing metal nanoparticles and our findings highlight the
interest to do so.
Received 7th March 2019,
Accepted 15th April 2019
DOI: 10.1039/c9dt01006h
rsc.li/dalton
metal atoms on the NP surface can be compared to the ligand
interactions with the metal centers in homogeneous catalysts,
Introduction
Metal nanoparticles (NPs) are largely investigated given their which is of paramount importance for stability and properties
unique properties and potential applications in multiple such as catalytic activity. Thus, ligands also allow tuning the
domains such as biology, medicine, microelectronics, optics surface properties of metal NPs through electronic or steric
or catalysis to cite only a few.1–3 In catalysis, the relevance of effects. Moreover, ligand-stabilized metal nanoparticles can be
metal NPs has been known for a long time due to their pres- applied to catalysis either as stable colloidal suspensions or
ence in heterogeneous catalysts.4,5 Metal NPs are attractive after deposition on the surface or confinement in the pores of
materials because of their high surface/volume ratios.6,7 With solid supports.14 The challenge is to find ligands able to stabil-
the recent advances in nanochemistry and particularly in the ize well-defined nanoparticles while controlling accessibility at
design of better defined metal NPs, nanocatalysis is now recog- the metal surface and reactivity.15 Compared to the investi-
nized as a full and rich part of catalysis.8–10
gation of facet dependency, the ligand influence on the cata-
Over the past 25 years ligand-stabilized metal nanoparticles lytic activity has been less intensively studied but recent
have attracted growing attention and numerous investigations results illustrate well the interest to do so.16–19 One aim lies in
are devoted to exploring the effect of the stabilizing agents on the design of better performing catalysts in terms of reactivity
the characteristics of NPs such as size, dispersion, morphology and selectivity in order to develop more efficient and eco-com-
and surface properties and also catalytic performance.11 Very patible chemical production.20
recent papers reported on the precise role of ligands and their
One efficient method to prepare well-controlled metal nano-
binding with the metal surface in NP size control.12,13 It is particles is the organometallic approach. This approach has
important to note that the interaction of the ligands with the proven powerful to provide controlled nanoparticles with
reproducible physical or chemical properties.21 Metal com-
plexes (organometallic or metal organic complexes) are decom-
aCNRS, LCC (Laboratoire de Chimie de Coordination), 205, route de Narbonne,
BP 44099, F-31077 Toulouse Cedex 4, France. E-mail: eric.deydier@iut-tlse3.fr,
eric.manoury@lcc-toulouse.fr, karine.philippot@lcc-toulouse.fr
bUniversité de Toulouse, UPS, INPT, F-31077 Toulouse Cedex 4, France
cIUT A Paul Sabatier, Département de Chimie, Avenue Georges Pompidou, CS 20258,
F-81104 Castres Cedex, France
posed under mild conditions to release metal atoms either in
the presence of polymers (steric barriers) or ligands (metal
surface coordination) as stabilizing agents. The choice of the
metal source is important in terms of the resulting metal
surface purity. Olefinic metal precursors are privileged
because their treatment under hydrogen liberates alkanes as
the only reaction byproducts that are inert towards the metal
dCNRS, CEMES (Centre d’Elaboration de Matériaux et d’Etudes Structurales),
29 rue J. Marvig, F-31055 Toulouse, France
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
surface. Therefore, metal NPs with clean surfaces capped with
the desired ligands are obtained. Numerous ruthenium nano-
Results and discussion
particles have been prepared this way, using [Ru(η6-C8H10) The first objective of this work was to take benefit of previous
((η-C8H12)] as the ruthenium source and a large panel of results in the organometallic synthesis of phosphine-stabilized
ligands as stabilizers. The ligands are either simple or sophis- rhodium nanoparticles, achieved from the [Rh(η3-C3H5)3]
ticated, and mainly contain amines, thiols, carbenes or phos- complex as the metal source and simple mono- or di-substi-
phines which were shown to influence the surface and catalytic tuted phosphines as stabilizers,26,28 in order to prepare a new
properties of the particles.22
family of Rh NPs by the same approach but using ferrocenyl
Rhodium particles can also be synthesized by the organo- phosphines. With ferrocene-based-phosphine-capped Rh NPs
metallic method, starting from [Rh(acac)(1,5-C8H12)],23 in hand, a second objective was to test their catalytic activity in
[Rh(µCl)(1,5-C8H12)]2,24 or [Rh(η3-C3H5)3].25 Phosphine a benchmark hydrogenation process, the hydrogenation of
ligands were shown to affect the reactivity of Rh NPs in styrene, in order to evaluate the influence of the capping
hydrogenation26,27 and hydroformylation catalysis.28 Apart ligands at the rhodium surface.
from usual ligands to stabilize metal NPs, one can envisage
All the ligands employed in this study (Scheme 1) are ferro-
to use metal-containing ligands. This can be of interest in cenyl diphenyl phosphines bearing different substituted
order to introduce novel properties such as redox surface methyl groups at the 2 position. L1 and L2 are both P,N biden-
modulation and also more functionality.29,30 One appli- date ligands.49 In L1 the amine is tertiary with an extra methyl
cation could be tandem catalysis for instance. One strategy substituent on the nitrogen atom, whereas L2 contains a sec-
to achieve this objective is to consider metal-containing ondary amine. Ligands L3 50 and L4 45 are respectively P,S and
ligands with functional groups able to interact with the P,O bidendate ligands. L5 is the simplest ligand of the series,
metal surface such as phosphine groups.31 To this end, we the only monodentate one, having a methyl group at the ferro-
chose to test ferrocene-based phosphine ligands because cenyl 2 position.
these compounds have several attractive properties: they are
The synthesis of the NPs was performed in one pot using
stable, they may feature planar chirality with potential inter- dichloromethane (CH2Cl2) as the solvent in order to ensure
est in materials chemistry,32 especially in asymmetric the solubility of both the [Rh(η3-C3H5)3] complex and the ferro-
catalysis,33–35 and they possess a redox-active ferrocene cenyl phosphine ligands. Thus, a Fischer–Porter reactor filled
moiety which may be exploited for switchable catalysis.36–38 with a CH2Cl2 solution containing both [Rh(η3-C3H5)3] and a
For instance, 1,2-disubstituted ferrocene phosphines play a given ferrocenyl ligand was pressurized with 3 bar of dihydro-
major role in several asymmetric catalysis applications. gen and then heated at 40 °C, under vigorous stirring
Various examples including bidendate (P,P),39 (P,N),40 (P, (Scheme 2). A change of color from a yellow homogeneous
S),41–43 (P,O)44–47 and (P, carbene)48 compounds have been solution to a black colloidal suspension was observed over a
reported, illustrating the great modularity of this family of few hours of stirring, but the reaction was kept running up to
ligands. To the best of our knowledge there is no previous 16 h to ensure complete decomposition of the precursor and
example of the use of such ligands for the stabilization of the formation of a homogeneous population of particles. For
metal nanoparticles.
the four ligands L1, L2, L3 and L4, a [ligand]/[Rh] ratio of
In the present work we have focused on a series of racemic 0.2 molar equivalents was used. These ligands, being bidentate,
ferrocenylphosphine ligands as stabilizing agents for rhodium offer an extra coordination through the second donor group,
nanoparticles obtained by
[Rh(η3-C3H5)3]. As will be described hereafter, a series of small which is monodentate, a molar [ligand]/[Rh] ratio of 0.4 was
rhodium nanoparticles in size range between 1.1 and applied for comparison purposes. These [ligand]/[Rh] ratios
1.7 nm have been obtained and these were shown to be active for appeared to be sufficient to yield stable colloidal suspensions
a
one-pot synthesis from for an effective [donor]/[metal] ratio of 0.4. For ligand L5,
a
a model catalytic reaction, namely the hydrogenation of styrene.
for several weeks under an argon atmosphere. This is particu-
Scheme 1 Ferrocenyl phosphine ligands used for the synthesis of RhNPs.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Scheme 2 On-pot synthesis of ferrocenylphosphine-capped Rh NPs.
larly important to note when considering that the L5 ligand bottom). Such a phenomenon was previously observed for
has only one –PPh2 coordinating group. Indeed, when PPh3 PPh3-stabilized RhNPs and explained as the result of a strong
was used as a stabilizer, a minimum [ligand]/[Rh] ratio of 1 coordinating effect of the phosphine ligand.26
was necessary to reproducibly yield stable colloidal suspen-
For the Rh@L2 powder, WAXS (Wide-Angle X-ray Scattering)
sions of small Rh NPs (ca. 1.3 nm).26 For comparison, the data revealed well-crystallized particles and confirmed their fcc
bidentate Ph2P-(CH2)4-PPh2 ligand led to stable Rh NPs at a crystalline structure as for the bulk metal. From the RDF, the
[ligand]/[Rh] ratio of 0.3 molar equivalents, which amounts to coherence length reached 3.5–4.0 nm, much more than the
a [donor]/[Rh] ratio of 0.6. Consequently, the bidendate phos- average diameter of ca. 1.7 0.5 nm determined from the TEM
phino-ferrocenyl ligands used here behave as efficient stabil- images (Fig. 3 top). However, the envelope of the RDF, related
izers at lower [ligand]/[Rh] ratios compared to the previously to the size and shape of the objects, is not consistent with
studied similar mono- and diphosphane ligands. The purifi- single-size spherical NPs but suggests a mixture with a
cation of the particles was performed by precipitation after majority of relatively small NPs (ca. 2.0 nm) and a small pro-
addition of cold pentane to the concentrated colloidal suspen- portion of much larger ones (up to 4.0 nm). This may be
sions. The generated black precipitates were isolated by related to a partial coalescence of the small NPs occurring in
decanting off the solvents, washing and drying under vacuum, the agglomerated parts of the sample. For the Rh@L5 sample
leading to dried nanoparticle powders.
(Fig. 3 bottom), the RDF (Radial Distribution Function)
TEM images recorded from samples prepared by deposition indicated a much smaller coherence length of ca. 1.3 nm,
of a few drops of the crude colloidal suspensions onto a which is in good agreement with the size measured by TEM
carbon-covered copper grid showed well-dispersed NPs for all (1.2 0.6 nm). These particles, however, appeared less crystal-
the systems studied. They present spherical shapes and quite line and displayed a very different pattern in reciprocal space
narrow size distribution, in a size range between 1 and 2 nm from that of the fcc network, but in good agreement with the
(Fig. 1), with mean diameters as follows: 1.5
0.6 nm for manganese-β structure (ICSD 41775, PDF 33-887) already
Rh@L1 NPs, 1.7 0.5 nm for Rh@L2 NPs, 1.3 0.7 nm for observed for very small Rh NPs.26,51 These WAXS results
Rh@L3 NPs, 1.1 0.5 nm for Rh@L4 NPs, and 1.2 0.6 nm suggest that ligand L5 has a stronger influence than ligand L2
for Rh@L5 NPs.
on the crystallization state of the particles.
The L2- and L5-stabilized nanoparticles were investigated in
Purified samples of L2- and L5-stabilized Rh NPs were also
greater detail in order to learn more about the effect of a analyzed by Fourier transform infrared (FT-IR) spectroscopy
second donor function. Thus, Rh@L2 NPs and Rh@L5 NPs before and after exposure to a CO atmosphere (1 bar; 48 h) in
were also characterized by high resolution electron microscopy the solid state. This investigation provided information on the
(HREM). In addition to isolated nanoparticles, the HREM coordination ability of CO molecules on the rhodium surface.
images of the L2-stabilized Rh NPs also revealed the presence Similar analyses have been previously reported for Ru52 and
of large agglomerates (Fig. 2 top), which appeared to be com- Rh53 NPs. The obtained spectra (KBr pellets) are given in Fig. 4
posed of small and spherical particles gathered together. The and 5 for the L2- and L5-stabilized Rh NPs, respectively. In the
isolated particles displayed crystalline regions with interplanar case of Rh@L5 NPs, the FT-IR spectrum recorded after 48 h of
distances of 0.22 nm, in good agreement with the fcc structure reaction with CO (Fig. 4b) revealed the presence of complex
(face-centered cubic) previously observed for similarly pre- bands in the region between 2050 cm−1 and 1550 cm−1 in
pared Rh nanoparticles and for bulk rhodium.26 In the case of addition to those already present before CO exposure (Fig. 4a).
L5-stabilized Rh NPs no agglomerates were observed. These The large and intense band at ca. 1990 cm−1 can be attributed
latter NPs were poorly crystallized with only a few of them dis- to the stretching vibrations of CO ligands adopting a terminal
playing crystalline planes. Moreover, the measured Rh–Rh coordination mode. The higher frequency shoulder at ca.
distances did not fit with that of fcc bulk rhodium (Fig. 2 2020 cm−1 is attributable to the CO ligands bonded in a
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 1 (a) TEM images and corresponding size distributions of Rh NPs stabilized with L1, (b) L2, (c) L3, (d) L4, and (e) L5.
geminal or multicarbonyl mode. Finally the band at 1800 cm−1
In summary, the FT-IR data suggest that the L2 ligand has a
can be attributed to the standard bridging coordination mode. stronger steric effect on the RhNP surface covering than the L5
The high intensity of the band at 1990 cm−1 indicates that one. This can be attributed to the different coordination
terminal CO molecules are abundant. It is generally accepted modes at the NP surface which can result from the different
that this type of coordination occurs on the edges and corners bonding modes (bidentate vs. monodentate). Consequently, a
of the particle surface while the bridging coordination mode greater fraction of Rh free sites are probably available in the
occurs on the extended faces of the particles.54,55 Given that case of Rh@L5 NPs than in the case of Rh@L2 NPs.
ligand L5 allows the CO coordination in all possible coordi-
Rhodium nanoparticles stabilized with ligands L2 and L5
nation modes and that the quantity of terminal CO is high, were then evaluated in the catalytic hydrogenation of styrene,
one can conclude that the metal surface remains relatively in order to assess the ligand effect in this simple model reac-
accessible with this ligand.
tion. Hydrogenation reactions were performed at room temp-
Conversely, exposure of Rh@L2 to CO yields only one new erature in an autoclave under a 30-bar H2 pressure, using iso-
and very low intensity band at 1924 cm−1 (Fig. 5), suggesting a propanol as a solvent. The activity was determined by GC ana-
much more limited access for the CO molecules to the metal lysis after 24 h of reaction. The obtained results are summar-
surface in this case. This effect may result from a tighter L2 ized in Table 1.
coordination, probably induced by the bidentate character of
In all cases, the styrene substrate reached full conversion.
L2 and possibly also by a larger covering of the particle surface In the absence of a base and with a 0.7% or 0.2% molar
by L2 relative to L5.
percent of L2-stabilized Rh NPs (entries 1 and 2, respectively),
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Fig. 2 HREM images of L2-stabilized Rh NPs (top) and L5-stabilized Rh NPs (bottom) at different magnifications.
Fig. 3 WAXS measurements (solid line) for L2-stabilized Rh NPs (top) and L5-stabilized Rh NPs (bottom) in reciprocal space (left, with reference dia-
grams for Rh in the fcc and manganese-β structures, dashed line) and real space (right).
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 4 (a) FT-IR spectrum of L5-stabilized Rh NPs and (b) FT-IR spectrum of L5-stabilized Rh NPs after 48 h under 1 bar of CO.
Fig. 5 (a) FT-IR spectrum of purified L2-stabilized Rh NPs and (b) FT-IR spectrum of L2-stabilized Rh NPs after 48 h under 1 bar of CO.
ethylbenzene (P1) was the major product but formation of the ethyl benzene reduction was carried out under similar con-
complete hydrogenation product (ethylcyclohexane, P2) was ditions (entry 5), yielding 6% of P2 (similar to the results
also observed. The ethylcyclohexane yield increased from 2 to obtained in entry 1). These reactions were repeated in the pres-
8% by increasing the catalyst loading from 0.2 to 0.7 mol%. ence of excess NaOMe (7–8 equivalents relative to Rh, entries
The putative product of selective arene hydrogenation, vinylcy- 3, 4 and 6). While the styrene consumption remained com-
clohexane, was never observed. These results suggest a sequen- plete, the formation of the total reduction product P2 was
tial reaction with initial reduction of the vinyl function to no longer observed. Thus, the base prevents the arene
ethylbenzene, followed by a slower arene ring reduction to hydrogenation favoring the selective formation of
ethylcyclohexane. To confirm this hypothesis, an independent ethylbenzene (P1).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Table 1 Styrene and ethyl-benzene hydrogenation using L2- and L5-stabilized Rh NPs under 30 bar of H2 at R.T
Entry
Catalyst
Substrate
nSubstrate
S/Ca
Catalystb (%)
nbase/nRh
Styrene (%)
P1 (%)
P2 (%)
TON
1
2
3
4
5
6
L2 Rh NPs
Styrene
Styrene
Styrene
Styrene
Ethyl-benzene
Ethyl-benzene
0.0005
0.0015
0.0005
0.0015
0.0005
0.0005
146
439
146
439
144
144
0.7
0.2
0.7
0.2
0.7
0.7
0
0
7
8
0
7
0
0
0
0
—
–
92
98
>99
>99
94
8
2
<1
<1
6
146
439
146
439
144
144
>99
<1
7
8
9
10
11
12
L5 Rh NPs
Styrene
Styrene
Styrene
Styrene
Ethyl-benzene
Ethyl-benzene
0.0005
0.0015
0.0005
0.0015
0.0005
0.0005
252
756
252
756
247
247
0.4
0.1
0.4
0.1
0.4
0.4
0
0
14
16
0
0
0
0
0
—
—
90
97
98
>99
81
99
10
3
2
<1
19
1
252
756
252
756
247
247
15
a Substrate to rhodium molar ratio. b mol%.
Similar results were observed when using the L5-stabilized those stabilized by the bidentate ligand L2 in arene hydrogen-
Rh NPs. Reactions performed with 0.4 and 0.1 molar % of Rh ation of ethylbenzene, in accordance with the lower efficiency
(entries 7 and 8) gave 10% and 3% of the fully hydrogenated of L5 to cover the nanoparticle surface. Deeper investigations
product (P2), and 90% and 97% of ethylbenzene (P1), respect- will be necessary for a better understanding of the reaction
ively. The P2 yield increased at higher catalyst loading, as mechanism and product selectivity.
observed with the L2-stabilized Rh NPs. When NaOMe was
added (entries 9 and 10) the P2 yield dramatically decreased.
A comparison of entries 5 and 11 shows that Rh@L5 exhi-
bits a greater activity than Rh@L2 for the arene ring hydrogen-
ation, which can be correlated with the FT-IR suggestion of a
Experimental part
General and materials
Chemicals were purchased as follows: dichloromethane
(CH2Cl2) and pentane from SDS; RhCl3·xH2O from Johnson
Matthey; [CH2vCHCH2MgCl], P(NMe2)3 and BH3·THF (1 M
solution in THF) from Sigma Aldrich; and H2 from Air
Liquide. Solvents were purified before use by filtration on an
adequate column of a MBraun purification apparatus and
handled under an argon atmosphere. All reagents were used
without purification. Reagents and solvents were degassed
before use according to a freeze–pump–thaw process. All oper-
ations concerning the synthesis of the Rh precursor and Rh
nanoparticles or the preparation of samples for characteriz-
ation were carried out using standard Schlenk tubes and
Fischer–Porter bottle techniques or in a glove-box (MBraun)
under an argon atmosphere.
more accessible surface for the former catalyst. Deeper
mechanistic investigations will be necessary to understand
how the presence of the base prevents arene reduction.
Conclusion
Ferrocene-based ligands, namely ferrocenyldiphenylpho-
sphines, have been reported here for the first time as stabil-
izers for the synthesis of metal nanoparticles. The Rh nano-
particles formed with these ligands have a well-controlled size
and are quite small (1–2 nm range). The FT-IR study after
exposure to a CO atmosphere has shown that the surface cover-
ing effect of the bidentate L2 ligand is greater than that of the
monodentate L5 ligand. These NPs were found catalytically
active in the hydrogenation of a model substrate, namely
Characterization techniques
styrene, under very mild reaction conditions in spite of the Transmission electron microscopy observations at low (TEM)
surface coverage by the ligands. In addition, we have discov- and high resolution (HREM) were made at the “Centre de
ered that NaOMe greatly inhibits the hydrogenation of the microcaractérisation Raymond Castaing” in Toulouse
phenyl ring. This inhibition is, to the best of our knowledge, (UMS-CNRS 3623). TEM grids were prepared by drop-casting of
unprecedented. No significant differences in selectivity were the crude colloidal suspension in CH2Cl2 onto a holey carbon-
observed using mono- and bidentate ligands, but the NPs coated copper grid under an argon atmosphere inside a glove-
stabilized by the monodentate ligand L5 are more active than box. TEM images were obtained using either a JEOL 1100 elec-
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
tron microscope operating at 100 kV with a point resolution of
Synthesis of phosphino-ferrocenyl ligands. Ligands L1 and
4.5 Å or a JEOL JEM 1400 operating at 120 kV with a point L2,49 L3 50 and L4 45 were prepared according to published
resolution of 4.5 Å. HTEM observations were made with a methods. Racemic (2-(diphenylthiophosphino)ferrocenyl)
JEOL JEM 2010 electron microscope working at 200 kV with a methanol (1) was synthesized by optimizing a literature pro-
point resolution of 2.3 Å and in EDS (Energy Dispersive X-ray cedure from N,N-dimethylaminomethylferrocene.57,58
Spectroscopy) mode. Statistical size distributions were deter- Synthesis of (2-(diphenylthiophosphino)ferrocenyl)methane
mined by analyzing TEM images with the Image J tool soft- 2. (2-(Diphenylthiophosphino)ferrocenyl)methane 2 was syn-
ware, counting mean diameters of >200 non-touching particles thesized from alcohol
and assuming a spherical form. Size distributions are quoted procedure:59
as the mean diameter the double of the standard deviation
(σ) which corresponds to a 95% confidence level. Fourier Fast
Transform treatments were carried out with the Digital
Micrograph Software (Version 1.80.70).
1 by adaptation of a published
Wide-angle X-ray scattering (WAXS) measurements were per-
formed at the CEMES-CNRS in Toulouse. Samples in the solid
state were sealed in 1.5 mm diameter Lindemann glass capil-
laries. The samples were irradiated with graphite monochro-
matized molybdenum Kα (0.071069 nm) radiation and the
X-ray scattering intensity measurements were performed using
a dedicated two-axis diffractometer. Radial distribution func-
tions (RDF) were obtained after Fourier transformation of the
corrected and reduced data.
Infrared spectroscopy. Fourier Transform Infrared (FT-IR)
spectra of L2- and L5-stabilized RhNPs were recorded in the
range 4000–400 cm−1 under an argon atmosphere using a
PerkinElmer GX2000 spectrometer installed inside a glovebox.
First, the spectra of the purified NPs (KBr pellets prepared in
the glove box) were recorded. Second, powders of purified
nanoparticles were exposed to a CO atmosphere (1 bar; 48 h)
inside a Fischer–Porter reactor. After evacuation of excess CO,
novel FT-IR spectra were recorded (KBr pellets).
ICP OES and elemental analyses. ICP OES analyses were per-
formed at the “Service d’analyse élémentaire” of LCC-CNRS in
Toulouse using a Thermo Scientific iCAP 6300 DUO spectro-
meter with a 3-channel, 12-roller pump, and a 27.12 MHz solid
state RF plasma generator. Mineralization of the NPs was per-
formed in a refluxing aqua regia solution.
In a Schlenk tube, alcohol 1 (500 mg, 1.16 mmol) was dis-
solved in 6 mL of anhydrous THF under argon. A BH3·THF 1 M
solution in THF (5 mL, 5 mmol) was then added with a
syringe. The reaction mixture was stirred at THF reflux for
70 h. After cooling to r.t., the Schlenk tube was opened to air
and water was slowly added (warning: an abundant gas evol-
ution occurred), leading to the formation of an orange precipi-
tate. After THF evaporation on the vacuum line, 10 mL of di-
chloromethane was added for full dissolution of the orange
precipitate. The remaining aqueous solution was then
extracted two times with 5 mL of dichloromethane. The com-
bined organic solutions were dried on anhydrous sodium
sulfate and purified, after evaporation, by flash chromato-
graphy on silicagel using pentane/dichloromethane (50/50, v/v)
as the eluent to yield, after solvent evaporation, 470 mg (98%
yield) of compound 2 as a yellow solid.
31P NMR (162 MHz, CDCl3), δ: 42.3. 1H NMR (400 MHz,
CDCl3), δ: 2.17 (s, 3H, CH3), 3.67 (m, 1H, subst Cp), 4.22 (m,
1H, subst Cp), 4.34 (s, 5H, Cp), 4.43 (m, 1H, subst Cp),
7.35–7.55 (m, 6H, Ar), 7.63 (m, 2H, Ar), 7.81 (m, 2H, Ar). 13C
{1H} NMR (100 MHz, CDCl3), δ: 14.3 (CH3), 68.2 (d, JC–P = 10.5
Hz, subst Cp), 70.7 (s, Cp), 74.1 (d, JC–P = 13.2 Hz, subst Cp),
74.34 (d, JC–P = 96.1 Hz, quat Cp), 74.36 (d, JC–P = 9.9 Hz, subst
Cp), 88.7 (d, JC–P = 12.4 Hz, quat Cp), 128.0 (d, JC–P = 12.4 Hz,
Ar), 128.3 (d, JC–P = 12.4 Hz, Ar), 131.15 (d, JC–P = 2.6 Hz, Ar),
131.18 (d, JC–P = 2.6 Hz, Ar), 131.9 (d, JC–P = 10.9 Hz, Ar), 132.0
(d, JC–P = 10.5 Hz, Ar), 133.7 (d, JC–P = 86.0 Hz, quat Ar), 134.8
(d, JC–P = 86.4 Hz, quat Ar). HRMS (ESI + ) m/z: 417.0522
(M + H, 100%; calcd for C23H22FePS: 417.0529).
Synthesis methods
Synthesis of the [Rh(η3-C3H5)3] complex. [Rh(η3-C3H5)3] was
prepared according to a slightly improved method from the lit-
erature.56 A THF solution (140 mL) of RhCl3·xH2O (1.5 g;
7.2 mmol) was reacted with allylmagnesium chloride (36 mL
of a THF solution at 2 mol L−1; 71.6 mmol) under an inert
atmosphere with vigorous stirring at −10 °C for 1 h and then
at 10 °C for 16 h. During this time, the solution color changed
from red to yellow. Solvent removal under reduced pressure
led to a solid, which was dissolved in pentane leading to a
yellow solution. After filtration through Celite, pentane was
evaporated under reduced pressure and the resulting yellow
solid was purified by sublimation at 40 °C. Yellow crystals
(660 mg, 41% yield) of [Rh(η3-C3H5)3] were collected and
Synthesis of (2-(diphenylphosphino)ferrocenyl)methane L5.
stored under an argon atmosphere at low temperature. 1H (2-(Diphenylphosphino)ferrocenyl)methane L5 was syn-
NMR and elemental data were found identical to those from thesized from compound 2 by deprotection of the phosphino
the literature.
group following a published procedure:60 in a Schlenk tube
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
under argon, compound 2 (470 mg, 1.13 mmol) was dissolved
in 40 mL of toluene. The reaction mixture was stirred at
Acknowledgements
toluene reflux overnight and purified by flash chromatography Financial support by the Centre National de la Recherche
on silicagel under argon using pentane/dichloromethane (CNRS) and the University Paul Sabatier – Toulouse is grate-
(50/50, v/v) as the eluent to yield, after solvent evaporation, fully acknowledged. M. Ibrahim thanks the University Paul
421 mg (97% yield) of compound 3 as a yellow solid.
Sabatier – Toulouse for a MESR PhD grant. M.-M. Wei thanks
31P NMR (162 MHz, CDCl3), δ: −21.7. 1H NMR (400 MHz, the CNRS and the Midi-Pyrénées Region for a PhD grant.
CDCl3), δ: 2.08 (s, 3H, CH3), 3.62 (m, 1H, subst Cp), 4.07 (s,
5H, Cp), 4.20 (m, 1H, subst Cp), 4.38 (m, 1H, subst Cp),
7.2–7.15 (m, 2H, Ar), 7.3–7.25 (m, 3H, Ar), 7.35–7.45 (m, 3H,
References
Ar), 7.55–7.60 (m, 2H, Ar). 13C{1H} NMR (100 MHz, CDCl3), δ:
14.0 (d, JC–P = 10.0 Hz, CH3), 68.3 (s, subst Cp), 69.7 (s, Cp),
70.4 (d, JC–P = 3.6 Hz, subst Cp), 72.3 (d, JC–P = 3.8 Hz, subst
Cp), 75.7 (d, JC–P = 4.3 Hz, quat Cp), 89.4 (d, JC–P = 25.0 Hz,
quat Cp), 127.8 (s, Ar), 128.14 (d, JC–P = 7.7 Hz, Ar), 128.17 (d,
JC–P = 5.9 Hz, Ar), 129.0 (s, Ar), 132.2 (d, JC–P = 18.0 Hz, Ar),
134.9 (d, JC–P = 20.7 Hz, Ar), 137.6 (d, JC–P = 8.8 Hz, quat Ar),
139.7 (d, JC–P = 10.9 Hz, quat Ar).
Synthesis of ferrocenylphosphine-capped Rh NPs. In a
general procedure, 150 mg of [Rh(η3-C3H5)3] (0.66 mmol) were
introduced under an argon atmosphere in a Fischer–Porter
reactor. After 0.5 h under vacuum at r.t. and cooling to 193 K, a
CH2Cl2 solution (150 mL) containing 0.2 molar equivalents of the
chosen ligand (L1, L2, L3 or L4) or 0.4 molar equivalents of L5
was added. The reaction medium was then allowed to warm up to
r.t. before adding dihydrogen (3 bar) and then introduced in an
oil bath preheated at 313 K for 16 h. After cooling to r.t., the
remaining dihydrogen was vented and the volume of the solution
was reduced to approximately 10 mL under reduced pressure
1 Physics and Chemistry of Metal Cluster Compounds, ed.
L. J. de Jongh, Kluwer Academic Publishers, Dordrecht,
1994.
2 Introduction to Nanotechnology, in Nanoscale Materials in
Chemistry, ed. K. J. Klabunde, John Wiley & Sons, Inc.,
2001.
3 Nanoparticles: From Theory to Application, ed. G. Schmid,
Wiley-VCH, Weinheim, 2nd edn, 2012.
4 J. E. Mondloch, E. Bayram and R. G. Finke, J. Mol. Catal. A:
Chem., 2012, 355, 1–38.
5 F. Schüth, Angew. Chem., Int. Ed., 2014, 53, 8599–8604.
6 Concepts in Nanocatalysis, ed. K. Philippot and P. Serp,
Wiley-VCH, Weinheim, 2013.
7 A. Roucoux, J. Schulz and H. Patin, Chem. Rev., 2002,
10(102), 3757–3778.
8 Nanocatalysis, ed. U. Heiz and U. Landman, Springer, 2007.
9 Nanoparticles in catalysis, ed. D. Astruc, Wiley-VCH,
Weinheim, 2008.
before its transfer into a solution of deoxygenated cold pentane 10 Metal Nanoparticles for Catalysis: Advances and Applications,
(100 mL), generating the product as a black precipitate. After fil- ed. T. Tao, RSC, 2014.
tration and pentane washing processes the RhNPs were collected 11 M. J. L. Tschan, O. Diebolt and P. W. N. M. van Leeuwen,
in the form of a fine powder. ICP OES analysis: for L2-stabilized
RhNPs: Rh: 29.3 wt%; for L5-stabilized RhNPs: Rh: 22.7 wt%.
Top. Catal., 2014, 57, 1054–1065.
12 S. Mozaffari, W. Li, C. Thompson, S. A. Ivanov, S. Seifert,
B. Lee, L. Kovarik and A. M. Karim, Nanoscale, 2017, 36(9),
13772–13785.
13 J. E. Mondloch, S. Ozkar and R. G. Finke, ACS Omega, 2018,
11(3), 14538–14550.
14 N. J. S. Costa and L. M. Rossi, Nanoscale, 2012, 4, 5826–
5834.
15 Y. Yuan, N. Yan and P. J. Dyson, ACS Catal., 2012, 2, 1057–
1069.
16 L. Yin and M. A. El-Sayed, J. Phys. Chem. B, 2001, 37, 8938–
8943.
17 R. R. Dykeman, N. Yan, R. Scopelliti and P. J. Dyson, Inorg.
Chem., 2011, 50(3), 717–719.
18 D. Gonzalez-Galvez, P. Nolis, K. Philippot, B. Chaudret and
P. van Leeuwen, ACS Catal., 2012, 2(3), 317–321.
19 S. A. Stratton, K. L. Luska and A. Moores, Catal. Today,
2012, 183(1), 96–100.
20 H.-J. Freund and G. A. Somorjai, Catal. Lett., 2015, 145, 1–
2.
Catalysis studies
In a glove-box, a Schlenk tube was charged with 12 mg of
rhodium NPs. Then, 15 mL of dry degassed isopropanol and
500 µL (2.2 mmol) of dodecane (internal reference) were added
to generate a stock of the colloidal suspension of the catalyst.
A portion of the catalyst suspension (1.5 mL) was transferred
into a 5 mL glass vial (pre-charged with the desired amount of
the styrene or ethylbenzene substrate and NaOMe), which was
then placed into a stainless steel autoclave. The reaction vessel
was pressurized with 30 bar of H2 and stirred with a magnetic
bar for 24 h at r.t. The reaction was stopped by the release of
the H2 pressure and quenched with 2 mL of CH2Cl2, followed
by filtration through silica gel. The product was finally ana-
lyzed by GC (Shimadzu 2014) using a SLB™ 5 ms Capillary
Column.
21 C. Amiens, D. Ciuculescu-Pradines and K. Philippot, Coord.
Chem. Rev., 2016, 308, 409–432.
22 K. Philippot, P. Lignier and B. Chaudret, Organometallic
Ruthenium Nanoparticles and Catalysis, in Ruthenium in
Conflicts of interest
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
catalysis 2014, ed. C. Bruneau and P. H. Dixneuf, Wiley 42 H. Pellissier, Tetrahedron, 2007, (63), 1297–1330.
VCH, Weinheim, Top. Organomet. Chem., 2014, vol. 48, 43 R. Malacea and E. Manoury, in “Phosphorus Ligands in
pp. 319–370.
Asymmetric Catalysis”, ed. A. Börner, Wiley-VCH,
Weinheim, Germany, 2008, pp. 749–784.
23 E. Ramírez-Meneses, Synthèse et caractérisation de nanopar-
ticules métalliques à base de rodhium, platine et palladium, 44 N. Mateus, L. Routaboul, J.-C. Daran and E. Manoury,
stabilisés par des ligands, University of Toulouse, 2004. J. Organomet. Chem., 2006, (691), 2297–2310.
24 S. Mavila, C. E. Diesendruck, S. Linde, L. Amir, R. Shikler 45 C. Audin, J.-C. Daran, E. Deydier, E. Manoury and R. Poli,
and N. Gabriel Lemcoff, Angew. Chem., Int. Ed., 2013, 52,
5767–5770.
25 E. Ramírez-Meneses, K. Philippot, M. A. Domínguez-
Crespo, M. Ibrahim, I. Betancourt, A. M. Torres Huerta,
C. R. Chim., 2010, (13), 890–899.
46 S. Bayda, A. Cassen, J.-C. Daran, C. Audin, R. Poli,
E. Manoury and E. Deydier, J. Organomet. Chem., 2014,
(772–773), 258–264.
A. Ezeta-Mejia and L. Palacios-Romero, J. Mater. Sci., 2018, 47 A. Karpus, O. Yesypenko, V. Boiko, R. Poli, J.-C. Daran,
53(12), 8933–8950.
26 M. Ibrahim, M. A. S. Garcia, L. Vono, M. Guerrero,
Z. Voitenko, V. Kalchenko and E. Manoury, Eur. J. Org.
Chem., 2016, (20), 3386–3394.
P. Lecante, L. Rossi and K. Philippot, Dalton Trans., 2016, 48 K. Yoshida and R. Yasue, Chem. – Eur. J., 2018, (24),
17782–17791. 18575–18586.
27 J. Llop Castelbou, E. Bresó-Femenia, P. Blondeau, 49 M.-M. Wei, M. García-Melchor, A. Lledós, C. Audin,
B. Chaudret, S. Castillón, C. Claver and C. Godard,
J.-C. Daran, R. Poli, E. Deydier and E. Manoury,
ChemCatChem, 2014, 6, 3160–3168.
Organometallics, 2012, (31), 6669–6680.
28 M. A. S. Garcia, M. Ibrahim, J. C. S. Costa, P. Corio, 50 L. Routaboul, S. Vincendeau, J.-C. Daran and E. Manoury,
E. V. Gusevskaya, E. N. dos Santos, K. Philippot and
L. M. Rossi, Appl. Catal., A, 2017, 548, 136–142.
29 K. M. Roth, A. A. Yasseri, Z. Liu, R. B. Dabke,
Tetrahedron: Asymmetry, 2005, (16), 2685–2690.
51 R. Choukroun, D. de Caro, B. Chaudret, P. Lecante and
E. Snoeck, New J. Chem., 2001, (25), 525–527.
V. Malinovskii, K.-H. Schweikart, L. Yu, H. Tiznado, 52 F. Novio, D. Monahan, Y. Coppel, G. Antorrena, P. Lecante,
F. Zaera, J. S. Lindsey, W. G. Kuhr and D. F. Bocian, J. Am.
Chem. Soc., 2003, 125(2), 505–517.
K. Philippot and B. Chaudret, Chem. – Eur. J., 2014, 5(20),
1287–1297.
30 X. Su and T. A. Hatton, Adv. Colloid Interface Sci., 2017, 244, 53 M. Ibrahim, Rhodium Based Mono- and Bi-Metallic
6–20.
Nanoparticles: Synthesis, Characterization and Application in
Catalysis, Université fédérale de Toulouse-Paul Sabatier,
2016.
31 J. De Tovar, N. Romero, S. Denisov, R. Bofill, C. Gimbert-
Suriñach, D. Ciuculescu-Pradines, S. Drouet, A. Llobet,
P. Lecante, V. Colliere, Z. Freixa, N. McClenaghan, 54 J. S. Bradley, J. M. Millar, E. W. Hill and S. Behal, J. Catal.,
C. Amiens, J. García-Antón, K. Philippot and X. Sala,
Mater. Today Energy, 2018, 9, 506–515.
1991, (129), 530–539.
55 J. S. Bradley, J. M. Millar, E. W. Hill, S. Behal, C. Klein,
A. Duteil and B. Chaudret, Chem. Mater., 1992, (4), 1234–
1239.
56 R. Beckhaus, in Synthetic Methods of Organometallic and
Inorganic Chemistry, ed. W. A. Herrmann, New York, 2000,
vol. 9, p. 38.
57 T. Hayashi, T. Mise, M. Fukushima, M. Kagotani,
N. Nagashima, Y. Hamada, A. Matsumoto, S. Kawakami,
M. Konishi, K. Yamamoto and M. Kumada, Bull. Chem.
Soc. Jpn., 1980, (53), 1138–1151.
58 E. Manoury and R. Poli, in the Series: Catalysis by Metal
Complexes (CMCO), Volume 36 (Phosphorus Chemistry:
Catalysis and Material Science Applications), ed. M.
Peruzzini and L. Gonsalvi, Springer Verlag, 2011, ( (36)),
pp. 121–149.
32 G. G. A. Balavoine, J.-C. Daran, G. Iftime, P. G. Lacroix,
E. Manoury, J. A. Delaire, I. Maltey-Fenton, K. Nakatani and
S. Di Bella, Organometallics, 1999, (18), 21–29.
33 W. Chen and H.-U. Blaser, in Phosphorus Ligands in
Asymmetric Catalysis, ed. A. Börner, Wiley-VCH, Weinheim,
2008, pp. 345–359.
34 R. Gómez Arrayás, J. Adrio and J. C. Carretero, Angew.
Chem., Int. Ed., 2006, (45), 7674–7715.
35 S. Toma, J. Csizmadiova, M. Meciarova and R. Sebesta,
Dalton Trans., 2014, (43), 16557–16579.
36 V. Blanco, D. A. Leigh and V. Marcost, Chem. Soc. Rev.,
2015, (44), 5341–5370.
37 A. Feyrer, M. K. Armbruster, K. Fink and F. Breher, Chem. –
Eur. J., 2017, (23), 7402–7408.
38 J. Choudhury, Tetrahedron Lett., 2018, (59), 487–495.
39 T. J. Colacot, Chem. Rev., 2013, (103), 3101–3118.
40 H. A. McManus and P. J. Guiry, Chem. Rev., 2004, (104),
4151–4202.
41 M. Mellah, A. Voituriez and E. Schulz, Chem. Rev., 2007,
(107), 5133–5209.
59 L. Routaboul, J. Chiffre, G. G. A. Balavoine, J.-C. Daran and
E. Manoury, J. Organomet. Chem., 2001, (637–639), 364–
371.
60 J. Lopez Cortes, O. Ramon, S. Vincendeau, D. Serra,
F. Lamy, J. C. Daran, E. Manoury and M. Gouygou,
Eur. J. Inorg. Chem., 2006, (24), 5148–5157.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019