At the frontier between heterogeneous and homogeneous catalysis: Hydrogenation of olefins and alkynes with soluble iron nanoparticles

Article abstract of DOI:10.1039/c0dt00177e

The use of non-supported Fe nanoparticles in the hydrogenation of unsaturated C-C bonds is a green catalytic concept at the frontier between homogeneous and heterogeneous catalysis. Iron nanoparticles can be obtained by reducing Fe salts with strong reductants in various solvents. FeCl₃ reduced by 3 equivalents of EtMgCl forms an active catalyst for the hydrogenation of a range of olefins and alkynes. Olefin hydrogenation is relatively fast at 5 bar using 5 mol% of catalyst. The catalyst is also active for terminal olefins and 1,1′ and 1,2-cis disubstituted olefins while trans-olefins react much slower. 1-Octyne is hydrogenated to mixtures of 1-octene and octane. Kinetic studies led us to propose a mechanism for this latter transformation where octane is obtained by two different pathways. Characterization of the nanoparticles via TEM, magnetic measurements and poisoning experiments were undertaken to understand the true nature of our catalyst.

Full text of DOI:10.1039/c0dt00177e

View Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
Bridging the gap in catalysis via
multidisciplinary approaches
Guest Editors: Christophe Coperet and Rutger van Santen
Université de Lyon, France and Eindhoven University of Technology, The Netherlands
Published in issue 36, 2010 of Dalton Transactions
Image reproduced with the permission of Dieter Vogt
Articles in the issue include:
Molecular understanding of alkyne hydrogenation for the design of selective catalysts
Javier Pérez-Ramírez, Blaise Bridier and Nuria Lopez
Dalton Trans., 2010, DOI: 10.1039/C0DT00010H
Molecular weight enlargement–a molecular approach to continuous homogeneous catalysis
Michèle Janssen, Christian Müller and Dieter Vogt, Dalton Trans., 2010,
DOI: 10.1039/C0DT00175A
Structure Determination of Zeolites and Ordered Mesoporous Materials by Electron
Crystallography
Xiaodong Zou, Junliang Sun, Dalton Trans., 2010, DOI: 10.1039/C0DT00666A
Metal-Catalyzed Immortal Ring-Opening Polymerization of Lactones, Lactides and Cyclic
Carbonates
Noureddine Ajellal, Jean-François Carpentier, Clémence Guillaume, Sophie M. Guillaume, Marion
Helou, Valentin Poirier, Yann Sarazin and Alexander Trifonov, Dalton Trans., 2010, DOI:
1
0.1039/C001226B
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
At the frontier between heterogeneous and homogeneous catalysis:
hydrogenation of olefins and alkynes with soluble iron nanoparticles†
a
b
a
c
a
Claudine Rangheard, C e´ sar de Juli a´ n Fern a´ ndez, Pim-Huat Phua, Johan Hoorn, Laurent Lefort* and
a
Johannes G. de Vries*
Received 19th March 2010, Accepted 30th July 2010
DOI: 10.1039/c0dt00177e
The use of non-supported Fe nanoparticles in the hydrogenation of unsaturated C–C bonds is a green
catalytic concept at the frontier between homogeneous and heterogeneous catalysis. Iron nanoparticles
can be obtained by reducing Fe salts with strong reductants in various solvents. FeCl reduced by 3
3
equivalents of EtMgCl forms an active catalyst for the hydrogenation of a range of olefins and alkynes.
Olefin hydrogenation is relatively fast at 5 bar using 5 mol% of catalyst. The catalyst is also active for
terminal olefins and 1,1¢ and 1,2-cis disubstituted olefins while trans-olefins react much slower.
1
-Octyne is hydrogenated to mixtures of 1-octene and octane. Kinetic studies led us to propose a
mechanism for this latter transformation where octane is obtained by two different pathways.
Characterization of the nanoparticles via TEM, magnetic measurements and poisoning experiments
were undertaken to understand the true nature of our catalyst.
Introduction
homogeneous and heterogeneous could be intrinsically active in
the hydrogenation of unsaturated C–C bonds. Soluble transition
metal nanoparticles have attracted a lot of interest from the
From an industrial point of view, hydrogenation catalysts based
on iron are very attractive. Iron is abundant on earth and conse-
quently its compounds are a few orders of magnitude cheaper than
those of other metals more commonly used in hydrogenation such
as Pd, Pt, Rh or Ir. Additionally, iron complexes are usually non-
toxic and can be handled and disposed of easily. Currently there is a
lot of renewed interest in the use of iron compounds as catalysts for
11
catalytic community in the last few decades. Their activity is
known to be very high under mild conditions because of their
very large surface area compared to heterogeneous catalysts. An
additional advantage is that they can be more easily recovered
compared to truly homogeneous catalysts. Iron nanoparticles have
already found use in a wide range of applications including mag-
netic fluids, catalysts for carbon nanotubes formation, magnetic
resonance imaging (MRI) contrast agents, nickel–iron batteries,
1
various transformations. However, outside the Fischer–Tropsch
2
3
process and the Haber–Bosch ammonia synthesis, the use of
iron in heterogeneous hydrogenation is rather limited. For the
hydrogenation of unsaturated C–C bonds, there are only a few
12
catalysts and sorbents for environmental remediation. However,
their use in catalysis has remained only marginal and centered
4
publications where high temperatures and pressures are used.
13
on C–C formation reactions. Additionally, we anticipated that
In addition, RANEYꢀR -iron, Urushibara iron and iron metal
Fe nanoparticles could be prepared via a simple procedure with
cheap stabilizers or even without stabilizers, thus warranting an
extremely cost-efficient catalyst.
obtained from the decomposition of Fe(CO)
5
were used for the
5
hydrogenation of alkynes. Examples of efficient homogeneous
hydrogenation of C–C double or triple bonds with Fe complexes
are also scarce. Early trials focused on Ziegler–Natta type of
In an earlier communication, we reported the synthesis of
soluble Fe–NPs prepared via Grignard reduction and their use
6
7
catalysts where Fe salts were reduced by AlR
3
. In the 1970’s, iron
14
in the hydrogenation of olefins and alkynes. In this article, we
carbonyl complexes have been used as poorly active hydrogenation
present a systematic study aimed at understanding how the method
of NPs synthesis as well as the hydrogenation conditions affect
the catalytic activity of nanoparticles. Kinetic studies during the
alkyne hydrogenation led us to propose a mechanism. Characteri-
zation of the nanoparticles by TEM, magnetic measurements, and
chemical poisoning experiments are presented.
8
9
10
catalysts. More recently, Fe complexes with phosphine or imine
ligands have been used. The latter catalysts have not only shown
high activities in the hydrogenation of simple alkenes, but also
N– and O-containing olefins. Instead of searching for new ligands
to impart hydrogenation activity to homogeneous Fe complexes,
we decided to explore the use of soluble Fe nanoparticles. Our
hope was that these peculiar entities at the frontier between
Results and discussion
a
DSM Innovative Synthesis B.V., a unit of DSM Pharma Chemicals,
P. O. Box 18, 6160, MD Geleen, The Netherlands
Synthesis of the Fe nanoparticles and their catalytic activity
b
Laboratorio di Magnetismo Molecolare, INSTM - Dipartimento di
Chimica, Universit a` degli Studi di Firenze, Polo Scientifico - Via della
Lastruccia 3, 50019, Sesto Fiorentino (FI), Italy
DSM Advanced Chemical Engineering Solutions, P. O. Box 18, 6160,
MD Geleen, The Netherlands
A number of synthetic methods for the preparation of Fe
12g
nanoparticles (Fe–NPs) have been reported, such as the thermal
c
15
decomposition of iron carbonyl complexes, the hydrogenation of
16
17
iron amides and the chemical reduction of iron salts. In our
first trials, we decided to use this last method. Using our high
†
Electronic supplementary information (ESI) available: Additional exper-
imental details. See DOI: 10.1039/c0dt00177e
8
464 | Dalton Trans., 2010, 39, 8464–8471
This journal is © The Royal Society of Chemistry 2010

a
a
Table 1 Influence of the reducing agent in the hydrogenation activity
Table 2 Influence of the iron precursor on the hydrogenation activity
Conversion
norbornene (%)
Iron
precursor
Conversion
norbornene (%)
Entry
Reducing agent
Entry
EtMgCl (eq/Fe)
1
2
3
4
5
6
7
8
9
Me
3
Al
0
50
1
2
3
4
5
6
7
8
FeCl
Fe(acac)
3
3
3
3
3
2
3
2
3
2
100
100
100
70
100
100
100
100
Et Al
3
MeLi
n-BuLi
PhLi
MeMgCl
EtMgCl
i-PrMgCl
PhMgBr
EtMgCl (1 eq/Fe)
EtMgCl (2 eq/Fe)
EtMgCl (4 eq/Fe)
100
100
100
100
100
100
100
0
FeCl
FeCl
2
2
FeI
FeI
2
2
Fe(OAc)
Fe(OAc)
2
2
a
Conditions: Reduction Fe salt: Fe salt (0.05 mmol), reducing agent, THF
1 mL), R.T., 15 min; Hydrogenation: norbornene (1 mmol), THF (3 mL),
1
1
1
0
1
2
(
0
100
2
R.T., 10 bar H , 30 min).
a
3
Conditions: Reduction Fe salt: FeCl (0.05 mmol), 3 eq. reducing agent,
THF (1 mL), R.T., 15 min; Hydrogenation: norbornene (1 mmol), THF
3 mL), R.T., 10 bar H , 30 min.
and Grignards reagents allowed the formation of a catalyst giving
full conversion of the norbornene to norbornane. We observed a
(
2
remarkable difference between AlMe
3
(Entry 1) and AlEt (Entry
3
1
8
throughput experimentation equipment, we were able to quickly
test a wide range of Fe salts (FeCl , Fe(Acac) , Fe(NO , Fe(SO ),
FeCl , FeBr , FeI , Fe(OAc) ) in combination with 11 different sta-
bilizers (PEG, PVP, SDS, primary amines, ammonium bromide)
2) although the same dichotomy was not observed for the Grignard
reagents (Entry 6,7). F u¨ rstner published a proposal regarding the
mechanismof the reductionofironchloride salts byEtMgBr based
3
3
3
)
3
4
2
2
2
2
24
on the observation of released ethane and ethylene. He reports a
difference of reactivity in cross-coupling reactions between methyl
reduced by NaBH
4
in EtOH. The resulting Fe species were tested
in the hydrogenation of norbornene at 80 C under 20 bar of H
Most of the combinations did not show any hydrogenation activity.
Only when using FeI or FeCl , small amounts of norbornane were
detected but the results were difficult to reproduce on larger scale.
In 2006, during a study on the coupling of alkyl halides with aryl
magnesium bromides with Fe complexes, Bedford et al. noticed
that independently of the homogeneous Fe catalysts used, the
reaction mixture would immediately turned black upon addition
of the Grignard reagent. A transmission electron microscopy
image of the reaction mixture revealed the presence of Fe–NPs
thus pointing towards a new method for the preparation of Fe–
◦
18
2
.
and ethyl alkylating agents. In agreement with the observations
25
of Bogdanovi c´ , he attributes this difference to the possibility
(with EtLi or EtMgCl) or impossibility (with MeLi or MeMgCl)
for the metal to undergo b-hydride elimination in its reduction
process. We do not observe such a difference as both MeLi and
MeMgCl as well as PhMgBr lead to an active catalyst. In the case
of EtMgCl, the effect of the ratio of Grignard reagent relative
to Fe was probed. Using 2 equivalents of Grignard reagents or
less, no hydrogenation activity was obtained while 3 equivalents
or more give full conversion. This minimum ratio EtMgCl : Fe
of 3 : 1 is consistent with the reduction of Fe(+III) into Fe(0). In
2
2
13a
NPs, i.e. the reduction of Fe halides with a Grignard reagent.
Immediately after the discovery of the Ziegler–Natta catalyst in
the 1960’s, several groups studied the reaction of Fe salts with AlR
a set of control experiments, MgCl
2
, EtMgCl, EtMgCl/MgCl
2
and FeCl /MgCl were tested in hydrogenation. In all cases, no
3
2
3
hydrogenated product was observed. At this stage, we decided to
use EtMgCl as the standard reducing agent in the rest of our study.
We then investigated the influence of the iron precursors on the
activity of the Fe–Np’s. Five different Fe salts were reduced by
EtMgCl in THF and then tested in norbornene hydrogenation
19
and other “alkylating” agents and some even tested the activity
7,20
of the obtained Fe species in hydrogenation. However, none of
these reports clearly mention the presence of Fe–NPs. Identically,
Fe complexes were extensively used in cross-coupling reaction
where they were contacted with Grignard and other alkylating
Currently, there is no definitive consensus on the
mechanism of this transformation where catalytic cycles involving
(Table 2). FeCl
the other iron salts FeCl
addition of EtMgCl, all the reaction mixtures turned black and no
remaining solid material was present. In contrast to FeCl , a ratio
3
and Fe(acac)
3
are entirely soluble in THF while
1a,21
reagents.
2
, FeI
2
and Fe(OAc) are not. After
2
22
22a,23
24
Fe(+I)/Fe(+III), Fe(0)/Fe(+II)
or even Fe(-II)/Fe(0) have
3
been proposed. Considering our initial goals to test Fe nanoparti-
cles in hydrogenation, the recent report of Bedford, and the early
successes met with Ziegler–Natta type catalysts in hydrogenation,
EtMgCl : Fe of 2 : 1 was sufficient to prepare an active catalyst
from Fe(+II) precursors (Entries 4,6). This is again in agreement
with the reduction of Fe(+II) into Fe(0). FeCl was chosen as our
3
we decided to switch from NaBH
4
to other reductants.
standard iron precursor due to its high solubility in THF.
Various solvents (tetrahydrofuran, toluene, heptane, dioxane,
ethyl acetate, diethylether) were tested for the preparation of the
The reduction of FeCl was carried using different reducing
3
agents such as trialkyl aluminiums, alkyl lithiums, Grignard
reagents, tetraalkyllead and tetraalkyltin. The reactions were
performed using 3 equivalents of reducing agents in THF under
inert atmosphere. In most cases, the color of the solution changed
instantaneously from yellow to dark brown/black. The suppos-
edly obtained Fe–NPs were tested in norbornene hydrogenation
catalyst. FeCl is entirely soluble in tetrahydrofuran, dioxane,
3
ethylacetate and diethylether, but only partially soluble in toluene
and heptanes at [Fe] = 0.05 M (see supplementary information†).
Upon treatment with 3 eq. of EtMgCl, all the solutions turned
instantaneously brown/black. Formation of a deposit occurred
overnight in all solvents except for toluene and THF. However,
full conversion to norbornane (Reaction conditions: 5 mol% Fe–
(
Table 1).
Gratifyingly, many reducing agents convert FeCl into an active
3
catalyst for the hydrogenation of norbornene (Table 1). All lithium
NPs, [Alkene] = 0.33 M, 3 mL solvent, 10 bar H
2
, R.T.) was
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 8464–8471 | 8465

nevertheless obtained in all solvents except in EtOAc and heptane.
Interestingly, the Fe–NPs prepared in THF could be dried and
entirely redispersed in the same solvent.
14
As already reported, the FeCl /EtMgCl system is active in
3
the hydrogenation of a range of olefins (Fig. 1) and alkynes.
Aliphatic linear terminal, gem-disubstituted and linear cis-olefin
are fully converted. Cyclic olefins lacking cycle strain and trans-
olefins are more difficult to hydrogenate and require higher
temperature to give full conversion. More bulky substrates such
as tri- and tetrasubstituted olefins were not hydrogenated even at
◦
1
00 C. Equally, no reaction was observed towards anisole, nitro-
benzene, acrylate and amido-acrylate. In the latter two cases, this
may be due to the incompatibility of our catalyst with esters as
already indicated by its low activity in EtOAc. Here, we decided
to monitor the progress of the 1-octyne hydrogenation reaction
over a period of a few hours in order to gain more insight into
the selectivity. Three experiments with different settings were
Fig. 2 Concentration 1-octyne, 1-octene and n-octane vs. time at high
-1
catalyst concentration (Reaction conditions: 0.9 g l Fe (0.017 M),
◦
[1-Octyne] = 0.33 M, THF, 5 bar H
2
, 40 C).
◦
performed in an autoclave at 40 C under 5 bar of hydrogen
(
Fig. 2 to 4). Under the reaction conditions only the hydrogenated
products, i.e. 1-octene and n-octane were formed and no by-
products were observed.
Fig. 3 Concentration 1-octyne, 1-octene and n-octane vs. time at low
-1
catalyst concentration (Reaction conditions: 0.2 g l Fe (0.0034 M),
◦
[
1-Octyne] = 0.067 M, THF, 5 bar H
2
, 40 C).
3
Fig. 1 Scope of the FeCl /EtMgCl hydrogenation catalyst (Reaction
conditions: 5 mol% Fe, [Alkene] = 0.33 M, THF, 20 bar H
values in parentheses at 100 C, 18 h).
2
, R.T. except
◦
As can be seen from Fig. 2, the catalyst becomes active 10 min
after the introduction of H . The reason for this induction period
2
remains unclear. At this point, the reaction order in 1-octyne
follows zero-order behavior. The formation of n-octane and 1-
octene from 1-octyne is simultaneous and the ratio of their
respective rates is constant. When all 1-octyne has reacted, the
hydrogenation of 1-octene proceeds much faster and appears to
be first order in 1-octene. This suggests that the active sites of the
catalyst are predominantly occupied by the alkyne, thus retarding
the hydrogenation of 1-octene. Consequently, in the presence
of 1-octyne, 1-octene is prevented from binding to the catalyst
and the initially observed n-octane must be produced via direct
hydrogenation of 1-octyne without release and re-adsorption of
Fig. 4 Concentration 1-octyne, 1-octene and n-octane vs. time starting
with 1-octyne and 1-octene simultaneously (Reaction conditions: 0.9 g l
-1
26
Fe (0.017 M), [1-Octyne] = 0.33 M, [1-Octene] = 0.33 M, THF, 5 bar H ,
2
◦
40 C).
is also constant when 1-octyne is still present. Interestingly, the
rate of hydrogenation of 1-octene after complete consumption
of 1-octyne is much more affected by the lower concentrations
used in this experiment, becoming even slower than the rate of
hydrogenation of 1-octyne. This behaviour is consistent with the
proposed orders of reactions, i.e. 0 for the alkyne hydrogenation
and 1 for the alkene hydrogenation.
1
-octene.
At a lower catalyst concentration, the rate of 1-octyne disap-
pearance is reduced but the reaction order appears to be also zero
Fig. 3). The ratio of the formation rates of 1-octene and n-octane
(
8
466 | Dalton Trans., 2010, 39, 8464–8471
This journal is © The Royal Society of Chemistry 2010

a
Next to the experiments, a simple Langmuir-Hinselwood type
model was developed (see supplementary information) to verify
some of the hypotheses regarding the mechanism. For all ex-
periments the initial induction periods were not included in the
data used to tune the model as the start-up phenomena were too
complicated to be described by a simple model. A good fit between
the model and the experimental data could be obtained only if the
equilibrium constant for the reversible binding of 1-octyne to the
Table 3 Selectivities of the bimetallic catalyst in 1-octyne hydrogenation
Entry
Metal nanoparticles
1-Octene : Octane
1
2
3
4
5
6
7
Fe
Ni
Co
Cu
1 : 0.42
1 : 0.16
1 : 0.32
No activity
1 : 0.26
1 : 0.34
No activity
Fe : Ni (1 : 1)
Fe : Co (1 : 1)
Fe : Cu (1 : 1)
4
catalyst is set to be 10 higher than for 1-octene (see curves on
Fig. 2 and 3). Further corroboration of this was obtained by an
experiment in which we hydrogenated 1-octyne in the presence of
a
Reaction conditions: 5 mol % metal, [1-Octyne] = 0.33 M, THF, 5 bar
2
H
, R.T.
1
-octene.
When the hydrogenation was performed with an equal concen-
The induction period observed in the initial phase can be caused
tration of 1-octyne and 1-octene, the induction period was slightly
longer than in the absence of 1-octene (Fig. 4). During this period,
a small amount of 1-octene hydrogenation also appears to take
place as the initial formation rate of n-octane is slightly higher
than the disappearance rate of 1-octyne. However, after 30 min
by a number of reasons. One possible explanation would be a
further decrease in size of the nanoparticles either by binding to the
substrate or to hydrogen. Gates observed that supported iridium
29
clusters decrease in size upon addition of alkenes. Ageing the
Fe–NPs in presence of 1-octyne slightly increases their activity (see
supplementary information, Figure S3†). Additionally, ageing the
nanoparticles in absence of 1-octyne also changes their activities
but in this case, different behaviours are obtained depending
1
-octyne hydrogenation completely dominates and no conversion
of 1-octene is found as long as 1-octyne is present. When 1-octyne
is completely converted the hydrogenation rate of 1-octene is equal
to the experiment with similar catalyst concentration (Fig. 2).
We propose that in the presence of 1-octyne, the initially
observed n-octane must be produced via direct hydrogenation of
on the concentration of the FeCl solution used to prepare the
3
nanoparticles. TEM measurements after the different treatments
of the nanoparticles could help to establish an eventual structure–
activity relationship.
In order to suppress the hypothetic direct pathway to octane,
we decided to turn towards bimetallic nanoparticles. Selective
poisoning of some of the active sites of a catalyst by a second metal
is a common method to improve the selectivity. Our bimetallic
nanoparticles were prepared by adding a second cheap metal
1
-octyne without release and re-adsorption of 1-octene. A possible
mechanism for this transformation is depicted in Scheme 1. After
exposure to H , the Fe–NPs have formed either some hydrides
and/or some dihydrogen species. The presence of hydrogen on the
Fe–NPs was proven by H titration experiments (2.5 hydrogen
2
2
atom per exposed iron atom), which we describe later in this
article. However, the exact binding mode of the hydrogen atoms is
unknown at this stage. For convenience, we write a mixed hydride-
dihydrogen species (1) based on the mechanism decribed by
Bianchini and Oro for alkyne hydrogenation with a homogeneous
halide (NiCl
2
, CoCl
2
, CuCl
2
) to FeCl in a 1 : 1 ratio and reducing
3
both metals with 3 eq of EtMgCl. The resulting bimetallic catalysts
were tested in the hydrogenation of 1-octyne (Fig. 5 for activities;
Table 3 for selectivities).
9a
iron complex. We assume that insertion of the alkyne into the
Fe–H bond occurs leading to a Fe-alkenyl complex (2) that forms
1
-octene via reductive elimination (Path 1). Additionally, we can
also imagine that in the presence of H , intermediate (2) rearranges
2
27
to form an iron hydride alkylidene (3). Such iron carbenes, which
in some instances are formed by reaction of Fe complexes with
28
alkynes are known. The Fe alkylidene can be hydrogenated to
an iron alkyl compound (4). Upon further reaction with H , the
2
latter compound forms n-octane and regenerates the starting Fe
species.
Fig. 5 Concentration of 1-octyne vs. time for various monometallic and
bimetallic catalysts (Reaction conditions: 5 mol % total metal content,
[
2
1-Octyne] = 0.33 M, THF, 5 bar H , R.T.).
As can be seen in Fig. 5, the monometallic Ni and Co nanopar-
ticles are much more active towards 1-octyne hydrogenation than
Fe–NPs. Cu–NPs and Fe–Cu NPs (prepared from CuCl , not
2
Scheme 1 Proposed mechanism for the direct hydrogenation of 1-octyne
to n-octane (Fe denotes an iron atom at the surface of the nanoparticles).
shown) are inactive in our reaction conditions. Bimetallic Fe–Co–
NPs have almost the same activity as Co-NPs while Fe–Ni-NPs
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 8464–8471 | 8467

display an intermediate activity between both single metal NPs.
As long as 1-octyne is present, the ratio 1-octene:octane stays
constant as already observed for Fe–NPs. Both active bimetallic
catalysts (FeNi, FeCo) exhibit intermediate selectivities compared
to the corresponding monometallic components (Table 3). These
are promising first results that show that the selectivity of the
hydrogenation can be modified using polymetallic NPs.
Characterization of the Fe catalyst
Having prepared an active hydrogenation catalyst based on Fe,
we then decided to proceed with its characterization and the
14
identification of the active Fe species. As already reported,
transmission electron microscopy of a solution of FeCl treated
3
with EtMgCl confirmed the presence of Fe nanoparticles (NPs)
with an average size of 2.67 ± 0.60 nm (Fig. 6).
Fig. 7 (a) Hysteresis loops measured at the temperature of 3 K and 40 K;
b) temperature dependence of the Zero-field-cooling and field cooling
magnetizations.
(
Fig.
FeCl
6
TEM micrographs of Fe nanoparticles obtained from
32
effects that give rise to the non-saturation of the field dependent
magnetization. These two results indicate the exclusive grown
of metallic Fe nanoparticles. The large magnetic moment in
nanoparticles is one desirable property for many applications but
often the magnetic moment is smaller due to oxidation of the
surface or to the chemical interaction with the capping agent. Free
3
/EtMgCl.
Unsurprisingly, EDX showed that the solids are composed of
iron, magnesium and chlorine. Although we do not use a stabilizer,
visual inspection of our Fe–NPs after 1 night in THF does not
reveal any precipitate. We assume that MgCl
2
and possibly THF
30
-
33
act as stabilizer. Finke recently reported that even BF
4
is a better
Fe clusters with similar or smaller sizes than 2–3 nm exhibit even
stabilizer of NPs than ethereal solvents. It thus seems highly likely
that the chloride ions of MgCl will bind to the outer rim of the
Fe–NPs. In order to probe the influence of MgCl on the activity
of the catalyst in hydrogenation, we added various amounts (from
.5 to 20 eq/Fe) of anhydrous MgCl to freshly prepared Fe–
NPs and tested their activity in the hydrogenation of norbornene
in the standard reaction conditions (R.T., 10 bar H , 30 min).
No influence of added MgCl was observed since in all cases,
norbornene was fully converted to norbornane. Advantageously,
if MgCl is effectively acting as a stabilizer, even an excess of it
higher magnetizations but their transfer to physical or chemical
supports is difficult. Bulk like magnetizations were also observed
by Chaudret et al. in Fe capped nanoparticles synthesized by
2
34
2
hydrogenation of Fe[N(SiMe
larger (7 nm).
3
)
2
2
] but the average particle size is
0
2
Fig. 7a also shows that at low temperatures K the hysteresis loop
exhibits both coercive field (150 Oe) and magnetization at zero field
(remanence) while at 40 K these properties are zero. This change of
magnetic behavior is a consequence of thermal demagnetization
effects. At low temperatures, the magnetic state is defined as
blocked while at higher temperatures the nanoparticles exhibits
2
2
2
does not prevent the substrate from reaching the active site of the
catalyst as is sometimes the case with more traditional stabilizers.
To further characterize the Fe–NPs, their magnetic proper-
ties were measured using a Quantum Design superconducting
quantum interference device (SQUID) magnetometer. In Fig. 7a,
the hysteresis loops, in terms of magnetization per gram of Fe,
measured at 3 K and 40 K are represented. The density of
magnetization reached at 3 K is 224 ± 5 emu/gr, value in excellent
35
superparamagnetism behaviour. The temperature dependence of
the ZFC magnetization exhibits a maximum at 10 K (Fig. 7b)
corresponding to the temperature at which most nanoparticles
changefromblockedtosuperparamagnetic. Moreover above20K,
the temperature at which the FC curve joins the ZFC curve, all
the nanoparticles are superparamagnetic. The change from the
blocking regime to the superparamagnetic regime is an additional
evidence of the nanometric nature of the particles. Because of
their small size, the thermally driven demagnetization leads to a
relaxation process occurring faster than the measurement time.
The description of this relaxation time in terms of the Arrhenius
31
accordance with the one of bulk bcc Fe(0), 220 emu/gr. The
presence of iron oxides is excluded as their magnetizations, in
the best of the cases (for magnetite and maghemite), are less
31
than half this value. Moreover the value of the magnetization
above 20 kOe is well saturated. This excludes nanostructural
features like the presence of ions and/or oxide nanoparticles
with the uncompensation of the antiferromagnetic order or size-
35,36
expression
for this type of materials and for the SQUID
measurements is given by the following expression: 25 k
B
B
T =
K
ef·V
m
, where T
B
is the blocking temperature, V
m
the average
8
468 | Dalton Trans., 2010, 39, 8464–8471
This journal is © The Royal Society of Chemistry 2010

particle volume, Kef the effective magnetic anisotropy, k the
B
6
-3
3
Boltzmann constant. Kef is equal to 2 ¥ 10 ergcm if V
m
is 19 nm
(
considering the lognormal distribution from the TEM histogram)
3
6
and T
B
= 11 K. The Kef value is larger than the one of bulk bcc Fe
indicating an enhancement of the anisotropy. Such an effect can
be due either to the local enhancement of the spin–orbit coupling
caused by the breaking of the crystalline symmetry at the surface
32c,33
of the nanoparticles
surface Fe–Cl bonds.
or to the electronic localization in some
Although TEM and magnetic measurements showed the pres-
ence of nanoparticles, one can not exclude that the active hydro-
genation species are molecular Fe complexes. Selective poisoning
37
Fig. 8 Influence of protic additives in the hydrogenation activity of the
Fe catalyst.
tests are commonly used to identify the nature of the true catalyst.
Hg is a well known poison for nanoparticles due to the formation
of an amalgam while the molecular metal complexes remains
unaltered. This property of Hg was first exploited by Whitesides
our Fe nanoparticles. In these experiments, the nanoparticles are
first exposed to 10 bar of hydrogen for 1 h at room temperature.
38
and co-workers. The experiment consists in adding Hg(0) to
the reaction mixture and the suppression of catalytic activity is
a strong evidence for catalysis by nanoparticles. In our case, the
conversion of norbornene to norbornane was reduced to 40%
instead of 100% when the hydrogenation was performed with Fe–
NPs that had been vigorously stirred for 1 h with Hg(0). This
intermediate result is inconclusive. However, one report mentions
that iron may not form a stable alloy with Hg(0) which could
explain the residual activity. Another poisoning study consists in
using ligands that strongly bind to a metal center such as thiophene
and triphenylphosphine that can block access of the substrate to
the active site. If a catalyst can be poisoned by a sub-stoichiometric
amount of poison (ꢀ1eq per metal atom), it can be concluded that
the true catalyst are nanoparticles since in this case, only a fraction
of the metal atoms, i.e. the ones on the surface, are active. On the
other hand if the catalyst is poisoned with one or more equivalent
of poisoning ligands, it is generally concluded that the catalyst is
After removal of H
2
, an atmosphere of nitrogen is introduced
(
three cycles of high pressure nitrogen followed by a purge),
norbornene is added to the Fe nanoparticles and the amount
of norbornane is determined by gas chromatography. Different
amounts of Fe nanoparticles (0.0125 to 0.175 mmol Fe) were
contacted with the same amount of norbornene (0.1 mmol). The
amount of norbornane obtained vs. the amount of Fe is plotted
in Fig. 9. The slope of the straight line fitting the experimental
points is 0.46, indicating that there is about 1 H per total atoms
of Fe. Assuming that our nanoparticles are monodispersed with
an average diameter of 2.7 nm as measured by TEM, we can
approximate the amount of H per atom of Fe on the surface to
39
2
.45 or about 5 H per 2 atoms of Fe. To account for this odd
number, one can assume the presence of a bridging H.
a homogeneous metal complex. In our case, thiophene or PPh
0.1–1 eq/Fe with PPh , 0.1–5 eq/Fe with thiophene) did not alter
3
(
3
the activity of our system in norbornene hydrogenation. Sulfur
containing substrates are notoriously difficult to hydrogenate with
most heterogeneous catalysts. Even homogeneous catalysts are
40
seriously retarded by the presence of thioethers. Consequently,
we took advantage of this last result and demonstrated that our
Fe catalyst is able to hydrogenate a sulfur containing olefin, i.e.
2
1
-allyl-thiophene. Using 5 mol% of Fe–NPs in THF, at R.T. and
0 bar of H , this substrate was fully hydrogenated after 30 min. We
2
2
Fig. 9 Determination of the amount of H/H in the catalyst.
are currently testing our catalyst with other S- and P-containing
substrates.
To a certain extent, our Fe–NPs were also remarkably resistant
towards protic additives. The catalyst was pretreated with various
Conclusion
amount of degassed H
2
O during 15 min prior to the hydrogenation
O/Fe are necessary to
We have shown that iron nanoparticles, prepared via reduction of
iron salts in THF with alkyl Grignards or alkyl lithiums are good
catalysts for the hydrogenation of alkenes and alkynes. A substrate
containing a thioether was hydrogenated without any retardation.
Kinetic measurements of the octyne hydrogenation led to the
interesting observation that the alkyne binds much stronger to
the surface of the iron nanoparticles than the alkene, leading us to
propose a mechanism where alkane formed during the first phase
of the hydrogenation when alkyne is still present is not caused by
hydrogenation of alkene but rather results from an isomerisation
of the initial formed alkenyl iron compound to an alkylidene iron
compound, which is further hydrogenated to the alkane. Although
(
Fig. 8). Up to 50 equivalents of H
2
completely eliminate the activity of the catalyst in norbornene
hydrogenation. The same holds true with the less protic MeOH. An
explanation for this behavior may be that highly hygroscopic
MgCl
2
may scavenge H O and protect to a certain extent Fe(0)
2
from oxidation. Only two equivalents of acetic acid are necessary
to completely deactivate the catalyst. In this case, the brown/black
solution of Fe–NPs becomes quickly yellow corresponding to the
redissolution of the Fe(0) species as iron acetate species.
Following the methodology described by Chaudret et al., we
investigated the presence of hydrogen or hydrides on the surface of
41
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 8464–8471 | 8469

the presence of Fe(0) nanoparticles was proven by TEM, EDX
and by magnetic measurements, we are still unable to ascertain
whether they are the true active species in hydrogenation. We are
currently pursuing the further characterization of the Fe–NPs and
are extending their use to the hydrogenation of more challenging
substrates.
the Premex A96 Reactor. The hydrogenated products were then
analyzed by GC.
Hydrogenation of 1-octyne in the autoclave
In the glovebox, 1-octyne (2.57 g, 23.3 mmol) and THF (44.5 mL)
were added to a 150 ml Parr autoclave. 23.5 mL of the catalyst
solution (0.05 M), prepared from 0.138 g of FeCl
3
(189.3 mg,
Experimental procedures
1
.17 mmol) and EtMgCl (1,76 ml of a 2.0 M solution) in THF
was added at once (S/C = 20). The vessel was purged 3 times
General procedures
with nitrogen and 3 times with hydrogen and kept under 5 bar
◦
All manipulations were performed under a nitrogen atmosphere in
an M. Braun glove box. Anhydrous solvents (THF, toluene, hep-
tanes, ethyl acetate, DCM) from the M. Braun Solvent Purification
System were used. The compounds 1-octyne, norbornene and the
other substrates were obtained from Aldrich and used as received.
Premex A 96 reactor (small scale reaction) and Parr autoclave
of hydrogen at 40 C under mechanical stirring at 800 rpm for
4 h. Samples were taken every 5 or 10 min and the hydrogenation
products were analyzed by GC.
Acknowledgements
(
large scale reaction) were used for hydrogenation reactions. GC
We thank the European Union for its financial support of this
research which was carried out as part of the Research Training
Network (R)evolutionary Catalysis (MRTN-CT-2006 035866).
were performed on a Hewlett-Packard 5890 Gas Chromatography
Series II using a CP-Sil-5CB Chrompack capillary column (50 m
¥
0.32 mm ¥ 1.2 mm) or CP-Sil-5CB Chrompack capillary column
(
25 m ¥ 0.25 mm ¥ 0.25 mm). Transmission electron microscopy
References
coupled with EDX (Energy Dispersive X-ray Spectrometer) was
performed at Paul Sabatier University of Toulouse. The magnetic
properties of the iron nanoparticles in THF were measured
using a Quantum Design superconducting quantum interference
device (SQUID) magnetometer at Laboratorio di Magnetismo
Molecolare, INSTM, Firenze. The samples were placed into the
magnetometer under inert atmosphere 5 min after their synthesis.
Temperature-dependent zero-field-cooled (ZFC) and field-cooled
1 (a) W. M. Czaplik, M. Mayer, J. Cvengro sˇ and A. J. von Wangelin,
ChemSusChem, 2009, 2, 396; (b) B. D. Sherry and A. F
u¨ rstner, Acc.
Chem. Res., 2008, 41, 1500; (c) C. Bolm, J. Legros, J. Le Paih and L.
Zani, Chem. Rev., 2004, 104, 6217; (d) A. F u¨ rstner and R. Martin,
Chem. Lett., 2005, 34, 624; (e) A. Correa, O. G. Mancheno and C.
Bolm, Chem. Soc. Rev., 2008, 37, 1108.
2
3
R. Schl o¨ gl, Angew. Chem., Int. Ed., 2003, 42, 2004.
W. Arabczyk, I. Jasinska and R. Jedrzejewski, Catal. Commun., 2009,
1
0, 1821.
(
FC) magnetization measurements were obtained for a range of
4
(a) Z. Lu, C. Wang, D. Liang and L. Lin, Shiyou Huagong, 1991, 20,
454; (b) Y. Nitta, S. Matsugi and T. Imanaka, Chem. Express, 1989,
temperatures between 2.5 K and 140 K (below the melting point of
the THF) with an applied magnetic field of 10 Oe. The hysteresis
loops were recorded at 3 K and at 40 K for a maximum applied
magnetic field of 50 kOe. The dia and paramagnetic contributions
of the sample holder and THF were measured and subtracted to
the measurement. The density of magnetization was calculated
considering the amount of iron in the synthesis.
4
, 547; (c) Y. M. Bondarev, Y. S. Mardashev, A. D. Mokrushin, V. F.
Blokhin and I. V. Razumovskaya, Zh. Fiz. Khim., 1985, 59, 2343.
(a) S. Nishimura in Handbook of Heterogeneous Catalytic Hydrogena-
tion for Organic Synthesis, J. Wiley & Sons, Inc., 2001, pp. 165 and
references within; (b) A. F. Thompson, Jr. and S. B. Wyatt, J. Am.
Chem. Soc., 1940, 62, 2555; (c) R. Paul and G. Hilly, Bull. Soc. Chim.,
5
1
939, 6, 218.
6
7
S. Enthaler, K. Junge and M. Beller in Iron catalysis in Organic
Chemistry. Reactions and applications, ed. B. Plietker, Wiley-VCH,
Weinheim, 2008, pp. 129-136 and references within.
Standard procedure to prepare Fe-NPs from FeCl
3
and EtMgCl
(a) M. F. Sloan, A. S. Matlack and D. S. Breslow, J. Am. Chem. Soc.,
1
963, 85, 4014; (b) Y. Takegami, T. Ueno and T. Fujii, Bull. Chem.
In the N glovebox, FeCl (0.1622 g, 1.0 mmol) was dissolved in
2
3
Soc. Jpn., 1965, 38, 1279; (c) N. F. Noskova, A. Zh. Kazimova, K. K.
Kambarova, S. R. Savel e´ v and N. L. Melamud, Zh. Org. Khim., 1992,
28, 1352.
dry THF (20.0 mL). A 2.0 M solution of EtMgCl (2.50 mL) in
THF was added via syringe. The reaction mixture turns instan-
taneously from yellow to brown/black. The reaction mixture was
vigorously stirred for 30 min. Iron nanoparticles were used without
further purification. The synthesized catalyst was characterized
by transmission electron microscopy coupled with EDX (Energy
Dispersive X-ray Spectrometer) and magnetic measurements using
a Quantum Design superconducting quantum interference device
8
9
(a) B. R. James, Homogeneous Hydrogenation, John Wiley & Sons, New
York, 1973, pp. 64–72; (b) C. U. Pittman Jr., R. C. Ryan, J. McGee
and J. P. O’Connor, J. Organomet. Chem., 1979, 178, C43; (c) M. A.
Schroeder and M. S. Wrighton, J. Am. Chem. Soc., 1976, 98(2), 551;
(d) Y. Tajima and E. Kunioka, J. Org. Chem., 1968, 33, 1689.
(a) C. Bianchini, A. Meli, M. Peruzzini, P. Frediani, C. Bohanna, M. A.
Esteruelas and L. A. Oro, Organometallics, 1992, 11, 138; (b) E. J. Daida
and J. C. Peters, Inorg. Chem., 2004, 43, 7474.
(
SQUID) magnetometer.
10 (a) Q. Knijnenburg, A. D. Horton, H. Van der Heijden, A. W. Gal and
P. H. M. Budzelaar, to Stichting voor de Technische Wetenschappen,
Int. Pat. Appl., WO03042131, 2003; (b) S. C. Bart, E. Lobkovsky and
P. J. Chirik, J. Am. Chem. Soc., 2004, 126, 13794; (c) R. J. Trovitch, E.
Lobkovsky, E. Bill and P. J. Chirik, Organometallics, 2008, 27, 1470.
Hydrogenation of alkenes or alkynes (substrate screening)
To a 5 mL reaction vial containing a stirring bar, an alkene or
an alkyne (1.0 mmol) in THF (1.0 mL) was added. 1.0 mL of the
catalyst solution (a) prepared as described above (1.0 mL, [Fe]
1
1 (a) A. Roucoux and K. Philippot, in Handbook of Homogeneous
Hydrogenation, ed. J. G. de Vries and C. J. Elsevier, Wiley-VCH,
Weinheim, 2007, vol. 1, pp. 217-256; (b) A. Roucoux, A. Nowicki and
K. Philippot, in Nanoparticles in Catalysis, ed. D. Astruc, Wiley-VCH,
Weinheim, 2008, pp. 349-388; (c) D. Astruc, F. Lu and J. Ruiz Aranzaes,
Angew. Chem., Int. Ed., 2005, 44, 7852; (d) L. Dur a´ n Pach o´ n and G.
Rothenberg, Appl. Organomet. Chem., 2008, 22, 288; (e) H. B o¨ nnemann
=
0.05 M, S/C = 20) was added to the reaction mixture under
nitrogen. The reaction mixture was then vigorously stirred in the
presence of hydrogen at 20bar at room temperature for 18 h in
8
470 | Dalton Trans., 2010, 39, 8464–8471
This journal is © The Royal Society of Chemistry 2010

and R. M. Richards, Eur. J. Inorg. Chem., 2001, 2455; (f) A. Roucoux,
J. Schulz and H. Patin, Chem. Rev., 2002, 102, 3757; (g) J. D. Aiken III
and R. G. Finke, J. Mol. Catal. A: Chem., 1999, 145, 1.
24 A. F u¨ rstner, R. Martin, H. Krause, G. Seidel, R. Goddard and C. W.
Lehmann, J. Am. Chem. Soc., 2008, 130, 8773.
25 (a) B. Bogdanovi c´ and M. Schwickardi, Angew. Chem., Int. Ed., 2000,
39, 4610; (b) L. E. Aleandri, B. Bogdanovi c´ , P. Bons, C. D u¨ rr, A.
Gaidies, T. Hartwig, S. C. Huckett, M. Lagarden, U. Wilczok and R. A.
Brand, Chem. Mater., 1995, 7, 1153; (c) G. Siedlaczek, M. Schwickardi,
U. Kolb, B. Bogdanovi c´ and D. G. Blackmond, Catal. Lett., 1998, 55,
67.
1
1
2 (a) K. T. Wu, Y. D. Yao, C. R. C. Wang, P. F. Chen and E. T. Yeh,
J. Appl. Phys., 1999, 85, 5959; (b) E. W. Wong, M. J. Bronikowski,
M. E. Hoenk, R. S. Kowalczyk and B. D. Hunt, Chem. Mater., 2005,
1
7, 237; (c) Y. W. Jun, Y. M. Huh, J. S. Choi, J. H. Lee, H. T. Song,
S. Kim, S. Yoon, K. S. Kim, J. S. Shin, J. S. Suh and J. Cheon, J. Am.
Chem. Soc., 2005, 127, 5732; (d) S. Mornet, S. Vasseur, F. Grasset and
E. Duguet, J. Mater. Chem., 2004, 14, 2161; (e) F. Li, C. Vipulanandan
and K. K. Mohanty, Colloids Surf., A, 2003, 223, 103; (f) W. X. Zhang,
J. Nanopart. Res., 2003, 5, 323; (g) D. Huber, Small, 2005, 1, 482.
3 (a) R. B. Bedford, M. Betham, D. W. Bruce, S. A. Davis, R. M. Frost
and M. Hird, Chem. Commun., 2006, 1398; (b) Y. Moglie, E. Mascar o´ ,
F. Nador, C. Vitale and G. Radivoy, Synth. Commun., 2008, 38, 3861;
26 We verified that the direct hydrogenation of 1-octene also follows a first
order kinetic (see supplementary information†).
27 (a) G. Dazinger and K. Kirchner, Organometallics, 2004, 23, 6281;
(b) V. Guerchais, J. Chem. Soc., Chem. Commun., 1990, 534; (c) M.
Brookhart and W. B. Studabaker, Chem. Rev., 1987, 87, 411.
28 (a) A. Klose, J. Hesschenbrouck, E. Solari, M. Latronico, C. Floriani,
N Re, A. Chiesi-Villa and C. Rizzoli, J. Organomet. Chem., 1999, 591,
45; (b) A. I. F. Ven aˆ ncio, L. M. D. R. S. Martins and A. J. L. Pombeiro,
J. Organomet. Chem., 2003, 684, 315; (c) G. Albertin, P. Agnoletto
and S. Antoniutti, Polyhedron, 2002, 21, 1755; (d) S. K. Russell, E.
Lobkovsky and P. J. Chirik, J. Am. Chem. Soc., 2009, 131, 36.
29 A. Uzun and B. C. Gates, Angew. Chem., Int. Ed., 2008, 47,
9245.
(
c) C. He, G. Tian, Z. Liu and S. Feng, Org. Lett., 2010, 12, 649; (d) See
also ref. 11g.
1
1
1
1
4 P.-H. Phua, L. Lefort, J. A. F. Boogers, M. Tristany and J. G. de Vries,
Chem. Commun., 2009, 3747.
5 D. Farrell, S. A. Majetich and J. P. Wilcoxon, J. Phys. Chem. B, 2003,
1
07, 11022.
6 O. Margeat, F. Dumestre, C. Amiens, B. Chaudret, P. Lecante and M.
Respaud, Prog. Solid State Chem., 2005, 33, 71.
30 (a) L. S. Ott and R. G. Finke, Inorg. Chem., 2006, 45, 8382; (b) L. S.
Ott and R. G. Finke, Coord. Chem. Rev., 2007, 251, 1075.
31 B. D. Cullity and C. D. Graham in Introduction to Magnetic Materials,
Wiley, John & Sons, Incorporated, 2009, pp.183.
32 (a) J. M. D. Coey, Phys. Rev. Lett., 1971, 27, 1140; (b) R. H. Kodama
and A. E. Berkowitz, Phys. Rev. B: Condens. Matter Mater. Phys., 1999,
59, 6321; (c) J. Bansmann, S. H. Baker, C. Binns, J. A. Blackman, J.-
P. Bucher, J. Dorantes-D a´ vila, V. Dupuis, L. Favre, D. Kechrakos, A.
Kleibert, K.-H. Meiwes-Broer, G. M. Pastor, A. Perez, O. Toulemonde,
K. N. Trohidou, J. Tuaillon and Y. Xie, Surf. Sci. Rep., 2005, 56,
189.
33 (a) I. M. L. Billas, J. A. Becker, A. Ch aˆ telain and W. A. de Heer, Phys.
Rev. Lett., 1993, 71, 4067; (b) M. L. Tiago, Y. Zhou, M. M. G. Alemany,
Y. Saad and J. R. Chelikowsky, Phys. Rev. Lett., 2006, 97, 147201.
34 F. Dumestre, B. Chaudret, C. Amiens, P. Renaud and P. Fejes,, Science,
2004, 303, 821.
7 (a) H. B o¨ nnemann, W. Brijoux, R. Brinkmann, R. Fretzen, T. Joussen,
R. Koppler, B. Korall, P. Neiteler and J. Richter, J. Mol. Catal., 1994,
8
6, 129; (b) G. N. Glavee, K. J. Klabunde, C. M. Sorensen and G. C.
Hadjipanayis, Inorg. Chem., 1995, 34, 28; (c) K. C. Huang and S. H.
Ehrman, Langmuir, 2007, 23, 1419.
1
1
2
8 Our high throughput experimentation setup is composed of
a
liquid handling robot (Zinsser Lissy: http://www.zinsser-
analytic.com/Catalogue/exeProduct/?id=67) placed within a glovebox
and a parallel hydrogenation reactor (Premex: http://www.premex-
reactorag.ch/index.php?page=464) where up to 96 reactions can be
2
carried out at the same time (Pressure max (100 bar H , 5 mL scale).
9 (a) Y. Takegami, T. Ueno and T. Fujii, Bull. Chem. Soc. Jpn., 1966, 39, 1;
b) A. R. Brodskii, V. F. Vozdvizhenskii, N. F. Noskova, A. S. Khlystov,
O. M. Pakhorukova and D. V. Sokolskii, J. Organomet. Chem., 1979,
78, 325; (c) A. R. Brodskii, V. F. Vozdvizhenskii, N. F. Noskova,
(
1
35 (a) J. L. Dormann, D. Fiorani and E. Tronc, Adv. Chem. Phys., 1997,
98, 283; (b) X. Batlle and A. Labarta, J. Phys. D: Appl. Phys., 2002, 35,
R15; (c) R. Skomski, J. Phys.: Condens. Matter, 2003, 15, R841.
36 C. de Juli a´ n Fern a´ ndez, Phys. Rev. B: Condens. Matter Mater. Phys.,
2005, 72, 054438.
A. S. Khlystov, O. M. Pakhorukova and D. V. Sokolskii, J. Organomet.
Chem., 1979, 179, 139.
0 (a) Y. Takegami, T. Ueno and T. Fujii, Bull. Chem. Soc. Jpn., 1969, 42,
1
(
1663; (b) W. R. Kroll and N. J. Linden, US patent 3 323 902, 1967;
c) California Research Corporation Patent 1 390 570, 1965.
1 (a) M. Kumada and J. K. Kochi, J. Am. Chem. Soc., 1971, 93, 1487;
b) J. K. Kochi, Acc. Chem. Res., 1974, 7, 351.
2 (a) R. S. Smith and J. K. Kochi, J. Org. Chem., 1976, 41, 502; (b) S. M.
Neumann and J. K. Kochi, J. Org. Chem., 1975, 40, 599; (c) R. B.
Allen, R. G. Lawler and H. R. Ward, J. Am. Chem. Soc., 1973, 95,
37 J. A. Widegren and R. G. Finke, J. Mol. Catal. A: Chem., 2003, 198,
2
2
317.
(
38 G. M. Whitesides, M. Hackett, R. L. Brainard, J. P. P. M. Lavalleye,
A. F. Sowinski, A. N. Izumi, S. S. Moore, D. W. Brown and E. M.
Staudt, Organometallics, 1985, 4, 1819.
39 F. Lihl, Z. Metallk., 1953, 44, 160.
1
3
692; (d) R. G. Lawler and P. Livant, J. Am. Chem. Soc., 1976, 98,
710; (e) J. Kleimark, A. Hedstr o¨ m, P.-F. Larsson, C. Johansson and
40 A. F. Meindertsma, M. M. Pollard, B. L. Feringa, J. G. de Vries and
A. J. Minnaard, Tetrahedron: Asymmetry, 2007, 18, 2849.
41 J. Garcia-Ant o´ n, M. Rosa Axet, S. Jansat, K. Philippot, B. Chaudret,
T. Pery, G. Buntkowsky and H-. H. Limbach, Angew. Chem., Int. Ed.,
2008, 47, 2074.
P.-O. Norrby, ChemCatChem, 2009, 1, 152.
3 G. Cahiez, V. Habiak, C. Duplais and A. Moyeux, Angew. Chem., Int.
Ed., 2007, 46, 4364.
2
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 8464–8471 | 8471