On the use of amine-borane complexes to synthesize iron nanoparticles

Article abstract of DOI:10.1002/chem.201204574

The effectiveness of amine-borane as reducing agent for the synthesis of iron nanoparticles has been investigated. Large (2-4 nm) Fe nanoparticles were obtained from [Fe{NACHTUNGTRENUNG(SiMe₃)₂}₂]. Inclusion of boron in the nanoparticles is clearly evidenced by extended X-ray absorption fine structure spectroscopy and M?ssbauer spectrometry. Furthermore, the reactivity of amine-borane and amino-borane complexes in the presence of pure Fe nanoparticles has been investigated. Dihydrogen evolution was observed in both cases, which suggests the potential of Fe nanoparticles to promote the release of dihydrogen from amine-borane and amino-borane moieties. Copyright

Full text of DOI:10.1002/chem.201204574

FULL PAPER
DOI: 10.1002/chem.201204574
On the Use of Amine–Borane Complexes To Synthesize Iron Nanoparticles
Frꢀdꢀric Pelletier,[a] Diana Ciuculescu,[a] Jean-Gabriel Mattei,[b, c] Pierre Lecante,[c]
Marie-Josꢀ Casanove,[c] Nader Yaacoub,[d] Jean-Marc Greneche,[d]
Carolin Schmitz-Antoniak,[e] and Catherine Amiens*[a]
Abstract: The effectiveness of amine–
borane as reducing agent for the syn-
thesis of iron nanoparticles has been
investigated. Large (2–4 nm) Fe nano-
particles were obtained from [Fe{N-
ACHTUNGTRENNUNG(SiMe3)2}2]. Inclusion of boron in the
nanoparticles is clearly evidenced by
extended X-ray absorption fine struc-
ture spectroscopy and Mçssbauer spec-
trometry. Furthermore, the reactivity of
amine–borane and amino–borane com-
plexes in the presence of pure Fe nano-
particles has been investigated. Dihy-
drogen evolution was observed in both
cases, which suggests the potential of
Fe nanoparticles to promote the re-
lease of dihydrogen from amine–
borane and amino–borane moieties.
Keywords: boranes · dehydrogena-
tion · inclusion compounds · iron ·
nanoparticles
Introduction
pending on the nature of the amine. As recently reviewed,[5]
the range of metal nanoparticles synthesized that way spans
from noble metals, such as gold or iridium, to more electro-
positive ones, such as cobalt, nickel, or iron. Amine–borane
derivatives are thus reported as promising mild reducing
agents. In previous work, we investigated the use of diiso-
propylamine–borane (AeB) to produce bimetallic FeRh and
CoRh nanoparticles[6] in an integrated two-step process: the
reaction between AeB and an amido iron or cobalt complex
generated diisopropylamino–borane (AoB) and the dihydro-
gen quantity necessary to reduce the Rh precursor on top of
the first-formed Fe or Co seeds. Separation in time of the
formation of the iron seeds and of the rhodium deposit fa-
vored a core–shell chemical distribution in the bimetallic
nano-objects. The final size of the so-formed nanoparticles
was around 1.9 nm, which suggested extremely small inter-
mediate iron or cobalt seeds that we could only partially
characterize at the time.[7] It is well established that ele-
ments such as cobalt, iron, and nickel display a huge tenden-
cy for boron incorporation when the nanoparticles are pre-
pared by borohydride reduction of metal salts.[8] Similarly,
reaction of FeCl3 with ammonia–borane led to the forma-
tion of FeB nanoparticles.[9] As boron inclusion leads to im-
portant changes in the magnetic properties of the nanoparti-
cles and also in their reactivity, it is thus important to scruti-
nize the possible incorporation of boron, especially in the
case of iron, which constitutes the cheapest magnetic and
catalytic material.
With the development of nanosciences, new strategies to
reach advanced nanomaterials have been sought. In particu-
lar, due to their unique electronic properties, metal nanopar-
ticles are key nanomaterials for many applications in cataly-
sis,[1] electronics,[2,3] or even the biomedical field.[4] Conse-
quently, considerable research effort has been devoted to
the invention of cheap and reliable synthesis routes. In this
context, in the last 10 years, amine–borane derivatives have
been explored for the synthesis of metal nanoparticles. They
proved to work in both organic and aqueous solutions, and
even in the solid state, with tunable reducing activity de-
[a] F. Pelletier, Dr. D. Ciuculescu, Prof. C. Amiens
CNRS, LCC (Laboratoire de Chimie de Coordination)
205 route de Narbonne, 31077 Toulouse (France) and
Universitꢀ de Toulouse, UPS, INPT, LCC, 31077 Toulouse (France)
Fax : (+33)561553003
E-mail: catherine.amiens@lcc-toulouse.fr
[b] J.-G. Mattei
CNRS, Centre dꢁElaboration des Matꢀriaux et dꢁEtudes Structurales
BP 94347, 29 rue J. Marvig, 31055 Toulouse (France) and
Universitꢀ de Toulouse, UPS, 31055 Toulouse (France)
[c] J.-G. Mattei, Dr. P. Lecante, Dr. M.-J. Casanove
Universitꢀ de Pau et des Pays de lꢁAdour, CNRS, IPREM
ECP, UMR 5254, 64053 Pau 09 (France)
[d] Dr. N. Yaacoub, Dr. J.-M. Greneche
LUNAM Universitꢀ du Maine
To shed some light on this issue, we investigated the syn-
thesis of iron nanoparticles from AeB more thoroughly. The
procedure was strongly inspired by that in reference [6]
apart from the absence of additional ligand during the syn-
thesis, which might have interfered during characterization.
Notably, given the bulkiness of the two isopropyl substitu-
ents on nitrogen, the corresponding AoB does not form
dimers or polymeric byproducts in solution,[10] again facili-
Institut des Molꢀcules et Matꢀriaux du Mans IMMM
UMR CNRS 6283, 72085 Le Mans (France)
[e] Dr. C. Schmitz-Antoniak
Fakultꢂt fꢃr Physik and
Center for Nanointegration Duisburg-Essen (CeNIDE)
Universitꢂt Duisburg-Essen
Lotharstrasse 1, 47048 Duisburg (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204574.
Chem. Eur. J. 2013, 19, 6021 – 6026
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6021

tating the characterization steps. Verifying the presence of
boron inside ultrafine particles that are often poorly crystal-
lized is not an easy task. Herein, a combination of techni-
ques (extended X-ray absorption fine structure spectroscopy
(EXAFS) and 57Fe Mçssbauer spectrometry) has been used
to reveal the presence of boron.
Results
The main characteristics of the samples—reaction conditions
and morphology—are reported in Table 1. TEM images of
the nanoparticles and size histograms can be found in the
Supporting Information (Figure SI1).
Table 1. Preparation conditions and sizes for samples 1–9.
Fe sample
Conditions
Size [nm][a]
1
AeB (2 equiv)
3–3.5
2
3
4
5
AeB/RT (0.16 equiv) then H2
AeB/RT (0.5 equiv) then H2
AeB/RT (1 equiv) then H2
AeB/RT (1.6 equiv) then H2
AoB (1 equiv)
2.8 (0.5)
3.4 (0.7)
3.5 (0.7)
4 (0.9)
2.9 (0.7)
1.8 (0.3)
1.8 (0.4)
2.2 (0.5)
6
7[b]
8
H2
sample 7, AeB (1 equiv)
sample 7, AoB (1 equiv)
9
[a] From a Gaussian fit of the size distribution observed by TEM (s
given in brackets) for nonagglomerated samples. [b] From ref. [16].
Figure 1. a) Mçssbauer spectra recorded at 10 K, 0 T or 10 K, 8 T. b) Hy-
perfine field distributions Beff =effective field, Bhyp =hyperfine field for
Direct formation of nanoparticles in the presence of AeB or
AoB: Sample 1 was obtained by following an already pub-
lished procedure,[6] by reacting a stoichiometric amount of
*
*
sample 1; : 10 K, 0 T, : 10 K, 8 T. Beff =effective field, Bhyp =hyperfine
field.
AeB with [Fe{N
nanoparticles obtained were strongly agglomerated with an
estimated diameter in the range 3–3.5 nm.
(SiMe3)2}2] according to Scheme 1. The
bauer spectra recorded at 10 K on sample 1 are illustrated in
Figure 1 (similar ones were obtained for sample 4, not
shown here, and give rise to the same conclusions). The
zero-field Mçssbauer spectrum consists of a weakly asym-
metrical magnetic sextet with broadened and overlapped
lines; one does consider a distribution of hyperfine fields
linearly correlated to that of isomer shift.
Samples 2–5 were obtained by a two-step procedure. In
The distribution exhibits a prevailing Gaussian-like peak
centered at around 27 T (with a mean isomer shift value of
d=0.28 mmsÀ1) and a small peak located at about 39 T (d=
0.32 mmsÀ1; about 2–3 Fe at%), whereas the lower-field
peak results from a fitting artifact (corresponding to inter-
mediate lines). In the presence of an external magnetic field
oriented parallel to the g beam, it can be clearly observed
that the intermediate lines disappear, which allows one to
conclude that the magnetic moments are ferromagnetically
coupled. In addition, the distribution of effective fields is ex-
actly shifted by 7.2 T from the hyperfine field distribution,
which suggests that the ferromagnetic domains are weakly
canted with respect to the external magnetic field. From the
hyperfine field and isomer shift values, the main contribu-
tion is attributed unambiguously to some FeB amorphous
alloy in which the B content is close to 25 at% and the
structure consists of a dense polytetrahedral atomic packing
(for comparison the mean values of hyperfine field and
Scheme 1. Reaction between AeB and [Fe{NACTHNUTRGNEUNG
(SiMe3)2}2].[6]
the first step, 0.16, 0.5, 1.1, or 1.6 equivalents of AeB, re-
spectively, were reacted with [Fe{N(SiMe3)2}2] followed by a
AHCTUNGTRENNUNG
hydrogenation step at 1108C designed to convert the un-
reacted iron precursor. TEM analysis shows better-dispersed
nanoparticles with an average diameter of 2.8, 3.4, 3.5, and
4 nm, respectively, for samples 2–5. Samples 1 and 4 were
further analyzed by Mçssbauer spectrometry. Typical Mçss-
6022
www.chemeurj.org
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6021 – 6026

Amine–Borane Complexes
FULL PAPER
isomer shift reported for Fe75B25 amorphous alloys are
27 T[11–13] and 0.25 mmsÀ1,[12–14] respectively). The mean hy-
perfine field value is consistent with that of Fe surrounded
by B whereas the width of the distribution is consistent with
a chemically homogeneous system; indeed, a nonhomogene-
ous distribution of B within the metallic matrix with B-rich
and B-poor regions would give rise to a broadened and non-
Gaussian hyperfine field distribution. The second minor
contribution associated with a larger isomer shift value can
be assigned to Fe species with lower electron density, that is,
having suffered a partial oxidation process.[15]
Sample 6 was prepared by reacting 1 equivalent of AoB
with [Fe{NACHTUNGTRENNUNG(SiMe3)2}2], at 1108C to ensure completion of the
reaction in a reasonable time. The nanoparticles display an
average size of 2.9 nm. Samples 1 and 4–6 all display polyte-
trahedral atomic packing, as shown by wide-angle X-ray
scattering (WAXS) investigation with coherence lengths of
1.2, 1.6, 1.7, and 1.1 nm for samples 1 and 4–6 (Figure SI3 in
the Supporting Information), respectively.
Figure 2. EXAFS oscillations weighted with k3 and the corresponding
Fourier transforms (FTs) for samples 1 and 7 and oxidized sample 7, that
is, after exposure to air. Note the different scaling for the oxidized
sample 7. Due to the EXAFS phase shift, the radial distance obtained
after Fourier transformation is not the geometric distance between
atoms.
Action of AeB and AoB on preformed Fe nanoparticles: As
a reference sample (hereafter labeled sample 7) pure iron
nanoparticles with diameters of 1.8 nm were produced ac-
cording to an already published procedure avoiding the use
analyzed (Figure 3). It can be seen that the oxide contribu-
tion is enhanced. Additionally, the shape of the FT related
to sample 1 now looks significantly different from that of
sample 7 in the range between 0.2 and 0.3 nm.
of boron derivatives: hydrogenation of [Fe{N
A
For the k1-weighted data, it is not possible to obtain a sim-
ilar FT for these two samples by just scaling one of the FTs
by a constant factor. The contribution of oxygen has already
decreased in this region of radial distances, which might in-
dicate the presence of another light element at similar radial
distances to the regular lattice sites of the polytetrahedral
Fe structure. However, it is not possible to further specify
the type of backscattering atoms in a straightforward
manner by using the standard Fourier-based analysis. A pos-
sible approach to identify these atoms is given by the wave-
let transform (WT). Due to the different k dependence of
backscattering amplitude of Fe, B, and O (see Figure SI4 in
the Supporting Information), the WT offers the possibility
to visualize the chemical environment of absorbing atoms
One equivalent of AeB (sample 8) or AoB (sample 9) was
reacted with nanoparticles of sample 7. In both cases dihy-
drogen evolution was observed upon addition of the boron
derivative to the iron nanoparticles in solution. The diame-
ter of the nanoparticles was checked at the end of the reac-
tion. No variation could be detected given the error inherent
to the analysis of TEM images at this scale, thus ruling out
any significant atom redistribution or coalescence process.
Measurements of the EXAFS at the FeK absorption edge
were carried out on the most significant samples, namely
samples 1, 4, and 6–9, to extract the crystallographic struc-
ture and type of backscattering atoms around Fe. As an ex-
ample, Figure 2 shows the k3-weighted EXAFS oscillations
for sample 1 compared with sample 7 (pure Fe nanoparticles
produced by hydrogenation) and the oxidized sample 7 after
exposure to air. In the lower panel of Figure 2, the corre-
sponding Fourier transforms (FTs) are presented. Note that
due to the EXAFS phase shift, the radial distance is not the
geometric distance between atoms. Despite the slightly dif-
ferent amplitudes, the FT calculated from the EXAFS of
sample 1 is quite similar to that of sample 7 and close to
that of amorphous Fe,[17] thus indicating the polytetrahedral
packing already observed by WAXS. The two peaks at small
radial distances (below 0.16 nm) may indicate traces of
oxide. EXAFS data and FTs of samples 1, 4, and 6–9 are
shown in Figures SI5 and SI6, respectively, in the Supporting
Information.
However, the k3 weighting of EXAFS data suppresses
contributions of light elements to the total EXAFS signal,
as discussed in the Supporting Information. To shift the
focus to the light elements incorporated in the sample, the
k1-weighted EXAFS and the corresponding FTs were also
Figure 3. EXAFS oscillations weighted with k1 and the corresponding
FTs for samples 1 and 7 and oxidized sample 7, that is, after exposure to
air. Note the different scaling for the oxidized sample 7. Due to the
EXAFS phase shift, the radial distance obtained after Fourier transfor-
mation is not the geometric distance between atoms.
Chem. Eur. J. 2013, 19, 6021 – 6026
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6023

C. Amiens et al.
itself proceeds either from atom addition at the surface of
the seeds or from aggregation of a few seeds to form a
nanoparticle, and it is often difficult to tell which mechanism
is at play. As such, evidence of the aggregation process can
only be obtained when the synthesis is carried out at mild
temperatures. Indeed, elevated temperature favors recon-
struction of the possible crystallographic defects present in
the metal lattice and coalescence of aggregated seeds into a
well-crystallized nanoparticle is efficient; thus any trace of
possible aggregation of the initial seeds is lost. However,
when the synthesis is carried out at mild temperatures, it is
possible to trap this intermediate stage at which grain boun-
daries are still present between the initial seeds. For exam-
ple, Pt nanoparticles coalesce in solution at room tempera-
ture into multiply twinned wormlike structures, with inner
grain sizes close to the size of the initial nanoparticles.[19]
The nanoparticles in sample 1 present a coherence length
(1.6 nm) much shorter than their overall diameter (3–
3.5 nm). Given the high tendency for agglomeration ob-
served from TEM images, this discrepancy between the
length of the crystalline domains and that of the nanoparti-
cles can be explained as the result of growth governed by
coalescence of seeds. This notion is in agreement with the
quasi-instantaneous reaction between AeB and [Fe{N-
Figure 4. b–d) WT compared with the a) FT of the k1-weighted EXAFS
data for sample 1. Due to the EXAFS phase shift, the radial distance ob-
tained after wavelet transformation is not the geometric distance be-
tween atoms.
(SiMe3)2}2] at a temperature as low as À938C (melting point
of toluene), which is bound to induce a burstlike nucleation
(i.e., burstlike consumption of the iron precursor under stoi-
chiometric conditions). In the absence of long-chain stabiliz-
ing agents, aggregation and coalescence of the seeds into
larger polycrystalline nanoparticles would naturally follow.
As a comparison, when the synthesis was carried out in the
presence of tetramethylpiperidine, it afforded well-dispersed
nanoparticles of average size (1.7 nm).[6] The aggregation
process is further supported in this specific case by the
atomic packing adopted by the seeds (polytetrahedral in
nature), the formation and growth of which was already sug-
gested to progress through aggregation of small units.[20]
As AoB derivatives have already been reported for the
synthesis of metal nanoparticles,[21] we also investigated the
use of AoB to produce Fe nanoparticles. The average size
measured in this case is 3 nm, close to the average size ob-
served in sample 4, with crystalline domains not exceeding
1.1 nm. This evidences once again that coalescence processes
are at play during the growth of the particles.
Surprisingly, in sample 7, which contains reference iron
nanoparticles prepared by direct hydrogenation of [Fe{N-
AHCTUNGTREN(GUNN SiMe3)2}2], the average size of the nanoparticles (1.8 nm) is
identical to that of the crystalline domains (within experi-
mental error). This finding indicates that the use of borane
derivatives most probably promotes coalescence of the iron
seeds, as it favors the formation of nanoscale iron–boron
metallic glass (see below), which is an amorphous material
with densely packed polytetrahedral units.
directly without any structural assumptions or further data
treatment. Figure 4 displays the result for sample 1 obtained
by this method (see Figure SI7 in the Supporting Informa-
tion for the WT of EXAFS data of samples 1, 4, and 6–9).
For comparison, the FT for this sample is presented in Fig-
ure 4a. The FT is equivalent to the sum over all k values of
the WT. Here, the k-weighted EXAFS data have been used
for the WT to emphasize the light elementsꢁ contributions
to the overall signal. To better visualize the B content in the
samples, the contribution of pure Fe and Fe surrounded by
O or N (Figure 4c) to the total WT (Figure 4a) was sub-
tracted.[18] In the difference signal, the B contribution is thus
well observed (Figure 4d). The two maxima reflect the com-
plex k dependence of the B backscattering amplitude (see
Supporting Information for details). In all samples but
sample 7, B is found in iron first neighbours shell (Fig-
ure SI7 in the Supporting Information). This EXAFS study
only gives qualitative evidence of the presence of B. Howev-
er, a rough estimation of the B content by comparison with
the WT of EXAFS measured on an Fe2B reference sample
yields (15Æ5) at% B in sample 1. All other samples studied
by means of EXAFS in this work have less B included.
Discussion
Growth mechanism (samples 1–5, with 6 and 7 as controls):
It is well established that the final size of nanoparticles de-
pends first on the nucleation step; each nucleus develops
into seeds that can grow further into nanoparticles. Growth
Samples 2 to 5 were produced in a two-step process ini-
tially designed to control the size of the nanoparticles. AeB
was added under substoichiometric conditions to produce
iron seeds upon which more iron atoms were supposed to
6024
www.chemeurj.org
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6021 – 6026

Amine–Borane Complexes
FULL PAPER
deposit during reduction of the excess iron precursor. How-
ever, despite the fact that the iron precursor was successfully
and completely decomposed at the end of the reaction, the
average diameter of the nanoparticles was similar to that in
sample 1 (within experimental error) although with an en-
larged size distribution. These observations are in agreement
with the fact that coalescence of the seeds is a dominant
growth process even when they are produced in limited
quantity (low concentration of seeds).
the nanoparticles of sample 1 most probably occurs at the
early stage of nanoparticle formation, that is, at the nuclea-
tion stage, at least before coalescence of the seeds. This is
further supported by the chemical homogeneity of the
sample revealed by Mossbauer spectrometry.
These results also suggest that pure Fe nanoparticles
might be efficient catalysts for dehydrogenation of AeB and
AoB derivatives. As a promising source of dihydrogen, de-
hydrogenation of amine–borane complexes is the subject of
intense research.[22] In water-free conditions,[23] it has been
shown in certain cases that metal nanoparticles form during
the catalytic dehydrogenation of amine–borane by organo-
metallic complexes.[24] Further work has shown the catalytic
efficiency of nanoclusters and particles in organic solutions,
especially those of Rh[25] and Ru.[26] However, even though
some iron complexes have been recently studied for the
photoactivated dehydrogenation of AeB,[27] Fe nanoparticles
have only been investigated under hydrolytic conditions.[28]
To the best of our knowledge, the closest related system is
that published by He et al.,[9] who reported the use of FeB
nanoparticles for the thermolysis of ammonia borane in the
solid state. Notably, the fate of AoB has been much less
studied than that of AeB.[29] However, recent studies have
Origin of B inclusion: Reduction of [Fe{NACTHNUTRGNEG(UN SiMe3)2}2] by
AeB and AoB derivatives led to samples 1–6. Surprisingly,
the presence of boron in close proximity to iron atoms has
been evidenced independently of reaction conditions. More
specifically, boron inclusion was clearly demonstrated by
Mçssbauer spectrometry in samples 1 and 4. Given the
mechanism depicted in Scheme 1, the use of AeB as a re-
ducing agent should not give B inclusion directly. Indeed,
trapping of the byproducts of the reaction at low tempera-
ture indicated the formation of a stoichiometric amount of
AoB.[6] However, the presence of AoB during warming of
the reacting medium might very well be at the origin of the
observed boron content. Thus, to ascertain the origin of
boron inclusion, we investigated the reactivity of preformed
pure iron nanoparticles with AeB and AoB.
À
shown the double activation of the B H bonds of the sim-
[30]
À
plest amino–borane NH2 BH2,
and borylene complexes
Upon addition of AeB to a solution of preformed iron
nanoparticles (sample 7) at room temperature, fast dihydro-
gen evolution was observed evidencing the efficiency of
these iron nanoparticles for dehydrogenation of AeB, as de-
picted in step 1, Scheme 2. However, the reactivity does not
stop with the release of AoB, as incorporation of B is re-
vealed from EXAFS measurements.
In a second experiment, AoB was reacted with the pre-
formed iron nanoparticles. Here again, dihydrogen evolution
was observed, thus evidencing the activity of the iron nano-
particles towards dehydrogenation of the amino–borane
moiety (step 2, Scheme 2). EXAFS investigation again indi-
cated the incorporation of boron in the nanoparticles
(sample 9). Based on these observations, the presence of
boron in the nanoparticles of sample 1 is attributed to a side
reaction involving AoB, the byproduct of the reaction, and
the iron nanoparticles. Moreover, as boron inclusion is less
pronounced when AoB is reacted with preformed nanopar-
ticles than when it is produced in situ, inclusion of boron in
have been demonstrated to be possible intermediates in
their dehydrogenation.[31] Such intermediates might form at
the surface of iron nanoparticles, but it is not clear how
B inclusion can proceed from there. The study of the dehy-
drogenation of AeB and/or AoB catalyzed by iron nanopar-
ticles is clearly outside the scope of this article, but the re-
sults reported herein show that iron nanoparticles might be
a cheap alternative of low toxicity, capable of generating
more than one equivalent of H2 per AeB unit.
Conclusion
The use of diisopropylamine–borane (AeB) as a cheap, mild
reducing agent for the synthesis of iron nanoparticles has
been investigated. Ultrafine iron particles (2–4 nm in size)
have been produced under various reaction conditions, with
the concomitant release of amino–borane and dihydrogen.
By a combination of complementary techniques we deter-
mined that the nanoparticles contained a non-negligible
amount of boron most probably resulting from the reaction
of the amino–borane byproduct at the surface of the first-
formed nanoparticles. This highlights the difficulty encoun-
tered in the search for alternative reducing agents for the
synthesis of pure iron nanoparticles, especially emphasizing
the role of the reaction byproducts. Up to now, dihydro-
gen[16] and amines[32] have been the sole reactants affording
pure iron nanoparticles. This study opens a route for the
preparation of iron boride nanoparticles and evidences the
activity of iron nanoparticles for dehydrogenation of both
AeB and AoB derivatives.
Scheme 2. Proposed steps for boron incorporation in reference iron nano-
particles (sample 7).
Chem. Eur. J. 2013, 19, 6021 – 6026
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6025

C. Amiens et al.
[4] R. R. Arvizo, S. Bhattacharyya, R. A. Kudgus, K. Giri, R. Bhatta-
charya, P. Mukherjee, Chem. Soc. Rev. 2012, 41, 2943.
[5] S. Babu Kalidindi, U. Sanyal, B. R. Jagirdar, ChemSusChem 2011, 4,
317.
[6] N. Atamena, D. Ciuculescu, G. Alcaraz, A. Smekhova, F. Wilhelm,
A. Rogalev, B. Chaudret, P. Lecante, R. E. Benfield, C. Amiens,
Chem. Commun. 2010, 46, 2453–2455.
[7] A. Glaria, PhD thesis, Universitꢀ Paul-Sabatier, Toulouse (France),
2007.
[8] G. N. Glavee, K. J. Klabunde, C. M. Sorensen, G. C. Hadjipanayis,
Inorg. Chem. 1995, 34, 28.
[9] T. He, J. Wang, G. Wu, H. Kim, T. Proffen, A. Wu, W. Li, T. Liu, Z.
Xiong, C. Wu, H. Chu, J. Guo, T. Autrey, T. Zhang, P. Chen, Chem.
Eur. J. 2010, 16, 12814.
[10] L. Pasumansky, D. Haddenham, J. W. Clary, G. B. Fisher, C. T. Gor-
alski, B. Singaram, J. Org. Chem. 2008, 73, 1898.
Experimental Section
General synthesis methods: All operations (material preparation, sam-
pling, and packaging) were carried out by using standard Fischer–Porter
bottle techniques and a glovebox under argon, commercial reactants, and
carefully dried and degassed solvents.
Compound 1: [Fe{NACHTUNGTRENNUNG(SiMe3)2}2] (100 mg, 0.266 mmol) was dissolved in tol-
uene (10 mL, H2O <3 ppm) in a Fischer–Porter bottle (200 mL) in a glo-
vebox. The homogeneous green solution was frozen under liquid nitrogen
and iPr2NHBH3 (2.2 equiv) in toluene (67 mg, 0.583 mmol in 4 mL tol-
uene) was transferred through a cannula into the Fischer–Porter bottle.
The mixture was allowed to warm to room temperature to afford a dark
solution. The mixture was further stirred overnight, and then toluene was
evaporated to afford a black sticky solid.
Compounds 2–5: [Fe{NACHTUNGTRENNUNG(SiMe3)2}2] (200 mg, 0.532 mmol) was dissolved in
toluene (18 mL, H2O <3 ppm) in a Fischer–Porter bottle (200 mL) in a
glovebox. The homogeneous green solution was frozen under liquid ni-
trogen and iPr2NHBH3 (0.16, 0.5, 1.1, or 1.6 equiv, respectively) in tol-
uene (4 mL) was transferred through a cannula into the Fischer–Porter
bottle. The mixture was allowed to warm to room temperature to afford
a dark brown solution. The solution was then pressurized under H2
(3 bar) and stirred at 1108C overnight. Then toluene was evaporated to
afford a black sticky solid.
[11] G. Barault, J.-M. Greneche, Solid State Commun. 1995, 96, 155.
[12] C. L. Chien, D. Musser, E. M. Gyorgy, R. C. Sherwood, H. S. Chen,
F. E. Luborsky, J. L. Walter, Phys. Rev. B 1979, 20, 283.
[13] C. L. Chien, K. M. Unruh, Phys. Rev. B 1982, 25, 5790.
[14] J. Dubois, G. L. Caer, J. P. Sꢀnateur, Solid State Commun. 1982, 43,
777.
[15] Mçssbauer experiments performed on both samples at 300 and 77 K
after a few weeks of aging clearly reveal the presence of ferric spe-
cies, thus demonstrating that an oxidation process occurs and con-
firming our conclusion.
Compound 6: [Fe{NACHTUNGTRENNUNG(SiMe3)2}2] (100 mg, 0.266 mmol) was dissolved in tol-
uene (18 mL, H2O <3 ppm) in a Fischer–Porter bottle (200 mL) in a glo-
vebox. The homogeneous green solution was frozen under liquid nitrogen
and iPr2NBH2 (1 equiv) diluted in toluene (30 mg, 0.266 mmol in 4 mL
[16] L.-M. Lacroix, S. Lachaize, A. Falqui, T. Blon, J. Carrey, M. Re-
spaud, F. Dumestre, C. Amiens, O. Margeat, B. Chaudret, P. Le-
cante, E. Snoeck, J. Appl. Phys. 2008, 103, 07D521.
toluene) was transferred through
a cannula into the Fischer–Porter
[17] G. J. Long, D. Hautot, Q. A. Pankhurst, D. Vandormael, F. Grand-
jean, J. P. Gaspard, V. Briois, T. Hyeon, K. S. Suslick, Phys. Rev. B
1998, 57, 10716.
bottle. The mixture was allowed to warm to room temperature to afford
a dark brown solution. The solution was then stirred at 1108C overnight.
Evaporation of toluene afforded a black sticky solid.
[18] The contribution of pure Fe and Fe surrounded by O or N is ob-
tained by a linear combination of the WT of EXAFS data on pure
Fe (sample 7) and an oxidized sample (sample 7 after exposure to
air), as explained in the Supporting Information.
[19] E. Ramirez, L. Eradꢅs, K. Philippot, P. Lecante, B. Chaudret, Adv.
Funct. Mater. 2007, 17, 2219.
[20] F. Dassenoy, M. J. Casanove, P. Lecante, M. Verelst, E. Snoeck, A.
Mosset, T. O. Ely, C. Amiens, B. Chaudret, J. Chem. Phys. 2000, 112,
8137.
[21] C. Barriꢅre, G. Alcaraz, O. Margeat, P. Fau, J. B. Quoirin, C.
Anceau, B. Chaudret, J. Mater. Chem. 2008, 18, 3084.
[22] C. W. Hamilton, R. T. Baker, A. Staubitz, I. Manners, Chem. Soc.
Rev. 2009, 38, 279.
[23] Hydrolytic dehydrogenation of AeB and AoB moieties is not con-
sidered herein, as the reaction conditions and thus mechanisms are
most probably completely different from what can be expected in
the water- and oxygen-free conditions used in this study.
[24] J. L. Fulton, J. C. Linehan, T. Autrey, M. Balasubramanian, Y. Chen,
N. K. Szymczak, J. Am. Chem. Soc. 2007, 129, 11936–11949.
[25] T. e. Ayvali, M. Zahmakiran, S. Ozkar, Dalton Trans. 2011, 40, 3584.
[26] M. Zahmakiran, T. e. Ayvali, K. Philippot, Langmuir 2012, 28, 4908.
[27] J. R. Vance, A. P. M. Robertson, K. Lee, I. Manners, Chem. Eur. J.
2011, 17, 4099.
[28] J.-M. Yan, X.-B. Zhang, S. Han, H. Shioyama, Q. Xu, Angew. Chem.
2008, 120, 2319; Angew. Chem. Int. Ed. 2008, 47, 2287.
[29] G. Alcaraz, S. Sabo-Etienne, Angew. Chem. 2010, 122, 7326; Angew.
Chem. Int. Ed. 2010, 49, 7170.
[30] V. Pons, R. T. Baker, N. K. Szymczak, D. J. Heldebrant, J. C. Line-
han, M. H. Matus, D. J. Grant, D. A. Dixon, Chem. Commun. 2008,
6597.
Compounds 7–9: [Fe{NACHTUNGTRENNUNG(SiMe3)2}2] (1.128 g, 3 mmol) was dissolved in me-
sitylene (60 mL, H2O <1 ppm) in a Fischer–Porter bottle (500 mL). The
green solution was pressurized under H2 (3 bar) and stirred overnight in
a 1508C oil bath. Then the black solution was distributed equally into
three Fischer–Porter bottles (200 mL). The first solution was evaporated
directly and recovered with a few spatulas of poly(2,6-dimethyl-1,4-phe-
nylenoxide) (PPO) to afford a black solid as a reference, sample 7. The
other two solutions were charged respectively with 1 equivalent of
iPr2NHBH3 (115 mg, 1 mmol) or iPr2NBH2 (113 mg, 1 mmol) with re-
spect to the iron content. After reaction they were treated the same way
as sample 7 to afford black solids (samples 8 and 9, respectively).
General measurement methods: Full details of TEM, WAXS, Mçssbauer
spectrometry, and EXAFS investigations are reported in the Supporting
Information.
Acknowledgements
We thank the EC for support through the Nanotrain 2 SUDOE program,
and through FP7/2007-2013 (grant no. 226716 at the light source facilities
DORIS III, HASYLAB/DESY). DESY is a member of the Helmholtz
Association (HGF). We thank E. Welter and R. Benfield for assistance
on Beamline C. This work also benefited from financial support from the
Agence Nationale pour la Recherche (ANR-09-BLAN-0002-01). C.S.-A.
thanks the Alexander von Humboldt Foundation for support through the
Feodor Lynen program, the EC (Nanoalloy COST-STSM, MP0903
Action), and University Paul-Sabatier for grants. We especially thank
Prof. M. Respaud for fruitful discussions.
[31] G. Alcaraz, U. Helmstedt, E. Clot, L. Vendier, S. Sabo-Etienne, J.
Am. Chem. Soc. 2008, 130, 12878.
[32] A. Meffre, S. Lachaize, C. Gatel, M. Respaud, B. Chaudret, J. Mater.
Chem. 2011, 21, 13464.
[1] Nanoparticles and Catalysis (Ed: D. Astruc), Wiley-VCH, Wein-
heim, 2007.
[2] G. Konstantatos, E. H. Sargent, Nature Nanotechnology 2010, 5, 391.
[3] S. Magdassi, M. Grouchko, A. Kamyshny, Materials 2010, 3, 4626.
Received: December 23, 2012
Published online: March 19, 2013
6026
www.chemeurj.org
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 6021 – 6026