Nickel phosphide nanocatalysts for the chemoselective hydrogenation of alkynes

Article abstract of DOI:10.1016/j.nantod.2011.12.003

Well-defined 25 nm nickel phosphide nanoparticles act as a colloidal catalyst for the chemoselective hydrogenation of terminal and internal alkynes. Cis-alkenes are obtained in mild conditions with good conversion and selectivity. The phosphorus inserted in the Ni-P nanoparticles is critical for the selectivity of the nanocatalyst. Mechanistic investigations using isotope labeling provide insight on the reactants interaction with the nanoparticles surface. They pinpoint the occurrence of CH bond cleavage in terminal alkynes during the reaction.

Full text of DOI:10.1016/j.nantod.2011.12.003

Nano Today (2012) 7, 21—28
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/nanotoday
RAPID COMMUNICATION
Nickel phosphide nanocatalysts for the
chemoselective hydrogenation of alkynes
Sophie Carencoa,b, Antonio Leyva-Pérezc, Patricia Concepciónc,
Cédric Boissièrea, Nicolas Mézaillesb,∗,1, Clément Sancheza,∗,
Avelino Cormac,∗,2
a Laboratoire de Chimie de la Matière Condensée, Collège de France, UPMC, CNRS UMR7574, Place Berthelot, 75005 Paris, France
b Laboratoire Hétéroéléments et Coordination, Ecole Polytechnique, CNRS UMR7653, Route de Saclay, 91128 Palaiseau, France
c Instituto de Tecnología Química, Universidad Politécnica de Valencia-Consejo Superior de Investigaciones Científicas, Avda. de
los Naranjos s/n, 46022 Valencia, Spain
Received 5 September 2011; received in revised form 30 November 2011; accepted 22 December 2011
Available online 31 January 2012
Summary Well-defined 25 nm nickel phosphide nanoparticles act as a colloidal catalyst for
the chemoselective hydrogenation of terminal and internal alkynes. Cis-alkenes are obtained
KEYWORDS
Nanocatalysis;
in mild conditions with good conversion and selectivity. The phosphorus inserted in the Ni—P
Chemoselective
nanoparticles is critical for the selectivity of the nanocatalyst. Mechanistic investigations using
hydrogenation;
isotope labeling provide insight on the reactants interaction with the nanoparticles surface.
Nickel phosphide;
They pinpoint the occurrence of C H bond cleavage in terminal alkynes during the reaction.
Alkynes
© 2012 Elsevier Ltd. All rights reserved.
hydrogenation;
Core—shell
nanoparticles
The selective hydrogenation of alkynes to alkenes is a rel-
evant industrial reaction with applications in fields such as
polymerization (removal of phenylacetylene in the styrene
feedstocks) [1] and fine chemical synthesis. As a cheaper
and greener alternative to the Pd-based catalysts such as
Lindlar and other Pd-poisoned catalysts [2], and more
recently Pd-alloyed nanoparticles [3], nickel-based hetero-
geneous catalysts such as Nickel Raney have been developed
[4]. However, they often show a lack of selectivity, yielding
mixtures of alkenes and alkanes, as olefins are also hydro-
genated [5—7].
The Ni-rich phosphides, having a metallic character
[8], have been used as heterogeneous catalysts mainly in
hydrotreating reactions. An early use of bulk Ni2P was
reported for the reduction of nitrobenzene [9], but was
not studied further. Besides, it was shown in the 1980s
and 1990s that phosphorus-containing amorphous nickel
catalysts exhibited attenuated hydrogenation activities
compared with Nickel, due to the electronic withdrawing
∗
Corresponding author. Tel.: +33 1 44 27 15 01;
fax: +33 1 44 27 15 04.
E-mail addresses: nicolas.mezailles@polytechnique.edu
(N. Mézailles), clement.sanchez@upmc.fr (C. Sanchez),
acorma@itq.upv.es (A. Corma).
1
Tel.: +33 1 69 33 44 02; fax: +33 1 69 33 44 40.
Tel.: +34 96 3877800; fax: +34 96 387 7809.
2
1748-0132/$ — see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.nantod.2011.12.003

22
S. Carenco et al.
effect of the phosphorus on the nickel, although the Ni/P
ratios were very high in these pioneering studies [10,11].
In the next decade, efforts concentrated on the crystal-
lized phases, as a wide variety of Ni—P structures had been
obtained, with stoichiometry ranging from Ni3P to NiP3 [12].
Their oxidation states and electronic properties strongly
depend on the Ni/P ratio [13]. Among the Ni-rich phases
(Ni3P, N i 12P5, Ni5P4, and Ni2P), the Ni2P phase has been most
studied, as it is highly stable in particular upon sulfur expo-
sure, which makes it very efficient for hydrodesulfurization
processes [14,15].
Scheme 1 Reaction conditions for the hydrogenation of
phenylacetylene catalyzed by Ni2P nanoparticles.
Interestingly, Oyama and Lee demonstrated that the
Ni2P surface catalyzes the hydrodesulfurization of the most
difficult substrate, namely the sterically hindered 4,6-
dimethyldibenzothiophene, through a hydrogenation route
that selectively involves one of the two nickel sites in the
hexagonal Ni2P lattice [16]. This hydrogenation is usually dis-
favored over the direct desulfurization of other substrates.
In that study, the experimental conditions were strenuous
(31 bar and 340 ◦C) yet in the typical range of the stud-
ies dealing with Ni—P heterogeneous catalysts. From an
external point of view, it indicates that crystalline Ni2P
could be a suitable catalyst for classical hydrogenation reac-
tions of more reactive substrates, under milder reaction
conditions. Very recently, it has been shown that TiO2-
promoted or CeO2-promoted bulk Ni2P is able to hydrogenate
phenylacetylene into styrene and ethylbenzene under softer
reaction conditions (150 ◦C, 10 bars) [17]. While the selec-
tivity for this simple substrate is interesting (up to 90%
selectivity for styrene), the operating conditions (high tem-
perature, high pressure) preclude the use of this process for
the transformations of polyfunctional or fine chemicals.
In the present work, the activity and selectivity of
well-defined Ni2P nanoparticles as a colloidal solution are
investigated for the chemoselective hydrogenation of a wide
range of alkynes under very mild reaction conditions (85 ◦C,
6 bar of H2) that are compatible with the presence of vari-
ous functional groups in the reactants. High conversions of
various alkynes are obtained in a few hours with a very good
selectivity (up to 100%) and excellent chemoselectivity. A
mechanistic study provides new insights for the selectivity
in the cis-alkene derivatives.
Additionally, the reaction also worked in THF at the lower
temperature of 60 ◦C, but slower: after 9 h of reaction, the
conversion reached only 70%, with a selectivity of 90%. For
further studies, dioxane was adopted as the solvent and the
temperature was set to 85 ◦C, as this temperature was prone
to yield better conversion than at 60 ◦C.
Recent studies indicate that molecular species leached
from the nanoparticles can be responsible for the catalytic
activity of nanoparticles [19]. In order to study this possibil-
ity, a filtration test was done: the reaction was stopped after
4 h at a conversion of 83%, the H2 pressure was released and
the nanoparticles were removed by centrifugation. Then,
the corresponding filtrate was heated back to 85 ◦C under
6 bar of fresh H2. No evolution of the composition mixture
was observed in the following hours, indicating that the Ni2P
nanoparticles and no leached metal, if any, were responsible
for the catalytic activity. To confirm the integrity of the cat-
alyst throughout the reaction, TEM and XRD analyses of the
isolated solid catalyst were performed. The used catalysts
were identical in size and shape to the initial nanoparticles
(see ESI).
A high selectivity for a single hydrogenation process was
observed for a variety of aryl- and alkyl-terminal alkynes
(Table 1). The reaction was found to be tolerant to vari-
ous functional groups including halides (entries 1 and 2),
alcohols (entry 3) and amines (entry 4). In order to further
The hydrogenation of a model alkyne, phenylacetylene,
was achieved under mild conditions in a batch reactor using
well-defined Ni2P nanoparticles. The spherical, monodis-
perse in size (25 nm diameter) and fully crystallized Ni2P
nanoparticles were obtained by reacting Ni monodispersed
nanoparticles with 1/8 P4, in a fully reproducible and
gram-scale synthesis (see ESI) [18]. The nanoparticles are
dense and single-crystalline, as shown before [23]. Reac-
tion conditions for the catalytic experiments are depicted
in Scheme 1.
Fig. 1 shows that phenylacetylene conversion of 98% was
reached within 6 h, with a very good selectivity for styrene
(96%). From this first set of observations, it appears that
the Ni2P catalyst more easily hydrogenates the alkyne func-
tion of the phenylacetylene than the alkene function of the
resulting styrene (Fig. 1). It must be noted that a two-step
reaction occurred where most of the alkyne was hydro-
genated in 6 h before the resulting alkene began to convert
to alkane after ca. 20 h (this feature will be discussed
later). Hence, a high alkene yield can easily be managed.
Figure 1 Hydrogenation of phenylacetylene catalyzed by Ni2P
nanoparticles: composition of the reaction mixture at different
times. Insert: Corresponding conversion of phenylacetylene and
selectivity for styrene. The lines are a guide to the eye.

Nickel phosphide nanocatalysts for the chemoselective hydrogenation
23
Table 1 Scope of the reaction.
Entry
Alkyne
Selectivity (%)
100
Yielda (%)
95
1
2
3
4
96
80
78
72
[60]
24
5
95
[76]b
6
7
8
85
82
33
46b
23 cis, 5 transc
[30]d
9
10
11
43
89
93
3
63 cis, 4 trans
[76] cise
12
100
19 cis
13
14
15
100
100
59
9 cis
3 cis
[52]f
The yield refers to the alkene. Reaction time is 14 h.
a
GC yield. Between brackets, isolated yield.
4% internal alkene hydrogenated.
b
c
Competitive hydrogenation of the terminal alkene: 26%.
d
Substrate (0.07 mmol), dioxane (0.12 mL), 0.1 equiv. of Ni2P, 2 bar of H2, 85 ◦C.
Catalyst: 0.2 equiv.
e
f
The alkene yield corresponds to diene (non isomerized, no enyne, as attested by the fragments analysis of the MS spectrum); 0.4 equiv.
of catalyst (0.2 equiv. per triple bond).
examine the alkyne-to-alkene selectivity, different enyne
substrates were prepared and tested. Firstly, a terminal
aryl alkyne was selectively hydrogenated to the corre-
sponding styrene derivative in the presence of an internal
trans-alkene in high yield (Table 1, entry 5). The same sat-
isfactory result was obtained when an alkyl counterpart was
used (Table 1, entry 6). A retinol derivative containing a
conjugated ␲-system of five alkenes with a single terminal
alkyne was hydrogenated to give a reasonable 30% yield of
the all-alkenyl compound (Table 1, entry 8). In contrast, the
introduction of nitro groups dramatically reduced the hydro-
genation of the alkyne (Table 1, entry 9), first because of its
electron withdrawing effect on this particular alkyne, sec-
ond because it competes with the alkyne as a substrate of

24
S. Carenco et al.
the hydrogenation reaction. Indeed, the hydrogenation of
the NO2 moiety, leading to yielding the amino derivative,
was observed [20].
In a second series of experiments, internal alkynes
that are less prone to hydrogenation due to steric and
electronic effects [21] were also investigated. When the
positions of the alkyne and alkene moieties were inter-
changed in the diaryl derivative (Table 1, entry 7, to
be compared with entry 5), a good selectivity was still
observed, although the conversion was lower. With Ni2P
as a nanocatalyst, other internal alkynes could be hydro-
genated to the corresponding cis-alkenes (Table 1, entries
10—14). The conversion remained good for alkynes with
limited steric hindrance (entries 10—11), while more hin-
dered alkynes reacted poorly, but always with a good
selectivity for the alkene (entries 12—14). Finally, an inter-
nal aryldiyne was doubly hydrogenated to the diene, with
a lower, though reasonable, selectivity (Table 1, entry
15). Altogether, the Ni2P nanoparticles catalyst exhibit
an interesting catalytic selectivity, combined with good
conversions, for the hydrogenation of multifunctional sub-
strates.
Nanoparticles catalysts can easily be separated from the
reaction mixture in order to be recycled. For this pur-
pose, phenylacetylene was chosen as the substrate and
the reaction was stopped after 6 h. The nanoparticles were
separated by centrifugation, washed with THF to remove
the remaining organic compounds, and reused in two addi-
tional runs of 6 h for phenylacetylene hydrogenation. A small
drop of ca. 10% was observed in terms of selectivity, but
the overall conversion remained very high (see ESI), even
slightly higher than in the first run. The nanocatalyst was
thus found to be recyclable in principle, even though indus-
trial application would require much more cycles to be
done.
As noted above, the hydrogenation of phenylacetylene
exhibited an induction period of ca. 2 h. This delay seemed
to diminish when using recycled nanoparticles. To shed light
on this phenomenon, complementary experiments were per-
formed by introducing sequentially H2 and the alkyne (H2
first or the alkyne first) with an interval of 2 h. Both exper-
iments did not modify the induction time, meaning that
neither preliminary reductive conditions, nor preliminary
alkyne introduction under stirring, promoting ligand dis-
placement [22], can suppress the induction period. Both H2
and stirring in the presence of the alkyne are thus necessary
to activate the nanocatalyst.
The very good activities that were observed in hydro-
genation plead for the existence of active medium and/or
poorly coordinated Ni sites in the active catalyst. The exact
reaction mechanism depends on their nature. Their direct
characterization in the colloidal solution was however out
of reach due to the complexity of the reaction medium
(solvent, ligands and reactants, all in liquid phase and in
interaction with the surface). The nature of the surface
could not be predicted either using the description of the
possible surfaces from crystallographic data. Indeed, the
nanoparticles are spheres and not polyhedrons even though
they are single-crystals [23]: a wide variety of surface sites
is therefore present. As a consequence, a series of experi-
ments was designed to get qualitative insights on the nature
Figure 2 Conversion and styrene yield over time for pheny-
lacetylene and 1-d-phenylacetylene. The lines are a guide to
the eye.
of the active sites, and phenylacetylene was adopted as a
substrate model.
In order to gain information about the limiting step of
the reaction, the hydrogenation of 1-d-phenylacetylene was
monitored over time. Both sets of data (phenylacetylene and
1-d-phenylacetylene hydrogenations) are plotted in Fig. 2.
A kinetic isotope effect of 2.4 was observed. This value
is not straightforward to interpret since low-coordinated
sites could interact with the incoming alkyne in different
manners, and since most of the elemental steps of the
hydrogenation reaction were reversible. However, it indi-
cated that the C(sp) H bond is involved in the limiting step
of the hydrogenation reaction, which could be possibly the
formation of a ␴-alkenyl bond between the alkyne and the
surface.
Additionally, cross-reactions involving two different
alkynes (1-d-phenylacetylene and p-chlorophenylacetylene)
were performed (Scheme 2). The analysis of the final
mixture showed the presence of significant amounts of
chloro-derivatives that were labeled with deuterium (see
ESI). This observation suggested that some dissociative
adsorption of the alkynes occurred on the most unsaturated
sites of the surface, resulting in H/D scrambling. This mode
of adsorption is also known to result in polycondensation and
highly coordinatively unsaturated sites deactivation, and
could thus also be partly responsible for the existence of
an induction time, which was discussed before.
Scheme 2 Cross-reaction between deuterated and non-
deuterated alkynes. The duration of the reaction was chosen
to be sufficiently long (46 h) to yield full conversion of both
alkynes.

Nickel phosphide nanocatalysts for the chemoselective hydrogenation
25
Figure 3 (Left) Schematic structures of the nanoparticles, and a TEM picture of the core—shell Ni3.5P nanoparticles; (Middle)
Conversion and (Right) selectivity obtained with Ni2P, N i 3.5P and Ni nanoparticles. The lines are a guide to the eye.
As mentioned before, it was reported in the 1980s that
phosphorus could act as a surface modifier by withdrawing
electrons from the Ni centers [10,11]. Additionally, P atoms
could poison the surface by blocking the most unsaturated
sites, in the same way that lead was used to poison Pd cat-
alysts [2]. As a consequence, the most unsaturated sites
would be blocked and the hydrogenation would work through
the Horiuti—Polanyi mechanism on moderately unsaturated
Ni sites. The very good cis-selectivity observed experimen-
tally (Table 1) correlates well with this hypothesis.
However, typical Ni/P ratios were in the range 4—10 in
these studies, higher than in our study. The role of phospho-
rus in the metal phosphide structure was therefore assessed
by varying the Ni/P ratio and comparing the corresponding
catalytic activity, keeping all the other parameters con-
stant: nanoparticles mean size, shape and ligands set.
Pure-Ni nanoparticles and Ni3.5P nanoparticles were then
compared with Ni2P nanoparticles for the hydrogenation of
phenylacetylene. The so-called Ni3.5P nanoparticles exhibit
in fact a core—shell structure: a Ni2P monocrystallized core
surrounded by a 2 nm polycrystallized shell of Ni (Fig. 3,
left), as described in a previous work [23]. Core—shell struc-
ture have indeed be found to greatly influence the outcome
of catalytic reactions [24]. XPS analysis was performed to
get information on the surface state. The surface composi-
tion of the air-exposed nanocatalysts revealed the presence
of (a) oxidized species in all samples in addition to a NiP
phase (Ni2p3/2BE = 853.4 eV) [25] in the Ni2P nanocatalysts;
(b) Ni0 (Ni2p3/2BE = 852.8 eV) in the pure Nickel nanoparti-
cles and (c) the coexistence of both Ni0(Ni2p3/2BE = 852.6 eV)
and NiP (Ni2p3/2BE = 853.5 eV) in the Ni3.5P sample, in agree-
ment with a core—shell structure.
expected in a self-selective reaction, that is, the remain-
ing alkynes displaced the alkenes on the catalysts surface
because of their stronger interaction with the active sites
[26]. In this situation, as soon as all alkynes had been hydro-
genated, the alkenes started to react and the selectivity
dropped, as observed experimentally (Fig. 3, right). In the
Ni3.5P nanoparticles, the effect of phosphorus was thus com-
patible with the one observed in the previous studies on
amorphous Ni—P nanoparticles [10]: it poisoned the surface
and avoided direct formation of alkanes by blocking the most
active sites and/or limiting the presence of hydrides in the
vicinity of the activated alkyne. As Ni3.5P did initially exhibit
a Ni2P—Ni core—shell structure, the phosphorus poison came
either from the core, by temperature-promoted diffusion
in the nanoparticles [27], either from the amorphous grain
boundaries of the crystallized Ni domains, or it was already
present at the nanoparticles very surface after their reaction
with white phosphorus.
Besides, the comparison between Ni3.5P and Ni2P cata-
lysts confirmed an unexpected feature of the latter (Fig. 3,
right), already mentioned above: Ni2P did not hydrogenate
alkenes for several hours even though no more alkynes were
present (see Fig. 1). A fortunate consequence was that the
reaction could be easily stopped at high conversion and high
selectivity. However, selectivity of the Ni2P nanocatalysts
was therefore not solely ruled by kinetics (the competi-
tion of alkynes and alkenes). Here, active sites in Ni2P were
not able to hydrogenate alkenes in a few hours, pinpointing
that phosphorus played a different role than in the Ni3.5P
nanocatalyst.
Gas-phase isotopic hydrogen—deuterium exchange
experiments were performed in order to investigate the
activity of the catalysts versus H2 activation and the con-
centration of hydride species, which remains adsorbed on
the catalysts surface, at a temperature close to the reaction
temperature. In these experiments, the nanoparticles were
exposed to a H2/D2 feed (molar ratio 1/1) at defined tem-
peratures in the 25—100 ◦C range. HD once formed on the
catalyst surface, could be desorbed toward the gas phase,
where it was detected by mass spectroscopy, or remained
adsorbed on the surface of the catalyst, depending on the
interaction strength of the hydride species with the surface
The hydrogenation of phenylacetylene was faster for the
two new catalysts than for Ni2P, mainly because the induc-
tion period was absent with both samples (see Fig. 3, Middle
inset). Apart from this effect, the initial rate was in the
same range. Moreover, the Ni/P ratio strongly impacted the
selectivity of the reaction. Using nickel nanoparticles, the
styrene was quickly hydrogenated to ethylbenzene. This sec-
ondary process was slower with Ni3.5P, which exhibited an
intermediate behavior between Ni and Ni2P. The selectiv-
ity curve for the Ni3.5P nanoparticles followed the shape

26
S. Carenco et al.
could now bring interesting perspectives to fine chemical
synthesis and to homogeneous catalysis in general.
Nickel
Ni3.5
Ni2P
P
Experimental
Catalysis setup
A typical reaction of hydrogenation was performed as fol-
low: In a 3 mL glass reactor equipped with a manometer
and washed with aqua regia before use, 1 mL of dry diox-
ane was added. 0.07 equiv. of Ni2P nanoparticles were
added (0.02 mmol, 3.0 mg, the stoichiometry is based on
the molecular formula ‘‘Ni2P’’, M = 148 g/mol), along with
1 equiv. of phenylacetylene (0.3 mmol, 30.6 mg, 32.9 ␮L).
The air was purged and 6 bar of H2 were added (correspond-
ing to ca. 3 equiv.). The reactor was placed in an oil bath
preheated at 85 ◦C and magnetically stirred. For the reac-
tion involving Ni nanoparticles and Ni3.5P nanoparticles, the
comparison was done by keeping the number of Ni mole con-
stant (i.e. 0.04 mmol of Ni). At the end of the reaction, the
pressure was released, and the composition of the mixture
was analyzed by GC—MS. For isolated yields, the volatiles
were removed under vacuum and the residue was weighted
and analyzed by NMR.
25
50
T (°C)
100
Figure 4 HD formation versus H2 consumption at selected
temperatures (25, 50, 100 ◦C) for the three nanocatalysts.
sites. H2 activation at several temperatures was lower on
the Ni2P sample compared to the Ni3.5P and Ni nanocatalysts
(see ESI). This behavior could explain the induction period
observed in the catalytic behavior of the Ni2P catalysts. On
the other hand, at 100 ◦C a higher HD release toward the
gas phase, versus H2 consumption, was observed for the
Ni2P sample (Fig. 4). These experiments confirmed a low
H/D storage on the Ni2P catalysts surface, a feature that is
likely due to the presence of high amounts of phosphorus,
in way that is reminiscent of the ability of carbide phases
to limit the formation of hydrides in Pd nanocatalysts
[28]. The low concentration of hydride species on the Ni2P
catalyst surface could thus explain their higher selectivity
in the hydrogenation process.
Substrates synthesis
Undec-5-en-1-yne [30] (Table 1, entry 6), p-(1-phenyl-
ethynyl)styrene [31] (entry 7) and 2-(phenylethynyl)aniline
[32] (entry 12) were prepared according to reported
procedures. p-(1E-Styryl)phenylacetylene (Table 1, entry
5), and 2-((1E,3E,5E,7E)-3,7-dimethyldeca-1,3,5,7-tetraen-
9-yn-1-yl)-1,3,3-trimethylcyclohex-1-ene (Table 1, entry 8)
were synthesized using new procedures. See ESI for detailed
procedures and characterizations.
In conclusion, the relevance of nickel phosphide as a
cheap and mild catalyst for the selective hydrogenation of
alkynes to cis-alkenes has been demonstrated for a wide
range of substrates. Nanoscaling of the catalyst provided
a high surface/volume ratio, which favored good activi-
ties in mild conditions of temperature (85 ◦C) and pressure
(6 bars). In the design of the catalytic nanoparticles, phos-
phorus loading in the nickel phosphide nanoparticles was
shown to be critical for the selectivity of the catalyst. More-
over, two regimes could be differentiated concerning the
phosphorus effect. With a high mean Ni/P ratio of 3.5, phos-
phorus blocked the most unsaturated sites as a poison and/or
slightly withdrawn electron density from the active Ni sites,
in agreement with previous studies: the selectivity was ruled
by kinetics. With a low Ni/P ratio of 2 and a crystallized
structure, the phosphide phase was found not to be sig-
nificantly active in alkene hydrogenation anymore, in the
mild conditions that were investigated here. The Ni2P crys-
tallized nanoparticles were thus found to have their own
intrinsic properties as a catalyst. Indeed, they appear to be
similar to metal-alloyed nanoparticles such as PdxGa [29],
by forming highly covalent metal-rich structures that seem
to limit the formation of hydrides and could thus find appli-
cations in chemoselective hydrogenation processes. Overall,
the metal phosphide, which deeply contributed to the design
of robust and active heterogeneous hydrotreating catalysts,
Gas-phase H/D exchange
Hydrogen/deuterium (H/D) exchange experiments were car-
ried out in a flow reactor at 25, 50, 100, 150, 200 and 250 ◦C.
The feed gas consisted of 4 mL/min H2, 4 mL/min D2 and
18 mL/min argon. Reaction products (H2, HD and D2) were
analyzed with a mass spectrometer (Omnistar, QMG 220M1),
and their values given in mA. Prior to the measurements,
the samples (50 mg diluted in 100 mg CSi) were activated in
flowing Argon (10 mL/min) at 150 ◦C for 120 min. After cool-
ing down to room temperature, the gas feed was changed to
the reactant gas composition. A delay of ca. 50 min was keep
between temperature steps in order to let the temperature
equilibrate.
XPS measurement
were performed on a SPECS spectrometer with a MCD-9
detector using a monochromatic Al(K␣ = 1486.6 eV) X-ray
source. Spectra were recorded using an analyzer pass energy
of 50 V, an X-ray power of 200 W and under an operating pres-
sure of 10−9 mbar. Spectra treatment has been performed
using the CASA software. Binding energies (BE) were refer-
enced to the C1s peak at 285.5 eV.

Nickel phosphide nanocatalysts for the chemoselective hydrogenation
27
[22] S. Carenco, C. Boissière, L. Nicole, C. Sanchez, P. Le Floch, N.
Mézailles, Chem. Mater. 22 (2010) 1340—1349.
Acknowledgements
[23] S. Carenco, X.-F. Le Goff, J. Shi, L. Roiban, O. Ersen, C.
Boissière, C. Sanchez, N. Mézailles, Chem. Mater. 23 (2011)
2270—2277.
The DGA, the Ecole Polytechnique, the CNRS and the UPMC
are acknowledged for financial support.
[24] X.-Y. Yang, Y. Li, G. Van Tendeloo, F.-S. Xiao, B.-L. Su, Adv.
Mater. 21 (2009) 1368—1372.
[25] NIST XPS database: http://srdata.nist.gov/xps/selEnergyType.
aspx.
Appendix A. Supplementary data
[26] D. Duca, L. Liotta, G. Deganello, J. Catal. 154 (1995) 69—79.
[27] R.-K. Chiang, R.-T. Chiang, Inorg. Chem. 46 (2007) 369—371.
[28] M.W. Tew, M. Janousch, T. Huthwelker, J.A. van Bokhoven, J.
Catal. (2011), doi:10.1016/j.jcat.2011.06.025.
[29] (a) L. Shao, W. Zhang, M. Armbrüster, D. Teschner, F. Girgsdies,
B. Zhang, O. Timpe, M. Friedrich, R. Schlögl, D.S. Su, Angew.
Chem. Int. Ed. (2011), doi:10.1002/anie.201008013;
(b) M. Armbruster, G. Wowsnick, M. Friedrich, M. Heggen, R.
Cardoso-Gil, J. Am. Chem. Soc. 133 (2011) 9112—9118.
[30] A. Leyva, X. Zhang, A. Corma, Chem. Commun. 33 (2009)
4947—4949.
Supplementary
cle can be found, in the online version, at
doi:10.1016/j.nantod.2011.12.003.
data
associated
with
this
arti-
References
[1] B.A. Wilhite, M.J. McCready, A. Varma, Ind. Eng. Chem. Res.
41 (2002) 3345—3350.
[2] Á. Molnár, A. Sárkány, M. Varga, J. Mol. Catal. A: Chem. 173
(2001) 185—221.
[3] S. Dominguez-Dominguez, A. Berenguermurcia, D. Cazor-
laamoros, A. Linaressolano, J. Catal. 243 (2006) 74—81.
[4] (a) F. Alonso, I. Osante, M. Yus, Tetrahedron 63 (2007) 93—102;
(b) D. Savoia, E. Tagliavini, C. Trombini, A. Umani-Ronchi, J.
Org. Chem. 46 (1981) 5340—5343;
(c) V. Polshettiwar, B. Baruwati, R.S. Varma, Green Chem. 11
(2009) 127;
(d) J. Choi, N.M. Hoon, Tetrahedron Lett. 37 (1996) 1057—1060.
[5] G.B. Bond, Metal-Catalysed Reactions of Hydrocarbons,
Springer, 2005.
[6] A. Dhakshinamoorthy, K. Pitchumani, K. Tetrahedron, Letter 49
(2008) 1818—1823.
[7] (a) F.M. Bautista, J.M. Campelo, A. Garcia, D. Luna, R.A.
Quiros, A.A. Romero, Catal. Lett. 52 (1998) 205—213;
(b) S. Abello, D. Verboekend, B. Bridier, J. Perezramirez, J.
Catal. 259 (2008) 85—95.
[31] T. Ishizone, G. Uehara, A. Hirao, S. Nakahama, Macromolecules
31 (1998) 12, 3764.
[32] C. Koradin, W. Dohle, A.L. Rodriguez, B. Schmid, P. Knochel,
Tetrahedron 59 (2003) 1571—1587.
Sophie Carenco did her undergraduate stud-
ies at the Ecole Polytechnique in Palaiseau
(France) from 2004 to 2007. She received
her Master degree of Chemistry at the Ecole
Polytechnique in 2008. She is currently com-
pleting her Ph.D. in Chemistry at the UPMC
in Paris under the supervision of Prof. Clé-
ment Sanchez (Collège de France) and Dr.
Nicolas Mézailles (Ecole Polytechnique). Her
main interests are the design of new synthetic
routes toward metal phosphide nanoparticles
and their applications in the field of materials sciences (lithium
batteries, catalysis). She was awarded the European Young Chemist
Award (first prize at Ph.D. student level) of the EuCheMS in 2010 in
Nuremberg.
[8] C. Stinner, Z. Tang, M. Haouas, T. Weber, R. Prins, J. Catal. 208
(2002) 456—466.
[9] N. Sweeny, C. Rohrer, O. Brown, J. Am. Chem. Soc. 80 (1958)
799—800.
[10] Y. Okamoto, Y. Nitta, I. Toshinobu, S. Teranishi, J. Catal. 64
(1980) 397—404.
[11] Y. Chen, Catal. Today 44 (1998) 3—16.
Antonio Leyva-Pérez was born in Sevilla,
Spain, in 1974. He studied Chemistry at
the Universidad de Valencia and recieved a
Ph.D. degree with honors by the Universidad
Politécnica de Valencia in 2005, working on
heterogeneous catalysis under the guidance
of Prof. Hermenegildo García. He moved for
post-doctoral studies to the group of Prof.
Steven V. Ley at The University of Cambridge,
U.K., working in the total synthesis of natu-
ral products. After two years, he joined the
group of Prof. Avelino Corma at the Instituto de Tecnología Química
(ITQ) in Valencia. His current research involves the development of
metal-catalyzed organic reactions.
[12] For a review, see: H.G. Von Schnering, W. Hoenle, Chem. Rev.
88 (1988) 243—273.
[13] E. Bekaert, J. Bernardi, S. Boyanov, L. Monconduit, M.-L. Dou-
blet, M. Meˇıneˇıtrier, J. Phys. Chem. C 112 (2008) 20481—20490.
[14] A. Nelson, M. Sun, A. Junaid, J. Catal. 241 (2006) 180—188.
[15] For a review, see: S.T. Oyama, T. Gott, H. Zhao, Y.-K. Lee,
Catal. Today 143 (2009) 94—107.
[16] S.T. Oyama, Y. Lee, J. Catal. 258 (2008) 393—400.
[17] X. Li, Y. Zhang, A. Wang, Y. Wang, Y. Hu, Catal. Commun. 11
(2010) 1129—1132.
[18] S. Carenco, I. Resa, X.-F. Le Goff, P. Le Floch, N. Mézailles,
Chem. Commun. (2008) 2568—2570.
[19] (a) See for instance: L.D. Pachón, G. Rothenberg, Appl.
Organomet. Chem. 22 (2008) 288—299;
Patricia Concepción Heydorn, tenured Sci-
entist of the Spanish National Research
Council (CSIC) since 2002, is actually work-
ing at the Instituto de Tecnología Química
in Valencia, Spain. Her research area is
focused on catalyst characterization under
reaction conditions by means of ‘‘in situ’’
Infrared and Raman spectroscopy in order
to develop structure—activity relationships.
(b) J.-S. Chen, A.N. Vasiliev, A.P. Panarello, J.G. Khinast, Appl.
Catal. A: Gen. 325 (2007) 76—86;
(c) S.S. Soomro, F.L. Ansari, K. Chatziapostolou, K. Köhler, J.
Catal. 273 (2010) 138—146.
[20] This second reaction is in agreement with an early report
by N. Sweeny et al. about the conversion of nitrobenzene to
aminobenzene that was achieved in much stronger conditions
(ca. 375◦). See Ref. [9].
[21] C.A. Brown, R.A. Coleman, J. Org. Chem. 44 (1979) 2328—2329,
13.
She is mainly involved on supported metal
nanoparticles, metal oxide catalyst and zeo-
lites. She is author of more than 70 publications and 3 patents on
different aspects of inorganic, organic and material chemistry.

28
S. Carenco et al.
Cédric Boissière is CNRS Research fellow
at Laboratory Chimie de la Matière Conden-
sée (Collège de France, UPMC, CNRS). After
a Ph.D. at the European Institute of Mem-
branes in 2001 (Montpellier-France), he did
post-doctoral work in biomineralisation of
phosphate materials at the Bristol’s School of
Chemistry (GB). Today he is currently working
on both fundamental and industrial devel-
opment of hierarchical functional inorganic
and hybrid materials made by coupling sol-gel
Clément Sanchez is Professor at Collège
de France (Paris) and leads the Laboratory
Chimie de la Matière Condensée (Collège
de France, UPMC, CNRS). After
a Ph.D.
at the University of Paris VI in 1981, he
did post-doctoral work at the University
of Berkeley. Today, he currently leads
a
research group working on sol—gel chemistry
and physical properties of nanostructured
porous and non-porous inorganic and hybrid
organic—inorganic materials shaped as mono-
chemistry and process.
liths, microspheres and films. He has awarded by many scientific
French and international prizes.
Nicolas Mézailles, after graduating from
Ecole Nationale Supérieure de Chimie de
Toulouse as chemical engineer, he obtained
Avelino Corma Canos was born in Moncó-
far, Spain in 1951. He studied Chemistry at
a Ph.D. degree in 1997 from Purdue Univer-
the Universidad de Valencia (1967—1973),
sity (IN, USA) under the direction of Prof.
and received his Ph.D. at the Universi-
Clifford Kubiak. He came back to France to
dad Complutense de Madrid in 1976. He
receive post doctoral training at the Ecole
did postdoctoral work in the Department of
Polytechnique in the group of Prof. Fran¸cois
chemical engineering at the Queen’s Univer-
Mathey. He was recruited by CNRS in 1998 in
sity (Canada, 1977—79). He is director of the
the ‘‘UMR 7653’’ and has worked with Prof.
Instituto de Tecnología Química (UPV-CSIC) at
P. Le Floch since then. He was promoted to
the Universidad Politécnica de Valencia since
Research Director in 2008. His ongoing research interests span the
use of geminal dianions as precursors for the synthesis of metal
carbene fragments; ligand synthesis and coordination; the mech-
anisms of nanoparticle synthesis; metal-phosphides nanoparticles;
and more recently transition metal activation of N2.
1990. His current research field is catalysis,
covering aspects of synthesis, characterization and reactivity in
acid—base and redox catalysis. Avelino Corma is co-author of more
than 700 articles and 100 patents on these subjects.