Palladium nanoparticles stabilised by cinchona-based alkaloids in glycerol: Efficient catalysts for surface assisted processes

Article abstract of DOI:10.1039/c6ra19230k

Palladium nanoparticles (PdNPs) were synthesised and fully characterised, both in solution and the solid state, using naturally-occurring cinchona-based alkaloids in neat glycerol. These nano-systems were stable under reaction conditions, finding applications in hydrogenation and hydrodehalogenation processes, as a result of their surface-like behaviour. Their reactivity was improved in relation to that involving PdNPs stabilised by phosphines and also by Pd/C as a heterogenous catalyst, mainly in terms of recyclability. In particular, the colloidal palladium catalyst stabilised by quinidine was highly efficient to promote the hydrodechlorination of aromatic compounds under low dihydrogen pressure. These original catalysts found applications in the synthesis of secondary and tertiary amines including N-substituted anilines, by means of one-pot tandem Pd-catalysed methodologies under smooth conditions. In all of these processes, glycerol performed a crucial function as a liquid support for the immobilisation of nanoparticle-based catalysts, allowing both the stabilisation of the nano-catalysts and easy recycling of the catalytic phase.

Full text of DOI:10.1039/c6ra19230k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Palladium nanoparticles stabilised by cinchona-
based alkaloids in glycerol: efficient catalysts for
surface assisted processes†
Cite this: RSC Adv., 2016, 6, 93205
a
a
b
a
a
*
A. Reina, C. Pradel, E. Martin, E. Teuma and M. Gomez
´
Palladium nanoparticles (PdNPs) were synthesised and fully characterised, both in solution and the solid
state, using naturally-occurring cinchona-based alkaloids in neat glycerol. These nano-systems were
stable under reaction conditions, finding applications in hydrogenation and hydrodehalogenation
processes, as a result of their surface-like behaviour. Their reactivity was improved in relation to that
involving PdNPs stabilised by phosphines and also by Pd/C as a heterogenous catalyst, mainly in terms of
recyclability. In particular, the colloidal palladium catalyst stabilised by quinidine was highly efficient to
promote the hydrodechlorination of aromatic compounds under low dihydrogen pressure. These
original catalysts found applications in the synthesis of secondary and tertiary amines including N-
substituted anilines, by means of one-pot tandem Pd-catalysed methodologies under smooth
conditions. In all of these processes, glycerol performed a crucial function as a liquid support for the
immobilisation of nanoparticle-based catalysts, allowing both the stabilisation of the nano-catalysts and
easy recycling of the catalytic phase.
Received 29th July 2016
Accepted 20th September 2016
DOI: 10.1039/c6ra19230k
www.rsc.org/advances
as cyclometallated complexes, also lead to the formation of
PdNPs mostly at high temperatures, which can give rise to
a surface reactivity.6 The various plausible scenarios can be
Introduction
Palladium represents one of the most efficient and versatile
metal-based catalysts, nding applications in a large variety of
processes (couplings, carbonylations, hydrogenations.) under
different material states, including molecular complexes,
nanoparticles and extended surfaces,1 and permitting immo-
bilisation on both solid supports2 and liquid phases.3 This
structural variety leads to different kinds of activations, a reason
for which palladium is the most used metal in organic
synthesis.4
In the last decades, huge efforts have been done in the
preparation of metal nanoparticles and in particular in palla-
dium nanoparticles (PdNPs), controlling their size, composition
and morphology, parameters able to directly impact on the
further catalytic reactivity.5 From a mechanistic point of view,
some studies seem to prove that nano-sized starting catalytic
precursors can act as a reservoir of small clusters or atoms,
probably operating under homogeneous regime. But catalytic
processes involving dened organometallic compounds, such
adjust modifying reaction conditions, nature of stabilisers,
solvent. principally when PdNPs are homogeneously
dispersed in a liquid phase. Contrary to common organic
solvents, structurally organised ones (for instance,
imidazolium-based ionic liquids7 or polyols8) by intermolecular
hydrogen bonds or p–p stacking assemblies, can make easy
the stabilisation of nanoparticles, avoiding their agglomeration
and in consequence favouring a heterogeneous reactivity.9
However because of the relative weak interactions between
solvents and metal surfaces, the addition of stabilisers other
than solvents is in general required for both inducing activity
and/or selectivity in the catalytic processes and stabilising the
nano-sized state.10
In a sustainable frame, the design of eco-friendly catalysts
represents a major concern. Our experience in the development
of catalytic processes in glycerol involving metal nano-
particles,11,12 in particular those using PdNPs stabilised by
phosphines,12 led us to work with stabilisers coming from the
biomass, such as cinchona-based alkaloids, well-known for
their use in stereoselective reactions, mainly in enantioselective
hydrogenations where nanoparticles are supported in different
kinds of materials.13 In contrast to the strong interactions
between P-donor ligands and metal surface, primarily by dative
s/p Pd–P interactions, cinchona ligands can interact with the
surface through N-quinoline atom or by p interaction through
the aromatic ring,14 disfavouring the formation of dened
a
´ ´
´
´
Laboratoire Heterochimie Fondamentale et Appliquee (LHFA), Universite de Toulouse,
UPS, CNRS UMR 5069, 118, route de Narbonne, 31062 Toulouse cedex 9, France.
E-mail: gomez@chimie.ups-tlse.fr
b
´
´
´
Departamento de Quımica Inorganica, Facultad de Quımica, Universidad Nacional
´
´
´
Autonoma de Mexico, Av. Universidad 3000, 04510 Mexico DF, Mexico
† Electronic supplementary information (ESI) available: General experimental
part, Tables S1 and S2, Fig. S1–S19, Schemes S1–S4 and 1H and 13C NMR
spectra of the products obtained in the catalytic reactions are included. See
DOI: 10.1039/c6ra19230k
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 93205–93216 | 93205

View Article Online
RSC Advances
Paper
Ligand exchange reaction at the surface of palladium
nanoparticles
In a Schlenk ask under argon, 1 equivalent of dodecanethiol
(0.05 mmol; 10.12 mg) were added to 1 mL of colloidal solution
(PdIa or PdId) and stirred for 24 h. Ligands a or d were extracted
form the glycerol phase with dichloromethane (10 ꢁ 3 mL);
dichloromethane was removed under vacuum. The mixture
1
extracted was analysed by H NMR.
Hydride titration of palladium nanoparticles
Isolated PdIa or PdId (0.02 mmol, 2.1 mg) were added to
a Young NMR tube under argon at room temperature contain-
ing 0.6 mL of THF-d8 and maleic anhydride (0.01 mmol, 1 mg)
in the presence of an external standard (0.01 mmol – 113 mg of
cyclooctane). The corresponding mixtures were monitored by
1H NMR.
Scheme 1 Synthesis of palladium nanoparticles PdxL in glycerol using
cinchona-based alkaloids (a–d) as stabilisers ([Pd] ¼ 10ꢂ2 mol Lꢂ1).
General procedure for Pd-catalysed hydrogenation in glycerol
coordination complexes, and in consequence, privileging the
stabilisation of nanoparticles more than molecular species.
In this context, we decided to explore the catalytic reactivity
of palladium nanoparticles prepared in glycerol in the presence
of naturally-occurring cinchona-based alkaloids, in particular
the optically pure stereoisomers cinchonidine, cinchonine,
quinine and quinidine (Scheme 1). We chose hydrogenations
and hydrodehalogenations because of their presumed hetero-
geneous pathway. To the best of our knowledge, no precedents
in the literature using palladium nanoparticles stabilised by
cinchona ligands as colloidal catalysts have been previously
reported.15
In a Fisher–Porter bottle (from 1 to 5 bar) or an autoclave (5 to 20
bar), the appropriate substrate (1 mmol for 1 mol% of catalyst; 5
mmol for 0.5 mol% or 10 mmol for 0.1 mol%) was added to 1
mL of preformed nanoparticles in glycerol under argon. The
reaction mixture was put under vacuum and then pressurised
with H2 at the convenient pressure, heated up to the desired
temperature, stirred for the appropriate time and then cooled
down to room temperature. Organic products were extracted
from glycerol with dichloromethane (5 ꢁ 3 mL) dichloro-
methane was removed under vacuum. All the products were
previously reported and identication was done by comparison
1
of their GC-MS data and H and 13C NMR spectra with the re-
ported data.
Experimental
Synthesis of palladium nanoparticles in glycerol
General procedure for Pd-catalysed hydrodehalogenation in
glycerol
0.05 mmol of palladium precursor (11.2 mg for Pd(OAc)2; 14.1
mg for [PdCl2(cod)]; 22.8 mg [Pd2(dba)3]) and 0.05 mmol of
ligand (14.7 mg for cinchonidine (a) and cinchonine (b); 16 mg
for quinine (c) and quinidine (d); 28.4 mg for TPPTS) were
dissolved in 5 mL of glycerol and stirred under argon in
a Fisher–Porter bottle at room temperature until complete
dissolution. The system was then pressurized under 3 bar of
dihydrogen and stirred at 80 ꢀC for 18 h. A black colloidal
solution was then obtained.
In a Fisher–Porter bottle (from 1 to 5 bar H2) or an autoclave (5
to 20 bar H2), the appropriate aryl-halide (1 mmol for 1 mol% of
catalyst; 5 mmol for 0.5 mol%; 10 mmol for 0.1 mol%) was
added to 1 mL of preformed nanoparticles in glycerol under
argon (total amount of palladium: 0.01 mmol). If necessary, 1.2
equivalents of KOH (66 mg) were also added. The reaction
mixture was put under vacuum and then pressurised with H2 at
the convenient pressure, heated up to the desired temperature,
stirred for the appropriate time and then cooled down to room
temperature. Organic products were extracted from glycerol
with dichloromethane (5 ꢁ 3 mL); dichloromethane was
removed under vacuum. All the products were previously re-
ported and identication was done by comparison of their GC-
Isolation of palladium nanoparticles from the glycerol
solution
Aer synthesis, PdNPs in glycerol were transferred to a centri-
fugation tube and 2 mL of ethanol were added. Centrifugation
was carried out at 600 rpm for 5 h and then the solution was
separated by decantation. This process was repeated 3 times
1
MS data and H and 13C NMR spectra with the reported data.
General procedure for Pd-catalysed synthesis of secondary
and tertiary amines
until complete removal of glycerol. The remaining black powder
ꢀ
was then dried under vacuum at 80 C overnight. Elementary In a Fisher–Porter bottle the appropriate primary or secondary
analysis (palladium load determined by ICP-AES): PdIa: Pd amine (1 mmol) and aldehyde (1 mmol) were added to 1 mL of
26.5%, C 57%, N 7%, H 5.5%; PdId: Pd 25%, C 56%, N 6.5%, H preformed nanoparticles in glycerol under argon (total amount
5.6%.
of palladium: 0.01 mmol). The reaction mixture was put under
93206 | RSC Adv., 2016, 6, 93205–93216
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Table 1 Hydrogenation of 4-phenylbut-3-en-2-one catalysed by
vacuum and then pressurised with H2 at the convenient pres-
sure, heated up to the desired temperature, stirred for the
appropriate time and then cooled down to room temperature.
Organic products were extracted from glycerol with dichloro-
methane (5 ꢁ 3 mL); dichloromethane was removed under
vacuum. All the products were previously reported and identi-
PdNPs capped by cinchona-based stabilisersa
1
cation was done by comparison of their GC-MS data and H
and 13C NMR spectra with the reported data.
Entry
Stabiliser
Conv. (yield)b (%)
General procedure for Pd-catalysed synthesis of N-substituted
anilines
1
2
3
4
a
b
c
d
dH
d
d
99 (96)
79 (74)
99 (90)
99 (98)
99 (97)
22 (21)
99 (95)
87 (85)
In a Fisher–Porter bottle nitrobenzene (1 mmol) and the
appropriated aldehyde (1 mmol) were added to 1 mL of pre-
formed nanoparticles in glycerol under argon (total amount of
palladium: 0.01 mmol). The reaction mixture was put under
vacuum and then pressurised with H2 at the convenient pres-
sure, heated up to the desired temperature, stirred for the
appropriate time and then cooled down to room temperature.
Organic products were extracted from glycerol with dichloro-
methane (5 ꢁ 3 mL); dichloromethane was removed under
vacuum. All the products were previously reported and identi-
5
6c
7d
8e
TPPTS
a
Results from duplicated experiments. Reaction conditions: 1 mmol of
1 and 1 mL of the corresponding catalytic glycerol solution of PdNPs
(10ꢂ2 mol Lꢂ1, 0.01 mmol of total Pd). Determined by GC and GC-
b
c
d
MS using decane as internal standard. 0.1 mol% Pd load. PdNPs
(1.9 mg) at solid state and redispersed in glycerol used as catalyst.
e
TPPTS ¼ tris(3-sulfophenyl)phosphine trisodium salt.
1
cation was done by comparison of their GC-MS data and H
and 13C NMR spectra with the reported data.
S1 in the ESI†) showed that [Pd2(dba)3] mainly led to agglomer-
ation and in the case of quinidine, black precipitate was formed
instead of colloidal suspensions. In contrast, Pd(II) precursors
gave homogeneous black solutions constituted by well-dispersed
nanoparticles, in particular starting from Pd(OAc)2. Therefore,
PdNPs with the four alkaloids were prepared using this palla-
dium salt. All of them yielded colloidal solutions, showing
a trend to aggregate, except for PdId which gave a well-dispersed
system (Fig. 1). However PdIa, PdIb and PdIc gave better
dispersions aer centrifugation (isolation of nanoparticles at the
solid state) and re-dispersion again in glycerol (Fig. S1 in the
ESI†). It is important to underline that when PdNPs at solid state
were redispersed in ethanol, agglomerates and bigger nano-
particles were observed (Fig. S2 in the ESI†). These facts unde-
niably evidence the role of glycerol in the dispersion of
nanoparticles, avoiding their aggregation, probably favoured by
General procedure for recycling of the catalytic phase
Aer extraction of the organic products, the catalytic phase was
ꢀ
put under vacuum at 60 C for two hours and then the appro-
priated reagents were added under argon to the catalytic phase.
The reaction mixture was then pressurised with H2 at the
convenient pressure, heated up to the desired temperature,
stirred for the appropriate time and then cooled down to room
temperature. Organic products were extracted from glycerol
with dichloromethane (5 ꢁ 3 mL); dichloromethane was
removed under vacuum. This process is repeated for each of the
different runs. All the products were previously reported and
identication was done by comparison of their GC-MS data and
1H and 13C NMR spectra with the reported data.
Results and discussion
Synthesis and characterisation of palladium nanoparticles
We were interested in the synthesis of palladium nanoparticles
(PdNPs) containing optically pure cinchona-based stabilisers in
glycerol. These sustainable adsorbates coming from the
biomass, exhibit the possibility to interact with both the solvent
through hydrogen bonds and the metallic surface by the quin-
oline moiety (p-bonded and/or N-lone pair bonded).14 Based on
our previous works,11,12 we explored the formation of PdNPs in
the presence of the corresponding alkaloid (a–d), starting with
Pd(II) and Pd(0) precursors (Pd(OAc)2, I; [PdCl2(cod)], II)
ꢀ
([Pd2(dba)3], III), under hydrogen atmosphere (3 bar) at 80 C
overnight (Scheme 1).
Cinchonidine (a) and quinidine (d) were selected as model
stabilisers to choose the most appropriate metal precursor on
the basis of their better catalytic activity (see below Table 1).
TEM analyses of the corresponding colloidal dispersions (Table Fig. 1 TEM images in glycerol corresponding to PdIa–PdId.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 93205–93216 | 93207

View Article Online
RSC Advances
Paper
ꢀ
ꢀ
palladium took place. At lower temperatures (40 C and 60 C
instead of 80 ꢀC), colloidal solutions were attained and their
TEM analyses revealed the formation of well-dispersed nano-
particles (Fig. S4 in the ESI†), but being less homogeneous in
size comparing with those obtained at 80 ꢀC (at 60 ꢀC: dmean
¼
1.7 ꢃ 0.4 nm; at 40 ꢀC: dmean ¼ 1.9 ꢃ 0.5 nm). However at higher
temperature (100 ꢀC) big agglomerates were mainly formed.
Concerning the metal concentration, lower concentrations (in
relation to starting conditions, 10ꢂ2 mol Lꢂ1) led to bigger
nanoparticles (for [Pd] ¼ 10ꢂ3 mol Lꢂ1, 3.0 nm; for [Pd] ¼ 10ꢂ4
mol Lꢂ1, 2.4 nm); at higher concentration (0.1 mol Lꢂ1), parti-
cles tended to be aggregated (Fig. S5 in the ESI†). This behav-
iour is in contrast to the general trend observed using common
organic solvents (THF, toluene), where metal concentrations are
oen lower than 10ꢂ3 mol Lꢂ1 17 what is in accordance with the
,
supramolecular arrangement of glycerol, which favours the
dispersion of nanoparticles.16
With the aim of checking the stability of alkaloids under
synthesis conditions, we carried out a ligand exchange reaction
with dodecanethiol using both PdIa and PdId in glycerol, taking
advantage of the strong interaction between sulphur-based
derivatives and metal surfaces (Fig. S6 in the ESI†).18
Cinchona-based stabilisers were recovered (ca. 90% yield) by
biphasic extraction as a mixture of the corresponding alkaloid
and its dihydro-cinchona partner, it means hydrocinchonidine
(aH) and hydroquinidine (dH), due to the anticipated hydro-
genation of the vinyl group under dihydrogen atmosphere (a/aH
¼ 1.3/1; d/dH ¼ 2/1, Fig. S6 in the ESI†).19 No modication of the
quinoline fragment was observed, preserving the adsorption of
the ligand at the metallic surface through p- or N-lone pair
bound interactions and in consequence avoiding agglomera-
tion during the PdNPs synthesis.14
It is expected that this structural change on the cinchona
ligands triggers minor effect on the stabilisation of PdNPs and
in the subsequent catalytic activity. Actually, PdNPs capped by
hydroquinidine (named PdIdH) exhibited spherical dispersed
nanoparticles (mean diameter: 2.5 ꢃ 0.9 nm), but tending to be
more agglomerated than those prepared from quinidine (Fig. S8
in the ESI†). Moreover, no differences were observed in hydro-
genation and hydrodebromination reactions in comparison
with the system PdId (see Catalytic applications section and
Scheme S1 in the ESI†).
The composition of PdIa and PdId at solid state was deter-
mined by ICP (Pd) and elemental analysis (C, H, N) (see
Experimental section), indicating a Pd/alkaloid ratio of 2.2 and
2.4 for PdIa and PdId, respectively. With the purpose of esti-
mating the content of hydrides, PdIa and PdId at solid state
were dispersed in THF-d8, in the presence of one equivalent of
maleic anhydride; the corresponding mixtures were monitored
by 1H NMR (cyclooctane used as internal standard), without
observing any evolution aer the rst hour.20 For both mate-
rials, a Pd/H ratio of ca. 2/1 was found (Fig. S9 in the ESI†). With
this analytical data and the mean diameter found by TEM
analysis (Fig. 1 and S1 in the ESI†), nanoclusters of idealized
formula Pd309L129H155 and Pd147L61H74 can be proposed for
PdIa and PdId, respectively (L means the corresponding
cinchona stabiliser).21
Fig. 2 HR-TEM micrographs (left) for PdIa (a and b) and PdId (c and d)
at solid state (a and c) and in glycerol (b and d), with the corresponding
EDX analyses (right). Crystallographic planes spots observed by Fast
Fourier Transform are inserted (yellow squares represent the zone
where FFT was applied).
its supramolecular arrangement due to the hydrogen bond
network,16 in agreement with our previous work using TPPTS
(tris(3-sulfophenyl)phosphine trisodium salt) as stabiliser.12a,b
Taking into account the better dispersion of PdId together
with the better results obtained in catalysis (see below), the
formation of these PdNPs in glycerol was privileged to study the
inuence of different parameters on the formation of colloidal
systems (hydrogen pressure, Pd/ligand ratio, temperature,
metal concentration). In the absence of dihydrogen, no reduc-
tion of palladium acetate occurred, in contrast to that observed
in the absence of alkaloid or in the presence of TPPTS as sta-
biliser, giving in both cases precipitation of black palladium.12
Using a lower dihydrogen pressure (1 bar instead of 3 bar), only
agglomerates were observed (Fig. S3 in the ESI†).
When the ligand amount in relation to metal decreased (Pd/
ligand ¼ 1/0.3 or 1/0.5), an immediate precipitation of black
93208 | RSC Adv., 2016, 6, 93205–93216
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
IR spectra showed the presence of the alkaloid in the isolated binding energies for Pd(0) (Fig. 3). In contrast to the PdNPs iso-
PdIa and PdId (Fig. S10 in the ESI†). No important differences lated at solid state (see above), two binding energies were
were observed between the free cinchona-based ligand and the observed in the region of N 1s, probably due to both types of
corresponding PdNPs, pointing probably to an N-lone pair alkaloid molecules present in the colloidal solution, those linked
interaction between the stabiliser and the metallic surface more to the surface and those solvated by glycerol. Actually, we ana-
than a p-binding coordination.14
lysed by 15N NMR neat quinidine both in the absence and in the
Powder XRD analyses of PdIa and PdId exhibited the peaks presence of glycerol. As expected, two signals were observed
corresponding to the fcc Pd(0) structure, without observing corresponding to the nitrogen atoms from quinoline and qui-
crystalline Pd(II) phases (Fig. S11 in the ESI†). HR-TEM at solid nuclidine moieties in deuterated methanol. However in a mixture
state also evidenced this structure (Fig. 2). The surface compo- of CD3OD/glycerol (1/1) only the 15N chemical shi correspond-
sition and the electronic state of the PdNPs were studied by XPS ing to quinoline was observed, probably due to the enlargement
spectroscopy. XPS survey spectra for both solid materials PdIa of the signal corresponding to the quinuclidine nitrogen by
and PdId showed the presence of palladium, carbon, oxygen hydrogen bond interaction with glycerol (Fig. S16 in the ESI†).13g
and nitrogen (Fig. S12 in the ESI†). The high-resolution spectra This behaviour could explain the two types of molecules consis-
in the binding energy region corresponding to palladium (Pd tent with the signals observed by XPS in the N 1s region.
3d3/2 and Pd 3d5/2) showed the merely presence of Pd(0), in
agreement with the data obtained by XRD (Fig. S13 in the
Catalytic applications
ESI†).22 For the N 1s binding region, only one signal was
Due to the surface reactivity of PdNPs,6,24 we were interested in
observed, consistent with the XPS analyses of neat ligands (i.e.
the evaluation of their behaviour in dihydrogen-based
processes, such as hydrogenation of unsaturated functional
groups (alkenes, alkynes, imines, nitro-based substrates) and
hydrodehalogenation reactions, because of their industrial
applications.
the electronic differences between both nitrogen atoms, N-
quinuclidine and N-quinoline, could not be distinguished by
XPS in agreement with the reported data23) (Fig. S14 in the ESI†).
With the aim of elucidating the structure of PdNPs in solu-
tion, PdIa and PdId dispersed in glycerol (colloidal solutions)
were also directly analysed by HR-TEM, EDX and XPS taking
advantage of the negligible vapour pressure of glycerol. HR-TEM
and EDX analyses proved both the fcc Pd structure for the
nanoclusters dispersed in glycerol and the presence of the sta-
biliser (Fig. 2).
Hydrogenation reactions. We chose the hydrogenation of 4-
phenylbut-3-en-2-one (1) as benchmark reaction to compare the
reactivity of the different PdNPs stabilised by the corresponding
alkaloids in glycerol (entries 1–5, Table 1). For all of them, only
4-phenylbutan-2-one (1H) was exclusively obtained. No impor-
tant differences in reactivity among the PdNPs catalysts were
found, except for cinchonine (b, entry 2), which gave lower
conversion. The partially hydrogenated quinidine dH (see
above) led to the same reactivity trend than its vinyl-based
partner d (entry 5 vs. 4), indicating that the structural modi-
cation on the stabiliser d does not trigger a different catalytic
behaviour of the corresponding PdNPs. PdId was active even at
very low load of catalyst (0.1 mol%) under the same conditions
(1 bar H2, 80 ꢀC, 2 h; entry 6). PdId at solid state was also used as
catalytic precursor, nding the same catalytic activity than for
the colloidal system (entry 7 vs. 1). PdNPs stabilised by the
phosphine TPPTS12a,b gave slightly lower conversion and yield
than for the cinchona capped nanoparticles (entry 8 vs. entries
1, 3–5 and 7).
XPS analyses of the colloidal solutions evidenced the absence
of Pd(II) species (for the survey XPS spectra, see Fig. S15 in the
ESI†); actually, HR-XPS spectra only showed the corresponding
Fig. 4 Diagram showing the recycling of the catalytic phase PdId for
Fig. 3 High-resolution XPS spectra for PdIa and PdId in glycerol: (a) Pd the hydrogenation of 4-phenylbut-3-en-2-one (Table 1). Figures
3d binding energy region; (b) N 1s binding energy region.
indicate yields of 1H (determined by GC).
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 93205–93216 | 93209

View Article Online
RSC Advances
Paper
Table 2 Hydrogenation of compounds containing C]C or C^C bonds catalysed by PdIda
Entry
1
Substrate
Product
Pd mol% (pH2)
0.1c (1)
Conv. (yield)b (%)
>99 (99)
2
3
0.1c (1)
1 (1)
94 (90)
90 (3/1)d
4
1 (3)
99 (94)e (99/1)
5
6
1 (1)
1 (1)
97 (94)
>99 (96)
7
8
1 (10)
1 (1)
85 (79)
>99 (94)
9
0.1c (1)
0.5g (1)
0.1c (1)
48 (46)f
>99 (95)
98 (90)h
10
11
12
13
14
15
1 (3)
>99 (90)i
>99 (96)i
>99 (96)
47 (41)
0.5g (1)
0.5g (1)
0.1c (1)
16
1 (3)
>99 (97)
93210 | RSC Adv., 2016, 6, 93205–93216
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Table 2 (Contd.)
Entry
Substrate
Product
Pd mol% (pH2)
0.5g (1)
Conv. (yield)b (%)
45 (40)j
17
a
Results from duplicated experiments. Reaction conditions: 1 mmol of substrate (2–12) and 1 mL of the catalytic glycerol solution of PdId (10ꢂ2 mol Lꢂ1
,
0.01 mmol of total Pd). See Table S2 in the ESI for the results using PdIa as catalyst. b Determined by GC and GC-MS using decane as internal standard. c 10
mmol of substrate for 1 mL of the catalytic solution (0.01 mmol of total Pd). d In brackets, 4H/4I ratio. e For 12 h. f At 30 ꢀC for 24 h. g 5 mmol of substrate
for 1 mL of the catalytic solution (0.01 mmol of total Pd). h 5% of 9H was detected. i Cyclooctane used as internal standard. j At 35 ꢀC for 6 h.
surfactants (soaps, detergents, cosmetics) or lubricants).
Remarkably, no esterication products with glycerol were
detected and the extraction of the corresponding hydrogenated
compound (stearic acid) from the glycerol catalytic phase (by
means of a biphasic extraction, adding dichloromethane) easily
worked (entry 8). The reaction also ran under very smooth
conditions (0.1 mol% PdId, 30 ꢀC, 24 h, 1 bar H2; entry 9). These
catalysts were also applied in the hydrogenation of terminal (9)
and internal (10–12) alkynes. Tuning the catalytic conditions,
Scheme
2 Pd-catalysed hydrogenation of nitro-arenes by PdId
we could selectively achieve the reduction towards the corre-
sponding alkene or alkane. Actually, at low palladium load (0.1–
0.5 mol% depending on the alkyne) and low dihydrogen pres-
sure (1 bar) the corresponding alkene was chemoselectively
formed (entries 11, 13, 15 and 17). At relative harsher conditions
(0.5–1.0 mol% Pd and 1–3 bar H2), the fully hydrogenated
products were only obtained (entries 10, 12, 14 and 16). The
NOESY-NMR analysis of 12HE revealed the exclusive formation
of the (Z)-stereoisomer (Fig. S18 in the ESI†), indicating a dihy-
drogen syn addition on the C^C bond, as expected for a Pd
surface-like mechanism. Unluckily, no asymmetric induction
could be observed for the hydrogenation of prochiral
substrates, such as ethyl pyruvate, dimethyl itaconate or iso-
phorone (Fig. S19 in the ESI†).
nanoparticles in glycerol ([Pd] ¼ 10ꢂ2 mol Lꢂ1). Conversions and yields
were determined by GC and GC-MS, using decane as internal standard
(a denotes the use of 0.5 mol% PdId).
The catalytic glycerol phase PdId was recycled four times
showing the same catalytic behaviour. Aer the activity decrease
on the 5th run, the catalytic solution was treated under H2 (3
bar) overnight and then reused, recovering its activity (Fig. 4).
Palladium was no detected in the isolated 4-phenylbutan-2-one
aer the different runs (the palladium load was lower than the
detection limit by ICP-AES, i.e. <0.01 ppm). TEM analyses aer
the rst and fourth runs did not evidence any sign of agglom-
eration, remaining the nanoparticles well dispersed but
showing a bigger size in relation to the starting catalytic solu-
tion (ca. 2.0 nm – Fig. S17 in the ESI† – vs. 1.4 nm – Fig. 1).
Using PdId as catalyst (with PdIa, a similar catalytic behav-
iour could be observed; see Table S2 in the ESI†), compounds
containing C]C or C^C bond were efficiently reduced
(Table 2). Therefore, non-functionalised alkenes, both conju-
gated (2, 3) and non-conjugated (4) C]C bonds, were hydro-
genated. Substrates 2 and 3 reacted using only 0.1 mol% of
catalyst (entries 1–2). In the case of 1-dodecene (4), the iso-
merisation towards the internal alkene 2-dodecene could be
also observed (entry 3); working at higher pressure (3 bar) and
longer time (12 h) dodecane was merely isolated (entry 4). Pol-
yfunctionalised C]C based substrates such as cinnamonitrile
(5) and ethylcrotonate (6) led to the hydrogenation of the cor-
responding C]C bond (entries 5 and 6, respectively). However
for the endocyclic C]C bond hydrogenation of coumarin (7),
higher pressure (10 bar H2) was required to get ca. 80% yield of
the corresponding hydrogenated product, 7H (entry 7).
Anilines could be easily obtained by hydrogenation of the
corresponding nitro-arene starting materials, 13–16 (Scheme 2).
The reaction seems to be more sensitive to steric than elec-
tronic effects on the substrate. In fact, nitrobenzene (13) and 4-
chloronitrobenzene (14) were more active than 4-nitro-
acetophenone (15) and 4-nitroanisole (16), getting 95% of the
corresponding aniline, 13H and 14H, using only 0.5 mol% PdId.
Working under higher hydrogen pressure (10 bar H2) and
ꢀ
temperature (100 C) some extent of hydrodechlorination was
detected (ca. 10% of formation of nitrobenzene) for 4-chloro-
nitrobenzene (see below).
This scope proves that PdNPs stabilised by cinchona-based
alkaloids cannot reduce ketone, ester or carboxylic acid func-
tions. For the tested aromatic ketones (17–19), only acetophenone
gave the corresponding secondary alcohol, 1-phenylethanol,
leading to less than 30% conversion working under harsher
ꢀ
conditions (80 C under 20 bar H2 for 24 h) (Scheme S2 in the
ESI†); no formation of the corresponding acetal derivative by
reaction of the carbonyl group with glycerol was detected. We also
studied the ability of these catalysts on the hydrogenation of
We also evaluated the hydrogenation of oleic acid (entry 8),
because of the industrial interest of saturated fatty acids (e.g. as
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 93205–93216 | 93211

View Article Online
RSC Advances
Paper
Table 3 Pd-catalysed synthesis of secondary and tertiary amines from
carbonyl and amine reagents under hydrogen pressurea
Conv. (yield)b
Entry
1
R1
Amine
Product
(%)
Scheme 3 One-pot multi-step synthesis of N-substituted anilines
catalysed by PdNPs in glycerol.
OMe (22)
Cy-NH2 (e)
93 (91)
In order to get a deeper insight in the synthesis of primary
and secondary amines from nitro-compounds, we envisaged the
one-pot procedure by reaction of amine and carbonyl-based
reactants in the presence of PdNPs under hydrogen atmo-
sphere, taking advantage of the lack or low reactivity of ketones
and aldehydes compounds (Table 3). This procedure obviously
involves the in situ formation of an imine (condensation
between primary amine and aldehyde or ketone) or iminium
(condensation involving a secondary amine) intermediate.25 We
previously proved the efficiency of PdIa and PdId on the
reduction of imines using N-benzylideneaniline (24) as model
imine-based substrate (Scheme S3 in the ESI†). We chose 4-
methoxybenzaldehyde (22) as model substrate (entries 1–3,
Table 3), which was not hydrogenated with the catalysts used
(see above), using primary (cyclohexylamine, e) and secondary
(piperidine and morpholine, f and g respectively) amines.
Conversions and yields were high, but somewhat lower for
secondary amines than for cyclohexylamine (entries 2–3 vs. 1).
The same trend was observed for benzaldehyde and 4-(tri-
uoromethyl)benzaldehyde (entries 4–5 and 6–7), proving that
electronic effects are no signicant. In the case of cyclohexyl-
amine, the corresponding tertiary amines were not formed,
even using an excess of aldehyde (entries 1, 4 and 5); this
behaviour points to the fact that the secondary amines 20e–22e
are sterically demanding enough to avoid the formation of the
corresponding iminium intermediate (entries 1, 4 and 5), in
contrast to that observed for piperidine and morpholine
(entries 2–3 and 6–7). We also evaluated the reactivity involving
ammonia (h) instead of organic amines (entries 8–10). In this
case, we observed an inuence of the aldehyde nature. Using
benzaldehyde (20) and 4-methoxybenzaldehyde (22), a mixture
of the corresponding secondary and tertiary amines were ob-
tained (entries 8 and 9), while with 4-(triuoromethyl)benzal-
dehyde (21), the secondary amine 21h was exclusively formed in
high yield (91%, entry 10). This behaviour proves that aldehydes
containing electron-withdrawing groups favour the formation
of the corresponding imine, but the iminium intermediate is
disfavoured as can be expected (Scheme S4 in the ESI†).
Unfortunately ketones were not reactive.
2
3
OMe (22)
OMe (22)
81 (77)
77 (73)
4
5
H (20)
Cy-NH2 (e)
Cy-NH2 (e)
99 (96)
98 (95)
CF3 (21)
6
7
H (20)
78 (76)
80 (76)
CF3 (21)
62 (40)d
91 (56)e
95 (91)
c
8
9
H (20)
NH3 (h)
c
OMe (22)
CF3 (21)
NH3 (h)
10f
NH3 (h)
c
a
Results from duplicated experiments. Reaction conditions: 1 mmol of
aldehyde (20–22), 1 mmol of the corresponding amine (e–h) and 1 mL of
the catalytic glycerol solution of PdId (10ꢂ2 mol Lꢂ1, 0.01 mmol of total
Pd), under 3 bar of H2 at 100 ꢀC for 2 h. b Determined by GC and GC-MS
using decane as internal standard. c 32% w/w aqueous solution NH3 (55
mL, 1 mmol). d 20% of tris(benzyl)amine was also obtained. e 34% of the
corresponding tris(benzyl)amine was also obtained, N(4-OMe-C6H4)3.
f
For 12 h.
aldehydes (20–23). Only benzaldehyde was hydrogenated to give
benzyl alcohol (60% yield) but in this case under relative smooth
conditions (Scheme S2 in the ESI†). In contrast to acetophenone,
the ve-membered acetal coming from benzaldehyde and glycerol
was obtained as by-product in ca. 30% yield.
Following this line-up, we could prepare N-substituted
anilines by one-pot three-steps sequential process;26 it means
the reduction of nitrobenzene, then the formation of the cor-
responding imine by condensation with the appropriate
93212 | RSC Adv., 2016, 6, 93205–93216
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Table 4 Pd-catalysed one-pot synthesis of N-substituted anilines Table
5
Pd-catalysed hydrodebromination processes under
from nitrobenzene and aldehydes under hydrogen pressurea
hydrogen pressurea
Conv.b
(%)
Products N-aniline/imineb
(yields, %)
Entry
1
R
Ph (20)
>99c
Entry
Catalyst
Substrate
pH2 (bar)
Conv. (yield)b (%)
1
2
PdIa
PdId
28
28
3
3
25 (15)
99 (94)
99 (94) 2nd run
97 (93) 3rd run
92 (89) 4th run
99 (91)
<5
40 (36)
<5
<5
2
3
4-CF3-C6H4 (21)
4-OMe-C6H4 (22)
97
3
PdIdH
PdTPPTS
PdId
PdId
PdId
PdId
PdId
PdId
PdId
PdId
PdId
PdId
PdId
PdId
Pd/C
28
28
28
29
30
31
29
30
31
32
33
32
33
31
31
3
3
3
3
3
4c
5d
6
>99
7
8
3
<5
9e
10
10
10
3
67 (65)
89 (86)
68 (65)
38 (32)
<5
77 (71)
10 (nd)f
98 (95)
97 (94)
45 (40) 2nd run
98 (94)
10e
11e
12
13
14
15
16g
17h
4
Bu (23)
86d
3
10
10
3
a
Results from duplicated experiments. Reaction conditions: 1 mmol of
aldehyde (20–23), 1 mmol of nitrobenzene and 1 mL of the catalytic
glycerol solution of PdId (10ꢂ2 mol Lꢂ1, 0.01 mmol of total Pd), under
3
18g,h
Pd/C
31
3
b
3 bar of H2 at 100 ꢀC for 2 h. Determined by GC and GC-MS using
c
d
a
decane as internal standard. 20% of Bn–OH. ca. 5% of N,N-
bis(pentyl)-N-phenyl-amine.
Results from duplicated experiments. Reaction conditions: 1 mmol of
aryl-halide (28–33) and 1 mL of the corresponding catalytic glycerol
solution (10ꢂ2 mol Lꢂ1, 0.01 mmol of total Pd), under 3 or 10 bar of
b
H2 at 100 ꢀC for 6 h. Determined by GC and GC-MS using decane as
c
internal standard. TPPTS ¼ tris(3-sulfophenyl)phosphine trisodium
d
e
f
g
salt. At 80 ꢀC. For 24 h. nd ¼ not determined. For 2 h in the
aldehyde, followed by its hydrogenation to give the N-
substituted aniline (Scheme 3).
h
presence of 1.2 mmol of KOH. Pd/C 10 wt 10% (Aldrich).
We applied this strategy using aromatic- (20–22) and alkyl-
(23) aldehydes (Table 4). We mainly observed in all the cases the
formation of the anticipated secondary amine (24–27, entries 1–
4). The corresponding imine (24im–27im) could be also detec-
ted, in particular for 4-(triuoromethyl)benzaldehyde, up to
20% (entry 2). In the case of pentanal, two by-products were
detected together with N-pentyl-N-phenylamine 27 (72%); in
addition to the imine 27im (10%), the tertiary amine N,N-bis(-
pentyl)-N-phenyl-amine was also formed (5%, entry 4), in
contrast to the behaviour observed for the secondary benzyl-
amines 24–26.
impact. Catalytic hydrodehalogenation mediated by dihydrogen
represents a sustainable approach as alternative of incinera-
tion, involving cleavages of C–X bonds which can lead to less
toxic and recyclable materials.27
We selected 4-bromobenzonitrile for evaluating the ability of
our catalysts in the hydrodehalogenation process (Table 5).
Under 3 bar H2 pressure at 100 ꢀC for 6 h using 1 mol% of
catalyst (entries 1–4), we could observe that the best results were
achieved with PdId and PdIdH: full conversion with more than
90% yield in benzonitrile (entries 2–3), in contrast to the low
reactivity observed using PdIa as catalyst (entry 1). PdId could be
recycled four times without loss of reactivity (entry 2). PdNPs
stabilised by TPPTS were inactive (entry 4).12a,b Lower
Hydrodehalogenation processes. We were also interested in
Pd-catalysed hydrodehalogenation of haloarenes, important
concern in the treatment of wastes. Halogen compounds are
typically xenobiotic which presume a negative environmental
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 93205–93216 | 93213

View Article Online
RSC Advances
Paper
95% were achieved with PdId except for 37 which gave only 12%
conversion (Scheme 4); PdIa was much less active (Scheme 4). In
contrast to the behaviour observed for hydrodebromination, Pd/
C was less efficient than PdId giving up to 48% of phenol in the
absence of base; in the presence of KOH, only 10% conversion
was achieved (for conditions, see Scheme 4). These results
points to a poisoning of the heterogeneous catalyst by the base
in glycerol, oppositely to the benecial effect triggered on the
nanocatalyst PdId.
Conclusions
In summary, we prepared palladium nanoparticles capped by
cinchona-based ligands in neat glycerol. Important effects in
size and dispersion of PdNPs were observed, depending on the
palladium precursor involved and also the nature of stabiliser.
The most homogeneously dispersed systems showing a small
size (ca. 1.5–2.0 nm) were those stabilised by cinchonidine (a)
and quinidine (d), using palladium acetate as metal precursor,
PdIa and PdId respectively. The characterisation, both at solid
state and in the glycerol phase, evidenced that only Pd(0) was
present in any case. Cinchona-based ligands are linked at the
metal surface in solution and in solid state, probably adopting
a relative weak interaction with the metal surface through N-
lone pair of the quinolone ring more than a p-binding coordi-
nation, in agreement with the literature.14 In glycerol phase,
both solvated quinidine and quinidine linked to the metal
surface were identied by XPS and 15N NMR analyses. PdIa and
PdId dispersed in glycerol were successfully applied in
dihydrogen-based processes, such as hydrogenation of unsat-
urated functional groups (alkenes, alkynes, imines, nitro-based
substrates) and hydrodehalogenation of halo-aromatic
compounds, taking advantage of their surface-like reactivity.
Glycerol played an important role as support for the immobi-
lisation of PdNPs, permitting an easy recycling of the catalytic
phase. The versatile behaviour of these catalysts, mainly PdId,
allowed the synthesis of secondary and tertiary amines, and N-
substituted anilines by a one-pot methodology without isolation
of intermediates.
Scheme
4 Hydrodechlorination catalysed by PdNPs in glycerol.
Figures represent conversions and yields (in brackets) using PdId as
catalyst; in italics, data corresponding using PdIa (nd ¼ not deter-
mined). aIn the absence of KOH, 48% conversion (45% yield).
temperatures affected markedly the conversion (entry 5 vs. 2).
Under smooth conditions, 4-bromoanisole (29), 4-bromoaniline
(30) and 4-bromophenol (31) did not react (entries 6–8),
obtaining the corresponding debrominated compound under
harsher conditions (10 bar of H2 for 24 h), in moderate to good
yields (65–86% yields, entries 9–11). Using 2-bromobenzonitrile
and 2-bromoaniline as substrates, the activity was negligible, as
consequence of the steric hindrance (entries 12–13). The pres-
sure increase (10 bar H2) gave 77% conversion towards benzo-
nitrile (entry 14), but only 10% of conversion was observed for
the hydrodebromation of 2-bromoaniline (entry 15), indicating
that electro-donor groups slowed down the process, as observed
for 29–31 in comparison with 4-bromobenzonitrile (entries 9–11
vs. 2).
With the aim of improving the hydrodehalogenation under
relative low H2 pressure, we then decided to add a base in the
medium in order to trap the hydrogen bromide formed as
concomitant product.28 Under these basic conditions, 95% yield
of phenol was obtained from 4-bromophenol working under 3
bar H2 at 100 ꢀC for 2 h (entry 16 vs. 8). For comparative
purposes, the dehydrodebromination of 4-bromophenol (31)
was conducted using the heterogeneous Pd/C catalyst both in
the absence and in the presence of base (entries 17 and 18,
respectively), leading to the same reactivity (95% yield of
phenol), likewise to PdId (entry 16). But the activity fell aer
reuse of the heterogeneous catalyst (entry 17), in contrast to the
colloidal catalyst (entry 2). It is important to mention that the
stabiliser (quinidine, d) remains unmodied aer catalysis, as
checked by 1H NMR (for catalytic conditions, see entry 2 of
Table 5).29
Further studies involving metal-based nanoparticles in
glycerol containing naturally-occurring stabilisers are currently
developed for applications in organic synthesis.
Acknowledgements
Financial support from the Centre National de la Recherche
´
Scientique (CNRS), the Universite de Toulouse 3 – Paul
Sabatier, the Associated International Laboratory LIA-LCMMC
and CONACYT FOINS-246968 are gratefully acknowledged.
Authors thank L. Calvo for his helpful discussions concerning
XPS data. A. R. thanks CONACYT for a PhD grant.
More challenging substrates are chloro-arenes. At 100 ꢀC
under 10 bar H2 for 24 h, 4-chlorobenzonitrile (36), 4-chlor-
oanisole (37), 4-chloroaniline (38) and 4-chlorophenol (39) were
inactive using both PdIa and PdId (1 mol%) as catalysts, in the
absence of base. However in the presence of KOH and working
under smoother conditions (100 ꢀC, 3 bar H2, 2 h), yields up to
Notes and references
1 For a selected contribution, see: J. Tsuji, Palladium reagents
and catalysts, John Wiley & Sons, Sussex, 2004.
93214 | RSC Adv., 2016, 6, 93205–93216
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
2 For selected reviews, see: (a) R. Ciriminna, V. Pandarus,
91; (c) A. Roucoux, J. Schulz and H. Patin, Chem. Rev.,
2002, 102, 3757.
´
A. Fidalgo, L. M. Ilharco, F. Beland and M. Pagliaro, Org.
´
Process Res. Dev., 2015, 19, 755; (b) S. Paul, M. M. Islam 11 F. Chahdoura, C. Pradel and M. Gomez, ChemCatChem,
and S. M. Islam, RSC Adv., 2015, 5, 42193; M. Opanasenko, 2014, 6, 2929.
P. Stepnicka and J. Cejka, RSC Adv., 2014, 4, 65317. (c) 12 (a) F. Chahdoura, C. Pradel and M. Gomez, Adv. Synth.
´
´
P. Taladriz-Blanco, P. Herves and J. Perez-Juste, Top. Catal.,
Catal., 2013, 355, 3648; (b) F. Chahdoura, S. Mallet-Ladeira
´
2013, 56, 1154; (d) M. O. Sydnes, Curr. Org. Synth., 2011, 8,
881; (e) I. Beletskaya and V. Tyurin, Molecules, 2010, 15, 4792.
3 For selected contributions, see: (a) D. J. Gavia and Y.-S. Shon,
and M. Gomez, Org. Chem. Front., 2015, 2, 312; (c)
F. Chahdoura, I. Favier, C. Pradel, S. Mallet-Ladeira and
M. Gomez, Catal. Commun., 2015, 63, 47.
´
ChemCatChem, 2015, 7, 892; (b) F. Chahdoura, I. Favier and 13 (a) P. J. Collier, J. A. Iggo and R. Whyman, J. Mol. Catal. A:
´
M. Gomez, Chem.–Eur. J., 2014, 20, 10884; (c) S. Navalon,
Chem., 1999, 146, 149; (b) M. F. Dumont, S. Moisan,
C. Aymonier, J.-D. Marty and C. Mingotaud,
Macromoleucles, 2009, 42, 4937; (c) E. Schmidt, W. Kleist,
F. Krumeich, T. Mallat and A. Baiker, Chem.–Eur. J., 2010,
16, 2181; (d) J. Keilitz, S. Nowag, J.-D. Marty and R. Haag,
Adv. Synth. Catal., 2010, 352, 1503; (e) F. Hoxha,
E. Schmidt, T. Mallat, B. Schimmoeller, S. E. Pratsinis and
A. Baiker, J. Catal., 2011, 278, 94; (f) Y. Huang, S. Xu and
V. S.-Y. Lin, ChemCatChem, 2011, 3, 690; (g) C. Richter,
K. V. S. Ranganath and F. Glorius, Adv. Synth. Catal., 2012,
354, 377; (h) F. Kirby, C. Moreno-Marrodan, Z. Baan,
B. F. Bleeker, P. Barbaro, P. H. Berben and P. T. Witte,
ChemCatChem, 2014, 6, 2904; (i) Z. Guan, S. Lu and C. Li, J.
Catal., 2014, 311, 1; (j) T. T. Si Bui, Y. Kim, S. Kim and
H. Lee, RSC Adv., 2015, 5, 75272; (k) T. S. B. Trung, Y. Kim,
S. Kang, S. Kim and H. Lee, Appl. Catal., A, 2015, 505, 319.
M. Alvaro and H. Garcia, ChemCatChem, 2013, 5, 3460; (d)
J. D. Scholten, B. C. Leal and J. Dupont, ACS Catal., 2012,
2, 184; (e) M. H. G. Prechtl, J. D. Scholten and J. Dupont, in
Palladium Nanoscale Catalysts in Ionic Liquids: Coupling and
Hydrogenation Reactions (Chapter 16) from Ionic Liquids:
Applications and Perspectives, ed. A. Kokorin, InTech, 2011,
pp. 393–414.
4 For selected contributions, see: (a) R. Rossi, F. Bellina,
M. Lessi, C. Manzini, G. Marianetti and L. A. Giulia, Curr.
Org. Chem., 2015, 19, 130; (b) B. M. Trost, Tetrahedron,
2015, 71, 5708; (c) Y. Liu, S.-J. Han, W.-B. Liu and
B. M. Stoltz, Acc. Chem. Res., 2015, 48, 740; (d) D.-L. Mo,
T.-K. Zhang, G.-C. Ge, X.-J. Huang, C.-H. Ding, L.-X. Dai
and X.-L. Hou, Synlett, 2014, 25, 2686; (e) C. Deraedt and
D. Astruc, Acc. Chem. Res., 2014, 47, 494; (f) A. Balanta,
¨
C. Godard and C. Claver, Chem. Soc. Rev., 2011, 40, 4973; 14 (a) D. Ferri and T. Burgi, J. Am. Chem. Soc., 2001, 123, 12074;
´
¨
(b) D. Ferri, T. Burgi and A. Baiker, Chem. Commun., 2001,
(g) I. Favier, D. Madec, E. Teuma and M. Gomez, Curr. Org.
¨
Chem., 2011, 15, 3127.
1172; (c) D. Ferri, T. Burgi and A. Baiker, J. Catal., 2002,
5 For selected contributions, see: (a) A. Bej, K. Ghosh, A. Sarkar
and D. W. Knight, RSC Adv., 2016, 6, 11446; (b) I. Saldan,
Y. Semenyuk, I. Marchuk and O. Reshetnyak, J. Mater. Sci.,
210, 160; (d) A. Kraynov, A. Suchopar, L. D'Souza and
R. Richards, Phys. Chem. Chem. Phys., 2006, 8, 1321; (e)
A. Vargas and A. Baiker, J. Catal., 2006, 239, 220.
2015, 50, 2337; (c) V. P. Ananikov and I. P. Beletskaya, 15 Only platinum nanoparticles immobilised in imidazolium-
´
Organometallics, 2012, 31, 1595; (d) M. Perez-Lorenzo, J.
based ionic liquids in the presence of cinchonidine were
previously applied in enantioselective hydrogenations:
M. J. Beier, J.-M. Andanson, T. Mallat, F. Krumeich and
A. Baiker, ACS Catal., 2012, 2, 337.
Phys. Chem. Lett., 2012, 3, 167; (e) A. Fihri, M. Bouhrara,
B. Nekoueishahraki, J.-M. Basset and V. Polshettiwar,
Chem. Soc. Rev., 2011, 40, 5181; (f) D. Astruc, Inorg. Chem.,
2007, 46, 1884.
16 T. Kusukawa, G. Niwa, T. Sasaki, R. Oosawa, W. Himeno and
M. Kato, Bull. Chem. Soc. Jpn., 2013, 86, 351.
6 (a) N. T. S. Phan, M. V. Sluys and C. W. Jones, Adv. Synth.
´
Catal., 2006, 348, 609; (b) J. Durand, E. Teuma and 17 For a representative example, see: M. Gomez, K. Philippot,
´
`
M. Gomez, Eur. J. Inorg. Chem., 2008, 3577; D. Astruc, F. Lu
V. Colliere, P. Lecante, G. Muller and B. Chaudret, New J.
and J. R. Aranzaes, Angew. Chem., Int. Ed., 2005, 44, 7852.
Chem., 2003, 27, 114.
7 (a) C. S. Consorti, P. A. Z. Suarez, R. F. de Souza, R. A. Burrow, 18 For a selected work, see: H. I. Lee, S. H. Joo, J. H. Kim,
D. H. Farrar, A. J. Lough, W. Loh, L. H. M. da Silva and
J. Dupont, J. Phys. Chem. B, 2005, 109, 4341; (b) J. Dupont,
D. J. You, J. M. Kim, J.-N. Park, H. Chang and C. Pak, J.
Mater. Chem., 2009, 19, 5934.
J. Braz. Chem. Soc., 2004, 15, 341; (c) I. Favier, D. Madec 19 Commercially available cinchonidine (a) and quinidine (d)
´
and M. Gomez, in Metallic nanoparticles in ionic liquids –
contain less than 8% of the corresponding hydro-alkaloid
(a/aH ¼ 24/1 (4% aH); d/dH ¼ 12/1 (8% dH), see Fig. S7 in
the ESI†).
Applications in catalysis (Chapter 5) from Nanomaterials in
Catalysis, ed. P. Serp and K. Philippot, Wiley-VCH,
Weinheim, 2013, pp. 203–249.
20 I. Favier, P. Lavedan, S. Massou, E. Teuma, K. Philippot,
´
B. Chaudret and M. Gomez, Top. Catal., 2013, 56, 1253.
8 (a) A. J. Biacchi and R. E. Schaak, ACS Nano, 2011, 5, 8089; (b)
´
I. Favier, E. Teuma and M. Gomez, C. R. Chim., 2009, 12, 533. 21 The diameter of MNPs considering a compact packing (for
9 For a selected review, see: I. Geukens and D. E. De Vos,
Langmuir, 2013, 29, 3170.
10 For selected reviews, see: (a) C. Amiens, D. Ciuculescu-
Pradines and K. Philippot, Coord. Chem. Rev., 2016, 308,
409; (b) E. Gross and G. A. Somorjai, J. Catal., 2015, 328,
PdNPs, fcc structure) gives the number of layers and in
consequence the total number of metal atoms in the
nanocluster. Then for PdIa with a mean diameter of 2.1
nm, the nanoclusters are constituted by 4 shells of Pd
atoms leading to 309 atoms per nanoparticle; for PdId
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 93205–93216 | 93215

View Article Online
RSC Advances
Paper
(mean diameter of 1.4 nm), each nanoparticle is constituted 27 For a selected review, see: M. A. Keane, ChemCatChem, 2011,
by 3 shells, then 147 atoms per nanocluster see G. Schmid,
3, 800.
Chem. Rev., 1992, 92, 1709.
28 T. Hara, T. Kaneta, K. Mori, T. Mitsudome, T. Mizugaki,
K. Ebitani and K. Kaneda, Green Chem., 2007, 9, 1246.
´
22 A. M. Lopez-Vinaso, I. Favier, C. Pradel, L. Huerta,
´
´
I. Guerrero-Rıos, E. Teuma, M. Gomez and E. Martin, 29 The hydrodehalogenation of 4-bromophenol was carried out
Dalton Trans., 2014, 43, 9038.
at larger scale than that described for the rest of reactions
studied in this manuscript: a scale-up of three times,
meaning that 3 mL of catalytic solution of PdId was used,
with the aim of characterising the stabiliser by 1H NMR
aer catalysis. Conversion and yield were similar to those
obtained using 1 mL of catalytic solution (98% conversion
and 93% yield). Aer extraction of phenol, 3 mmol of 1-
dodecanethiol were added to the glycerol catalytic phase
under argon and stirring for 24 h, with the aim of
removing the stabiliser from the metal surface. Extraction
with dichloromethane from the glycerol phase permitted
to recover the ligand as a mixture of quinidine (d) and
hydroquinidine (dH), in a ratio 70/30 respectively (see
Fig. S20†), analogously to that obtained aer the synthesis
of PdId (see above in the section Synthesis and
characterisation of palladium nanoparticles).
23 For a contribution concerning the binding energies for N 1s
of organic compounds, see: W. J. Gammon, O. Kra,
C. Reilly and B. C. Holloway, Carbon, 2003, 41, 1917.
24 For some selected recent works, see: (a) V. V. Zhivonitko,
I. V. Skovpin, M. Crespo-Quesada, L. Kiwi-Minsker and
I. V. Koptyug, J. Phys. Chem. C, 2016, 120, 4945; (b)
K.-H. Dostert, C. P. O'Brien, F. Ivars-Barcelo,
S. Schauermann and H.-J. Freund, J. Am. Chem. Soc., 2015,
137, 13496; (c) J. Sauer and H.-J. Freund, Catal. Lett., 2015,
145, 109; (d) M. A. Mahmoud, R. Narayanan and M. A. El-
Sayed, Acc. Chem. Res., 2013, 46, 1795.
25 J. Wu, S. Lu, D. Ge and H. Gu, RSC Adv., 2015, 5, 81395.
26 For some selected contributions, see: (a) G. Guillena,
´
D. J. Ramon and M. Yus, Chem. Rev., 2010, 110, 1611; (b)
¨
S. Bahn, S. Imm, L. Neubert, M. Zhang, H. Neumann and
M. Beller, ChemCatChem, 2011, 3, 1853.
93216 | RSC Adv., 2016, 6, 93205–93216
This journal is © The Royal Society of Chemistry 2016