Palladium nanocatalysts in glycerol: Tuning the reactivity by effect of the stabilizer

Article abstract of DOI:10.1016/j.catcom.2017.10.004

Palladium nanoparticles (PdNPs) prepared in neat glycerol containing TPPTS (tris(3-sulfophenyl)phosphine trisodium salt) or cinchona-based alkaloids (cinchonidine, quinidine) as capping agents, were applied as catalysts in fluoride-free Hiyama couplings and conjugate additions with the aim of evaluating the influence of the stabilizer in the catalytic reactivity. Therefore, PdNPs stabilized by phosphine favored C–C cross-couplings, whereas those containing alkaloids showed enhanced suitability for C–C homo-couplings and conjugate additions. The metal/stabilizer coordination mode, i.e. Pd–P dative bond and π-π interaction between quinoline moiety and palladium surface, is certainly key for the stabilization of different active metallic species and then promoting distinctive catalytic pathways.

Full text of DOI:10.1016/j.catcom.2017.10.004

Accepted Manuscript
Palladium nanocatalysts in glycerol: Tuning the reactivity by
effect of the stabilizer
Antonio Reina, Alejandro Serrano-Maldonado, Emmanuelle
Teuma, Erika Martin, Montserrat Gómez
PII:
DOI:
Reference:
S1566-7367(17)30415-6
doi:10.1016/j.catcom.2017.10.004
CATCOM 5215
To appear in:
Catalysis Communications
Received date:
Revised date:
Accepted date:
20 July 2017
30 September 2017
3 October 2017
Please cite this article as: Antonio Reina, Alejandro Serrano-Maldonado, Emmanuelle
Teuma, Erika Martin, Montserrat Gómez , Palladium nanocatalysts in glycerol: Tuning the
reactivity by effect of the stabilizer. The address for the corresponding author was captured
as affiliation for all authors. Please check if appropriate. Catcom(2017), doi:10.1016/
j.catcom.2017.10.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Palladium nanocatalysts in glycerol: tuning the reactivity by effect of the
stabilizer
Antonio Reinaa, Alejandro Serrano-Maldonadob, Emmanuelle Teumaa, Erika Martinb, Montserrat
Gómeza,*
a Université de Toulouse, UPS, LHFA, 118 route de Narbonne, F-31062 Toulouse Cedex 9, France
and CNRS, LHFA UMR 5069, F-31062 Toulouse Cedex 9, France. Fax: +33 561558204; Tel: +33
561557738; e-mail: gomez@chimie.ups-tlse.fr.
b Departamento de Química Inorgánica, Facultad de Química, Universidad Nacional Autónoma de
México, Av. Universidad 3000, 04510 México DF, México.
Dedicated to the memory of Prof. Erika Martin (deceased, March 4th, 2017)

ACCEPTED MANUSCRIPT
ABSTRACT: Palladium nanoparticles (PdNPs) prepared in neat glycerol containing TPPTS (tris(3-
sulfophenyl)phosphine trisodium salt) or cinchona-based alkaloids (cinchonidine, quinidine) as
capping agents, were applied as catalysts in fluoride-free Hiyama couplings and conjugate additions
with the aim of evaluating the influence of the stabilizer in the catalytic reactivity. Therefore, PdNPs
stabilized by phosphine favored C-C cross-couplings, whereas those containing alkaloids showed
enhanced suitability for C-C homo-couplings and conjugate additions. The metal/stabilizer
coordination mode, i.e. Pd-P dative bond and interaction between quinoline moiety and palladium
surface, is certainly key for the stabilization of different active metallic species and then promoting
distinctive catalytic pathways.
Keywords: Palladium nanoparticles – Cinchona alkaloids – Phosphine – Glycerol – Hiyama couplings
– Conjugate additions

ACCEPTED MANUSCRIPT
1. Introduction
Palladium is one of the most used metals in catalysis, due to its versatility in inducing different types
of activations (C-halide bonds by oxidative addition; C-H bonds by metalation; small molecules
activation such as H2 or CO…) [1]. This multipurpose reactivity finds its origin in the palladium
ability for stabilizing different species exhibiting different oxidation states in the reaction medium:
well-defined mono-metallic complexes, small clusters, nanoparticles, in addition to the distinctive
reactivity of the extended metal surface [2,3]. The design of ligands and stabilizers together with the
fine-tuning of the reaction conditions, enables the formation of a large variety of metal-based
structures both at molecular and nanometric scale, involving dynamic equilibria among them [4]. This
behavior can lead to a dual reactivity for palladium nanoparticles (PdNPs), both molecular- and
surface-like, mainly when working under wet conditions, finding useful applications in one-pot multi-
step processes (e.g. C-C coupling followed by hydrogenation) [5,6]. Furthermore, the nature of
stabilizers can favor the formation of complexes or nanoclusters depending on their coordination
ability [7]. Thus, polymers, which in general do not present strong interaction with metals, promote
aggregated species (small clusters, nanoparticles) avoiding at the same time the formation of bulk
metal. On the contrary, ligands involving strong Lewis donor centers (such as P- or S-based structures)
can stabilize molecular complexes (both Pd(0) and Pd(II) species) using high ligand/metal ratios (at
low ligand/metal ratios nanoparticles are also favored). Furthermore, ligands offering the possibility to
interact with metal centers through aromatic moieties ( metal/ligand interaction), also lead to the
stabilization of metal nanoparticles [8,9].
With the purpose of developing sustainable processes, the catalyst recovery (Pd/ligand or stabilizer
systems) and in consequence its efficient immobilization, preserving both activity and selectivity,
become crucial factors to obtain catalyst-free target organic products. The immobilization of PdNPs on
solids has been extensively studied, involving a large variety of supports, e.g. carbon-based materials,
zeolites, metal oxides, polymers or magnetic composites, among the most used [10,11].
However, PdNPs immobilized on liquid phases have been less considered, apart from the developed
research with ionic liquids [12,13], including catalytic supported ionic liquid phases (SILP) [14-16]
and functionalized membranes with poly(ionic liquids) [17,18]. Looking for alternative solvents
showing suitable physico-chemical properties for synthesis and catalysis (e.g. high boiling point,
negligible vapor pressure, high solubilizing ability for organic and inorganic compounds) [19-21],
glycerol appeared us an appropriate choice taking into account that its H-bonded supramolecular
structure [22], permits to confine metal nanoparticles. Glycerol enables an efficient immobilization of
metal species, allowing at the same time a straightforward extraction of the corresponding organic
products from the catalytic phase in the absence of metal [23,24]. In addition, its low vapor pressure
permits direct analyses of the solution by TEM and XPS techniques [19-21]. Our previous work
proved that PdNPs in glycerol containing two different types of stabilizers, TPPTS (tris(3-
sulfophenyl)phosphine trisodium salt) and cinchona-based alkaloids (cinchonidine and quinidine),
were efficient catalysts in molecular- (Suzuki, Heck and Sonogashira couplings) [23] and surface-like
reactivity (hydrogenations and hydrodehalogenations) [24], respectively.
Herein, we have chosen the study of Hiyama coupling reactions with the aim of evaluating the
differences in terms of selectivity towards homo- or cross-coupling pathways, induced by PdNPs of
nearly the same size but capped by two different kinds of stabilizers, i.e. phosphine- and cinchona-
based ligands. Furthermore, Michael conjugate additions permitted the coordination chemistry study at
the palladium surface between stabilizers and reagents, revealing a positive catalytic effect in the case
of PdNPs stabilized by quinidine.
2. Experimental

ACCEPTED MANUSCRIPT
General details (materials and instrumental techniques), ligand exchange reactions at the palladium
nanoparticles surface, 31P NMR monitoring of the formation of palladium species during the oxidative
addition step in Hiyama couplings, and characterization of both palladium nanoparticles and organic
compounds obtained in the catalytic processes including their purification, are provided in the
Supporting Information.
2.1. General procedure for the synthesis of palladium nanoparticles in glycerol, PdA-PdC
0.05 mmol of Pd(OAc)2 (11.2 mg) and 0.05 mmol of ligand (28.4 mg for TPPTS (A); 14.7 mg for
cinchonidine (B); 16 mg for quinine (C)) 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. Nanoparticles both in glycerol solution and at solid state were fully characterized by
conventional techniques ((HR)TEM, IR, NMR, XPS, XRD, EA) [23,24].
2.2. General procedure for Pd-catalyzed Hiyama cross-coupling in glycerol
In a Schlenk tube, substrate (0.5 mmol), silane (0.6 mmol), base (1.5 mmol) and decane (71 mg, 0.5
mmol) as internal standard were added to 1 mL of preformed palladium nanoparticles (total amount of
palladium 0.01 mmol) under argon. The reaction mixture was heated at 80 °C and stirred for the
indicated time (3 – 24 h), and then cooled down to room temperature. The organic products were
extracted from glycerol with dichloromethane (5 x 3 mL). The obtained products were previously
described in the literature and their identification was carried out by comparison of their GC-MS data,
1H and 13C NMR spectra with the reported data (see Table S1 in the Supporting Information).
2.3. General procedure for Pd-catalyzed conjugate additions in glycerol
In a pressure cap tube, substrate (1 mmol), the corresponding amine, thiol or phosphine (1 mmol) and
decane (71 mg, 1 mmol) as internal standard were added to 1 mL of preformed palladium
nanoparticles (total amount of palladium 0.01 mmol) under argon. The reaction mixture was heated up
to the desired temperature (35 – 100 °C) and stirred for the indicated time (0.25 – 24 h), and then
cooled down to room temperature. Organic products were extracted from glycerol with
dichloromethane (5 x 3 mL). The obtained products were previously described in the literature and
13
their identification was done by comparison of their GC-MS data, 1H and C NMR spectra with the
reported data (see Table S1 in the Supporting Information).
2.4. Catalytic phase recycling
For each kind of reaction (i.e. Hiyama reactions and conjugate additions), the catalytic phase recycling
followed the same procedure. The first run was conducted as explained above (see 2.2 and 2.3
sections). After extraction of the organic products by a biphasic method using dichloromethane (5 x 3
mL), the glycerol phase was treated under vacuum at 50 ºC for 1.5 h, and then cooled down to room
temperature. Reagents were then added to the catalytic solution under argon as indicated above and the
corresponding procedure was repeated.
3. Results and discussion
3.1. Synthesis and characterization of palladium nanocatalysts
In the present study, preformed PdNPs synthesized from Pd(OAc)2 in the presence of the
corresponding stabilizer (A-C) in neat glycerol, under dihydrogen atmosphere, were used as catalysts,
based on our previous studies (Scheme 1) [23,24]. The resulting black colloidal solutions were
constituted by small (mean diameters in the range 1.4 – 2.1 nm), spherical and homogeneously
dispersed Pd(0) nanoparticles as evidenced by TEM and XPS analyses (Figs. S1 and S2 in the

ACCEPTED MANUSCRIPT
Supporting Information); PdNPs showed the same morphology in both colloidal solutions and solid
state (Figs. S3 and S4 in the Supporting Information).
Scheme 1 HERE
3.2. Michael-type conjugate additions catalyzed by PdNPs stabilized by cinchona-based alkaloids
Substrates involved in aza-, thio- and phosphino-Michael conjugate additions present a remarkable
coordination ability [25], which can trigger a displacement of the stabilizer present at the palladium
surface. This coordination competition between substrate and stabilizer can serve as a rational to
explain the catalytic activity induced by the different catalytic systems. Thus, strong stabilizer
coordination can lead to low catalytic activity; in contrast, labile stabilizers will be fast exchanged
with the reagents favoring the corresponding conjugate addition. However, in this latter case,
agglomeration issues can be observed due to the removal of the stabilizer from the surface, resulting in
a loss of activity after the first run.
With the aim of studying the plausible lability of quinidine due to its weaker interaction with the
palladium surface (by  interaction through the quinoline fragment) in comparison with TPPTS (by
Pd-P dative bonds), we carried out a NMR monitoring study corresponding to the ligand exchange
reaction between PdNPs stabilized by quinidine (PdC isolated at solid state by centrifugation from the
glycerol phase) and TPPTS in a mixture of THF-d8/D2O (1/1), using triphenylphosphine oxide and
cyclooctane as external standards (placed in a capillary in THF-d8; see Fig. S5 in the Supporting
Information). After 1 h, we observed the presence of free quinidine (as a mixture of quinidine and
dihydroquinidine resulting of the partial hydrogenation of the vinyl group [24]) up to 60% in relation
to the ligand present in the nanoparticles (0.01 mmol) and simultaneously, the signal disappearance of
free TPPTS caused by its approach to the palladium surface, up to 62% in relation to the added
phosphine (for details, see Experimental part in the Supporting Information); this means that ca. 60%
of the cinchona ligand was replaced by the added phosphine. However, when the ligand exchange was
done starting with palladium nanoparticles stabilized by phosphine, i.e. PdA, in the presence of free
quinidine, no exchange was observed, corroborating the robustness of the TPPTS bonded at the
palladium surface.
With the purpose of evidencing this effect by catalytic reactivity, we tested PdA, PdB and PdC as
catalytic precursors in the chosen benchmark base-free reaction between acrylonitrile (1) and
morpholine (a) (Table 1). No important differences were observed under harsh conditions (100 °C, 1
mol% Pd), giving more than 93% yield towards the desired product 1a (entries 1-3, Table 1); the
catalytic phases remained as black colloidal solutions. However, the catalytic system PdC (stabilized
by quinidine) was also highly active under smoother conditions, leading to 80% yield using 0.1 mol%
palladium load at 80 °C in 2 h (entry 4, Table 1); under these conditions, PdA, stabilized by TPPTS,
only afforded 50% conversion after 12 h of reaction (entry 5, Table 1). The reaction was monitored
during the first hour by GC-MS, evidencing that the catalyst remains active during the process and
showing a TOF of 1,440 h-1 after 15 min of reaction; the initial reaction rates (calculated between 6
and 12 min, meaning up to 25% conversion) were in the range 2.5x10-3 – 3.5x10-3 molL-1s-1 (see Fig.
S6 in the Supporting Information). Accordingly, PdC was efficiently recycled up to ten times without
significant activity loss (Fig. 1); ICP analyses indicated a negligible loss of palladium after each run
(0.004 – 0.01 ppm, determined by ICP-OES) and the absence of quinidine. To the best of our
knowledge, PdC represents the most efficient and recyclable catalyst for Michael-type additions
reported up to now [23,26].
Table 1 HERE
Figure 1 HERE

ACCEPTED MANUSCRIPT
The reaction scope involving primary (b,c) and secondary (d) amines, aqueous ammonia (e), thiols (f,
g) and secondary phosphines (h, i) is illustrated in Scheme 2, using PdC as catalytic precursor. In all
cases, the reaction proceeded selectively, leading to the expected products in high yields, except for
the alkyl thiol (g) and secondary phosphine (i) where moderate conversions (31-67%) and yields (26-
64%) were obtained. No asymmetric induction was prompted using methacrylonitrile (3) as prochiral
substrate using morpholine, cyclohexylamine and 4-methylthiophenol as Michael donor reagents (see
Table S2 and Fig. S7 in the Supporting Information). This result is in agreement with the de-
coordination of the cinchona-based stabilizer from the palladium surface that permits the conjugate
addition.
To sum up, the high activity shown by PdC compared with PdA agrees with the higher coordination
lability of the cinchona-based stabilizers than that exhibited by TPPTS. Quinidine can be de-
coordinated from the metal surface in the presence of donor compounds (such as N-, S- or P-based
reagents), favoring then the coordination of reagents at the palladium surface followed by the
conjugate addition, which can be otherwise obstructed in the presence of the bulky phosphine TPPTS.
The efficient recycling of the catalytic phase together with the absence of quinidine in the products
seem to indicate that the cinchona stabilizer coordinates again at the palladium surface after the
reaction is completed. Moreover, this catalytic behavior showed the positive role of palladium
nanoparticles in conjugate additions (for control tests, see Table S3 in the Supporting Information).
Scheme 2 HERE
3.3. Hiyama cross-couplings catalyzed by PdNPs in glycerol
It is very well-known that Pd-catalyzed C-C cross-couplings promoted by nanoparticles in wet phases,
work under homogeneous regime, i.e. preformed nanoparticles act as reservoirs of catalytically active
molecular species in particular for Heck reactions [5,27]. However, C-C homo-couplings can involve
activations of reagents at the metal surfaces through single electron transfer paths [28,29]. Taking into
account the different surface state caused by TPPTS and alkaloids B and C, we envisaged that
phosphines and alkaloids could favor different mechanisms for C-C couplings and therefore give
different products (i.e. homo-coupling or cross-coupling compounds) starting from the same reagents.
For this study, we chose the Hiyama C-C cross-coupling reaction because it represents a more
attractive way for the synthesis of biphenyls than other cross-couplings, such as Stille (involving
organotin reagents) or Suzuki (involving organoboron derivatives) reactions, due to the intrinsic
properties of organosilanes, such as stability, easy handling and low toxicity, together with the
straightforward conversion of the silicon-based by-products in innocuous silica waste [30].
Due to the scarce works reported in the literature using preformed PdNPs as catalytic precursors
[30,31], we first optimized the reaction conditions using PdA in glycerol for the catalytic reactions of
bromo-arenes [4-bromobenzotrifluoride (4Br), 4-bromonitrobenzene (5Br) and 4-bromobenzonitrile
(6Br)] and trimethoxyphenylsilane (j) as reagents (Table S4 in the Supporting Information). Under
these conditions, we studied the coupling between substrates containing both electron-withdrawing (4-
7) and electro-donor groups (8-10) (Table 2). For substrates exhibiting electron-withdrawing groups in
para-position (4-6), full conversions and excellent selectivity towards the corresponding cross-
coupling products (CC) were obtained for 4-bromobenzotrifluoride (4Br) and 4-bromobenzonitrile
(6Br) (entries 1 and 3, Table 2). At shorter times (3 h), the selectivity was preserved (entry 4, Table 2).
Special mention should be made in the case of 4-bromonitrobenzene (5Br), which reached almost
complete conversion in a short time (3 h), leading to a nearly equimolar mixture of 5j and 5HC, and in
consequence evidencing a strong competition between the transmetalation step (to give 5j) and the
homocoupling reaction (to give 5HC) (entry 2, Table 2). The monitoring of this reaction at 3 h and 24
h using a low catalyst load (1 mol%), indicated that the cross-coupling reaction is faster at high
concentrations of silylated reagent, but after ca. 50% conversion, the transmetalation step seems to
slow down, favoring the formation of the corresponding HC product (see entries 6-7, Table S4 in the

ACCEPTED MANUSCRIPT
Supplementary Material). In the case of 2-bromobenzonitrile (7Br), the steric hindrance of the CN
group undoubtedly prevents the oxidative addition of the arylbromide to the metal center and as a
result the reaction did not proceed (entry 5, Table 2). As expected, for substrates containing electron-
donor groups (entries 6-8, Table 2), yields were lower than those obtained with activated arylbromides
(entries 1-4, Table 2), and the selectivity towards the HC product was higher in particular for 4-
bromotoluene (entry 8, Table 2) where 10HC was mainly obtained. However, for 4-bromoanisole
(8Br), the CC product was privileged (entry 6, Table 2), probably due to the interaction of the
methoxide group with the palladium surface, thus favoring the oxidative addition step [33]. Chloro-
arenes 4Cl and 6Cl (entries 9-10, Table 2), despite containing electron-withdrawing substituents, only
gave low-to-moderate conversions. The hard activation of the C-Cl bond may lead to a different
mechanism than the conventional Hiyama cross-coupling pathway, mainly producing the
homocoupling product [34,35]. Analogously to trimethoxy(phenyl)silane (j), PdA catalyzed the
coupling between vinyltrimethylsiloxane (k) and 4-bromobenzotrifluoride (4Br) and 4-
bromobenzonitrile (6Br), showing a high activity and selectivity towards the CC product (> 90%
conversion after 3 h of reaction; see Scheme S1 in the Supporting Information); only small amounts of
oligomers were detected by GC (< 8%).
Table 2 HERE
It is important to highlight that for the reactions mainly giving the homocoupling product (HC), the
reaction medium remained as a black colloidal solution after catalysis, suggesting that the formation of
HC products arises from the surface reactivity of the palladium nanoparticles. In contrast, for the
reactions where the cross-coupling product (CC) was predominant, orange solutions were observed
once the process started, certainly due to the palladium leaching from the starting nanoparticles to give
molecular active complexes, specially favored by TPPTS.
These results indicate that for the substrates triggering a fast halo-arene oxidative addition, the
formation of homogeneous Pd(II) molecular species is preferred and then the cross-coupling process
takes place under a homogeneous regime. On the contrary, when the formation of Pd(II) species is
impeded (slow oxidative addition), in spite of the presence of phosphine, single electron transfer paths
are privileged [29], promoting a surface reactivity towards the homocoupling product. Actually, the
behavior of palladium in the presence of TPPTS was monitored by 31P NMR. Palladium acetate and
TPPTS in glycerol at 80 °C gave an orange solution, showing the apparition of a new signal at ca. 30
ppm (see Fig. S8 and Experimental section in the Supporting Information), evidencing the formation
of a Pd(II) complex, which decomposes in the presence of H2, giving a black colloidal solution of
PdNPs (with absence of NMR signals). Further addition of a bromo-arene, such as 4-
bromobenzotrifluoride, triggered the formation of Pd(II) complexes by oxidative addition on Pd(0),
observing again signals in the region 20-30 ppm; the resulting orange solution was stable at 80 °C for
long time (up to 24 h).
In the case of PdNPs stabilized by alkaloids (PdB and PdC) under the previously optimized
conditions, the Hiyama couplings only gave the corresponding homocoupling products (Table 3).
Curiously, PdB was more active than PdC (entries 2, 5 and 8 vs 3, 6 and 9, Table 3), probably due to
the stronger interaction of C with the palladium surface than that with B, due to the presence of the
methoxide group on the quinoline fragment [8,9,33]. This lack of reactivity of PdB and PdC to give
the expected Hiyama C-C cross-coupling products contrary to the efficiency exhibited by PdA,
correlates with the inactivity observed of PdB and PdC in other C-C cross-couplings using aryl
bromides as substrates (see Tables S5-S7 in the Supporting Information for Heck, Suzuki and
Sonogashira reactions, respectively).
Table 3 HERE

ACCEPTED MANUSCRIPT
This change on the reactivity (PdA vs PdB or PdC, Scheme 3) can be associated to both the different
type of interaction of stabilizers with the metallic surface and the oxidative addition step rate. It is
known that cinchona ligands privilege the  aromatic coordination through the quinoline fragment
[8,9,24], probably due to the weak N-Pd bond in contrast to the stronger P-Pd bond, because of the -
back donation of Pd to phosphorus. Thus, metal leaching and further stabilization of Pd(II) molecular
complexes are favored with phosphine-type ligands. In consequence, the cross-coupling path is
inhibited for PdNPs containing cinchonas, which usually operates under homogeneous regime
conditions. However, homocoupling processes are commonly favored at the surface of palladium
nanoparticles as previously described in the literature [28,29,34,35]. In the absence of any ligand as
stabilizer (i.e. no dative interactions between ligand and nanoparticle), PdNPs only favor the formation
of cross-coupling products (see Table S8 in the Supporting Information for related reported works).
Scheme 3 HERE
In conclusion, with this study, we proved how the capping agent permits to tune the catalytic reactivity
(for both Michael conjugate additions and Hiyama couplings), using the same type of PdNPs in terms
of size, morphology, and surface state, when the nature of the stabilizer is changed. The catalytic
results obtained in this paper not only prove the efficiency of PdNPs in different C-C and C-
heteroatom bond formation processes, but also offer a useful approach to rationalize the reaction
pathways using palladium nanoparticles as catalytic precursors.
Acknowledgements
Financial support from the Centre National de la Recherche Scientifique (CNRS), the Université de
Toulouse 3- Paul Sabatier, the Associated International Laboratory LIA-LCMMC and CONACYT
FOINS-246968 are gratefully acknowledged. A.R. and A.S.-M. thank CONACYT for PhD grants.
Supplementary data
Supplementary data to this article can be found online at http:

ACCEPTED MANUSCRIPT
References
[1] For a selected contribution, see: J. Tsuji, Palladium reagents and catalysts, John Wiley & Sons,
Sussex, 2004.
[2] B. M. Trost, Tetrahedron 71 (2015) 5708-5733.
[3] I. Favier, D. Madec, E. Teuma, M. Gómez, Curr. Org. Chem. 15 (2011) 3127-3174.
[4] For a selected contribution, see: R. Ciriminna, V. Pandarus, A. Fidalgo, L.M. Ilharco, F. Béland,
M. Pagliaro, Org. Process Res. Dev. 19 (2015) 755-768.
[5] S. Jansat, J. Durand, I. Favier, F. Malbosc, C. Pradel, E. Teuma, M. Gómez, ChemCatChem 1
(2009) 244-246.
[6] A M. López-Vinasco, I. Guerrero-Ríos, I. Favier, C. Pradel, E. Teuma, M. Gómez, E. Martin,
Catal. Commun. 63 (2015) 56-61.
[7] For a critical review concerning the role of different types of capping agents in the stabilization of
metal nanoparticles, see L. Starkey Ott, R.G. Finke, Coord. Chem. Rev. 251 (2007) 1075-1100.
[8] I. Favier, S. Massou, E. Teuma, K. Philippot, B. Chaudret, M. Gómez, Chem. Commun. (2008)
3296-3298.
[9] A. Vargas, A. Baiker, J. Catal. 239 (2006) 220-226.
[10] M.J. Climent, A. Corma, S. Iborra, M.J. Sabater, ACS Catal. 4 (2014) 870-891.
[11] M.B. Gawande, P.S. Brancoa, R.S. Varma, Chem. Soc. Rev. 42 (2013) 3371-3393.
[12] I. Favier, D. Madec, M. Gómez, in Chapter 5 “Metallic nanoparticles in ionic liquids.
Applications in catalysis” in “Nanomaterials in catalysis”, Eds. K. Philippot and P. Serp, Wiley-VCH,
Weinheim, 2013, pp 203-249.
[13] M.H.G. Prechtl, J.D. Scholten, J. Dupont, Molecules 15 (2010) 3441-3461.
[14] L. Luza, C.P. Rambor, A. Gual, F. Bernardi, J.B. Domingos, T. Grehl, P. Bruner, J. Dupont, ACS
Catal. 6 (2016) 6478-6486.
[15] H.-P. Steinruck, J. Libuda, P. Wasserscheid, T. Cremer, C. Kolbeck, M. Laurin, F. Maier, M.
Sobota, P. S. Schulz, M. Stark, Adv. Mater. 23 (2011) 2571-2587.
[16] L. Rodriguez-Perez, C. Pradel, P. Serp, M. Gómez, E. Teuma, ChemCatChem 3 (2011) 749-754.
[17] Y. Gu, I. Favier, C. Pradel, D.L. Gin, J.-F. Lahitte, R.D. Noble, M. Gómez, J.-C. Remigy, J.
Membrane Sci. 492 (2015) 331-339.
[18] V.W. Faria, D.G.M. Oliveira, M.H.S. Kurz, F.F. Gonçalves, C.W. Scheeren, G.R. Rosa, RSC
Adv. 4 (2014) 13446-13452.
[19] F. Chahdoura, I. Favier, M. Gómez, Chem. Eur. J. 20 (2014) 10884-10893.
[20] Y. Gu, F. Jérôme, Green Chem. 12 (2010) 1127-1138.
[21] A E. Díaz-Alvarez, J. Francos, B. Lastra-Barreira, P. Crochet, V. Cadierno, Chem. Commun. 47
(2011) 6208-6227.
[22] T. Kusukawa, G. Niwa, T. Sasaki, R. Oosawa, W. Himeno, M. Kato, Bull. Chem. Soc. Jpn. 86
(2013) 351-353.

ACCEPTED MANUSCRIPT
[23] F. Chahdoura, C. Pradel, M. Gómez, Adv. Synth. Catal. 355 (2013) 3648-3660.
[24] A. Reina, C. Pradel, E. Martin, E. Teuma, M. Gómez, RSC Adv. 6 (2016) 93205-93216.
[25] For a selected review, see: A. Gutnov, Eur. J. Org. Chem. (2008) 4547-4554 and references
therein.
[26] For a selected contribution involving PdNPs, see: M. Saha, A. K. Pal, Adv. Nanoparticles, 1
(2012) 61-70.
[27] N.T.S. Phan, M. Van der Sluys, C. W. Jones, Adv. Synth. Catal. 348 (2006) 609-679.
[28] S. Mukhopadhyay, G. Rothenberg, D. Gitis, H. Wiener, Y. Sasson, J. Chem. Soc., Perkin Trans. 2
(1999) 2481–2484.
[29] M. Zeng, Y. Du, L. Shao, C. Qi, X.-M. Zhang, J. Org. Chem. 75 (2010) 2556–2563.
[30] H.F. Sore, W.R.J.D. Galloway, D.R. Spring, Chem. Soc. Rev. 41 (2012) 1845-1866.
[31] B.C. Ranu, R. Dey, K. Chattopadhyay, Tetrahedron Lett. 49 (2008) 3430-3432.
[32] D. Srimani, A. Bej, A. Sarkar, J. Org. Chem. 75 (2010) 4296-4299.
[33] D. Sanhes, D. Raluy, S. Rétory, N. Saffon, E. Teuma, M. Gómez, Dalton Trans. 39 (2010) 9719-
9726.
[34] S. Xu, H. Sheng, S. Liao, D. Hu, D. Yu, T. Ye, Catal. Commun. 89 (2017) 121-124.
[35] X. Li, D. Li, Y. Bai, C. Zhang, H. Chang, W. Gao, W. Wei, Tetrahedron 72 (2016) 6996-7002.

ACCEPTED MANUSCRIPT
CAPTIONS FOR FIGURES, SCHEMES AND TABLES
Scheme 1. Synthesis of PdNPs containing TPPTS (PdA) and cinchona-based ligands, cinchonidine
(PdB) and quinidine (PdC).
Scheme 2. Compounds obtained by Pd-catalyzed conjugate additions from acceptor substrates (1, 2)
using different Michael donor reactants (a-i). Figures correspond to conversions (yields). Reaction
conditions: 1 mol% PdC, 100 °C, 2 h (or 24 h); conversions and yields determined by GC using
decane as internal standard.
Scheme 3. Schematic representation of the pathways for Hiyama couplings catalyzed by PdNPs based
on the nature of the stabilizer.
Fig. 1. Palladium-catalyzed C-N bond formation by direct reaction of acrylonitrile (1) and morpholine
(a). Diagram showing the recycling of the catalytic phase. Figures indicate yield of 1a after each
recycling (see entry 3 of Table 1 for reaction conditions).
Table 1. Pd-catalyzed C-N bond formation between acrylonitrile and morpholine by Michael-type
addition.
Table 2. Pd-catalyzed Hiyama cross-coupling reactions with aryl halides.
Table 3. Effect of the nature of stabilizers under Pd-catalyzed Hiyama cross-coupling conditions.

ACCEPTED MANUSCRIPT
Figure 1

ACCEPTED MANUSCRIPT
Scheme 1

ACCEPTED MANUSCRIPT
Scheme 2

ACCEPTED MANUSCRIPT
Scheme 3

ACCEPTED MANUSCRIPT
Table 1
Pd-catalyzed C-N bond formation between acrylonitrile
and morpholine by Michael-type addition.a
Entry
Catalyst
T
Conv. (yield)b
(%)
(load, mol%)
PdA (1)
(°C)
100
100
100
80
1c
2
>99 (98)
98 (93)
>99 (98)
85 (80)
50 (44)
PdB (1)
3
PdC (1)
4
5c
PdC (0.1)
PdA (0.1)
80
a
Results from duplicated experiments. Reaction
conditions: 1 mmol of acrylonitrile (1), 1 mmol of
morpholine (a) and 1 mL of the catalytic glycerol
solution of PdA, PdB or PdC (10-2 molL-1, 0.01 mmol
of total Pd) for 2 h. b Determined by GC using decane as
internal standard. c See reference [23].

ACCEPTED MANUSCRIPT
Table 2.
Pd-catalyzed Hiyama cross-coupling reactions with aryl halides.a
Entry Substrate
R
t (h)
Conv(%)b
Selectivity
CC/HC (%)b,c
1
2
3
4
5
6
7
8
9
10
4Br
5Br
6Br
6Br
7Br
8Br
9Br
10Br
4Cl
4-CF3
4-NO2
4-CN
4-CN
2-CN
24
3
>99
91
90/10 (4j/4HC)
57/43 (5j/5HC)
95/5 (6j/6HC)
100/0 (6j)
24
3
>99
42
24
<5
75
-
4-OCH3 24
92/8 (8j/8HC)
72/28 (9j/9HC)
11/89 (10j/10HC)
20/80 (4j/4HC)
0/100 (6HC)
4-SCH3
4-CH3
4-CF3
4-CN
24
24
24
24
78
86
36
6Cl
75
a
Results from duplicated experiments. Reaction conditions: 0.5 mmol of aryl
halide (4-10), 0.6 mmol of trimethoxy(phenyl)silane (j), 1.5 mmol of tBuOK
and 1 mL of the catalytic glycerol solution of PdA (10-2 molL-1, 0.01 mmol
b
of total Pd). Conversion and selectivity determined by GC using decane as
internal standard. c In brackets, labels corresponding to the products formed;
CC denotes cross-coupling product; HC denotes homocoupling product.

ACCEPTED MANUSCRIPT
Table 3.
Effect of the nature of stabilizers under Pd-catalyzed Hiyama cross-
coupling conditions.a
Entry PdL Substrate
R
Conv. (%)b
Selectivity
CC/HC (%)b,c
1
2
3
4
5
6
7
8
9
PdA
PdB
PdC
PdA
PdB
PdC
PdA
PdB
PdC
4Br
4Br
CF3
CF3
>99
92
26
75
24
9
90/10 (3j/3HC)
0/100 (3HC)
0/100 (3HC)
92/8 (7j/7HC)
0/100 (7HC)
n.d.d
4Br
CF3
8Br
OCH3
OCH3
OCH3
CH3
8Br
8Br
10Br
10Br
10Br
86
24
10
11/89 (9j/9HC)
0/100 (9HC)
n.d.d
CH3
CH3
a
Results from duplicated experiments. Reaction conditions: 0.5 mmol
of aryl halide, 0.6 mmol of trimethoxy(phenyl)silane (j), 1.5 mmol of
tBuOK and 1 mL of the catalytic glycerol solution of PdA, PdB or
PdC (10-2 molL-1, 0.01 mmol of total Pd). b Conversion and selectivity
c
determined by GC using decane as internal standard. In brackets,
labels corresponding to the products formed; CC denotes cross-
d
coupling product; HC denotes homocoupling product.
Only
homocoupling was detected.

ACCEPTED MANUSCRIPT
Graphical abstract

ACCEPTED MANUSCRIPT
HIGHLIGHTS





Palladiumnanoparticles dispersed in glycerol led to versatile catalysts
Stabilizercoordination modeat the metal surface tunes the catalyticreactivity
The TPPTS phosphine privileged C-Ccross-couplings
Cinchona-based stabilizers induced surface-likereactivity
Catalyticglycerol phase was efficiently recycled