Suzuki-Miyaura coupling under air in water promoted by polymer supported palladium nanoparticles

Article abstract of DOI:10.1016/j.molcata.2012.09.022

Suzuki couplings of aryl bromides were efficiently performed by a polymer supported palladium catalyst under air in water at 100°C without additives. In the case of activated aryl chlorides the reactions proceeded smoothly in the presence of a suitable phase transfer agent. The catalyst was active and recyclable for at least five times. Atomic absorption analyses revealed that the metal content in the polymeric support did not significantly decrease with the cycles while inductively coupled plasma analyses revealed that the palladium amount both in the mother liquors and in the organic products after reactions was lower than 500 ppb. The activity of the mother liquors has been investigated in detail. A transmission electron microscopy study of the supported catalyst before, during and after duty is also described.

Full text of DOI:10.1016/j.molcata.2012.09.022

Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Suzuki–Miyaura coupling under air in water promoted by polymer supported
palladium nanoparticles
Maria Michela Dell’Annaa,∗, Matilda Malia, Piero Mastrorillia,b, Antonino Rizzutia,
Chiara Ponzonic, Cristina Leonellic
a DICATECh, Politecnico di Bari, via Orabona 4, 70125 Bari, Italy
b Consiglio Nazionale delle Ricerche, Istituto di Chimica dei Composti Organometallici (ICCOM-CNR), via Orabona 4, 70125 Bari, Italy
c Dipartimento d’Ingegneria dei Materiali e dell’Ambiente, Università di Modena e Reggio Emilia, via Vignolese 905, 41125 Modena, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 July 2012
Received in revised form
13 September 2012
Accepted 24 September 2012
Available online 29 September 2012
Suzuki couplings of aryl bromides were efficiently performed by a polymer supported palladium catalyst
under air in water at 100 ◦C without additives. In the case of activated aryl chlorides the reactions pro-
ceeded smoothly in the presence of a suitable phase transfer agent. The catalyst was active and recyclable
for at least five times. Atomic absorption analyses revealed that the metal content in the polymeric sup-
port did not significantly decrease with the cycles while inductively coupled plasma analyses revealed
that the palladium amount both in the mother liquors and in the organic products after reactions was
lower than 500 ppb. The activity of the mother liquors has been investigated in detail. A transmission
electron microscopy study of the supported catalyst before, during and after duty is also described.
© 2012 Elsevier B.V. All rights reserved.
Keywords:
Suzuki–Miyaura coupling
Palladium supported nanoparticles
Green solvent
Recyclable catalyst
1. Introduction
We have recently prepared a supported palladium(II) complex
(in the following Pd-pol) by co-polymerization of Pd(AAEMA)2
Suzuki–Miyaura reactions between organoboronic acid and aryl
halides [1] play an important role in organic synthesis and find
wide application in laboratory and industries. Usually, this cross-
coupling reaction is catalyzed by soluble phosphane palladium
complexes in organic solvents but such a protocol suffers from the
difficulty in recovering and reusing the expensive homogeneous
catalysts. To overcome this drawback, much work has been devoted
to the development of reusable Pd heterogeneous catalysts [2–19].
In this regard, recyclable nanocatalysts [20–28] experienced great
prosperity in the past decade, because they combine high surface
area with excellent accessibility of the reactants to the active sites.
Another drawback of the classic Suzuki–Miyaura process is the
organic solvents disposal, which is the major problem of phar-
maceutical industries. With the aim to propose more sustainable
protocols, much effort has been devoted in trying to substi-
tute expensive, flammable and toxic organic solvents with water
[14,27,29–34]. In fact, the use of water as reaction medium has
several benefits, since it is cheap, non-toxic, non-flammable and
allows an easy recovery of the products due to the insolubility in
water of the majority of the Suzuki–Miyaura products.
[AAEMA− = deprotonated form of 2-(acetoacetoxy)ethyl methacry-
late] with suitable methacrylic co-monomers and cross-linkers
[35,36], which was found very active in palladium promoted reac-
tions such as hydrogenations, Heck and Stille reactions, as well as
asymmetric allylic alkylations [37–41]. Moreover, using Pd-pol as
promoter for the reductive amination of aldehydes, we observed
that the overall Pd(II) onto the insoluble support was reduced in situ
to Pd(0) under H2 atmosphere, forming supported Pd nanoparticles
of average size of about 5 nm, exhibiting high catalytic efficiency
and good recyclability [42].
Since Pd nanoparticles are known to facilitate the Suzuki reac-
tion and given that the supported Pd(II), which incidentally swells
well in water, could expectedly be reduced to Pd nanoparticles
by the excess of aryl boronic acid, we started a study to evalu-
ate the catalytic activity of Pd-pol for the Suzuki–Miyaura reaction
of phenyl boronic acid with aryl bromides and chlorides under
air in water. In fact, despite of the large number of reports con-
cerning the use of polymer supported catalysts for promoting the
Suzuki reaction in water, there are still margins of improvement.
First of all the supported catalysts are generally less active (Pd
mol% ≥ 0.1) than their homogeneous counterparts [12,14,16,43]
(with homogeneous catalysts sometimes homeopathic concentra-
tion of palladium are sufficient [9,44]). In some cases supported
catalysts suffer of considerable metal leaching and very often they
∗
Corresponding author. Tel.: +39 080 5963695; fax: +39 080 5963611.
E-mail address: mm.dellanna@poliba.it (M.M. Dell’Anna).
1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.09.022

M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
187
are not able to promote the coupling of the aryl chlorides (more
evaluable from an industrial point of view) [12,14,16]. Moreover,
most of the polymer supported catalysts are not active in neat
water (maybe for the swelling proprieties of the polymeric matrix),
needing for promoting the reaction, even with aryl bromides
and iodides, the presence of a conventional organic co-solvent
[21,23,29,45,46] and/or the addition of a phase transfer agent [29],
both disadvantageous from the environmental point of view. For
example, a very active polymer supported Pd catalyst such as the
one reported by Uozumi et al. (TON of up to 3,570,000) works in
water only in the presence of tetrabutylammonium fluoride [28].
It is noteworthy that even supported palladium catalysts based
on new polymeric materials such as covalent organic frameworks
(COFs) promote the Suzuki reaction in not environmentally friendly
conditions (p-xylene as the solvent, T = 150 ◦C, 0.5–1 Pd mol% with
aryl iodides and bromides) [47].
Since it is known that the properties of the polymeric support
playing a crucial role in catalysis [48] are: the swelling capability
(Pd-pol swells in water), the porosity (Pd-pol is macroporous) and
the mild reticular structure (Pd-pol is not too much cross-linked),
we deemed it worthwhile to test Pd-pol as a supported catalyst
for the Suzuki reaction in neat water. We found that Pd-pol was
active and recyclable: (i) with aryl bromides as substrates without
any additive or co-solvent, even using a Pd/substrate molar ratio of
1/10,000; (ii) with aryl chlorides, in the presence of a phase trans-
fer agent. Moreover, although the catalytic cycle is promoted by Pd
active species leached out from the insoluble support (which may
re-deposit onto it at the end of reaction), the corresponding cat-
alytic efficiencies are remarkable since the amounts of the leached
Pd species are at ppb levels at any stage of the reaction.
(30.0 m × 0.25 mm × 0.25 ␮m) capillary column coupled with a
mass spectrometer HP 5973. The products were identified by com-
parison of their GC–MS features with those of authentic samples or
1
by their NMR 1H and 13C{ H} spectra. Reactions were monitored
by TLC carried out on 0.25 mm silica gel coated glass plates using
UV light as visualizing agent or by GC–MS.
The microstructure of the polymeric matrix embedded Pd
nanoparticles was determined by TEM observations at accelera-
tion voltage of 200 kV (Model JEM 2010, Jeol, Akishima Tokyo,
Japan), equipped with X-ray energy dispersive spectroscopy (EDS).
The samples were prepared by dispersing the powders in distilled
water using an ultrasonic stirrer and then placing a drop of sus-
pension on a copper grid covered with a transparent polymer film,
followed by drying and carbon coating. The particle size distri-
butions were obtained by TEM image analysis using the ImageJ
software (freeware software: http://rsb.info.nih.gov/ij/). Each mor-
phology observation was accompanied by EDS confirmation for the
presence of Pd.
2.2. Catalytic runs
The following procedure was adopted for the standard reac-
tion catalyzed by Pd-pol (0.1 mol% Pd). A two necked round flask
was charged in air with a magnetic stir bar, Pd-pol (2.2 wt% Pd),
the aryl halide (1.0 mmol), phenyl boronic acid (1.5 mmol), K2CO3
(2.0 mmol), H2O (4.0 mL) and the system was refluxed at 100 ◦C
under vigorous magnetic stirring. In the case of aryl chlorides,
K2CO3 was replaced by KOH (2.0 mmol) and tetrabutylammonium
bromide (TBAB, 1.0 mmol) was added. The time to reach reaction
completion, reported in Table 2, was ascertained in the following
way: after 2 h reaction at 100 ◦C, the mixture was cooled down to
room temperature, diethyl ether (10 mL) was added to the sus-
pension and 0.1 mL of the organic phase was taken by means of
a syringe and analyzed by GC–MS after purification on a micro-
column filled with silica gel. If the GC–MS control showed no
residual aryl halide in solution, a new reaction was carried out in
the same conditions but for a shorter time. If the GC–MS control
showed residual aryl halide in solution, a new reaction was carried
out in the same conditions but for a longer time until two subse-
quent reactions gave the same result in terms of product yield. This
allowed us to ascertain the minimum time needed to reach reaction
completion. The cross-coupling products were isolated by extract-
ing the organic compounds from the water mixture with diethyl
ether (10 mL), washing the water phase with 2× 10 mL diethyl ether
and treating with brine (3× 10 mL) the collected organic layers. The
obtained organic phase was dried over Na2SO4 and then the solvent
was removed under reduced pressure to give the crude product,
which was purified by flash chromatography using petroleum ether
(b.p. 40–60 ◦C)–dichloromethane as eluent.
2. Experimental
2.1. General considerations
Tap water was de-ionized by ionic exchange resins (Millipore)
before use. All other chemicals were purchased from commercial
sources and used as received. Pd-pol was synthesized according to
literature procedure [42]. Palladium content in Pd-pol was assessed
after sample mineralization by atomic absorption spectrometry
using a Perkin–Elmer 3110 instrument. The experimental error
on the palladium percentage was 0.3. Palladium amount in the
mother liquors and in the organic products was assessed after sam-
ple mineralization by inductively coupled plasma-optical emission
spectroscopy (ICP-OES, Thermo ICAP 6000) analyses. For the corre-
sponding calibration curve the limit of detection was ca. 50 ppb. All
Pd analyses on the mother liquors, organic products and the used
Pd-pol were performed by carrying out the reactions on 5.0 mmol
scale.
Mineralization of the different matrices prior to Pd analyses
was carried by microwave irradiation with an ETHOS E-TOUCH
Milestone applicator, after addition of 12 mL HCl/HNO3 (3:1,v/v)
solution to each weighted sample. In the case of the mother liquor
analyses, 5.0 mL of sample was used. In the case of the organic
products, the diethyl ether solutions coming from extraction were
evaporated under vacuum and the solid residue was added of aqua
regia and submitted to microwave irradiation.
Microwave irradiation up to 1000 W was used, the temperature
being ramped from rt to 220 ◦C in 10 min and the sample being held
at this temperature for 10 min. After cooling to room temperature
the mother liquor and the organic products digested samples were
diluted to 50 mL, while the digested Pd-pol was diluted to 1000 mL
before submitting to metal analyses.
2.3. Catalyst recycling
2.3.1. “One-pot” reuse
A two necked round flask was charged in air with a magnetic
stir bar, Pd-pol (0.1 mol% Pd), the aryl halide (1.0 mmol), phenyl
boronic acid (1.5 mmol), K2CO3 (2.0 mmol), H2O (4.0 mL) and the
system was put in a thermostated bath at 100 ◦C under vigorous
stirring and reflux. After the indicated time, the mixture was cooled
down to room temperature. The minimum time needed to reach
reaction completion was ascertained as described in Section 2.2. At
reaction completion, the mixture comprised of water phase, sup-
ported catalyst and 10 mL diethyl ether (used for the monitoring)
was put under vacuum (rotavapor) until the organic solvent was
evaporated. Then, fresh aryl halide (1.0 mmol), phenyl boronic acid
(1.5 mmol) and K2CO3 (2.0 mmol) was added to the water mixture
(containing the catalyst and the products of the precedent run). The
GC–MS data (EI, 70 eV) were acquired on
a HP 7890
instrument using a HP-5MS cross-linked 5% PH ME siloxane

188
M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
Table 1
Effect of bases on the Suzuki reaction of 4-bromoacetophenone and phenylboronic acid in water under air.a
B(OH)
2
O
O
Pd-pol (0.1 mol%)
Br
+
base, H O
2
100°C, 2 h
Entry
Base
Yieldb (%)
1
2
3
4
5
NO
0
98
98
97
98
NaOH
CH3COOK
K3PO4
K2CO3
a
4-Bromoacetophenone (1.0 mmol), phenylboronic acid (1.5 mmol), base (2.0 mmol), H2O (4.0 mL), T = 100 ◦C, t = 2 h.
Isolated yield.
b
flask was put in a thermostated bath at 100 ◦C under vigorous stir-
ring and reflux for a subsequent cycle. Iteration of this procedure
was continued for five reuses of the catalyst.
4-bromoacetophenone (1.0 mmol), phenylboronic acid (1.5 mmol)
and K2CO3 (2.0 mmol) were added to the clear filtrate. The system
was stirred at 100 ◦C for 3 h, before being analyzed by GLC (Yield:
70%).
2.3.2. “Catalyst isolation” reuse
A two necked round flask was charged in air with a magnetic
stir bar, Pd-pol (0.1 mol% Pd), the aryl halide (5.0 mmol), phenyl
boronic acid (7.5 mmol), K2CO3 (10.0 mmol), H2O (20.0 mL) and the
all system was put in a thermostated bath at 100 ◦C under vigor-
ous stirring and reflux. After the minimum time needed to reach
reaction completion, the mixture was cooled down to room tem-
perature. The catalyst was recovered by filtration, washed with
water, acetone, THF and diethyl ether and dried under high vacuum.
The recovered catalyst was weighed and reused employing appro-
priate amounts of organic substrates and base, assuming that the
palladium content remained unchanged with the recycles. Iteration
of this procedure was continued for five reuses of the catalyst.
2.6. Hg poisoning test
The model reaction (0.1 mol% Pd) was left under stirring at
100 ◦C for 15 min. Then, it was hot filtered. GLC measurement per-
formed immediately after filtration showed the absence of the
coupling product in the reaction solution. Then Hg (500 equiv.
relative to Pd) was added at 100 ◦C and the mixture was left
under stirring for 5 min. Afterwards, fresh 4-bromoacetophenone
(1.0 mmol) and phenylboronic acid (1.5 mmol) were added and the
reaction mixture was stirred at 100 ◦C for 5 h and analyzed by GLC
(Yield: 40%).
3. Results and discussion
2.4. Catalytic activity of the mother liquor
3.1. Catalytic activity of Pd-pol in the Suzuki–Miyaura reaction in
water
At 30% conversion: the model reaction (0.1 mol% Pd) was left
under stirring at 100 ◦C for 15 min. Then, it was hot filtered.
GLC measurement performed immediately after filtration showed
the absence of the coupling product in the reaction solution.
Fresh 4-bromoacetophenone (1.0 mmol) and phenylboronic acid
(1.5 mmol) were added and mixture was refluxed at 100 ◦C under
stirring for 2 h, before being analyzed by GLC (Yield: 99%).
At 100% conversion: the model reaction (0.1 mol% Pd) was left
under stirring at 100 ◦C for 2 h. Then, it was hot filtered. GLC
measurement performed immediately after filtration showed the
absence of the coupling product in the reaction solution. Fresh
4-bromoacetophenone (1.0 mmol), phenylboronic acid (1.5 mmol)
and K2CO3 (2.0 mmol) were added and mixture was refluxed at
100 ◦C under stirring for 2 h, before being analyzed by GLC (Yield:
0%).
In order to optimize the reaction conditions, the coupling
of 4-bromoacetophenone (1.0 mmol) and phenyl boronic acid
(1.5 mmol) as a model reaction was carried out in water in the
presence of Pd-pol (0.1 mol% of Pd). The reaction required the pres-
ence of a base to proceed, and quantitative yield of the product was
obtained by using 2.0 mmol of bases such as sodium hydroxide,
potassium acetate, potassium phosphate or potassium carbonate
(Table 1). Considering its economical and environmental advan-
tages, we chose K2CO3 as the base for subsequent tests.
The scope of the reaction is reported in Table 2.
4-Bromoacetophenone was quantitatively arylated to 4-
phenylacetophenone using even a Pd loading as low as 0.01% mol
(entry 1 of Table 2). Interestingly, the time necessary to reach
reaction completion was not inversely proportional to the Pd
amount, being ca. 2 h in both cases (0.1 and 0.01% Pd, entry 5
of Table 1 and entry 1 of Table 2, respectively). For the same
substrates, the addition of tetrabutylammonium bromide (TBAB)
[49] to the reaction mixture shortened the reaction time to 15 min
(entry 2, Table 2).
2.5. Leaching test of Pd-pol in the absence of aryl halide
A two necked round flask was charged in air with a magnetic
stir bar, Pd-pol (0.1 mol% Pd), phenylboronic acid (1.5 mmol),
K2CO3 (2.0 mmol), H2O (4.0 mL) and the all system was put
in a thermostated bath at 100 ◦C under vigorous stirring and
reflux for 15 min. Then the mixture was hot filtered and fresh
4-bromoacetophenone (1.0 mmol) and phenylboronic acid
(1.5 mmol) were added to the clear filtrate. The system was stirred
at 100 ◦C for 2 h, before being analyzed by GLC (Yield: 99%).
Pd-pol (4.6 mg, 2.2 wt% Pd) was left under stirring in water
(4 mL) at 100 ◦C for 15 min. Then it was hot filtered and
The aryl bromides bearing electron-withdrawing groups in para
position, such as NO2 (entry 3) and CN (entry 4), provided the
desired products in quantitative yields with a reaction time of 3.5 h
and 4 h respectively. Moderate yields (ca. 60%) were obtained using
the less reactive aryl bromides with electron-donating groups,
such as 4-bromoanisole and 4-bromotoluene (entries 5 and 7).
Using KOH instead of K2CO3 as base and carrying out the coupling

M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
189
Table 2
Suzuki–Miyaura coupling reactions of aryl halides in water. Conditions: aryl halide (1.0 mmol), phenylboronic acid (1.5 mmol), Pd-pol, K2CO3 (2.0 mmol) in water (4.0 mL),
T = 100 ◦C.
B(OH)
2
Pd-pol
K CO H O
R
R
X
+
,
2
3
2
100°C
X = Br, Cl
Entry
Aryl halide
Pd load. (mol%)
0.01
t (h)
Product
Yielda (%)
97
1
2
2b
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.25
3.5
4
98
95
96
60
98
61
96
3
4
5
12
6b,c
12
12
12
7
8b,c
9
0.1
12
50
10b,c
0.1
12
98

190
M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
Table 2 (Continued )
B(OH)
2
Pd-pol
K CO H O
R
R
X
+
,
2
3
2
100°C
X = Br, Cl
Entry
Aryl halide
Pd load. (mol%)
t (h)
Product
Yielda (%)
S
11
0.1
12
30
12b,c
13b,c
14d
0.1
0.1
0.1
3
3
3
98
97
98
a
Isolated yield.
In the presence of TBAB (1.0 mmol).
KOH (2.0 mmol) was used instead of K2CO3 as base.
In DMF/water (2.0 mL/2.0 mL).
b
c
d
reaction in the presence of TBAB gave satisfactory yields in the
desired products (entries 6 and 8). A similar behavior was observed
with 1-bromonaphthalene as substrate which gave 50% yield using
K2CO3 as base (entry 9) and 98% yield using KOH and TBAB (entry
10). An unsatisfactory yield (30%) was obtained in the case of 3-
bromothiophene (entry 11), probably be due to its instability in
water or to the possible interaction between the hetero-element
and the palladium complex [32].
In order to test the feasibility of our protocol for challenging
substrates, we carried out the reaction of phenylboronic acid with
aryl chlorides. While poor yields (<10%) were achieved with 4-
chloronitrobenzene and 4-chloroacetophenone (in the presence of
K2CO3 and 0.4 mol% Pd-pol in water solvent at 100 ◦C), by using KOH
as the base in the presence of TBAB and 0.1 mol% Pd-pol resulted in
almost quantitative yields in the coupling products after 3 h reac-
tion (entries 12 and 13). Finally, carrying out the reaction between
4-chloroacetophenone and phenylboronic acid in DMF/water (1/1,
v/v) solvent, quantitative yield was achieved after 3 h in the pres-
ence of K2CO3 as the base, 0.1 mol% Pd-pol without the addition of
the phase transfer catalyst (entry 14).
addition of fresh 4-bromoacetophenone, phenylboronic acid and
K2CO3 to the reaction mixture after each cycle. The consumption of
4-bromoacetophenone in the reaction mixture after each catalytic
cycle was ascertained by GLC analyses (see Section 2). The results
are shown in Fig. 1, where turn over frequencies (TOFs) relative
to five subsequent reuses of the catalyst are reported. As apparent
from the histogram, the catalytic performance of Pd-pol remains
constant for the first two cycles (the time to reach reaction com-
pletion was 2 h), and decreases in the subsequent three reuses, the
time needed to reach reaction completion being 3 h for the third
cycle and 5 h for the fourth and fifth ones.
3.2. Catalyst reuse
The reusability of the catalyst was examined in the case of
the model reaction between 4-bromoacetophenone and phenyl-
boronic acid in the presence of K2CO3 in water at 100 ◦C. Tests
were carried out using a 0.1 mol% palladium amount for both “one
pot” [50,51] and “by catalyst isolation” consecutive recycling.
The “one pot” recyclability test was designed to avoid the
tricky separation of the small amount of the catalyst employed
(0.001 mmol Pd) from the reaction system. It consisted of the
Fig. 1. “One pot” recyclability test of Pd-pol (0.1 mol% of Pd) in the Suzuki reaction
between 4-bromoacetophenone and phenylboronic acid in the presence of K2CO3
in water at 100 ◦C.

M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
191
Table 4
Activity test of the hot filtered mother liquors of the model reaction.
Entry
1
Water solution
t (h)
Yield (%)a
Hot filtered during the first run
(conv. = 30% ca)
2
99
2b
3b
4b
5
Recycling of entry 1
5
5
5
3
0
0
0
Hot filtered at the end of the first run
Hot filtered at the end of the fifth run
Hot filtered during the third run
(conv. = 30% ca)
99
6b
7
Recycling of entry 5
5
5
0
40
Hot filtered during the first run
(conv. = 30% ca) and added of Hg
Hot filtered after 15 min stirring with
only Pd-pol
Hot filtered after 15 min stirring with
K2CO3 + PhB(OH)2 + Pd-pol
8c
9d
3
2
70
99
Cross coupling reaction between 4-bromoacetophenone (1.0 mmol), phenyl boronic
acid (1.5 mmol) in the presence of K2CO3 (2.0 mmol) and Pd-pol (0.1 mol% Pd) in
H2O (4.0 mL) at 100 ◦C. Instantly after the filtration, fresh 4-bromoacetophenone
(1.0 mmol) and phenylboronic acid (1.5 mmol) were added and the mixture was left
under stirring at 100 ◦C for the allotted time.
Fig. 2. “Catalyst isolation” recyclability test of Pd-pol (0.1 mol% of Pd) in the Suzuki
reaction between 4-bromoacetophenone and phenylboronic acid in the presence of
K2CO3 in water at 100 ◦C.
a
GLC yield with the internal standard method.
K2CO3 (2.0 mmol) added.
Neither reagent nor base was present in the reaction mixture before the hot
b
The lowering of the catalytic activity observed in the “one pot”
recycling procedure may be imputed in part to the large amount of
the organic product that accumulates in the reaction mixture with
the runs and that could hamper the superficial contact between
the active sites of the catalyst and the fresh added reagents [20].
Thus we carried out the recyclability tests in the standard way, i.e.
by isolating the solid catalyst after each cycle (“catalyst isolation”
procedure) by scaling up five-fold all reagents, in order to handle
a manageable amount of Pd-pol. The obtained results are reported
in Fig. 2, that shows a better recyclability of the supported catalyst
submitted to “catalyst isolation” procedure, with respect to that
observed for the “one pot” reusing test.
Atomic absorption analyses carried out on the pristine cata-
lyst and on Pd-pol recovered after the first run, the second run,
the fifth run as well as during reaction (at 30% of substrate con-
version) showed only a slight decrease of the metal content with
the cycles. The measurements were replicated three times with
different batches of Pd-pol (Table 3).
c
filtration.
d
The aryl halide was absent in the reaction mixture before the hot filtration.
constant in the mixture, i.e. the relevant conversion was negligi-
ble (entry 2, Table 4). In addition, catalytic tests were carried out
on the mother liquors recovered by hot filtration after the first and
after the fifth run. They were added of fresh 4-bromoacetophenone,
phenylboronic acid and K2CO3 and kept under stirring at 100 ◦C
for 5 h. No coupling product was detected, confirming the lack of
catalytic activity of the mother liquors after consumption of the
substrates (entries 3 and 4, Table 4).
The activity of the mother liquor hot filtrated at 30% conversion
and the inactivity of the mother liquor after substrate consumption
was ascertained also for the third cycle (entries 5 and 6 of Table 4),
the one for which a slight decrease of the TOF was observed (Fig. 2).
This indicates that the reaction mechanism was the same for all the
consecutive cycles.
In addition, ICP-OES analyses were performed both on the
mother liquors recovered by hot filtration after the first run, after
the fifth run and during reaction (at 30% of substrate conver-
sion), and on the respective organic products, separated from the
supported catalyst. In all cases the metal content was below the
detection limit (0.5 ppm).
3.3. Analysis of the mother liquors
To gain insights into the type of mechanism (homogeneous vs
heterogeneous) [52,53], we studied the catalytic activity of the
mother liquors during and after substrate consumption. The sup-
ported catalyst and the organics were hot filtered off from the
reaction mixture after 30 min reaction (the conversion was ca. 30%)
and the water solution was added of fresh reagents and stirred
at 100 ◦C. Quantitative yield in 4-phenylacetophenone was regis-
tered after 2 h (entry 1, Table 4), exactly the time necessary for
completion of the reaction in the presence of Pd-pol. Then, fresh
4-bromoacetophenone, phenylboronic acid and K2CO3 was added
to the reaction mixture which was left under stirring for more than
5 h at 100 ◦C. GLC measurements carried out before and after the
final 5 h reaction revealed that the amount of substrate remained
Tests carried out on the mother liquors indicate that palladium
species leached out from the insoluble support are the true catalyst
and that such palladium species deactivate as soon as the sub-
strates are consumed. The catalytic efficiency of the mother liquors
is therefore quite remarkable, considering that the metal amount
in solution is very small (<0.5 ppm). In other words, Pd-pol acts as a
reservoir of “homeopathic” amounts [44,54] of catalytically active
metal (soluble complexes and palladium nanoparticles) which is
released in solution in a controlled way. The steadiness of the time
required for completing the model reaction when using amounts
of Pd-pol corresponding to either 0.1 mol% Pd (entry 5 of Table 1)
or 0.01 mol% Pd (entry 1 of Table 2) is in accord with this hypoth-
esis. In fact, it is known that an higher concentration of “naked”
palladium nanoparticles in the reaction medium does not always
increase the catalytic activity. In some cases it was found that the
“naked” nanoparticles catalytic performances were improved by
lowering their loading in solution, thus preventing the formation
of inactive palladium black [55].
Table 3
Palladium amount on the supported catalyst recovered after different cycles of the
model reaction.
Pd-pol
Pd (wt%)
Batch 1
Batch 2
Batch 3
Pristine
2.2
2.0
2.0
1.9
1.9
1.4
1.4
1.2
1.2
1.2
5.7
5.6
5.7
5.7
5.5
Recovered during reaction (30% conv.)
Recovered after the first run
Recovered after the second run
Recovered after the fifth run
It has been demonstrated that the palladium leaching during the
Suzuki reaction is function of several parameters, such as temper-
ature, additives, choice of solvent, substrate, etc. [56]. In our case

192
M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
a factor playing a crucial role for the Pd leaching is the addition of
TBAB. In fact, as mentioned before, the model reaction carried out
with 0.1 mol% Pd in the presence of TBAB was eight times faster
(the time to reach completion passed from 2 h, without TBAB, to
15 min, with TBAB, entry 5 of Table 1 and entry 2 of Table 2) sug-
gesting that an eight-fold amount of extremely active palladium
species was leached out in solution when TBAB was present in the
reaction mixture. Consistently, a ca. 28% Pd leaching was observed
after the first run in the presence of TBAB.
The maintenance of catalytic activity upon several reuses can
be ascribed to two different processes: one in which the leached
palladium re-deposits onto the insoluble support, being prone to
re-dissolve in solution at the beginning of the subsequent cycle
[57–61] and one in which the leached palladium does not re-
deposit onto the resin, but fresh palladium is released in solution
at the beginning of each cycle. The very little amount of leached
palladium does not allow us to discriminate between these two
mechanisms (which may not be exclusive each other).
Several authors have suggested that under C C bond form-
ing reaction conditions the employed palladium(II) pre-catalysts
(soluble or supported ones) generate palladium nanoparticles
[43,62,63], which can leach in solution if immobilized onto an
insoluble matrix. These nanoparticles may in turn generate soluble
palladium(0) mononuclear (or dinuclear) complexes in equilib-
rium with them [62,64]. Also in the case of Pd-pol the reduction
of the Pd(II) pre-catalyst generated palladium nanoparticles, as
ascertained by TEM analyses (see below). This reduction was both
thermal (being the reaction temperature 100 ◦C) [65] and due to a
redox process (being the phenyl boronic acid a reductant).
In order to ascertain whether the true catalysts were soluble
mononuclear (or dinuclear) Pd(0) compounds [62] or colloidal pal-
ladium nanoparticles [64] leached out from the insoluble support in
very controlled way, we submitted the catalytically active mother
liquor of the model reaction to Hg poisoning test [66]. In fact,
although the utility of Hg as a selective surface poison is rather con-
tentious [53], the inhibition of reaction caused by mercury addition,
presumably due to surface adsorption/amalgam formation, is taken
as a proof of being palladium nanoparticles the catalytically active
species [64]. By adding a drop of mercury to the solution obtained
from hot filtration of the reaction between 4-bromoacetophenone
and phenylboronic acid at 30% conversion, a 40% conversion was
registered 5 h after the addition of fresh reagents (entry 7, Table 4).
The observed proceeding of the reaction, albeit with a lower rate,
in the presence of a very high excess Hg seems to indicate that
both molecular palladium species and palladium nanoparticles are
involved in the catalysis.
It has been reported that in the case of Au–Pd core–shell
nanoparticles, the base and the phenylboronic acid synergistically
cause the Pd leaching into solution, with formation of species active
in the Suzuki reaction [67]. In order to ascertain which reagents
sequestrates Pd from Pd-pol, we studied the Pd leaching without
any reagent as well as in the presence of K2CO3 and phenylboronic
acid, measuring the amount of the leached metal as a function of
the catalytic activity of the mother liquors.
Thus, leaving under stirring at 100 ◦C in water for 15 min fresh
Pd-pol and hot filtering the obtained mixture gave a mother liquor
which afforded, in the model reaction, 70% conversion after 3 h at
100 ◦C (entry 8 of Table 4). On the other hand, leaving under stirring
Fig. 3. Transmission electron micrographs and associated size distribution of matrix polymer embedded Pd nanoparticles: (a) pristine Pd-pol; (b) Pd-pol recovered during
the model reaction at 30% conversion; (c) Pd-pol recovered after the first run of the model reaction; (d) Pd-pol recovered after the fifth run of the model reaction.

M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
193
at 100 ◦C in water for 15 min fresh Pd-pol in the presence of phenyl-
boronic acid and K2CO3 and hot filtering the obtained mixture gave
a mother liquor which afforded, in the model reaction, quantitative
yield after 2 h at 100 ◦C (entry 9 of Table 4). These results show that,
for Pd-pol, the controlled palladium leaching occurs either in the
absence and in the presence of base and boronic acid, the extent
of leaching being higher in the presence of phenylboronic acid and
K2CO3.
after substrate consumption. The Pd leaching, which was enhanced
by the presence of additives such as TBAB, occurred both in the pres-
ence of base and boronic acid (in the absence of TBAB), and even
when Pd-pol was left under stirring in neat water at 100 ◦C. The TEM
study on Pd-pol recovered during and after reaction revealed that
during reaction the number density of Pd(0) nanoparticles, with
diameter of 2–6 nm, increases presumably due to reaction of sup-
ported Pd(II) with phenylboronic acid. Moreover, on passing from
the first to the fifth cycle, the mean Pd(0) nanoparticle dimension
became slightly larger, and their size distribution became broader.
Finally, we can conclude that the good recyclability of the solid cat-
alyst, the identical amount of metal measured on it after each run,
the retention of the nanoparticles size distribution on the insolu-
ble support with the re-cycles, and the absence of catalytic activity
of the mother liquors filtered off after substrate consumption can-
not always be taken as a proof of true heterogeneous nature of the
supported catalyst.
3.4. TEM characterization of the catalyst
This study was completed with the TEM characterization of
Pd-pol before, during and after duty, with the aim of ascertain-
ing whether the reaction cycles affected the morphology and the
dispersion of the palladium particles on the surface of the support.
TEM pictures of pristine Pd-pol and of the catalyst recov-
ered during reaction at 30% conversion, are reported in Fig. 3a
and b, respectively. The pristine catalyst was constituted mainly
by polymer-bound Pd(II) species (not visible in the TEM micro-
graphs) [42], together with a low number density of metallic
Pd nanoparticles (with diameter ranging between 4 and 8 nm).
This nanostructure, which reflected the way of preparation of the
pre-catalyst, was inferred from the image showing dark dots rep-
resenting the metallic particles on different shades of the polymers
[68]. Pd nanoparticles were embedded in the polymeric matrix,
and not just deposited on its surface, as justified by the presence
in the TEM image of some nanoparticles with different gray-scale
color and not all in the same focal plane. These nanoparticles were
likely formed during the thermal polymerization step of the catalyst
synthetic procedure [42].
Acknowledgments
The authors thank Dr. Roberto Terzano (Department of Agricul-
ture of University of Bari) for ICP-OES measurements and Italian
MIUR (project PRIN09 n. 2009LR88XR) for financial support.
References
[1] N. Miyaura, A. Suzuki, Chem. Rev. 95 (1995) 2457–2483.
[2] N.F. Leadbeater, M. Marco, Chem. Rev. 102 (2002) 3217–3274.
[3] S. Yi, D.H. Lee, E. Sin, Y.S. Lee, Tetrahedron Lett. 48 (2007) 6771–6775.
[4] A. Ohtaka, T. Teratani, R. Fujii, K. Ikeshita, O. Shimomura, R. Nomura, Chem.
Commun. 45 (2009) 7188–7190.
[5] D. Connolly, B. Twamley, B. Paull, Chem. Commun. 46 (2010) 2109–2111.
[6] S. Schweizer, J.M. Becht, C.L. Drian, Tetrahedron 66 (2010) 765–772.
[7] H. Yang, X. Han, Z. Ma, R. Wang, J. Liu, X. Ji, Green Chem. 12 (2010) 441–451.
[8] L.S. Sobjerg, D. Gauthier, A.T. Lindhardt, M. Bunge, K. Finster, R.L. Meyer, T.
Skrydstrup, Green Chem. 11 (2009) 2041–2046.
[9] A.K. Diallo, C. Ornelas, L. Salmon, J.R. Aranzaes, D. Austruc, Angew. Chem. Int.
Ed. 46 (2007) 8644–8648.
[10] R. Narayanan, M.A. El-Sayed, J. Am. Chem. Soc. 125 (2003) 8340–8347.
[11] S. Proch, Y. Mei, J.M.R. Villanueva, Y. Lu, A. Karpov, M. Ballauff, R. Kempe, Adv.
Synth. Catal. 350 (2008) 493–500.
The TEM images of Pd-pol recovered during the reaction at 30%
conversion (Fig. 3b) showed the formation of new Pd nanoparticles,
now ranging from 2 to 6 nm.
The TEM images of Pd-pol recovered after the first run and after
the fifth run (“one pot” recycling way) are reported in Fig. 3c and
d, respectively. On passing from the first to the fifth cycle, the
mean Pd(0) nanoparticle size became bigger, with diameters ran-
ging from 2–8 nm to 4–10 nm, and their size distribution appeared
broader. The nanostructure of the catalyst was substantially main-
tained during the subsequent cycles and the Pd nanoparticles
remained embedded to the polymer bulk.
The slight decrease in the catalytic activity of Pd-pol recorded
in the Suzuki–Miyaura coupling reaction after the second catalytic
cycle (Fig. 1) can hence be influenced by the increase of the sup-
ported nanoparticle sizes during catalysis which, may either render
the nanoparticles less prone to be leached in solution, or may result
in the leaching of bigger, and therefore less active, nanoparticles.
[12] A.C. Albéniz, N. Carrera, Eur. J. Inorg. Chem. 15 (2011) 2347–2360.
[13] M.K. Bhunia, S.K. Das, P. Pachfule, R. Banerjee, A. Bhaumik, Dalton Trans. 41
(2012) 1304–1311.
[14] M. Lamblin, L. Nassar-Hardy, J.-C. Hierso, E. Fouquet, F.-X. Felpin, Adv. Synth.
Catal. 352 (2010) 33–79.
[15] H.-J. Freund, Chem. Eur. J. 16 (2010) 9384–9397.
[16] A. Molnar, Chem. Rev. 111 (2011) 2251–2320.
[17] Z. Wang, G. Chen, K. Ding, Chem. Rev. 109 (2009) 322–359.
[18] P. Mastrorilli, C.F. Nobile, Coord. Chem. Rev. 248 (2004) 377–395.
[19] P. Mastrorilli, C.F. Nobile, Macromolecules containing metal and metal-like ele-
ments, in: A.S. Abd-El-Aziz, C.E. Carraher Jr., C.U. Pittman Jr., M. Zeldin (Eds.),
Metal-coordination Polymers, vol. 5, Wiley-Interscience, New York, 2005, pp.
227–255 (Chapter 9).
[20] A. Fihri, M. Bouhrara, B. Nekoueishahraki, J.-M. Basset, V. Polshettiwar, Chem.
Soc. Rev. 40 (2011) 5181–5203, and references therein.
[21] M. Cao, Y. Wei, S. Gao, R. Cao, Catal. Sci. Technol. 2 (2012) 156–163.
[22] A. Balanta, C. Godard, C. Claver, Chem. Soc. Rev. 40 (2011) 4973–4985, and
references therein.
[23] J. Wang, G. Song, Y. Peng, Tetrahedron Lett. 52 (2011) 1477–1480.
[24] P. Zhang, Z. Weng, J. Guo, C. Wang, Chem. Mater. 23 (2011) 5243–5249.
[25] S. Singha, M. Sahoo, K.M. Parida, Dalton Trans. 40 (2011) 7130–7132.
[26] A. Schatz, O. Reiser, W.J. Stark, Chem. Eur. J. 16 (2010) 8950–8967.
[27] N. Yan, C. Xiao, Y. Kou, Coord. Chem. Rev. 254 (2010) 1179–1218.
[28] Y.M.A. Yamada, S.M. Sarkar, Y. Uozumi, J. Am. Chem. Soc. 134 (2012)
3190–3198.
[29] V. Polshettiwar, A. Decottignies, C. Len, A. Fihri, ChemSusChem 3 (2010)
502–522, and references therein.
[30] Y. Yu, T. Hu, X. Chen, K. Xu, J. Zhanga, J. Huang, Chem. Commun. 47 (2011)
3592–3594.
[31] F. Amoroso, S. Colussi, A. Del Zotto, J. Llorca, A. Trovarelli, Catal. Commun. 12
(2011) 563–567.
[32] A. Fihri, D. Luart, C. Len, A. Solhy, C. Chevrinb, V. Polshettiwar, Dalton Trans. 40
(2011) 3116–3121.
4. Conclusions
The palladium supported Pd-pol behaves as a highly active and
selective catalyst for Suzuki–Miyaura cross-couplings of aryl bro-
mides and activated aryl chlorides. The most remarkable results
of the present investigations are the following: (i) the catalyst is
active at 100 ◦C in air, in an environmentally friendly solvent such
as water; (ii) Pd-pol is active in low amount (down to 0.01 mol%
Pd) and the presence of a transfer phase additive, such as TBAB,
is needed only for aryl chlorides; (iii) the catalyst can be recycled
at least five times without a significant loss of activity. The analy-
ses of the mother liquors permitted to establish that Pd-pol acts as
a reservoir of catalytically active species that leached out in a very
little amount for each cycle in solution (less than 500 ppb, that is the
requirements for pharmaceutical industry). The leached palladium
species may re-deposit onto the insoluble support after reaction
completion. When they remained in solution, they became inactive
[33] C. Zhou, J. Wang, L. Li, R. Wang, M. Hong, Green Chem. 13 (2011) 2100–2106.
[34] A.N. Marziale, D. Jantke, S.H. Faul, T. Reiner, E. Herdtweck, J. Eppinger, Green
Chem. 13 (2011) 169–177.

194
M.M. Dell’Anna et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 186–194
[35] M.M. DellAnna, P. Mastrorilli, C.F. Nobile, in: P.J.H. Scott (Ed.), Solid-Phase
Organic Syntheses, vol. 2, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2012,
pp. 79–86.
[36] M.M. Dell’Anna, M. Gagliardi, P. Mastrorilli, C.F. Nobile, J. Mol. Catal. A: Chem.
158 (2000) 515–520.
[50] O. Metin, F. Durap, M. Aydemir, S. Ozkar, J. Mol. Catal. A: Chem. 337 (2011)
39–44.
[51] J. Zhou, X. Li, H. Sun, J. Organomet. Chem. 695 (2010) 297–303.
[52] J.A. Widegren, R.G. Finke, J. Mol. Catal. A 198 (2003) 317–341.
[53] N.T.S. Phan, M. Van Der Sluys, C.W. Jones, Adv. Synth. Catal. 348 (2006) 609–679.
[54] F. Amoroso, S. Colussi, A. Del Zotto, J. Llorca, A. Trovarelli, J. Mol. Catal. A: Chem.
315 (2010) 197–204.
[37] M.M. Dell’Anna, A. Lofù, P. Mastrorilli, V. Mucciante, C.F. Nobile, J. Organomet.
Chem. 691 (2006) 131–137.
[38] M.M. Dell’Anna, P. Mastrorilli, C.F. Nobile, G.P. Suranna, J. Mol. Catal. A: Chem.
201 (2003) 131–135.
[39] M.M. Dell’Anna, P. Mastrorilli, F. Muscio, C.F. Nobile, G.P. Suranna, Eur. J. Inorg.
Chem. 8 (2002) 1094–1099.
[40] M.M. Dell’Anna, P. Mastrorilli, A. Rizzuti, G.P. Suranna, C.F. Nobile, Inorg. Chim.
Acta 304 (2000) 21–25.
[41] M.M. Dell’Anna, P. Mastrorilli, F. Muscio, C.F. Nobile, in: M. Anpo, M. Onaka,
H. Yamashita (Eds.), Science and Technology in Catalysis, Elsevier, Amsterdam,
2002, pp. 133–136.
[55] D. Astruc, F. Lu, J.R. Aranzaes, Angew. Chem. Int. Ed. 44 (2005) 7852–7872.
[56] S.S. Soomro, F.L. Ansari, K. Chatziapostolou, K. Kohler, J. Catal. 273 (2010)
138–146.
[57] M. Weck, C.W. Jones, Inorg. Chem. 46 (2007) 1865–1875.
[58] A. Biffis, M. Zecca, M. Basato, Eur. J. Inorg. Chem. 5 (2001) 1131–1133.
[59] F. Zhao, M. Shirai, Y. Ikushima, M. Arai, J. Mol. Catal. A 180 (2002) 211–219.
[60] S.S. Prockl, W. Kleist, M.A. Gruber, K. Kholer, Angew. Chem. Int. Ed. 43 (2004)
1881–1882.
[61] S. MacQuarrie, J.H. Horton, J. Barnes, K. McEleney, H.-P. Look, C.M. Crudden,
Angew. Chem. Int. Ed. 47 (2008) 3279–3282.
[62] D. Astruc, Inorg. Chem. 46 (2007) 1884–1894.
[42] M.M. Dell’Anna, P. Mastrorilli, A. Rizzuti, C. Leonelli, Appl. Catal. A: Gen. 401
(2011) 134–140.
[43] R.J. Kalbasi, N. Mosaddegh, J. Inorg. Organomet. Polym. 22 (2012)
404–414.
[44] L.P. Beletskaya, A.V. Chepakrov, Chem. Rev. 100 (2000) 3009–3066.
[45] A. Modak, J. Mondal, M. Sasidharan, A. Bhaumik, Green Chem. 13 (2011)
1317–1321.
[63] A.M. Trzeciak, J.J. Ziołkowski, Coord. Chem. Rev. 251 (2007) 1281–1293.
[64] A.F. Lee, P.J. Ellis, I.J.S. Fairlamb, K. Wilson, Dalton Trans. 39 (2010)
10473–10482.
[65] A.S. Rozenberg, A.A. Rozenberg, G.I. Dzardimalieva, A.D. Pomogailo, J. Colloid
67 (2005) 63–71.
[46] S. Schweizer, J.-M. Becht, C. Le Drian, Tetrahedron 66 (2010) 765–772.
[47] S.-Y. Ding, J. Gao, Q. Wang, Y. Zhang, W.-G. Song, C.-Y. Su, W. Wang, J. Am. Chem.
Soc. 133 (2011) 19816–19822.
[66] See for example: S.M. Nobre, A.L. Monteiro, J. Mol. Catal. A 313 (2009) 65–73.
[67] P.-P. Fang, A. Jutand, Z.-Q. Tiana, C. Amatore, Angew. Chem. Int. Ed. 50 (2011)
1–5.
[48] M. Benaglia, New J. Chem. 30 (2006) 1525–1533.
[49] M.T. Reetz, E. Westermann, Angew. Chem. Int. Ed. 39 (2000) 165–168.
[68] S. Sarkar, E. Guibal, F. Quignard, A.K. SenGupta, J. Nanopart. Res. (2012),
http://dx.doi.org/10.1007/s11051-011-0715-2.