Palladium(0) nanoparticles encapsulated in diamine-modified glycidyl methacrylate polymer (GMA-CHDA) applied as catalyst of Suzuki-Miyaura cross-coupling reaction

Article abstract of DOI:10.1039/b717995b

Cyclohexyldiamine-modified glycidyl methacrylate polymer (GMA-CHDA) in the form of gel-type beads was used to encapsulate Pd(0) nanoparticles 4-15 nm in diameter and applied as a new, reusable catalyst for the Suzuki-Miyaura cross-coupling reaction of 2- and 4-bromotoluene with phenylboronic acid. It was found that the precatalyst preparation methodology strongly influenced its catalytic activity. The best results (100% yield of the product) were obtained when GMA-CHDA was first treated with hydrazine (reducing agent for Pd(ii)) and next with PdCl₂ solution. The new catalyst acts heterogeneously, and the post-reaction solution after catalyst separation is not catalytically active, suggesting that there is no leaching. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique.

Full text of DOI:10.1039/b717995b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Palladium(0) nanoparticles encapsulated in diamine-modified glycidyl
methacrylate polymer (GMA-CHDA) applied as catalyst of
Suzuki–Miyaura cross-coupling reactionw
a
a
a
b
Anna M. Trzeciak,* Ewa Mieczyn
´
b
ska, Jo
´
b
zef J. Zio
´
$kowski, Wiktor Bukowski,
c
Agnieszka Bukowska, Jaros$aw Noworol and Janina Okal
´
Received (in Durham, UK) 21st November 2007, Accepted 10th January 2008
First published as an Advance Article on the web 5th February 2008
DOI: 10.1039/b717995b
Cyclohexyldiamine-modified glycidyl methacrylate polymer (GMA-CHDA) in the form of gel-
type beads was used to encapsulate Pd(0) nanoparticles 4–15 nm in diameter and applied as a
new, reusable catalyst for the Suzuki–Miyaura cross-coupling reaction of 2- and 4-bromotoluene
with phenylboronic acid. It was found that the precatalyst preparation methodology strongly
influenced its catalytic activity. The best results (100% yield of the product) were obtained when
GMA-CHDA was first treated with hydrazine (reducing agent for Pd(II)) and next with PdCl
solution. The new catalyst acts heterogeneously, and the post-reaction solution after catalyst
separation is not catalytically active, suggesting that there is no leaching.
2
The activity of this catalyst increased when PPh was added;
3
Introduction
17
however, some leaching of Pd was observed.
The immobilization of metal catalysts on solid supports has
become a valuable tool for the efficient separation of organic
products and catalyst recycling. Polymers play a significant
role in this area, offering different ways of metal attachment to
the polymer matrix via covalent or non-covalent bonding,
through hydrogen bridges, as well as through ionic, hydro-
It is also possible to incorporate a Pd precatalyst into a
polymeric material during the synthesis of the polymer. This
method was applied to the synthesis of an encapsulated Pd
catalyst, subsequently commercialized as the Pd EnCat cata-
18
lyst, used for the Suzuki reaction with a phosphine additive.
Recent progress in the design, preparation and application of
polymer-supported Pd complexes is reviewed in ref. 19, where,
among other things, the preparation of various supported
Pd–phosphine complexes used in carbon–carbon bond
forming reactions, such as the Heck reaction, the Suzuki–
Miyaura coupling and p-allylic substitution is discussed.
In this work, we report on the application of diamine-
modified glycidyl methacrylate resin (GMA-CHDA) as a
support for the encapsulation of Pd nanoparticles used in
Suzuki–Miyaura cross-coupling. Glycidyl methacrylate poly-
mers, being easily modifiable materials, have been used, for
1
,2
phobic or fluorous interactions.
For the purposes of immobilizing catalysts in the form of
metal complexes, different polymers have been examined—both
soluble (linear, branched and hyperbranched) and insoluble,
3
cross-linked micro- and macroporous, and hybrid materials.
Typically, functional groups present in a polymer support
participate in bonding with metals. Palladium complexes an-
4
chored on polymers with mono- and diphosphines, N-hetero-
–6
7–10
cyclic carbene groups,
11,12
and dendrimers
have been
described. Pd nanoparticles are another kind of attractive
catalyst for C–C cross-coupling reactions, including the Suzuki–
20
instance, as supports for enzymes and sorbents for chroma-
21
1
3,14
Miyaura coupling.
To prevent the eventual aggregation of
tography. They can also be utilized for removing metal ions
22
nanoparticles, polyvinylpyrrolidone (PVP) has been used as a
or impurities from solutions and as biomaterials, e.g. protein
23
1
supporting medium. Such a Pd colloid (Pd/PVP) has also been
5
separation and purification agents,
and biodegradable
24
16
used as a catalyst precursor after immobilization on carbon.
hydrogels in drug control release processes.
Good results in Suzuki couplings, close to those obtained in
homogeneous systems, have been reported for Pd nanoparticles
supported on a hybrid organic/inorganic material, polyHIPE,
The polymers under study, with low levels of cross-linking
of three-component glycidyl methacrylate terpolymers (GMA
resins), were synthesized by the suspension polymerization
17
functionalized with amine or quaternary ammonium groups.
25
technique, and were found not only to be good supports
for Pd nanoparticles, but also to exhibit interesting swelling
characteristics. It was found that the gel-type beads, formed
from the mixture of glycidyl methacrylate, styrene and a
divinyl monomer (divinylbenzene, mono-, di- or triethylene-
glycol dimethacrylates), in the presence of a cyclohexanol/
n-octanol mixture, were able to enlarge their volume many
times over under the influence of thermodynamically favour-
able (compatible) solvents and return to their initial volume
a
b
c
Faculty of Chemistry, University of Wroc!aw, 14 F. Joliot-Curie St.,
0-383 Wroc!aw, Poland
Faculty of Chemistry, Rzeszo´w University of Technology, al.
Powstan´co´w Warszawy 6, 35-959 Rzeszo´w, Poland
Institute of Low Temperature and Structure Research Polish
Academy of Sciences, PO Box 937, 50-950 Wroc!aw, Poland
5
w Dedicated to Professor Karel Mach in recognition of his outstanding
contribution to organometallic chemistry.
1
124 | New J. Chem., 2008, 32, 1124–1130
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
when a thermodynamically unfavourable (incompatible) sol-
vent was added.
Table 1 Swelling characteristics of GMA and GMA-CHDA resins
a
Swelling degree /g g
À1
Solvents
An increase in bead volume indicates that molecules of
compatible solvents are able to penetrate deeply into the
polymer network and solvate polymer chains. These pheno-
mena facilitate access of potential reagents or other species to
GMA resin GMA-CHDA resin
CH
O after CH
THF
O after THF
2
Cl
2
3.67
3.65
2.30
0.84
1.75
1.70
—
2.56
2.35
1.60
0.59
1.35
1.23
0.52
0.31
2.04
0.42
0.55
0.47
0.63
0.65
0.50
0.66
0.94
1.41
2.02
2.35
1.18
H
2
2 2
Cl
2
6
essentially all segments of the polymer network. Therefore,
we decided to use the selected GMA resin to encapsulate Pd
nanoparticles in the polymer network and test the obtained
catalysts in Suzuki–Miyaura cross-coupling reactions.
H
2
DMF
H O after DMF
2
b
4
N
2
H
after DMF
b
DMF after N
,4-Dioxane
2
H
4
—
1
2.35
0.49
0.54
—
—
—
0.75
0.85
—
0.77
1.79
2.70
1.29
MeCN
MeOH
Toluene
Experimental
Polymer support synthesis
2
2
-Propanol
-Propanol–H
2
O (1 : 1 v/v)
Synthesis of the diamine gel-type polymer support for Pd
immobilization was carried out in a two-stage procedure. In
the first stage, a mixture of glycidyl methacrylate (GMA; 20
mol%), styrene (S; 77 mol%) and divinylbenzene (DVB,
mixture of isomers, B80%; 3 mol%) was polymerized in
suspension, according to a procedure described in ref. 25.
The monomers were diluted in a mixed solvent (cyclohexanol/
n-octanol, 4 : 1 w/w) in a 1 : 1 w/w proportion. Spherical
microbeads of the GMA resin were obtained. After being
washed and dried under reduced pressure, the beads were
sieved and analyzed. A fraction 150–250 mm in diameter was
used for further investigations. The epoxy group content in the
H
2
O
Ethylene glycol
Ethylene glycol–H O (1 : 1 v/v)
2
CH
CH
CH
2
2
2
Cl
Cl
Cl
2
–MeOH (1 : 2 v/v)
2
–MeOH (1 : 1 v/v)
2
–MeOH (2 : 1 v/v)
2
THF–H O (1 : 1 v/v)
a
Swelling degree was calculated as: SD = (m
s
À m
d d s d
)/m ; m , m =
b
weights of the swelled and dry resins, respectively. 64% aqueous
solution of N
H .
2 4
2
0 min and then filtered. Catalyst P1 was washed three times
À1
with DMF, then with MeOH, left for 15 min in MeOH,
beads, amounting to B1.3 mmol g , was determined using a
modification of Jay’s method.
2
7
filtered and dried in a vacuum.
À1
IR (KBr, cm ): 1730 m (nCO), 1656 vs (nCOO).
In the second stage, the GMA resin was reacted with a
5
-fold excess of trans-1,2-diaminocyclohexane (CHDA) in
3
Catalyst P2. For the preparation of catalyst P2, the same
DMF (10 cm per gram of the GMA resin) at 70 1C. The
reaction was carried out under argon in thermostatic glass
reactors placed on a horizontal shaker. After 20–24 h, the
diamine-modified resin, GMA-CHDA, was filtered and
washed carefully several times, first with DMF, and then with
CH Cl and MeOH, and finally dried in a vacuum at 40 1C. Its
solutions were used as for P1, however PdCl was added to the
2
polymer before hydrazine.
3
.25 g of GMA-CHDA polymer in 2.5 cm of DMF was
0
3
heated to 60 1C. In another vessel, to 2.5 cm of the PdCl
2
3
solution in DMF pre-heated to 60 1C, was added 0.5 cm of
the hydrazine solution. The resulting black–green solution was
stirred for ca. 5 min and transferred to the first vessel contain-
ing GMA-CHDA; the color changing immediately to black.
After stirring for 20 min at 60 1C, catalyst P2 was filtrated,
washed three times with DMF, then with MeOH, left for
2
2
analysis (Kiejdahl’s method) indicated a nitrogen content of
À1
B2 mmol g . The nitrogen content was also confirmed by
elemental analysis: N, 2.66; C, 79.67; H, 7.91%.
À1
IR (KBr, cm ): 1730 vs (nCO), 1604 s (nCOO).
1
5 min in MeOH, and then filtered and dried in a vacuum.
À1
Test of polymer swelling
IR (KBr, cm ): 1730 m (nCO), 1656 vs (nCOO).
3
Swelling of the resin was determined using 2.5 cm syringes
equipped with a polypropylene filter. A resin sample of 0.1 g
was treated with a selected solvent, and the solvent excess was
removed by syringe piston after establishing a swelling equili-
brium. The weights of dry and swelled resins were compared to
calculate the swelling degree (SD) (Table 1).
Catalyst P3. For the preparation of catalyst P3, the same
solutions were used as for P1.
3
To 0.5 g of GMA-CHDA polymer was added 5 cm of
3
DMF, and the mixture stirred for 20 min. Next, 5 cm of the
PdCl solution in DMF was added, and the mixture was left
2
stirring for 24 h. The polymer was filtered, washed three times
3
with DMF, dried, transferred to another vessel with 5 cm of
Preparation of palladium catalysts (P1, P2, P3)
Catalyst P1. To 0.5 g of GMA-CHDA polymer was added 5
3
DMF and stirred for 20 min. The mixture was then warmed to
3
3
cm of DMF, and the mixture stirred for 20 min. Next, 0.5 cm
60 1C and 0.5 cm of the hydrazine solution added; the color of
3
of the hydrazine solution, prepared by dissolving of 0.2 cm of
3
0% aqueous hydrazine solution in 10 cm DMF, was added,
the polymer changing to black. After 20 min of stirring at
60 1C, catalyst P3 was filtered, washed three times with DMF,
then with MeOH, left for 15 min in MeOH, and then filtrated
and dried in a vacuum.
8
and the mixture was heated to 60 1C. After temperature
stabilization, 5 cm of a PdCl solution in DMF, containing
3
2
3
.00832 g of PdCl in 1 cm DMF, was added, and the color of
À1
0
IR (KBr, cm ): 1730 m (nCO), 1656 vs (nCOO).
In modified procedures regarding the P3 catalyst:
2
the solution changed to black. The mixture was stirred for
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1124–1130 | 1125

View Article Online
a
Table 2 Structural data of Pd catalysts (P1, P2, P3 and Pd/PVP) and the yield of 4-methylbiphenyl obtained in Suzuki coupling in ethylene glycol
g
Catalyst Pd content (%)
Nanoparticle size from Nanoparticle size from TEM data before reaction /nm
XRD data/nm
Before
reaction reaction reaction
After
Before
After
reaction
d
e
Mean (arithmetic)
Variance
Median
Yield of 4-Me-biphenyl (%)
b
P1
P2
P3
3.5
3.5
3.5
3.5
3.0
2.0
—
4.8
5.0
15.0
10.5
3.3
5.2
10.6
—
4.7 (5.2)
4.8
5.7
2.2 (3.9)
1.8
4.4
4.6 (4.9)
4.5
5.4
90
42
0–15
79
f
Pd/PVP 2.7
—
—
—
a
b
Reaction conditions: [Pd] = 1% mol, [4-bromotoluene] = 0.172 g, [BPh(OH)
remained after reaction with P1 (catalyst separated) produced 1% of 4-methylbiphenyl. Arithmetic mean d =
2
] = 0.184 g, base = Cs
c
2
CO
3
, 80 1C, 3 h. The solution that
P
P
n
i
i
d /
i
n ; n = total number of
P
P
d
2
2
2
e
particles, n
i
= number of particles of diameter d
i
.
Variance s = (n
which there are equal numbers of particles smaller and larger (centre of the nanoparticle size distribution). Different results were obtained with
d
i
À
d
i
)/n(n À 1) (describes the width of the distribution). Size for
f
g
the P3 catalyst obtained by impregnation according to the modified procedures. Parameters of the P1 catalyst after the Suzuki reaction are given
in parentheses.
(
(
a) The time of impregnation was prolonged to 5 d (P3a).
b) After impregnation, the Pd polymer was dried, and the
complete reduction of the supported Pd. Finally, the sample
À6
was de-gassed at 120 1C for 2 h under a vacuum of 10 Torr
reduction with hydrazine was performed after 3 d (P3b).
c) After impregnation, the Pd polymer was not separated
and then cooled down in a vacuum to room temperature.
Firstly, the total uptake of hydrogen (adsorption at the Pd
surface and absorption in the bulk) was measured at room
temperature at pressures ranging from 90 to 180 Torr. An
equilibration time of 1 h was allowed for each adsorption
(
from the solution, and hydrazine was added to the mixture
heated at 65 1C (P3c).
All four types of P3 catalyst (P3, P3a, P3b and P3c)
À6
exhibited similar, very low catalytic activities (Table 2).
point. After evacuating the catalyst sample at 10 Torr for
3
0 min at the same temperature, a second isotherm, represent-
Pd/PVP (stabilized colloid). Pd colloid stabilized on PVP
ing only hydrogen absorbed in the Pd bulk (b-hydride), was
measured in the same way.
was prepared according to ref. 28.
3
To 10 cm of water was added 0.5 g of PVP (M = 10 000),
w
3
and the mixture was stirred 20 min. Next, 0.01 cm of
Suzuki–Miyaura reaction (Scheme 3)
hydrazine hydrate (80%) was added, the mixture stirred for
3
3
To a carefully deoxygenated Schlenk tube (40 cm ) in an N
2
2
0 min and warmed to 60 1C. Then, 5 cm of a Pd solution
3
atmosphere was added 0.031 g of P1, P2 or P3 catalyst
(
prepared by dissolving 0.5 g of PdCl
2
in 100 cm of water
containing 1.5 cm of concentrated HCl) was added. After
5 min of stirring at 60 1C, the colour of the solution changed
3
(containing ca. 1 mg Pd), 1.5 mmol of phenylboronic acid,
2
mmol of base, 1 mmol of 4-bromotoluene or 2-bromoto-
1
3
luene, and 5 cm of solvent (usually ethylene glycol). The
Schlenk tube was placed in an oil bath and heated for 3 h with
magnetic stirring. After the reaction was complete, the reac-
to dark brown. Evaporation to dryness produced dark
Pd/PVP colloid.
TEM measurements
tion mixture was cooled down to room temperature and
3
.075 cm of dodecane was added as an internal standard.
0
Morphology and microstructure were investigated by TEM
The reaction product was extracted with three portions of
3
hexane (4, 3 and 3 cm , respectively). The combined extracts
(
Philips CM-20 SuperTwin operating at 200 kV and providing
.25 nm resolution). Analyses of TEM images were made with
0
3
10 cm ) were analyzed by GC-MS.
(
2
9
the ImageJ program. Specimens for TEM were prepared by
dispersing a powder sample in MeOH and placing a droplet of
the suspension onto a copper microscope grid covered with
perforated carbon.
In recycling experiments, the reaction products were ex-
tracted from the reaction mixture by the above-described
procedure, and the portions of all the new reactants were
added to the post-reaction residue.
The Pd and Cl content in the samples were measured using
EDS spectroscopy (Edax PV 9800 spectrometer) in an SEM
microscope (Philips SEM 515).
Results and discussion
Adsorption of hydrogen
Synthesis and properties of GMA resin
Volumetric adsorption measurements were carried out in
conventional glass apparatus. About 0.5 g of the catalyst
was taken for each experiment. Prior to the adsorption
measurements, the P1 or P3 sample was de-gassed for
Divinylbenzene glycidyl methacrylate resin (GMA) was ob-
tained as a result of the suspension polymerization of a
mixture of glycidyl methacrylate, styrene and divinylbenzene
(Scheme 1). The resin, containing B1.3 mmol of epoxy groups
À1
3
0 min at room temperature, then reduced in H
at 120 1C for 1 h and de-gassed at the same temperature for
h. This treatment was repeated twice to make sure that the
Pd was reduced to its metallic form. According to the litera-
2
(250 Torr)
g
was then treated with an excess of a 1,2-diamine (ethyle-
nediamine (EDA) or trans-1,2-diaminocyclohexane (CHDA))
2
in DMF to obtain a diamine-modified polymer (Scheme 2).
3
1
An interesting property of GMA resin is its remarkable
swelling in CH Cl , THF, DMF and 1,4-dioxane solvents, but
3
0,31
ture,
a temperature of 120 1C is high enough to assure
2
2
1
126 | New J. Chem., 2008, 32, 1124–1130
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
chosen because it strongly interacted with the polymer, while
2
at the same time was a good solvent for PdCl . During the
preparation of the P1 catalyst, the GMA-CHDA resin was
first treated with a DMF solution of hydrazine, and then
PdCl dissolved in DMF was added. In the synthesis of the P2
2
Scheme 1 Copolymerisation of glycidyl methacrylate—formation of
GMA resin.
catalyst, the reactants were added in the reverse order, i.e.
freshly prepared Pd(0) nanoparticles were introduced to
GMA-CHDA. The third procedure was based on direct im-
2
5
not in water, MeOH or MeCN. The introduction of diamine
into the polymer network decreases the swelling degree
2
pregnation of PdCl onto the polymer, followed by the reduc-
tion of Pd(II) with hydrazine.
(
(
Table 1); however, it remains quite high when appropriate
compatible) solvents are used. Furthermore, the resin retains
It may be concluded from the IR spectra (in the region
À1
1
500–1750 cm ) of the polymer-supported Pd catalysts that,
a good swelling ability in mixed solvents, such as mixtures of
THF/water and CH Cl /MeOH. Consequently, by changing
most probably, Pd interacts with the oxygens of ester func-
tional groups. The most significant changes in the IR spectra
2
2
the proportion of the compatible and the uncompatible sol-
vent, the bead volume can be controlled, which makes it
possible to encapsulate Pd colloid particles in a network of
polymer-bound diamines.
À1
are visible in the region between 1500 and 1750 cm , includ-
ing in particular a decrease in the intensity of the band at
À1
1
730 cm , a nCO band very characteristic of the polymer, and
À1
the appearance of a new intensive nCO band at 1656 cm
Such changes were observed for all the polymer-supported Pd
.
Since the nature of the solvent may be important for the
effective attachment of Pd to the polymer, several experiments
with different solvents were performed to check the effective
replacement of one solvent by another one. These experiments
catalysts (P1, P2 and P3).
The Pd contents of all the three catalysts under study were
similar: 3.5% for the P1–P3 catalysts. It is, however, worth
noting that the method of preparation has an important effect
on catalyst composition, as can be illustrated by the Cl/Pd
atomic ratio determined by EDS spectroscopy, which
amounted to 0.62, 0.52 and 1.11 for P1, P2 and P3, respec-
tively. During the impregnation procedure, most of the
chlorine remains in the catalyst (P3).
showed that water as solvent cannot substitute CH
DMF in the polymer but effectively replaces THF.
2 2
Cl or
The swelling of the polymer with an aqueous solution of
hydrazine (64%) (used as a reducing agent for Pd(II) in the
catalyst synthesis procedure) was also tested and found to
differ considerably compared to pure water. Although hydra-
zine itself does not swell the polymer, it interacts with the
polymer strongly enough to diminish the swelling properties of
DMF. The polymer treated first with DMF and next with the
SEM studies of the GMA-CHDA resin before and after
2
impregnation with PdCl did not show any visible structural
changes. The resin beads have different shapes, regular sphe-
rical or irregular. The presence of Pd caused only small
changes, demonstrated by the polymer particle size decrease,
but without shape changes (Fig. 1).
same amount of an aqueous solution of hydrazine gave SD =
À1
0
.52 g g , which means that DMF molecules were dislodged
from the polymer by the water-containing hydrazine. Con-
versely, when the polymer was first treated with the hydrazine
The Pd nanoparticle size determined by XRD was 4.8 nm
for P1, 5 nm for P2 and 15 nm for P3. Similar results were also
obtained from TEM measurements. The measured histo-
grams, as well as their parameters, are given in Fig. 2 and
solution and then with DMF, a swelling parameter SD =
À1
0
.31 g g was obtained, which indicates that entry of DMF
into the polymer was blocked.
Characterization of supported palladium catalysts (P1, P2, P3)
The first experiments with Pd(0) nanoparticle immobilization
on GMA resin were not successful because only a very small
amount of Pd was bound very weakly. Much better results
were obtained for polymers modified with diamines, and
consequently the GMA-CHDA polymer was selected as the
support for further experiments.
When preparing the catalysts, the swelling properties of the
solvents were carefully considered, and finally DMF was
Fig. 1 SEM images of (a) GMA-CHDA polymer, (b) P1 catalyst, (c)
Scheme 2 Modifications of GMA-resin.
P2 catalyst and (d) P3 catalyst.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1124–1130 | 1127

View Article Online
Scheme 3 Suzuki–Miyaura reaction pattern.
tion) on pre-treated catalyst samples, and next to determine
the amount of hydrogen absorbed in the Pd bulk (b-hydride).
This was possible because Pd b-hydride decomposes easily
3
4
after short de-gassing at room temperature . The results
presented in Table 3 clearly show that only Pd in the P1
catalyst was able to form b-hydride, and therefore also to
adsorb hydrogen at the surface. In contrast, Pd nanoparticles
in the P3 catalyst exhibited no activity in Pd b-hydride
formation, indicating that the surface of the Pd particles was
totally blocked. This is in agreement with very low catalytic
activity of the P3 catalyst (Table 2).
Fig. 2 Pd particle size distribution: (a) P2, (b) P1, (c) P3 and (d) P1
after Suzuki–Miyaura reaction.
Table 2, respectively. Representative TEM images for samples
P1, P3 and P1 after the Suzuki reaction are presented in
Fig. 3(a)–(c). The TEM results obtained for P2 are summar-
ized in Fig. 4. It appears that the distribution of Pd on the
support is rather non-uniform (Fig. 4a). In addition to small,
individual particles a few nm in size, there are large aggregates
of particles. In a high-resolution image (Fig. 4b), it can be seen
that Pd nanoparticles of ca. 4 nm are crystalline and exhibit
Catalytic activity of Pd nanoparticles supported on
GMA-CHDA (P1, P2, P3) and Pd/PVP in Suzuki–Miyaura
cross-coupling
Catalytic tests on the Suzuki–Miyaura coupling reaction were
performed in ethylene glycol, which was selected as the best
solvent in preliminary screening experiments (Scheme 3). It is
very interesting to note a remarkable difference in the reaction
yields obtained with the P1, P2 and P3 catalysts. The highest
yield, 90% of biaryl, was obtained in a reaction catalyzed by
P1, 42% was produced by P2, whereas P3 was practically
inactive. A soluble Pd colloid, Pd/PVP, yielded 79% of
0
.24 nm lattice fringes corresponding to Pd(1 1 1). The SAED
(
Selected Area Electron Diffraction) pattern obtained from the
aggregate in Fig. 4a contains broad rings that could be
assigned to Pd (Fig. 4c).
4
-methylbiphenyl under the same conditions, slightly less than
It is known from the literature that hydrogen adsorbed onto
3
P1. The low activity of P2 can be explained by the presence of
large aggregates of nanoparticles, probably catalytically-inac-
tive. According to TEM results, the P1 and P3 catalysts have
similar structures. However, the lack of activity of P3 can be
caused by blocking of the active centres by polymers or other
molecules, as was shown by chemisorption tests (Table 3).
The catalytic activity of the P1 catalyst was also checked in
other solvents, as well as in the presence of different bases. The
highest yields (100 and 99%) were obtained in mixtures of
DMF and water (1 : 1), and 2-propanol and water (1 : 1), and
it is clear that the presence of water increased the reaction yield
compared to DMF (0%) and 2-propanol alone (32%). In
contrast, the addition of water to ethylene glycol or MeOH
1
supported Pd catalysts is only partly de-gassed at 120 1C.
Depending on the kind of Pd catalyst, about 60–80% of the
total adsorbed hydrogen remains on the surface of Pd parti-
cles. The total removal of hydrogen from the surface of
supported Pd catalysts is achieved during de-gassing at tem-
3
3
peratures as high as 300–400 1C. In the experimental condi-
tions we used, it was impossible to apply such high
temperatures, and consequently, after final de-gassing of the
P1 and P3 catalysts at 120 1C, the majority of the adsorbed
hydrogen (from the reduction procedure) remained at the Pd
surface. For that reason, hydrogen chemisorption was not
used to estimate the dispersion of Pd in the P1 and P3
catalysts. However, it was possible to determine the total
uptake of hydrogen at room temperature (adsorption/absorp-
Fig. 3 Representative TEM images of (a) P1, (b) P3 and (c) P1 after
Fig. 4 TEM results for the P2 sample: (a) general view, (b) high
Suzuki–Miyaura reaction.
resolution image and (c) SAED pattern.
1
128 | New J. Chem., 2008, 32, 1124–1130
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
Table 3 Adsorption of hydrogen by P1 and P3 catalysts
Catalyst
Total uptake of hydrogen/
mmol H
Hydrogen absorbed in
À1
À1
2
Pd bulk/mmol H /g cat.
2
g cat.
P1
P3
28.0
0.0
19.1
0.0
leads to a decrease in yield from 90 to 66% and from 87 to
3%, respectively. High yields were also obtained in ethylene
glycol and in 2-propanol–water mixtures at lower temperature
5
Fig. 5 The yield of 2-methylbiphenyl obtained with different amounts
(
60 1C).
The data in Table 4 do not make it possible to formulate a
2 3
of catalyst P1 in ethylene glycol at 80 1C with Cs CO base over 3 h.
simple correlation between the catalytic results and the swel-
ling properties of the solvent. However, the best solvents for
Suzuki coupling do not cause volume increase in the
GMA-CHDA resin.
P1 also catalyzes the Suzuki coupling of 2-bromotoluene
with phenylboronic acid with a yield of 88%. Some decrease of
reaction yield was observed when the amount of Pd was lower
than 0.37 mol% (Fig. 5). It was also checked that after
separation of the organic products, the glycol solution was
not catalytically active; used as a catalyst, it produced only 1%
of 2-methylbiphenyl. This provides evidence that only Pd
nanoparticles immobilized on the GMA-CHDA polymer are
responsible for the catalytic activity of the system. Additional
evidence of P1 catalyst stability was obtained from ICP
analysis of the Pd content in the solution after the Suzuki
reaction, which confirmed the lack of leaching into the solu-
tion (Pd content was below the detection limit).
Fig. 6 The yield of 4-methylbiphenyl obtained in recycling experi-
ments with catalyst P1 in ethylene glycol at 80 1C with Cs CO base.
2
3
a higher yield of the product in a recycling experiment, it is
better not to filter off the catalyst.
Fig. 6 summarizes the results of a recycling experiment
performed with the P1 catalyst in ethylene glycol. It can be
concluded that the second reaction is slower than the first one,
but still quite satisfactory results can be obtained after a
prolonged reaction time. Similarly, in the third reaction,
Conclusions
A new Pd catalyst obtained by the encapsulation of Pd(0)
nanoparticles in a GMA-CHDA polymer acts as a hetero-
geneous catalyst in Suzuki–Miyaura cross-coupling reactions
in different organic solvents. The lack of leaching was con-
firmed through analysis of the Pd content, as well as by the
observation that the post-reaction solution, when used as a
catalyst, is practically inactive. In contrast to the PolyHipe
8
6% of 4-methylbiphenyl was obtained after 24 h. To obtain
Table 4 Effect of solvent and base in a Suzuki–Miyaura coupling
reaction catalyzed by P1
a
17
catalyst, ours is highly active in the absence of phosphine or
other additives.
Solvent
Base
Cs CO
Yield (%)
Catalyst recycling is quite good, but it is complicated by the
presence of side products (mainly boron compounds) in the
reaction mixture that can block the active centres of the
catalyst.
Ethylene glycol
Ethylene glycol
Ethylene glycol
Ethylene glycol
Ethylene glycol
Ethylene glycol–H
MeOH
MeOH
MeOH–H
MeOH–H
Dioxane
Toluene
DMF
2
3
90
93
86
89
90
66
87
67
86
53
1
1
0
100
5
32
87
88
99
98
3 4
K PO
KOH
KOH
NaHCO
b
3
It was found that water has a very interesting effect on the
course of the Suzuki–Miyaura reaction. In most cases, the
addition of water led to an increase in biphenyl yield; however,
recycling of the catalyst from the reaction mixture led to a
total loss of activity. Similarly, washing the catalyst with water
caused its deactivation.
2
O (1 : 1 v/v)
Cs
Cs
2
CO
CO
3
2
3
NaHCO
3
2
O (3.5 : 1.5 v/v)
O (1.5 : 3.5 v/v)
Cs
Cs
Cs
Cs
Cs
Cs
Cs
2
CO
2
CO
2
CO
2
CO
2
CO
2
CO
2
CO
3
3
3
3
3
3
3
2
DMF–H
2
O (1 : 1 v/v)
2
2
2
2
2
2
-Propanol
-Propanol
-Propanol–H
-Propanol–H
-Propanol–H
-Propanol–H
Acknowledgements
KOH
KOH
KOH
KOH
KOH
The authors thank Prof. Leszek K e˛ pinski (Institute of Low
´
2
2
2
2
O (4.5 : 0.5 v/v)
O (3.5 : 1.5 v/v)
O (1 : 1 v/v)
Temperature and Structure Research, Polish Academy of
Sciences, Wroc"aw) for performing the TEM measurements
and helpful discussions, and to Tomasz Borkowski and Prze-
mys"aw Stach (Faculty of Chemistry, University of Wroc"aw)
for experimental assistance.
b
O (1 : 1 v/v)
a
Reaction conditions: [Pd] = 1% mol, [4-bromotoluene] = 0.172 g,
b
[
2
BPh(OH) ] = 0.184 g, 80 1C, 3 h. Reaction at 60 1C.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1124–1130 | 1129

View Article Online
The authors thank Marek Hojniak (Faculty of Chemistry,
University of Wroc"aw) for performing the GC-MS analyses.
Financial support from grant PBZ-KBN-116/T09/2004 is
gratefully acknowledged.
15. R. Narayanan and M. A. El-Sayed, J. Am. Chem. Soc., 2003, 12,
8
340.
1
6. R. Narayanan and M. A. El-Sayed, J. Catal., 2005, 234, 348.
7. A. Desforges, R. Backov, H. Deleuze and O. Mondain-Monval,
Adv. Funct. Mater., 2005, 15, 1689.
1
1
8. Ch. Ramarao, S. V. Ley, S. C. Smith, I. M. Shirley and
N. DeAlmeida, Chem. Commun., 2002, 1132.
19. Y. Uozumi, Top. Curr. Chem., 2004, 242, 77.
20. N. Mustan and E. A. H. Hall, Anal. Chim. Acta, 1999, 396, 171.
References
1
. Polymeric Materials in Organic Synthesis and Catalysis, ed.
M. R. Buchmeiser, Wiley-VCH, Weinheim, 2003.
. J. Horn, F. Michalek, C. C. Tzschucke and W. Bannwarth, Top.
Curr. Chem., 2004, 242, 43.
21. E. Sua
´
rez, B. Paredes, F. Rubiera, M. Rendueles, M. A. Villa-
az, Sep. Purif. Technol., 2002, 27, 1.
Garcıa and J. M. Dı
´
22. N. Bicak, D. C. Sherrington, S. Sungur and N. Tan, React. Funct.
Polym., 2003, 54, 141.
´
2
3
4
. R. Haag and S. Roller, Top. Curr. Chem., 2004, 242, 1.
. K. Grego"a, E. Framery, K. M. Pietrusiewicz and D. Sinou, Adv.
Synth. Catal., 2006, 348, 1728.
. Y. M. A. Yamada, K. Takeda, H. Takhashi and S. Ikegami,
J. Org. Chem., 2003, 68, 7733.
23. H. Zou, Q. Luo and D. Zhou, J. Biochem. Biophys. Methods,
2001, 49, 199.
24. G. Giammone, G. Pitarresi, G. Cavallaro, S. Buscemi and
F. Saiano, Biochim. Biophys. Acta, 1999, 1428, 29.
25. A. Bukowska, W. Bukowski and J. Noworol, J. Appl. Polym. Sci.,
´
5
6
7
. Q.-S. Hu, Y. Lu and Z.-B. Yu, J. Am. Chem. Soc., 2003, 125, 2856.
. P. G. Steel and C. W. T. Teasdale, Tetrahedron Lett., 2004, 45,
2006, 101(3), 1487.
26. S. C. Hong, U. Rief and M. O. Kristen, Macromol. Rapid
Commun., 2001, 22, 1447.
27. R. R. Jay, Anal. Chem., 1964, 36, 667.
8977.
8
. J.-H. Kim, J.-W. Kim, M. Shokouhimehr and Y.-S. Lee, J. Org.
Chem., 2005, 70, 6714.
28. A. Gniewek, J. J. Zio
"kowski, A. M. Trzeciak and L. K e˛ pinski,
´ ´
9. J.-W. Byun and Y.-S. Lee, Tetrahedron Lett., 2004, 45, 1837.
J. Catal., 2006, 239(2), 272.
1
1
1
0. J.-W. Kim, J.-H. Kim and D.-H. Lee, Tetrahedron Lett., 2006, 47,
745.
1. J. C. Garcia-Martinez, R. W. J. Scott and R. M. Crooks, J. Am.
Chem. Soc., 2003, 125, 11190.
2. R. van Heerbeek, R. C. J. Kamer, P. W. N. M. Van Leeuwen and
J. N. H. Reek, Chem. Rev., 2002, 102, 3717.
29. W. Rasband, ImageJ, U. S. National Institutes of Health, Bethes-
da, Maryland, USA, 1997–2005 (http://rsb.info.nih.gov/ij/).
30. T. B. Lin and T. C. Chou, Appl. Catal., 1994, 108, 7.
31. G. Prelazzi, M. Cerboni and G. Leofanti, J. Catal., 1999, 181, 73.
´ ´
32. W. Bukowski, A. Bukowska, J. Noworol and B. Mysliwiec,
Polimery, 2006, 51(11–12), 787.
4
1
1
3. D. Astruc, Inorg. Chem., 2007, 46, 1884.
4. D. Astruc, F. Lu and J. R. Aranzaes, Angew. Chem., Int. Ed.,
33. N. Krishnankutty and M. A. Vannice, J. Catal., 1995, 155, 312.
34. J. E. Benson, H. S. Hwang and M. Boudart, J. Catal., 1973, 30,
146.
2005, 44, 7852.
1
130 | New J. Chem., 2008, 32, 1124–1130
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008