The effect of stabilizers on the catalytic activity and stability of Pd colloidal nanoparticles in the Suzuki reactions in aqueous solution

Article abstract of DOI:10.1021/jp010904m

Transition metal nanoparticles used in catalysis in solution are stabilized by capping the surfaces that are supposed to be used for catalysis. Determining how these two properties, that is, the catalytic activity and stability of nanoparticles, change as

Full text of DOI:10.1021/jp010904m

8938
J. Phys. Chem. B 2001, 105, 8938-8943
The Effect of Stabilizers on the Catalytic Activity and Stability of Pd Colloidal
Nanoparticles in the Suzuki Reactions in Aqueous Solution†
Yin Li and Mostafa A. El-Sayed*
Laser Dynamics Laboratory, School of Chemistry and Biochemistry, Georgia Institute of Technology,
Atlanta, Georgia 30332-0400
ReceiVed: March 8, 2001; In Final Form: May 13, 2001
Transition metal nanoparticles used in catalysis in solution are stabilized by capping the surfaces that are
supposed to be used for catalysis. Determining how these two properties, that is, the catalytic activity and
stability of nanoparticles, change as different capping materials are used is the aim of this work. Pd nanoparticles
prepared by the reduction of metal salts in the presence of three different stabilizersshydroxyl-terminated
poly(amido-amine) (PAMAM) dendrimers (Gn-OH, where Gn represents the nth generation), block copolymer
polystyrene-b-poly(sodium acrylate) and poly(N-vinyl-2-pyrrolidone) (PVP)sare used as catalysts in the Suzuki
reactions in an aqueous medium to investigate the effects of these stabilizers on the metallic nanoparticle
catalytic activity and stability. The stability of the Pd nanoparticles is measured by the tendency of the
nanoparticles to give Pd black powder after the catalytic reaction. The Suzuki reaction is a good “acid test”
for examining the stability of these nanoparticles, as it takes place when refluxed at about 100 °C for 24 h.
The stability is found to depend on the type of the stabilizer, the reactant, and the base used in the reaction
system. Pd nanoparticles stabilized by block copolymer, G3 dendrimer, and PVP are found to be efficient
catalysts for the Suzuki reactions between phenylboronic acid (or 2-thiopheneboronic acid) and iodobenzene.
G4 dendrimer is found to be an effective stabilizer; however, strong encapsulation of Pd particles in the
dendrimer results in a loss of catalytic activity. The Suzuki reactions between arylboronic acids and bromoarenes
catalyzed by Pd nanoparticles result in byproducts due to the homo-coupling of bromoarenes. A summary of
the catalytic activity and stability of the Pd nanoparticles in these different systems is tabulated. As one
would expect, these two properties are anticorrelated, that is, the most stable is the least catalytically active.
Introduction
phine-mediated catalysis must be replaced by methods involving
either new types of ligands forming highly reactive but stable
complexes, or new forms of palladium able to operate in the
absence of tightly coordinated ligands. These new methods can
be regarded as phosphine-free catalysis, and the development
of new ligands has brought some promising results.5 One
important drawback of homogeneous catalysts is that they are
difficult to separate from reaction mixtures and recycle. To solve
this problem, studies concerning coupling reactions in aqueous
media using water-soluble phosphine ligands such as sulfonated
analogues of triphenylphosphine6 have been carried out. The
use of water as a reaction medium for transition-metal-catalyzed
reactions is very attractive for organic synthesis due to
environmental, economical, and safety reasons.
A great interest is placed on the use of metallic nanoparticles
for catalysis. In solution, the particles are always capped to
stabilize them against precipitation. In this paper, we examine
the stability of these particles and correlate the stability to the
catalytic activity. For this, we select a very critical “acid test”,
the Suzuki reaction. It takes place when the reactants are
refluxed at about 100 °C for 24 h. For capping materials, we
select three different systems, dendrimers of different genera-
tions, block copolymer, and linear polymer.
For the past three decades, palladium catalysts have played
an important role in the field of organic synthesis due to the
variety of transformations they are able to catalyze.1 One of
the major contributions of zerovalent palladium catalysts in
organic synthesis is the formation of carbon-carbon bonds
through many coupling reactions. The cross-coupling of aryl-
boronic acids with aryl halides, known as the Suzuki-type
reaction,2 and the coupling of aryl halides with alkenes, known
as the Heck reaction,3 are catalyzed by zerovalent palladium
species and represent important synthetic methods in organic
chemistry. Traditionally, these reactions are carried out in
organic solvents and catalyzed by various Pd/ligand systems.4
Phosphine-based palladium catalysts are generally used and, in
most cases, the reactions are restricted to bromo or iodoarenes.
The usage of chloroarenes is still a challenge due to economical
considerations. To obtain high reactivity, the common phos-
Transition metal nanoparticles are effective catalysts for
chemical transformations due to their large surface area and a
unique combination of reactivity, stability, and selectivity.7 A
common method to prepare metal particles involves the reduc-
tion of metal ions in the presence of stabilizers such as
surfactants and polymers. Crooks et al.8 have prepared
PAMAM dendrimer-encapsulated Pd, Pt, and other metal
nanoparticles that employ dendrimers as both template and
stabilizer. Pd particles thus prepared show high catalytic activity
for the hydrogenation of alkenes and are quite stable during
the reaction.8a,8c,8e Mayer et al.9 have reported that Pd nano-
particles prepared by reducing the metal precursors in the
presence of protective amphiphilic block copolymer polystyrene-
b-poly(methacrylic acid) are catalytically active for the hydro-
† Part of the special issue “Royce W. Murray Festschrift”.
10.1021/jp010904m CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/20/2001

Pd Colloidal Nanoparticles in the Suzuki Reactions
J. Phys. Chem. B, Vol. 105, No. 37, 2001 8939
genation of cyclohexene. Bronstein et al.10 have reported that
Pd colloids prepared in polystyrene-poly-4-vinylpyridine (PS-
b-P4VP) block copolymer micelles and deposited on Al2O3
exhibited high activity and selectivity in hydrogenation of the
triple bond of acetylene alcohols to the double bond of olefin
alcohols. It has been shown that palladium colloids on the
nanometer scale are effective catalysts for the Heck and Suzuki
reactions in organic solvent.11 However, poor stability of the
Pd particles under these coupling reaction conditions has been
reported.11a,11c
Recently, we reported12 that palladium nanoparticles stabilized
by PVP in colloidal aqueous solution are efficient catalysts for
some Suzuki cross-couplings. By using fluorescence spectros-
copy, it is shown that the initial rate of the Suzuki reactions
depends linearly on the concentration of Pd catalyst, giving
strong evidence that the catalysis occurs on the surface of the
Pd nanoparticles. One disadvantage of this reaction system is
the precipitation of Pd particles after reaction, leading to
decreased catalytic activity of the Pd colloidal solution.
The development of phosphine-free palladium nanoparticle
catalysis in aqueous media has brought about procedures with
the catalytic efficiency and simplification of techniques with a
positive impact on safety, but there are still many problems to
overcome. The first is how palladium nanoparticles function in
catalytic C-C bond formation reactions. The second is that the
deactivation of Pd catalysts occurs due to the formation of Pd
black. To increase the stability of Pd nanoparticles in catalytic
C-C bond formation reactions, it is necessary to prepare Pd
nanoparticles by using various kinds of stabilizers and to
investigate the effect of these stabilizers on the catalytic activity
and the stability of Pd nanoparticles.
OH(Pd2+)10, n ) 3 and 4). For G2 dendrimer-Pd particles, a
solution containing an average of 5 Pd2+ ions per dendrimer
(G2-OH(Pd2+)5) was prepared. Prior to reduction, the pH of
the dendrimer-Pd2+ solution was adjusted to 4.0 using 1 M HCl.
The solution was purged with N2 for 5 min. A 2 mL portion of
0.36 M NaBH4 aqueous solution (the concentration of NaBH4
added was 8 times that of metal ions) was then added, and the
solution was vigorously stirred in N2 for 1 h. The solution color
changed from yellow to dark brown.
Preparation of PVP-Pd Nanoparticles. The method to
prepare Pd nanoparticles in the presence of PVP is similar to
that we12 and Teranishi13 et al. reported previously. H2PdCl4
solution (2.0 mM) was prepared by mixing 0.6 mmol of PdCl2,
6.0 mL of 0.2 M HCl, and 294 mL of H2O. A mixture of 15
mL of a 2.0 mM H2PdCl4 solution, 21 mL of H2O, 14 mL of
ethanol (40 vol %), and 0.0333 g of PVP (PVP/Pd (monomeric
unit/metal ion) ) 10) was refluxed for 3 h in air. The solution
thus prepared had a dark brown color.
Preparation of Block Copolymer-Pd Nanoparticles. The
method to prepare Pd nanoparticles in the presence of PS-b-
PANa is similar to that of Mayer9 et al. reported previously
with some modifications. A sample of 0.0798 g PS-b-PANa
was added to 21 mL of H2O, and the solution was sonicated
for about 2 h until the copolymer almost dissolved. A 15 mL
portion of 2.0 mM H2PdCl4 solution (mass ratio of copolymer:
Pd metal ) 25:1) was then added, and the solution was sonicated
until the copolymer dissolved completely. A solution of 14 mL
of EtOH was added and the solution was refluxed in air for 3
h. The resulting dark brown solution indicated the reduction of
the Pd ions.
Transmission Electron Microscopy (TEM). Pd nanopar-
ticles were investigated by transmission electron microscopy
(TEM) on a JEM 100C transmission electron microscope. The
samples were prepared by placing a drop of the solution on
carbon-coated Cu grids and allowed to dry in air. The particle
size and standard deviation were determined by counting 200
particles from enlarged TEM images.
In the present paper, Pd nanoparticles are prepared by the
reduction of metal salts in the presence of PAMAM dendrimers,
block copolymer polystyrene-b-poly(sodium acrylate), and PVP.
The effect of these stabilizers on the catalytic activity and
stability of Pd nanoparticles in the Suzuki reactions in aqueous
medium is investigated.
Pd-Catalyzed Suzuki Cross-Coupling Reactions. Suzuki
coupling reactions were carried out by using different Pd
catalysts. In a typical procedure (e.g., coupling of 2-thiophenebo-
ronic acid with iodobenzene catalyzed by G3 dendrimer-Pd),
1.5% molar equiv of a dendrimer-Pd solution (100 mL) and 3
equiv (15 mmol) of Na3PO4•12H2O were added to 100 mL of
a 80% EtOH solution. Then, 1 equiv (5 mmol) of iodobenzene
and 1.5 equiv (7.5 mmol) of 2-thiopheneboronic acid were added
to the solution. The mixture was refluxed for 24 h and the extent
of reaction was monitored by TLC (thin-layer chromatography)
Experimental Section
Chemicals. Poly(amido-amine) (PAMAM) dendrimers (Gn-
OH, where n ) 2,3 and 4) were obtained in 10-20% methanol
solutions from Aldrich. Prior to use, methanol was removed by
rotary evaporation at room temperature. Palladium chloride
(PdCl2, 99%, Lancaster), Potassium tetrachloro-palladate (II)
(K2PdCl4, 99.99%, Aldrich), poly(N-vinyl-2-pyrrolidone) (PVP,
average molecular weight 40 000, Aldrich), polystyrene-b-poly-
(sodium acrylate) (PS-b-PANa, molecular weight (Mn) of PS
1800, Mn of PANa 42 500, Mw/Mn 1.06, Polymer Source),
sodium borohydride (NaBH4, 99%, Aldrich), hydrochloric acid
(HCl, Fisher), 2-thiopheneboronic acid (Aldrich), phenylboronic
acid (97%, Aldrich), iodobenzene (Eastman Kodak), bromoben-
zene (Fisher), 2-bromothiophene (98+%, Lancaster), triethyl-
amine (Aldrich), sodium phosphate (J. T.Baker), sodium acetate
(99+%, Fisher), and all solvents (from Aldrich) were used as
received. All solutions were prepared with doubly deionized
water.
Preparation of Dendrimer-Pd Nanoparticles. Pd nanopar-
ticles in the presence of PAMAM dendrimers with different
generations (G2, G3, and G4) were prepared according to the
method described by Crooks et al.,8 with some modifications.
The aqueous stock solutions of 3 mM K2PdCl4 and 0.1 mM
Gn-OH dendrimer (from solvent-free dendrimer) were prepared
separately. A solution of 30 mL of K2PdCl4 was mixed with
90 mL of dendrimer solution (10 Pd2+ ions per dendrimer, Gn-
1
and H NMR. After cooling to room temperature, the reaction
mixture was extracted with petroleum ether, and the organic
layers were combined and dried with anhydrous Na2SO4.
Purification by column chromatography (silica gel/petroleum
1
ether) gave the desired products confirmed by H NMR. The
other reactions followed similar procedures.
Results and Discussion
Preparation of Pd Nanoparticles. We prepare dendrimer-
Pd nanoparticles based on the method described by Crooks et
al.8 In their approach, Pd2+ ions are sorbed into OH or NH2-
terminated PAMAM dendrimers, where they complex strongly
with interior tertiary amine groups. Subsequent chemical reduc-
tion of the metal ions with BH4- results in dendrimer-
encapsulated Pd nanoparticles. Here, the dendrimers are used
both as templates and stabilizers.
In the present study, Pd particles are prepared by using various
generations of PAMAM dendrimers (Gn-OH, n ) 2, 3 and 4)

8940 J. Phys. Chem. B, Vol. 105, No. 37, 2001
Li and El-Sayed
stable for months, and no evidence for precipitation of Pd
particles is observed, also suggesting the formation of encap-
sulated Pd particles. By using G3 and G4 dendrimers as
stabilizers, the mean diameters of Pd nanoparticles are the same,
indicating that the Pd particle size is insensitive to the generation
of hydroxyl-terminated PAMAM dendrimers. G2 dendrimer
does not provide an effective protective action. When the molar
ratio of Pd2+ to dendrimer is either 10:1 or 5:1, many aggregates
of Pd colloids are observed, and stable Pd particles cannot be
obtained. The un-aggregated G2 dendrimer-Pd particles have a
mean diameter of 3.6 nm with quite wide size distribution (Table
1, TEM not shown here). This diameter (3.6 nm) is large
compared to that of dendrimer G2 (2.9 nm),15 indicating that
the Pd particles are not formed inside the dendrimer. The
difference in the protective action of G2, G3, and G4 dendrimers
may be due to their structures. Higher generation dendrimers
(G3 and G4) have closed, increasingly compact structures and
they are expected to provide effective protective action for the
prepared Pd particles. Lower generation dendrimers (G2) have
relatively open structures (G2 dendrimer is flat). A single G2
dendrimer cannot provide enough material to stabilize the
surface of one Pd particle, and the particle stabilization by G2
dendrimer can be considered to be similar to that by conven-
tional linear polymers: G2 dendrimers provide protective action
by adsorbing on the surface of Pd particles. This behavior has
been reported previously.16
Figure 1. TEM image of Pd nanoparticles stabilized by G4-OH
PAMAM dendrimer (Pd2+/G4-OH dendrimer ) 10).
TABLE 1: Mean Diameters of Pd Particles Stabilized by
Various Generations of Dendrimers
dendrimer
generation
mean diameter
(standard deviation) (nm)
[Pd2+]/[dendrimer]
G2
G3
G4
5
10
10
3.6 (0.8)a
1.4 (0.4)
1.4 (0.4)
Amphiphilic block copolymer PS-b-PANa consists of a
hydrophobic PS block and a hydrophilic block, which by itself
is a moderate to good stabilizer for metal particles. After
reduction of H2PdCl4 by refluxing alcoholic solution in the
presence of block copolymer, a clear dark-brown solution that
is stable for months is obtained. The TEM picture of block
copolymer-Pd nanoparticles (Figure 2) shows that the particles
are quite monodispersed, and the mean diameter of the Pd
particles is 3.0 ( 0.7 nm. From Figure 2 alone, it is hard to tell
whether the Pd nanoparticles are actually embedded within the
hydrophobic micelle core of block copolymer. However, it is
expected that the Pd nanoparticles are not located exactly in
the micelle core because there is very small tendency for the
hydrophilic Pd precursor H2PdCl4 to be accumulated inside the
hydrophobic micelle core. The principle for the particle
stabilization here is expected to be similar to the general steric
stabilization of colloidal metal nanoparticles such as homopoly-
mers and random copolymers.
a The values are determined by counting un-aggregated Pd nano-
particles from TEM.
as stabilizers. Our procedure involves mixing K2PdCl4 with
dendrimer. Before reduction, the pH of the solution is adjusted
to 4.0. A clear dark brown solution is obtained after reduction
of the composite with NaBH4. It should be mentioned that
without adjusting the pH of the solution, Pd precipitation is
always observed after reduction with NaBH4. This might be
attributed to the coordination of palladium ions with other free
ions outside the dendrimer, rather than the complexation of
palladium ions with tertiary amine groups inside the dendrimer
at higher pH. Reduction of palladium ions outside the dendrimer
results in the precipitation of Pd particles. Adjusting the solution
pH to 4.0 avoids the coordination of Pd2+ with other ions outside
the dendrimer. Accordingly, Pd2+ ions bind preferentially to
the interior tertiary amines and, upon reduction, the dendrimer-
encapsulated Pd particles are formed. It has been demonstrated
by Crooks8c and Tomalia14 et al. that at pH ) 4, the interior
tertiary amines (pKa ) 5.5)8f of the dendrimers are almost all
deprotonated and allow their coordination with Pd2+ ions.
The effect of dendrimer generation on the size and stability
of the Pd particles is investigated by comparing nanocomposites
prepared using G2, G3, and G4 PAMAM dendrimers. Figure 1
shows the TEM micrograph of Pd particles stabilized by G4
dendrimer. It can be seen that the dendrimer-encapsulated
particles are quite monodisperse, and the mean diameter of the
Pd particles (G4-OH(Pd10)) is 1.4 ( 0.4 nm, similar to that
reported previously.8a Table 1 gives the mean diameter with
the standard deviation of Pd nanoparticles in different generation
dendrimers. It can be seen from Table 1 that for G2, the particles
formed are larger than those obtained with G3 or G4. The mean
diameter of Pd particles in the presence of G3 or G4 dendrimer
is 1.4 nm, which is smaller than the diameter of dendrimer G3
(3.6 nm) and G4 (4.5 nm),15 indicating the formation of
dendrimer-encapsulated Pd nanoparticles. Both G3 and G4
dendrimers are found to be effective stabilizers when the molar
ratio of Pd2+ to dendrimer is 10:1. The Pd nanoparticles are
Catalytic Activity and Stability of Pd Nanoparticles in the
Suzuki Reactions. The Suzuki reactions are tested using
phenylboronic acid (or 2-thiopheneboronic acid) and iodoben-
zene in the presence of catalytic amounts of Pd nanoparticles
stabilized by various generations of PAMAM dendrimers.
Table 2 gives the reaction conditions and results for these
two reactions. It is seen that G4 dendrimer-encapsulated Pd
nanoparticles is not an efficient catalyst for the Suzuki reactions
under investigation (Table 2, entries 1-4). The product yield
is affected by the bases used in the reaction system (Table 2,
entries 2-4), but all these three commonly used bases result in
poor yields. No Pd black is observed after the reaction; however,
the color of the solution changes from dark brown before

Pd Colloidal Nanoparticles in the Suzuki Reactions
J. Phys. Chem. B, Vol. 105, No. 37, 2001 8941
TABLE 2: Product Yields for Suzuki Coupling Reactions Catalyzed by Dendrimer-Pd Particles (1.5 mol % of Metal) in 40%
EtOH. (Temp, Reflux; Time, 24 h)
formation
of Pd black
soln color
after reacn
entry
reacn
catalyst
base
yielda
1
2
3
4
5
6
7
8
9
(1)
(2)
(2)
(2)
(1)
(2)
(1)
(2)
(1)
(2)
G4-OH(Pd)10
G4-OH(Pd)10
G4-OH(Pd)10
G4-OH(Pd)10
G3-OH(Pd)10
G3-OH(Pd)10
yellow/green solnb
yellow/green solnb
G2-OH(Pd)5
Na3PO4
Na3PO4
NEt3
NaOAc
Na3PO4
Na3PO4
Na3PO4
Na3PO4
Na3PO4
Na3PO4
17%
20%
35%
< 5%
71%
90%
40%
80%
69%
70%
no
no
no
yes
no
yes
no
yes
yes
yes
light yellow
light yellow
light yellow
brown
yellow/green
yellow/greenc
yellow/green
yellow/greenc
yellow/greenc
yellow/greenc
10
G2-OH(Pd)5
a Isolated yield. b The clear yellow/green solution after extraction is used as the catalyst. See text for detail. c Solution color after filtering the Pd
precipitates.
solution, the following control experiment is performed. After
the reaction between phenylboronic acid and iodobenzene
catalyzed by G3 dendrimer-encapsulated Pd nanoparticles (Table
2, entry 5), the reaction mixture is extracted with petroleum
ether, where the organic components are separated from the
aqueous solution containing Pd catalyst. The resulting clear
yellow/green solution is used as the catalyst for the Suzuki
reactions, and it can be seen from Table 2 (entries 7 and 8) that
the yellow/green solution is still catalytically active in the Suzuki
reactions. It should be mentioned that for the reactions catalyzed
by yellow/green solution, no Pd black is formed in reaction (1)
whereas Pd black is observed in reaction (2).
The Suzuki reactions are also tested in the presence G2
dendrimer-Pd particles. As can be seen from Table 2 (entries 9
and 10), the G2 dendrimer-Pd particles are catalytically active
in the Suzuki reactions. The Pd black is formed during the
reaction, which is not surprising considering that G2 dendrimer
is not an effective stabilizer, as mentioned previously.
Dendrimer porosity is a function of generation; higher
generation materials are less porous and thus less likely to admit
the reactants to the interior metal nanoparticles.17 On the basis
of the investigation of the catalytic activity and stability of G2,
G3, and G4 dendrimer-Pd nanoparticles in the Suzuki reactions,
we come to the following conclusions. First, G4 PAMAM
dendrimers provide an effective protective action toward Pd
particles (no observable Pd black during the reaction), but strong
encapsulation of Pd particles within the dendrimers results in a
loss of catalytic activity; Second, G2 and G3 dendrimer-Pd
particles are efficient catalysts in the Suzuki reactions. G3
dendrimers also provide effective protective action to the
particles in the reaction between phenylboronic acid and
iodobenzene, but the stability of Pd catalyst depends on the
dendrimer generations employed as well as components in the
reaction mixtures. By using dendrimers with different genera-
tions, it is possible to control the stability of Pd particles, the
reaction rate and do selective catalysis. Finally, the presence of
sulfur on one of the reactants seems to play a role in the
formation of Pd black. Somehow, it seems to drive the particles
out of their capping nest.
The catalytic activity and stability of block copolymer-
stabilized Pd nanoparticles in the Suzuki reaction (2) is
investigated. It is seen from Table 3 that block copolymer-Pd
nanoparticles having a mean diameter of 3.0 nm show high
catalytic activity, comparable to PVP-stabilized Pd nanoparticles.
By using triethylamine as the base, no Pd precipitate is observed
during the reaction. By using sodium acetate as the base, Pd
nanoparticles precipitate out of the solution. The ionic strength
of the solution is increased upon the addition of inorganic salt
such as sodium acetate, resulting in the ineffective protective
Figure 2. TEM image of Pd nanoparticles stabilized by block
copolymer PS-b-PANa (mass ratio of copolymer:metal ) 25:1).
reaction to clear light yellow after the reaction. The solution
color is found to be pH-dependent. After the reaction, the
solution pH is ∼10. When the pH is adjusted to ∼2, the solution
color changes back to brown. It is worth noting that no color
change is observed when the pH of G4 dendrimer-encapsulated
Pd nanoparticle solution alone is adjusted to ∼10. This suggests
that during the reaction, the initially formed nanoparticles may
decompose to extremely fine particles, which are encapsulated
in the dendrimers more efficiently than the initially formed
particles, and/or the surface of the particles is oxidized into ionic
forms that complex strongly with interior tertiary amine groups.
In both cases, the surface of the particles is fully passivated,
resulting in a loss of catalytic activity. When the pH of solution
is adjusted to ∼2, the interior tertiary amines are protonated.
As a result, palladium precipitate out of the dendrimer. More
detailed studies of the identifications of reaction byproducts and
their interactions with Pd particles is currently underway.
G3 dendrimer-encapsulated Pd nanoparticles is found to be
an efficient catalyst for the Suzuki reactions (Table 2, entries 5
and 6). No evidence of Pd precipitation is observed in the
coupling reaction (1), whereas Pd black is observed in the
reaction (2) after refluxing the solution for 24 h. In both
reactions, the solution color changes from dark brown to yellow/
green during the reaction. The solution color is also found to
be pH-dependent. After the reaction, the pH is ∼10, and the
solution color changes back to brown by adjusting the pH to
∼2. To investigate the catalytic activity of clear yellow/green

8942 J. Phys. Chem. B, Vol. 105, No. 37, 2001
Li and El-Sayed
TABLE 3: Product Yields for Suzuki Coupling Reactions Catalyzed by Pd Nanoparticles (Temp, Reflux; Time, 12 h)
catalyst
(mol % of metal)
formation
of Pd black
entry
reacn
base
NEt3
NaOAc
NEt3
Na3PO4
Na3PO4
yielda
1
2
3
4
5
(2)
(2)
(2)
(1)
(2)
block copolymer-Pd (0.6%)b
block copolymer-Pd (0.6%)b
PVP-Pd (0.6%)b
85%
88%
80%
95%
92%
no
yes
yes
yes
yes
PVP-Pd (0.3%)c
PVP-Pd (0.3%)c
a Isolated yield. b The reactions are carried out in 3:1 CH3CN/H2O. c The reactions are carried out in 40% EtOH. From ref 12.
TABLE 4: Product and Byproduct Yields for Suzuki Coupling Reactions between Arylboronic Acids and Aryl Bromides
Catalyzed by Pd Nanoparticles (1.0 mol % of Metal) in 3:1 CH3CN/H2O. (Base, NEt3; Temp, Reflux; Time, 24 H)
suzuki product
(yield %)a
byproduct
(yield %)a
formation
of Pd black
entry
1
reacn
catalyst
(3)
block copolymer-Pd
2-phenyl thiophene
(16%)b
bithiophene
(16%)b
no
2
3
4
(4)
(3)
(4)
block copolymer-Pd
PVP-Pd
2-phenyl thiophene
(0%)
2-phenyl thiophene
(0%)
2-phenyl thiophene
(0%)
biphenyl
(13%)
bithiophene
(18%)
biphenyl
(10%)
no
a little
a little
PVP-Pd
a Isolated yield. b A mixture of thiophene benzene and bithiophene are obtained (yield: 32%) which cannot be separated by column chromatography
1
(silica gel/petroleum ether). The yield is determined by H NMR ratio of these two compounds.
action of block copolymer to Pd particles. By adding triethyl-
amine, further stabilization of Pd nanoparticles is obtained, in
addition to the protection from block copolymer. The stability
of the Pd particles depends on the components in the reaction
mixture. The results from Table 3 (entries 1 and 3) show that
block copolymer has a better protective action toward the Pd
particles than PVP, although the principle for the particle
stabilization here is expected to be similar to the general steric
stabilization of metal particles such as PVP, as mentioned earlier.
It is expected that the careful choice of polymer types, metal
precursors, as well as reaction conditions will lead to the
preparation of the desired catalyst type for special needs.
The catalytic activity and stability of Pd nanoparticles are
also investigated in the Suzuki reactions between arylboronic
acids and aryl bromides.
PVP-Pd nanoparticles as a catalyst, Pd particles precipitate out
of the solution during the reaction. Once again, this shows that
block copolymer has a better protective action toward the Pd
particles than PVP.
Conclusions
In the present paper, palladium nanoparticles are prepared in
the presence of three different stabilizers, and the catalytic
activity as well as the stability of Pd nanoparticles thus prepared
is investigated in the Suzuki reactions in aqueous solution. It is
found that the Pd nanoparticles stabilized by PVP, block
copolymer, and G3 dendrimer are efficient catalysts for the
Suzuki reactions between phenylboronic acid (or 2-thiophenebo-
ronic acid) and iodobenzene. The stability of Pd particles is
controlled by the type of the stabilizers, the bases, and the
solvents, as well as the reactants used in the reaction system.
G4 dendrimer is an effective stabilizer; however, the strong
encapsulation of Pd particles within the dendrimers results in a
loss of catalytic activity.
In the present study, the results of the Pd-nanoparticle-
catalyzed Suzuki reactions between arylboronic acids and aryl
bromides in aqueous medium are unsatisfactory, although it has
been reported11a that Pd clusters in nanometer scale operate as
an efficient catalyst in some Suzuki reactions between arylbo-
ronic acids and activated aryl bromides in organic solvent
system. A general limitation is the fact that nonactivated
bromoarenes do not react well.11a-c,11f-g Indeed, a general
solution to this problem remains a challenge in organic
chemistry. By carefully choosing stabilizers, bases, solvents,
reactants, and reaction conditions, it is possible that both the
catalytic activity and stability of Pd nanoparticles are obtained
for special reactions.
It is found that the reactions between arylboronic acids and
aryl bromides are much slower than the reactions between
arylboronic acids and aryl iodides. In addition, byproducts
resulting from the homo-coupling of aryl bromides are obtained
here. It can be seen from Table 4 that by using either block
copolymer-Pd or PVP-Pd nanoparticles as the catalyst, the
homo-coupling products from bromoarenes are always dominant
and the yield is low. This might be attributed to the very slow
rate-determining step during the Suzuki reaction between
arylboronic acid and aryl bromide, where the homo-coupling
reaction becomes competitive. It might be helpful to organic
chemists who are performing Suzuki reactions. When byprod-
ucts are obtained, sometimes by using bromoarenes, and are
hard to separate from the desired Suzuki products, iodoarenes
might be an alternative to obtain the desired pure products. By
using block copolymer-Pd nanoparticles as a catalyst, no
precipitation of Pd particles is observed. However, by using
Acknowledgment. We thank the support of the National
Science Foundation (Grant No. CHE-9727633) and the Georgia
Tech Microscopy Center for providing the research facilities
used in the characterization of these samples. Y.L. wishes to
gratefully acknowledge partial support of this project by the
Georgia Institute of Technology Molecular Design Institute,

Pd Colloidal Nanoparticles in the Suzuki Reactions
J. Phys. Chem. B, Vol. 105, No. 37, 2001 8943
(c) Chechik, V.; Zhao, M.; Crooks, R. M. J. Am. Chem. Soc. 1999, 121,
4910. (d) Zhao, M.; Crooks, R. M. Chem. Mater. 1999, 11, 3379. (e)
Chechik, V.; Crooks, R. M. J. Am. Chem. Soc. 2000, 122, 1243. (f) Crooks,
R. M.; Zhao, M.; Sun, L.; Chechik, V.; Yeung, L. K. Acc. Chem. Res.
2001, 34(3), 181.
under Prime Contract No. N00014-95-1-116 from the Office
of Naval Research.
References and Notes
(9) Mayer, A. B. R.; Mark, J. E. Colloid Polym. Sci. 1997, 275(4),
333.
(1) (a) Wilkinson, G., Ed. ComprehensiVe Organometallic Chemistry;
Pergamon Press: New York, 1982. (b) Tsuji, J., Ed. Palladium Reagents
and Catalysts; Wiley: New York, 1995. (c) Malleron, J. L.; Fiaud, J. C.;
Legros, J. Y., Ed. Handbook of Palladium-Catalysed Organic Reactions;
Academic Press: London, 1997.
(2) (a) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457. (b) Suzuki,
A. J. Organomet. Chem. 1999, 576, 147. (c) Wolfe, J. P.; Singer, R. A.;
Yang, B. H.; Buchwald, S. L. J. Am. Chem. Soc. 1999, 121, 9550.
(3) (a) Heck, R. F. Org. React. 1982, 27, 345. (b) Heck, R. F. In
ComprehensiVe Organic Synthesis; Pergamon Press: Oxford, 1991. (c) De
Meijere, A.; Meyer, F. E. Angew. Chem., Int. Ed. Engl. 1994, 33, 2379. (d)
Negishi, E.; Coperet, C.; Ma, S.; Liou, S. Y.; Liu, F. Chem. ReV. 1996, 96,
5.
(4) Suzuki, A. In Metal-Catalyzed Cross-Coupling Reactions; Dieder-
ich, F., Stang, P. J., Eds.; VCH: Weinheim, 1998; pp 49.
(5) (a) Herrmann, W. A.; Brossmer, C.; Oefele, K. Angew. Chem., Int.
Ed. Engl. 1995, 34, 1844. (b) Beller, M.; Fischer, H.; Herrmann, W. A.;
Oefele, K.; Brossmer, C. Angew. Chem., Int. Ed. Engl. 1995, 34, 1848. (c)
Bei, X.; Turner, H. W.; Weinberg, W. H.; G uram, A. S. J. Org. Chem.
1999, 64, 6797.
(6) (a) Casalnuovo, A. L.; Calabrese, J. C. J. Am. Chem. Soc. 1990,
112, 4324. (b) Genet, J. P.; Linquist, A.; Blart, E.; Mouries, V.; Savignac,
M. Tetrahedron Lett. 1995, 36, 1443. (c) Genet, J. P.; Savignac, M. J.
Organomet. Chem. 1999, 576, 305.
(10) (a) Sulman, E.; Bodrova, Yu.; Matveeva, V.; Semagina, N.;
Cerveny, L.; Kurtc, V.; Bronstein, L.; Platonova, O.; Valetsky, P. Appl.
Catal. A 1999, 176, 75. (b) Sulman, E.; Matveeva, V.; Usanov, A.;
Kosivtsov, Yu.; Demidenko, G.; Bronstein, L.; Chernyshov, D.; Valetsky,
P. J. Mol. Catal. A 1999, 146, 265.
(11) (a) Reetz, M. T.; Breinbauer, R.; Wanninger, K. Tetrahedron Lett.
1996, 37, 4499. (b) Reetz, M. T.; Lohmer, G. Chem. Commun. 1996, 1921.
(c) Beller, M.; Fischer, H.; Kuehlein, K.; Reisinger, C.-P.; Herrmann, W.
A. J. Organomet. Chem. 1996, 520, 257. (d) Klingelhoefer, S.; Heitz, W.;
Greiner, A.; Oestreich, S.; Foerster, S.; Antonietti, M. J. Am. Chem. Soc.
1997, 119, 10 116. (e) Le Bars, J.; Specht, U.; Bradley, J. S.; Blackmond,
D. G. Langmuir 1999, 15, 7621. (f) Ding, J. H.; Gin, D. L. Chem. Mater.
2000, 12, 22. (g) Yeung, L. K.; Crooks, R. M. Nano Lett. 2001, 1(1), 14.
(12) Li, Y.; Hong, X. M.; Collard, D. M.; El-Sayed, M. A. Org. Lett.
2000, 2(15), 2385.
(13) Teranishi, T.; Miyake, M. Chem. Mater. 1998, 10, 594.
(14) Ottaviani, M. F.; Montalti, F.; Turro, N. J.; Tomalia, D. A. J. Phys.
Chem. B 1997, 101, 158.
(15) Technical data from Dendritech, Inc. (Midland, MI).
(16) (a) Esumi, K.; Suzuki, A.; Aihara, N.; Usui, K.; Torigoe, K.
Langmuir 1998, 14, 3157. (b) Garcia, M. E.; Baker, L. A.; Crooks, R. M.
Anal. Chem. 1999, 71, 256. (c) Gro¨hn, F.; Bauer, B. J.; Akpalu, Y. A.;
Jackson, C. L.; Amis, E. J. Macromolecules 2000, 33, 6042. (d) Esumi,
K.; Nakamura, R.; Suzuki, A.; Torigoe, K. Langmuir 2000, 16, 7842.
(7) Bradley, J. S. In Clusters and Colloids: From Theory to Applica-
tions; Schmid, G., Ed.; VCH: Weinheim, 1994; pp523.
(8) (a) Zhao, M.; Crooks, R. M. Angew. Chem., Int. Ed. Engl. 1999,
38, 8(3), 364. (b) Zhao, M.; C rooks, R. M. AdV. Mater. 1999, 11(3), 217.
(17) Wells, M.; Crooks, R. M. J. Am. Chem. Soc. 1996, 118, 3988.