Palladium nanoparticles supported on polyvinylpyridine: Catalytic activity in Heck-type reactions and XPS structural studies

Article abstract of DOI:10.1016/j.jcat.2009.01.005

Palladium nanoparticles, obtained by metal vapour synthesis (MVS), were deposited on cross-linked polyvinylpyridine. The Pd/PVPy system showed high catalytic activity in the Heck C{single bond}C coupling reaction of iodo- and bromo-arenes (iodobenzene, bromobenzene, p-nitrobromobenzene, p-bromoacetophenone, p-(methoxy)bromobenzene) with alkyl acrylates (methyl acrylate, n-butyl acrylate, ethylhexyl trans-3-(4-methoxyphenyl)acrylate) at 100 °C-175 °C working under nitrogen atmosphere as well as in air. The catalyst is stable and the leaching of metal in solution is very low. When reused, the recovered Pd/PVPy maintains the catalytic activity of the pristine material. XPS structural studies performed on the starting catalyst as well as on the recovered one indicate the presence of a interaction between the basic nitrogen of the pyridine present in the polymer and the metal.

Full text of DOI:10.1016/j.jcat.2009.01.005

Journal of Catalysis 262 (2009) 287–293
Contents lists available at ScienceDirect
Journal of Catalysis
www.elsevier.com/locate/jcat
Palladium nanoparticles supported on polyvinylpyridine:
Catalytic activity in Heck-type reactions and XPS structural studies
Claudio Evangelisti a, Nicoletta Panziera a, Paolo Pertici b,∗, Giovanni Vitulli c, Piero Salvadori a,b, Chiara Battocchio d,
Giovanni Polzonetti d
a
Dipartimento di Chimica e Chimica Industriale, Università di Pisa, via Risorgimento 35, 56126 Pisa, Italy
Istituto per la Chimica dei Composti Organometallici, ICCOM-CNR, Sezione di Pisa, via Risorgimento 35, 56126 Pisa, Italy
Advanced Catalysts s.r.l., via Risorgimento 35, 56126 Pisa, Italy
Dipartimento di Fisica, Unità INSTM e CISDiC, Università Roma Tre, via della Vasca Navale 84, 00146 Roma, Italy
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Palladium nanoparticles, obtained by metal vapour synthesis (MVS), were deposited on cross-linked
polyvinylpyridine. The Pd/PVPy system showed high catalytic activity in the Heck C–C coupling reaction
of iodo- and bromo-arenes (iodobenzene, bromobenzene, p-nitrobromobenzene, p-bromoacetophenone,
p-(methoxy)bromobenzene) with alkyl acrylates (methyl acrylate, n-butyl acrylate, ethylhexyl trans-3-
(4-methoxyphenyl)acrylate) at 100 C–175 C working under nitrogen atmosphere as well as in air. The
catalyst is stable and the leaching of metal in solution is very low. When reused, the recovered Pd/PVPy
maintains the catalytic activity of the pristine material. XPS structural studies performed on the starting
catalyst as well as on the recovered one indicate the presence of a interaction between the basic nitrogen
of the pyridine present in the polymer and the metal.
Received 5 September 2008
Revised 29 December 2008
Accepted 5 January 2009
Available online 3 February 2009
◦
◦
Keywords:
Heck reaction
Palladium
Polyvinylpyridine
Metal vapour synthesis
X-ray photoelectron spectroscopy
© 2009 Elsevier Inc. All rights reserved.
1. Introduction
olite, silica, polymers, etc.) [6]. Unfortunately, in many cases, the
supported systems leach palladium under the reaction conditions
leading to highly active soluble palladium species. Only a few ex-
The palladium-catalyzed coupling of olefins with aryl or vinyl
halides, known as the Heck reaction [1], is one of the most im-
portant reactions for the formation of C–C bonds in organic syn-
thesis, from both the research and industrial points of view [2,3].
It enables the functionalization of sp2 olefinic carbon atoms with
a wide range of aryl and vinyl substrates in a single step under
mild conditions. Homogeneous catalytic systems for the Heck reac-
tion are generally palladium salts or organo–metallic complexes in
the presence of additional ligands, such as phosphines, which pre-
vent the formation of catalytically inactive “palladium black” [2,4].
However, the homogeneous systems present difficulties associated
with the limited reusability and the removal of homogeneous ma-
terials from the reaction mixture [5]. For these reasons, there has
recently been increasing interest, from both economic and envi-
ronmental points of view, in the development of heterogeneous
phosphine-free palladium catalysts which can be recovered from
the reaction mixture by simple filtration and subsequently reused.
Lately, many solid systems containing supported palladium have
been studied with emphasis on the role of the procedure for the
preparation of the catalyst and the kind of support (carbon, ze-
amples of leach-free heterogeneous palladium systems have been
reported; they include palladium salts on zeolites [7], whose be-
haviour depends on the catalyst pre-treatment and on the base
and solvent used, and palladium nanoparticles deposited on a par-
ticular layered double hydroxide [8] or on suitable functionalized
zeolites containing primary amino [9] and SH [10] groups.
Recently [11] we reported the preparation of palladium cata-
lysts by deposition of palladium nanoparticles, prepared by metal
vapour synthesis (MVS) [12], on a commercial polyvinylpyridine
crosslinked with divinylbenzene (PVPy). This system showed ap-
preciable catalytic activity in the Heck reaction using aryl iodide
and methyl acrylate as model substrates with an important con-
tribution to the overall catalytic activity from supported palladium
species. In this context it was emphasized that PVPy, thanks to its
affinity with palladium, could be a very interesting catalytic sup-
port to obtain leach-free catalysts in Heck reaction. In extension of
these interesting preliminary results, we report here an improved
method to prepare palladium on PVPy by MVS and a more com-
prehensive study of its catalytic activity in the Heck reaction with a
wide range of halide substrates. Moreover, in order to better under-
stand the nature of the interaction between palladium and PVPy,
a study of the structural features performed by X-Ray Photoelec-
tron Spectroscopy (XPS) is reported.
Corresponding author. Fax. +39 050 2219260.
*
E-mail address: pertici@dcci.unipi.it (P. Pertici).
0021-9517/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2009.01.005

288
C. Evangelisti et al. / Journal of Catalysis 262 (2009) 287–293
2. Experimental
2.3. Catalytic reactions
2.1. Materials and apparatus
2.3.1. Heck reaction between phenyl halide derivatives and alkyl
acrylates to substituted trans-cinnamates: general procedure
1-Methyl-2-pyrrolidinone (NMP) (9 ml), alkyl acrylate (10
All operations involving the MVS products were performed un-
der a dry argon atmosphere. The co-condensation of palladium
and the appropriate solvent was carried out in a static reactor
as previously described [11,12]. The solvated Pd atom solutions
were worked up under argon with the use of standard Schlenk
techniques. The amount of palladium in the solutions was de-
termined by atomic absorption spectrometry (AAS) in an electro-
chemically heated graphite furnace with a Perkin–Elmer 4100ZL
instrument. The limit of detection (lod) calculated for palladium
was 2 ppb. Solvents were purified by conventional methods, dis-
tilled and stored under argon. Aryl halides, alkyl acrylates, K2CO3,
CsCO3 were used as received from Aldrich. Triethylamine and
tri-n-propylamine (from Aldrich) were distilled and stored over
KOH pellets before use. The GLC analyses were performed on a
Perkin–Elmer Auto System gas chromatograph, equipped with a
flame ionization detector (FID), using a SiO2 column (BP-1, 12 m
× 0.3 mm, 0.25 μm) and helium as carrier gas. 1H- and 13C-
NMR spectra were measured on Varian Gemini 200 spectrometer
at 200 and 50.3 MHz, respectively, using chloroform-d as sol-
vent; chemical shifts are relative to internal Si(CH3)4. XPS was
performed on an instrument of our own design and construc-
tion, consisting of a preparation and an analysis UHV chamber,
equipped with a 150 mm mean radius hemispherical electron anal-
yser with a four-element lens system with a 16-channel detector
giving a total instrumental resolution of 1.0 eV as measured at the
Ag 3d5/2 core level. MgKα non-monochromatised X-ray radiation
(hν = 1253.6 eV) was used for acquiring core level spectra of all
samples (C 1s, Pd 3d, O 1s and N 1s). The energies were refer-
enced to the C 1s signal of the aromatic C atoms having a binding
energy BE = 284.70 eV. Atomic ratios were calculated from peak
intensities by using Scofield’s cross section values and calculated λ
factors [13]. Curve-fitting analysis of the C 1s, Pd 3d, O 1s and N
1s spectra was performed using Voigt profiles as fitting functions,
after subtraction of a Shirley-type background [14].
mmol), phenyl halide derivatives (5 mmol), base (5 mmol) and
−3
53 mg of Pd/PVPy catalyst (containing 5 × 10
mg atom of Pd)
were introduced under argon atmosphere into a 25-ml round-
bottomed, two-necked flask equipped with a stirring magnetic bar,
a bubble condenser, and a silicon stopper. The reaction mixture
was magnetically stirred at the required temperature (see Table 1,
Section 3.2). When needed, small samples of the reaction mix-
ture were taken from the stoppered side neck. For GC analysis the
samples were treated with either water or 0.5 M aqueous HCl.
The organic products were then extracted with diethyl ether, dried
over anhydrous sodium sulfate and analyzed by GLC. The coupling
product was purified by column chromatography (silica, solvent:
n-hexane/ethylacetate). The reaction was interrupted at the time
reported in Table 1 (Section 3.2).
Trans-4-nitrocinnamic acid n-butyl ester:
1H NMR (CDCl3, 200 MHz, ppm): δ 8.18 (d, 2H, J = 8.8 Hz), 7.65
(d, 1H, J = 16.4 Hz), 7.65 (d, 2H, J = 8.8 Hz), 6.50 (d, 1H, J =
16.0 Hz), 4.18 (t, 2H, J = 6.7 Hz), 1.65 (quint, 2H, J = 5.4 Hz),
1.42 (sextet, 2H, J = 7.5 Hz), 0.92 (q, 3H, J = 7.3 Hz); 13C NMR
(CDCl3, 50 MHz, ppm): δ 165.9, 148.4, 141.4, 140.5, 128.5, 124.0,
+
122.5, 64.8, 30.6, 19.1, 13.6; GC-MS (EI) m/z (%): 249 (M , 100).
Trans-4-methoxylcinnamic acid n-butyl ester:
1H NMR (CDCl3, 200 MHz, ppm): δ 7.61 (d, 1H, J = 16.0 Hz), 7.45
(d, 2H, J = 8.7 Hz), 6.87 (d, 2H, J = 8.8 Hz), 6.28 (d, 1H, J =
15.9 Hz), 4.18 (t, 2H, J = 6.8 Hz), 3.80 (s, 3H), 1.66 (quint, 2H,
J = 5.8 Hz), 1.41 (sextet, 2H, J = 7.6 Hz), 0.94 (t, 3H, J = 7.3 Hz);
13C NMR (CDCl3, 50 MHz, ppm): δ 167.3, 161.3, 144.1, 129.6, 127.2,
115.8, 114.3, 64.2, 55.3, 30.8, 19.2, 13.7; GC-MS (EI) m/z (%): 234
2.2. Preparation of palladium catalysts
2.2.1. Preparation of the solvated Pd atoms
−4
In a typical experiment, Pd vapour, generated at 10
Bar by
resistive heating of the metal (500 mg) in an alumina-coated tung-
sten crucible, was co-condensed at liquid nitrogen temperature
with a 1:1 mixture of mesitylene (30 ml) and 1-hexene (30 ml)
in a glass reactor described elsewhere [11,12]. The reactor cham-
+
(M , 100).
Trans-cinnamic acid n-butyl ester:
◦
ber was heated to the melting point of the solid matrix (−40 C),
and the resulting red-brown solution was siphoned and handled at
◦
low temperature (−30/−40 C) with the Schlenk tube technique.
For AAS, the metal-containing mesitylene/1-hexene solution (1 ml)
was heated over a heating plate in a porcelain crucible, in the pres-
ence of aqua regia (2 ml), four times and the solid residue was
dissolved in 0.5 M aqueous HCl. The palladium content of the sol-
vated metal solution was 2.8 mg/ml.
1H NMR (CDCl3, 200 MHz, ppm): δ 7.67 (d, 1H, J = 16.0 Hz), 7.49–
7.52 (m, 2H), 7.34–7.37 (m, 3H), 6.42 (d, 1H, J = 16.0 Hz), 4.20
(t, 2H, J = 6.7 Hz), 1.69 (quint, 2H, J = 7.2 Hz), 1.44 (sextet, 2H,
J = 7.5 Hz), 0.95 (t, 3H, J = 7.4 Hz); 13C NMR (CDCl3, 50 MHz,
ppm): δ 167.0, 144.5, 144.5, 134.5, 130.1, 128.8, 128.0, 118.3, 64.3,
2.2.2. Preparation of palladium on polyvinylpyridine, Pd/PVPy
(1 wt% Pd)
+
30.8, 19.2, 13.7; GC-MS (EI) m/z (%): 204 (M , 100).
Trans-methyl cinnamate:
The mesitylene-1-hexene Pd atom solution (36 ml, 101 mg Pd)
was added to a suspension of PVPy (10 g) in mesitylene (40 ml).
The mixture was stirred for 12 h at room temperature. The colour-
less solution was removed and the light-brown solid was washed
with pentane and dried under reduced pressure. The metal con-
tent of the Pd on PVPy catalyst (1 wt% Pd) was determined by AAS
analysis, as reported in Section 2.2.1.
1H NMR (CDCl3, 200 MHz, ppm): δ 7.64(d, 1H, J = 16.0 Hz), 7.51–
7.54 (m, 2H), 7.33–7.39 (m, 3H), 6.45 (d, 1H, J = 16.0 Hz), 3.78 (s,

C. Evangelisti et al. / Journal of Catalysis 262 (2009) 287–293
289
Scheme 1. Preparation of the Pd/PVPy catalyst by metal vapour synthesis technique.
3H); 13C NMR (CDCl3, 50 MHz, ppm): δ 167.0, 144.2, 144.5, 135.2,
128.9, 127.2, 118.3, 52.0; GC-MS (EI) m/z (%): 162 (M , 100).
mesitylene on the cooled wall of the reactor. On warming until
+
◦
◦
the solid matrix melts (−30 C to −40 C), a brown solution, the
so-called solvated palladium atoms solution, is obtained. It con-
tains palladium atoms and/or palladium nanoclusters, which are
solubilized and stabilized by the excess of the organic solvents act-
ing as weakly coordinating ligands. It behaves as a source of metal
nanoparticles. On warming the solution to room temperature in
the presence of PVPy, the nanoclusters agglomerate with subse-
quent deposition of palladium nanoparticles on PVPy (Scheme 1).
The mesitylene-1-hexene solvated Pd atom solution is stable
Trans-methyl 3-(4-acetylphenyl)acrylate:
1H NMR (CDCl3, 200 MHz, ppm): δ 7.88 (d, 2H, J = 8.3 Hz), 7.53
(d, 2H, J = 8.3 Hz), 7.44 (d, 1H, J = 16.2 Hz), 6.69 (d, 1H, J =
16.2 Hz), 2.52 (s, 3H), 2.31 (s, 3H); 13C NMR (CDCl3, 50 MHz, ppm):
δ 197.9, 197.1, 141.5, 138.6, 137.9, 128.9, 128.7, 128.2, 27.6, 26.6; GC-
◦
up to 0 C. It is more stable than the mesitylene solvated Pd
atom solution that has been employed previously in the prepa-
ration of the palladium on polyvinylpyridine catalyst, which de-
◦
composes at −30 C [11]. The high stability of the first solution
+
MS (EI) m/z (%) 188 (M , 100).
allows a more controlled deposition of the metal particles on the
PVPy.
Trans-4-nitrocinnamic acid methyl ester:
3.2. Palladium on polyvinylpyridine, Pd/PVPy, as a catalyst in the Heck
reaction of bromoarene and chloroarene derivatives with acrylates
1H NMR (CDCl3, 200 MHz, ppm): δ 8.21 (d, 2H, J = 8.8 Hz), 7.77
(d, 1H, J = 16.3 Hz), 7.65 (d, 2H, J = 8.8 Hz), 6.50 (d, 1H, J =
As previously reported [11] the Pd/PVPy catalyst, prepared from
the mesitylene solvated Pd atom solution, showed good catalytic
activity in the C–C coupling reaction of iodobenzene and methyl
acrylate, used as the standard reaction, giving selectively trans-
16.0 Hz), 3.85 (s, 3H); 13C NMR (CDCl3, 50 MHz, ppm): δ 166.6,
+
148.9, 141.3, 129.3, 124.7, 123.2, 51.5; GC-MS (EI) m/z (%): 207 (M
,
100).
◦
methyl cinnamate in 60% yield after 4 h at 75 C. The performance
ꢀ
Trans-4-methoxylcinnamic acid 2 -ethyl hexyl ester:
of the catalytic system Pd/PVPy, prepared following the improved
procedure as reported in Section 3.1, has been compared with that
discussed above and it has been also examined in the C–C cou-
pling reaction of both activated and deactivated bromoarene and
chloroarene derivatives with alkyl acrylates (Scheme 2).
The cross-coupling reactions have been performed at differ-
ent temperatures (100–175 C) in N-methylpyrrolidinone (NMP)
1H NMR (CDCl3, 200 MHz, ppm): δ 7.61 (d, 1H, J = 16.0 Hz), 7.45
(d, 2H, J = 8.7 Hz), 6.87 (d, 2H, J = 8.7 Hz), 6.30 (d, 1H, J =
15.9 Hz), 4.15 (m, 2H), 3.82 (s, 3H), 1.30–1.59 (m, 9H), 0.92 (m,
6H); 13C NMR (CDCl3, 50 MHz, ppm): δ 167.9, 161.5, 144.3, 130.0,
127.4, 116.1, 114.5, 67.2, 55.7, 39.3, 30.8, 29.4, 24.3, 23.4, 14.5, 11.3;
◦
as solvent using a molar ratio aryl halide/Pd = 100 or 1000
and an excess of alkyl acrylate with respect to aryl halide (mo-
lar ratio = 2) to avoid the formation of twofold coupling by-
products [2a]. The alkyl acrylates employed were methyl acry-
late, n-butyl acrylate and ethylhexyl acrylate. Different bases (Et3N,
nPr3N, K2CO3, CsCO3) were used; the base reacts with the hydro-
halic acid, formed in the course of the reaction, thus shifting the
equilibrium toward the products. The reaction mixture was eas-
ily separated from the solid catalyst by filtration and the organic
products were extracted with diethyl ether. Pure coupling prod-
ucts were obtained by column chromatography and characterized
by NMR spectroscopy and mass spectrometry. These compounds
were selectively obtained with trans stereochemistry. The results
are reported in Table 1.
+
GC-MS (EI) m/z (%): 290 (M , 100).
3. Results and discussion
3.1. Preparation of the polyvinylpyridine supported palladium catalyst,
Pd/PVPy
The polyvinylpyridine supported palladium catalyst, Pd/PVPy,
has been prepared by depositing Pd nanoparticles, obtained by
the metal vapour synthesis technique [12,15], on PVPy. Using the
MVS apparatus, palladium atoms, produced by resistive heating of
the metal under high vacuum, are co-condensed at liquid nitro-
The Pd/PVPy catalyst, prepared from a Pd/1-hexene-mesitylene
solution, showed high catalytic activity in the C–C coupling re-
◦
gen temperature (−196 C) with a 1:1 mixture of 1-hexene and

290
C. Evangelisti et al. / Journal of Catalysis 262 (2009) 287–293
Scheme 2. Heck reaction of aryl halide derivatives with alkyl acrylates.
Table 1
reaction conditions. The strongly electron-withdrawing nitro group
in para position to bromine should weaken the C–Br bond, so p-
nitrobromobenzene should, in principle, react more rapidly than
bromobenzene or other mono substituted bromoarenes. The reac-
Reaction between aryl halide derivatives and alkyl acrylates giving substituted trans-
cinnamates catalyzed by palladium on polyvinylpyridine, Pd/PVPy.a
Run
1
2
T
( C)
Time
(h)
Yield of
product 3
(%)b
Specific
activity
◦
◦
tion of p-nitrobromobenzene with methyl acrylate, at 125 C in the
−1
c
(h
)
presence of nPr3N, gives selectively methyl trans-4-nitrocinnamate,
after 6 h in 95% yield (SA = 158 h−1, run 2). Inorganic bases, such
as K2CO3 or Cs2CO3, which are often employed in the Heck reac-
tion, furnish lower yields of the C–C coupling product (47% after
6 h, TOF = 78 h−1, run 3; and 65% after 6 h, SA = 108 h−1, run 4,
1d
2
R = H, X = I
R
= CH3
= CH3
100
125
3
93 (90)
310
158
78
ꢀ
ꢀ
R = NO2
R
3
6
55
95 (93)
X = Br
3e
4f
R = NO2
X = Br
R
R
R
R
R
R
R
R
R
R
R
R
= CH3
= CH3
= CH3
125
125
150
150
125
150
150
175
175
3
6
25
47
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
◦
respectively). With nPr3N at 150 C the yield becomes 82% after 4 h
(SA = 205 h−1, run 5). At this temperature, using the high boiling
n-butyl acrylate, the yield is 90% after 4 h (SA = 225, entry 6) and
it is quantitative after ca. 6 h.
R = NO2
3
6
30
65
X = Br
108
205
5
R = NO2
X = Br
4
82
p-Bromoacetophenone is slightly less reactive than p-nitrobro-
◦
mobenzene. At 125 C its reaction with methyl acrylate in the pres-
ence of nPr3N gives trans-methyl 3-(4-acetylphenyl)acrylate in 65%
6
R = NO2
=
nBu
4
90 (87)
225
◦
yield after 8 h (SA = 82 h−1, run 7). At 150 C, using n-butyl acry-
X = Br
late, the reaction furnishes trans-butyl 3-(4-acetylphenyl)acrylate
ꢀ
7
R
= COMe
= CH3
4
8
36
65 (62)
in good yield (75%, SA = 188 h−1, run 8).
X = Br
82
Bromobenzene is less reactive than either p-nitrobromobenzene
or p-bromoacetophenone, the Heck reaction with n-butyl acry-
ꢀ
8
R
= COMe
=
=
=
=
nBu
nBu
nBu
nBu
4
75 (72)
10
188
X = Br
◦
late at 150 C giving n-butyl trans-cinnamate in low yield (10%
−1
9
R = H
4
25
38
after
4
h, SA = 25
h
, run 9). Raising the temperature to
X = Br
◦
175 C gives n-butyl trans-cinnamate in good yield after 24 h
−1
10
11g
12g
13g
R = H
4
24
17
90 (85)
(90%, SA = 38
h
, run 10). p-(Methoxy)bromobenzene was
X = Br
the least reactive of the bromo derivatives examined. The reac-
tion with n-butyl acrylate was performed using a molar ratio
R = OCH3
4
24
7
◦
X = Br
34 (30)
2
2
p-(methoxy)bromobenzene/Pd = 100. At 175 C, n-butyl trans-4-
methoxycinnamate was obtained in low yield (7%) after 4 h, and
R = OCH3
X = Br
= CH2CH(Et)(nBu) 175
24
24
24
37 (34)
20
in 34% yield after 24 h (SA = 2 h−1, run 11).
The cross-coupling reaction of p-(methoxy)bromobenzene with
2-ethylhexyl acrylate was also examined, because the product,
ethylhexyl trans-3-(4-methoxyphenyl)acrylate, is used as a sun-
R = NO2
=
=
nBu
nBu
175
175
–
–
X = Cl
14g
R = H
X = Cl
15
◦
screen agent in the cosmetic industry [16]. At 175 C, after 24 h,
ethylhexyl trans-3-(4-methoxyphenyl) acrylate was obtained in 37%
a
yield (SA = 2 h−1, run 12).
Reaction conditions: aryl halide (5 mmol); acrylate (10 mmol); base (5 mmol),
N(nPr)3 (unless otherwise stated); 1-methyl-2-pyrrolidinone, NMP (9 ml); catalyst
As far as the Heck reaction with chloroarene derivatives is con-
cerned, p-nitrochlorobenzene and chlorobenzene were examined
using a molar ratio substrate/Pd = 100. These compounds react
−3
53 mg (5 × 10
mg-atom Pd) (unless otherwise stated).
b
Yield of the substituted trans-cinnamate determined by GLC. The yield of the
isolated products are in parenthesis.
◦
c
with n-butyl acrylate at 175 C furnishing, after 24 h, respectively,
Calculated as moles of aryl halide converted/g-atom of metal per hour.
d
Base = Et3N.
n-butyl trans-4-nitrocinnamate and n-butyl trans-cinnamate in low
yield (20%, run 13; and 15%, run 14) and low specific activity. The
observed reactivity of the halo compounds, aryl iodide > aryl bro-
mide > aryl chloride is that expected considering the stability of
the carbon (sp2)-halogen bond and it is in agreement with the data
reported in the literature [17].
It is now well known that one of the principal disadvantages
associated with the use of Pd supported catalysts in the Heck re-
action is leaching of the metal into the solution, resulting in con-
tamination of the product with metal that is difficult to remove.
e
Base = K2CO3.
f
Base = Cs2CO3.
g
−2
mg-atom Pd.
Catalyst 5 × 10
action of iodobenzene with methyl acrylate to methyl trans-
cinnamate at 100 C in the presence of Et3N (SA = 310 h−1, run 1).
◦
It was more active than the analogous catalyst prepared from the
mesitylene-solvated Pd atom solution (SA = 168 h−1) [11].
For the reactions with bromoarene derivatives, p-nitrobromo-
benzene was chosen as the reference compound to establish the

C. Evangelisti et al. / Journal of Catalysis 262 (2009) 287–293
291
Table 2
XPS data for pure PVPy, pristine Pd-containing catalyst and recovered catalyst.
Sample
PVPy
Signal
C 1s
BE (eV)
FWHM (eV)
Ni(tot)/NN(tot)
284.70
286.20
288.51
399.26
401.61
2.08
2.08
2.08
1.77
1.77
34.00
N 1s
1.00
Na/Nb = 9.2
Sample
Signal
Starting catalyst
Recovered catalyst
BE (eV)
FWHM (eV)
Ni(tot)/NN(tot)
BE (eV)
FWHM (eV)
Ni(tot)/NN(tot)
Pd/PVPy
1%
C 1s
284.70
286.36
288.44
2.27
2.27
2.27
30.27
284.70
286.01
288.28
2.16
2.16
2.16
25.7
N 1s
O 1s
399.07
401.97
2.08
2.08
1.00
Na/Nb = 1.2
399.28
401.75
2.08
2.08
1.00
Na/Nb = 3.5
531.35
532.70
534.20
2.66
2.66
2.66
10.08
530.26
532.42
534.40
2.28
2.28
2.28
21.74
Pd 3d5/2
C 1s
335.27
337.21
1.90
1.90
1.56
Pd(0)/Pd(II)
7.2
335.36
337.19
2.06
2.06
0.29
Pd(0)/Pd(II)
6.3
Pd/PVP
5%
284.70
286.43
288.67
2.06
2.06
2.06
30.50
284.70
285.90
288.10
2.09
2.09
2.09
33.43
N 1s
O 1s
398.80
399.95
1.97
1.97
1.00
9.66
Na/Nb = 1.3
399.45
400.70
1.77
1.77
Na/Nb = 3.3
1.00
531.31
532.35
534.04
2.50
2.50
2.50
530.63
532.42
534.40
2.21
2.21
2.21
29.49
Pd 3d5/2
335.16
337.09
1.90
1.90
1.58
Pd(0)/Pd(II)
4.6
335.39
337.14
1.92
1.92
1.08
Pd(0)/Pd(II)
5.2
For these reasons leaching of the metal from Pd/PVPy catalyst has
been examined. The reaction between iodobenzene and methyl
acrylate and between p-nitrobromobenzene and n-butyl acrylate
(run 1 and run 6, respectively), which present high conversion and
different reaction temperature, have been examined in detail. In
appreciable decline of its catalytic activity indicating the high sta-
bility of Pd/PVPy system even in air.
4. Structural studies of the Pd/PVPy catalyst by X-ray
photoelectron spectroscopy (XPS)
◦
the case of run 1, performed at 100 C, the amount of Pd leached
To obtain information on the structural features of the Pd/PVPy
catalyst, X-ray photoelectron spectroscopy (XPS) studies have
been performed on the pristine catalyst as well as on the cata-
lyst recovered from the catalytic test. The system examined was
Pd/PVPy, containing 1 wt% Pd, employed in the reaction between
during the reaction (3 h) ranged between 0.5–0.9% of the total
available metal, corresponding to 0.4–0.6 ppm in solution; in the
case of run 6, performed at 150 C for 4 h, the Pd leached was
1.0–1.3%, corresponding to 0.6–0.7 ppm. Moreover, on the isolated
products, obtained by purification on silica column chromatogra-
phy, very low palladium amount (<0.5 ppm) was detected. It is
worth noting that these low Pd level satisfy specifications required
by the pharmaceutical industry regarding the final purity of the
products (Pd <2 ppm) [6a,18].
◦
◦
p-nitrobromobenzene and n-butyl acrylate at 150 C (run 6). The
Pd/PVPy catalyst, containing 5 wt% Pd, employed in the same re-
action, has been also studied to investigate the influence of the
amount of metal on the structure of the catalyst.
The C 1s, Pd 3d, N 1s and O 1s core level spectra have been
collected and analysed. The core level binding energy (BE) and full
width at half-maxima (FWHM) were analysed with particular at-
tention to the Pd 3d spin–orbit components, which are of major
interest for the assessment of the Pd–pyridine interaction in the
catalyst and for the investigation of its role in the catalytic pro-
cess. BE, FWHM and atomic ratio Ni(tot)/NN(tot) values observed
for both the pristine and the recovered catalysts are collected and
compared in Table 2, together with the XPS data collected on a
pure PVPy sample, which will be taken as reference in the follow-
ing discussion about the chemical structure of the organic moiety
of the composite Pd/PVPy catalysts.
All the XPS spectra were calibrated in energy using the main
component of the C 1s signal, attributed to aromatic carbons, at
284.70 eV. The N 1s spectra as detected for the pure PVPy, for
the pristine sample and after catalytic test, appeared quite sim-
ilar in structure due to the presence of multiple features. Two
components were evident from peak fitting analysis; the compo-
As far as the properties of the Pd/PVPy catalyst are concerned,
it is interesting to note that the Pd leaching is less than that
previously observed using the similar catalyst prepared from the
Pd/mesitylene solvated atom solution, even at lower reaction tem-
perature (1.3–1.9% at 75 C and 4.2–5.2% at 125 C) [11]. This
demonstrates the high stability of the Pd/PVPy catalyst obtained
by the modified procedure in which the Pd/mesitylene-1-hexene
solution is the source of metal nanoparticles.
The reactions between iodobenzene and methyl acrylate and
between p-nitrobromobenzene and n-butyl acrylate gave the same
results regarding conversion and Pd leaching in air as under argon
(run 1 and run 6), indicating that these processes do not require
the use of an inert atmosphere. For this reason, the experiments
on the recycle of the Pd/PVPy catalyst have been performed un-
der air. The catalysts in runs 1, 5, 6, 8, 11 and 12 were recovered
from the reaction mixture and reused in the same reactions show-
ing the same catalytic activity as that of the starting material. In
particular, the catalyst in run 1 was reused 5 times without an
◦
◦

292
C. Evangelisti et al. / Journal of Catalysis 262 (2009) 287–293
the dispersion process of Pd clusters in PVPy matrix. For compar-
ison, the Pd 3d5/2 signal for PdO is reported in the literature at
337.0 eV [21]. However, if PdO were present, we would expect also
to observe a O 1s signal at about 529.3 eV, which is not present.
On this basis, therefore, it is reasonable to exclude the presence of
palladium oxide. On the other hand, Pd 3d5/2 BE values associated
with Pd atoms covalently bonded to carbon were found at 338.2 eV
in Pd(II)-diethynylbenzene poly-ynes [23], and BE values of about
337.8 eV are reported in the literature for similar systems [18]. Al-
though this palladium species is present in only small amount, it
could play a role in the grafting of the palladium nanospheres. We
suggest that the low charge density Pd atoms detected in the com-
posite samples may interact with the nitrogen atoms of the PVPy
rings; in this picture, there may be an electrostatic interaction be-
tween the Pd atoms and the nitrogen atoms of the pyridine rings,
which will lead to an increase of the higher BE N 1s component
(Nb) ascribed to positively charged N atoms. In fact, as reported in
Table 2, the Nb component intensity is negligible in the pristine
PVPy sample (Na/Nb = 9.2) and grows in the composite Pd/PVPy
samples (Na/Nb ∼ 1.3). Semi-quantitative analysis of the XPS sig-
nals enabled the Pd(0)/Pd(II) atomic ratio to be estimated as 6.0:1
for the 1 wt% Pd and 5.0/1 for the 5 wt% Pd samples. The observed
discrepancy between the Pd(0)/Pd(II) atomic ratio values founded
for sample of different Pd% is, however, extremely small, suggest-
ing that the two samples are completely similar.
Both qualitative and semi-quantitative analysis performed on
the recovered catalysts are in excellent agreement with the data
already discussed for the pristine Pd/PVPy samples, as shown in
Table 2. Moreover, the Pd 3d spectrum of Pd/PVPy 1 wt% sample
as recovered from the catalytic test is displayed in Fig. 1, together
with the curve fitting results, and can be easily compared with the
same spectrum collected on the pristine catalyst, reported in the
same Fig. 1. The C 1s, N 1s, O 1s and Pd 3d spectra of the pristine
and the recovered samples do not show appreciable BE differences,
indicating that the chemical and electronic structure of the catalyst
is not modified by the catalytic process.
In conclusion, comparison of the XPS spectra of the pure PVPy
and Pd/PVPy samples show that the chemical structure of PVPy
is not appreciably affected by the presence of metallic Pd atoms.
Furthermore, from the comparison between pristine and recovered
catalysts, we assume that the catalytic process does not induce
permanent modification in the chemical and/or electronic struc-
ture of the catalyst; therefore, the material appears to be almost
completely and efficiently recovered unchanged after the catalytic
tests. Moreover, the, Pd(0)/Pd(δ+) atomic ratios as estimated by
the XPS semiquantitative analysis are 6.3/1 for the 1 wt% Pd and
5.2/1 for the 5 wt% Pd recovered samples. They show only a neg-
ligible variation and are consistent with the values calculated for
the pristine systems.
Fig. 1. XPS Pd 3d spectra of Pd/PVPy 1% both pristine and recovered from the cat-
alytic test.
nent at about 399.0 eV BE (referred as Na in Table 2) was as-
signed to the nitrogen atoms of the pyridine rings [19]; the peak
of lower intensity at 400.0–401.0 eV (Nb in Table 2), conversely,
was ascribed to positively charged N atoms. A similar structured
N 1s spectrum has already been observed for the polyvinylpyri-
dine sample itself, and is attributed in the literature to “strongly”
H-bonded pyridine (400.4–400.9 eV) [20] or protonated pyridine
(401.4–402.4 eV) [19,21]. As reported in Table 2, the C/NTOT atomic
ratios are in good agreement with the molecular stoichiometry ex-
pected and experimentally found for the pure PVPy, and are strictly
similar for the two composite Pd/PVPy samples (1 wt% and 5 wt%
Pd content), both pristine and after recovery from the catalytic test.
The Pd 3d spectrum shows the presence of a main contribu-
tion together with a minor component at higher BE values. Curve-
fitting analysis showed that the Pd 3d spectra of the pristine cat-
alysts resulted from two pairs of spin-orbit components, as shown
in Fig. 1 (sample Pd/PVPy 1 wt%). On the basis of previous mea-
surements, the Pd 3d5/2 peak found at BE = 335.5 eV has been
attributed to metallic palladium (Pd(0)) in agreement with litera-
ture reports [22]; the second Pd 3d5/2 signal occurring at higher
BE values (BE = 337.21 eV for sample 1 wt% Pd; BE = 337.09 eV
for sample 5 wt% Pd) is due to Pd atoms with lower charge den-
sity. The origin of such palladium species is not completely clear;
one possible hypothesis considers a perturbation occurring during
5. Conclusions
The Pd/PVPy system, prepared by the MVS technique, is an ef-
ficient and stable catalyst for the C–C coupling reaction of iodo-
and bromoarenes with alkyl acrylates. The corresponding trans-
cinnamates are obtained with yields in the range 35–95% at 100–
◦
175 C depending on the substituent present on the aromatic ring
of the aryl halide. The catalyst is particularly stable and the leach-
ing of palladium into the solution is very low. It works well also
in air, thus simplifying the work-up of the catalytic process. In ad-
dition, the catalyst can be recovered and reused without loss of
catalytic activity. XPS structural studies indicate the presence of
interactions between the basic nitrogen atoms of the pyridine moi-
eties present in the polymer and the metal. Such interactions are
present in the starting material as well as in the catalyst recovered

C. Evangelisti et al. / Journal of Catalysis 262 (2009) 287–293
293
after the catalytic reaction, thus accounting for the high stability
of this catalytic system.
(c) K. Okumura, K. Nota, K. Yoshida, M. Niwa, J. Catal. 231 (2005) 245;
(d) C.M. Crudden, M. Sateesh, R. Lewis, J. Am. Chem. Soc. 127 (2005) 10045;
(e) J.R. Ruiz, C. Jimenez-Sanchidrian, M. Mora, Tetrahedron 62 (2006) 2922;
(f) X. Jiang, J. Sclafani, K. Presad, O. Repic, T.J. Blacklock, Org. Process Dev. 11
(2007) 769.
These results suggest that the Pd/PVPy system may be more
generally useful, for example, in other Heck-type reactions of inter-
est to chemical industry. The very low palladium leaching observed
for Pd/PVPy is a particular advantage, in view of the high price of
the metal and because the products should be ready for use with-
out the need for cleaning with metal scavengers. It is also worth
highlighting the versatility of the MVS technique to prepare even
more stable and efficient catalysts.
[7] (a) M. Dams, L. Drijkoningen, B. Pauwels, G. Van Tendeloo, D.E. De Vos, P.A.
Jacobs, J. Catal. 209 (2002) 225;
(b) A. Corma, H. Garcia, A. Leyva, A. Primo, Appl. Catal. A Gen. 247 (2003) 41;
(c) A. Corma, H. Garcia, A. Leyva, A. Primo, Appl. Catal. A 257 (2004) 77.
[8] B.M. Choudary, S. Madhi, N.S. Chowdari, M.L. Kantam, B. Sreehdar, J. Am. Chem.
Soc. 124 (2002) 14127.
[9] S. Mandal, D. Roy, R.V. Chaudhari, M. Sastry, Chem. Mater. 16 (2004) 3714.
[10] K. Shimizu, S. Koizumi, T. Hatamachi, H. Yoshida, S. Komai, T. Kodama, Y. Ki-
tayama, J. Catal. 228 (2004) 141.
[11] A.M. Caporusso, P. Innocenti, L.A. Aronica, G. Vitulli, R. Gallina, A. Biffis,
M. Zecca, B. Corain, J. Catal. 234 (2005) 1.
Acknowledgment
The authors thank Dr. Emanuela Pitzalis (IPCF-CNR, Pisa, Italy)
for AAS analyses.
[12] (a) G. Vitulli, C. Evangelisti, A.M. Caporusso, P. Pertici, N. Panziera, S. Bertozzi,
P. Salvadori, in: B. Corani, G. Schmid, N. Toshima (Eds.), Metal Nanoclusters in
Catalysis and Materials Science: The Issue of Size-Control, Elsevier, Amsterdam,
2008, chap. 32;
(b) K.J. Klabunde, P. Timms, P.S. Skell, D. Ittel, Inorg. Synth. 19 (1979) 59.
[13] P. Swift, D. Shuttleworth, M.P. Seah, in: D. Briggs, M.P. Seah (Eds.), Practical
Surface Analysis by Auger and X-Ray Photoelectron Spectroscopy, Wiley, Chich-
ester, 1983, chap. 5, App. 3.
References
[1] (a) T. Mizoroki, K. Mori, A. Ozaki, Bull. Chem. Soc. Jpn. 44 (1971) 581;
(b) R.F. Heck, J.P. Nolley Jr., J. Org. Chem. 37 (1972) 2320.
[2] For recent reviews of Heck reaction, see: (a) S. Bräse, A. de Meijere, in: F. Die-
derich, P.J. Stang (Eds.), Metal-Catalyzed Cross-Coupling Reactions, Wiley–VCH,
Weinheim, 1999, chap. 3;
[14] D.A. Shirley, Phys. Rev. B 5 (1972) 4709.
[15] K.J. Klabunde, Free Atoms, Clusters and Nanoscale Particles, Academic Press,
San Diego, 1994.
(b) I.P. Beletskaya, A.V. Cheprakov, Chem. Rev. 100 (2000) 3009;
(c) A. Biffis, M. Zecca, M. Basato, J. Mol. Catal. A Chem. 173 (2001) 249;
(d) N.J. Whitcombe, K.K. Hii, S.E. Gibson, Tetrahedron 57 (2001) 7449;
(e) K.C. Nicolau, P.G. Bulger, D. Sarlah, Angew. Chem. Int. Ed. Engl. 44 (2005)
4442;
[16] J.G. De Vries, Can. J. Chem. 79 (2001) 1086.
[17] (a) R.F. Heck, Org. React. 27 (1982) 345;
(b) V.M. Wall, A. Eisenstadt, D.J. Ager, S.A. Laneman, Plat. Met. Rev. 43 (1999)
138.
[18] M. Prashad, Top. Catal. 6 (2004) 181.
(f) A.M. Trzeciak, J.J. Ziółkowski, Coord. Chem. Rev. 249 (2005) 2308;
(g) F. Alonso, I.P. Beletskaya, M. Yus, Tetrahedron 61 (2005) 11771.
[3] A. Zapf, M. Beller, Top. Catal. 19 (2002) 101.
[4] C. Amatore, A. Jutand, Acc. Chem. Res. 33 (2000) 314.
[5] C.E. Garrett, K. Prasad, Adv. Synth. Catal. 346 (2004) 889.
[6] See as example: (a) S. Klingelhöfer, W. Heitz, A. Greiner, S. Oestreich, S. Förster,
M. Antonietti, J. Am. Chem. Soc. 119 (1997) 10116;
[19] J. Lahaye, G. Nanse, Ph. Fioux, A. Bagreev, A. Broshnik, V. Stelko, Appl. Surf.
Sci. 147 (1999) 153.
[20] Y. Zubavichus, M. Zharnikov, Y. Yang, O. Fuchs, C. Heske, E. Umbach, A. Ulman,
M. Grunze, Langmuir 20 (2004) 11022.
[21] X. Zhou, S.H. Goh, S.Y. Lee, K.L. Tan, Appl. Surf. Sci. 126 (1998) 141.
[22] NIST X-ray Photoelectron Spectroscopy Database NIST Standard Reference
Database 20, Version 3.4.
(b) A.L. Dantas Ramos, P. da Silva Alves, D.A.G. Aranda, M. Schmal, Appl. Catal.
A 277 (2004) 71;
[23] I. Fratoddi, A. La Groia, C. Battocchio, M.V. Russo, J. Polym. Sci. Part. A Polym.
Chem. 45 (2007) 3311.