Heck reactions between aryl halides and olefins catalysed by Pd-complexes entrapped into zeolites NaY

Article abstract of DOI:10.1016/S0022-328X(99)00080-7

Palladium complexes entrapped into zeolite cages have been prepared and characterised by MAS-NMR. They exhibit a high activity towards the Heck reaction of aryl bromides with olefins for small palladium concentrations (0.1 mol%). The catalysts can be easily separated from the reaction mixture and reused after washing without loss in activity. Except for very large complexes, no limitation to the diffusion of educts in the zeolite cages was observed. As for the homogeneous Heck reaction, the electronic nature of the aryl bromides and the olefins has a dominating effect on the reaction yield and selectivity. The heterogeneous catalysts activate even aryl chlorides under standard reaction conditions.

Full text of DOI:10.1016/S0022-328X(99)00080-7

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 584 (1999) 16–26
Heck reactions between aryl halides and olefins catalysed by
Pd-complexes entrapped into zeolites NaY
Laurent Djakovitch *, Henrike Heise, Klaus K o¨ hler
Technische Uni6ersit a¨ t M u¨ nchen, Anorganisch-Chemisches Institut, Lichtenbergstrasse 4, D-85747 Garching b. Munich, Germany
Received 24 November 1998
Abstract
Palladium complexes entrapped into zeolite cages have been prepared and characterised by MAS-NMR. They exhibit a high
activity towards the Heck reaction of aryl bromides with olefins for small palladium concentrations (0.1 mol%). The catalysts can
be easily separated from the reaction mixture and reused after washing without loss in activity. Except for very large complexes,
no limitation to the diffusion of educts in the zeolite cages was observed. As for the homogeneous Heck reaction, the electronic
nature of the aryl bromides and the olefins has a dominating effect on the reaction yield and selectivity. The heterogeneous
catalysts activate even aryl chlorides under standard reaction conditions. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Heck olefination; Zeolite; Palladium; Aryl halides; Heterogeneous catalysis; Entrapped complexes
1
. Introduction
[4]. However, only sporadic reports have appeared in
the last 15 years for the activation of iodo- and bromo-
benzenes in solution. Anchored Pd-complexes [5], Pd/C,
The palladium-catalysed carbon–carbon bond for-
mation between aryl halides and olefins (Heck reaction)
discovered in 1971 became an excellent tool for the
synthesis of elaborated styrene derivatives due to its
tolerance of a wide variety of functional groups on both
partners [1].
Pd/MO [6], Pd/MCM-41 [7] were used. Generally no
x
data have been reported concerning the problem of
dissolution of active species (leaching). Recently, we
reported a study concerning the activation of aryl bro-
mides by Pd(0) and Pd(II) zeolites for the Heck reac-
tion [8]. Direct activation of chlorobenzenes by
heterogeneous systems has been reported, but only
under drastic reaction conditions (high temperature, no
solvent or participating solvent such as MeOH) [9].
Although, to the best of our knowledge, there is no
report about the Heck reaction with Pd-complexes en-
trapped in zeolites, their use as support should have the
following advantages: (1) complexes immobilised in
zeolite super-cages should have almost the same activity
as the free complexes in solution [10], and (2) the zeolite
microstructure (micro-reactor) could help to overcome
the problems of leaching using heterogeneous catalysts
in solution. In addition, zeolites are capable of stabilis-
ing intermediate active species retained in their cavities
and combine the shape-selectivity [11].
While the use of soluble palladium complexes to
promote this reaction is well established (high reactivity
and high selectivity) [1,2], homogeneous catalysts are
generally connected with problems of separation (purity
of the products), recovery and regeneration of the
catalysts. Such problems are of importance for indus-
trial applications (large-scale synthesis) where the costs
of the catalyst materials are of importance. Although
recent advances in the design of homogeneous catalysts
have been made, in addition to the separation problems
deactivation of the homogeneous catalysts by forma-
tion of inactive colloidal species is encountered at the
comparatively high reaction temperature [3].
The problems discussed could be principally min-
imised by a heterogeneously catalysed Heck reaction
Here we report for the first time that Pd-complexes
partially used in homogeneous catalysis could be immo-
bilised in zeolites to give highly active, selective and
*
Corresponding author. Tel.: +49-89-28913074; fax: +49-89-
2
8913473.
E-mail address: laurent.djakovitch@ch.tum.de (L. Djakovitch)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00080-7

L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
17
recoverable heterogeneous catalysts towards the Heck
reaction of aryl bromides with olefins. The phe-
nomenon of leaching of active species in the bulk
solution has been studied in detail.
As we can deduce from Fig. 1, limitations are en-
countered for most of the chosen complexes:
2
+
1. Concerning the [Pd(NH ) ] , we did not find any
3
4
limitation since its small size and its pseudo-spheri-
cal geometry allows it to pass through the zeolite
window in all directions and for all orientations.
. Concerning the [Pd(OAc) ] and [Pd(C H )Cl] com-
2
2
3
5
2
2. Results and discussion
plexes we found that according to their geometry, a
perpendicular orientation or an orientation with a
minimum angle of 56 and 42°, respectively, versus
the plane defined by the window is required, in
2.1. Catalyst preparation
order to pass through the window.
Analogous to the well known ion-exchange process
2
+
+
reported for the [Pd(NH ) ]
[12], we prepared the
3. Concerning the {Pd[P(o-C H CH ) (C H CH )]} ,
6 5 3 2 6 5 2
3
4
Pd-complex loaded zeolites by immobilisation of the
three different complexes: [Pd(OAc) ], [Pd(C H )Cl]
we found that only a few orientations allow it to
pass through the window due to its steric hindrance.
Experimentally, in accordance with these calculations,
we found that in order to obtain a reasonable amount
2
+
3
5
2
and {Pd[P(o-C H CH ) (C H CH )]} [2e,13] in the
6
5
3 2
6
5
2
zeolite super-cage.
Because the diameter of the zeolite channels is limited
for the Y zeolite: ¥=7.4 A) [14], we performed
of palladium in the zeolite, for the [Pd(OAc)
[Pd(C )Cl] complexes, 3-day exchange (giving 0.7
] and
2
(
,
3
H
5
2
molecular modelling calculations in order to check
whether the complexes can pass through the zeolite
window. The results obtained using the MM2 aug-
mented parameters [15] provided with the CAChe™
software from Oxford Molecular Ltd. [16] are reported
in Fig. 1.
and 0.5% Pd-catalyst, respectively) and for {Pd[P(o-
+
C H CH ) (C H CH )]} 9-day exchange experiments
6
5
3 2
6
5
2
(giving 0.3% Pd-catalyst) are required. The palladium
concentrations of the modified zeolites were determined
by AAS to be between 0.3 and 1.0% depending on the
complexes.
2
+
Fig. 1. Molecular modelling of the Pd-complexes ([Pd(NH ) ]
,
[Pd(OAc)2], [Pd(C H )Cl] and the palladacycle {Pd[P(o-
3 5 2
3
4
+
C H CH ) (C H CH )]} ) in NaY zeolite using the CAChe™ software with the extended MM2 molecular mechanics method.
6
5
3 2
6
5
2

18
L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
Scheme 1. Reaction conditions: 10 mmol bromofluorobenzene 1, 15 mmol styrene 2, 15 mmol NaOAc, 0.1 mol% [Pd]zeo., 8 ml DMAc, 140 or
100°C, 20 h.
MAS-NMR of the Pd-complex loaded zeolites have
Table 1 also shows that the activity of the complexes
immobilised in zeolite cages is not strongly dependent on
the temperature of the reaction (slightly lower yields
being observed at 100°C). These small differences could
indicate that different active species are formed at the two
temperatures, the species generated at 100°C being prob-
ably more active (but less stable, see later).
been recorded and confirmed that the immobilised com-
plexes are intact in the zeolite cages, and gave compat-
1
13
31
ible spectra ( H, C and P) to the solution NMR
obtained for the corresponding free complexes (see
Section 4).
In order to confirm the above interpretations of the
catalytic results as a steric effect, it is necessary to
compare these results with those of the free Pd-complexes
in solution (homogeneous catalysts without zeolite).
Table 1 shows that the Pd-complexes immobilised in
zeolite have almost the same activity as the free Pd-com-
plex when no steric hindrance dramatically decreases the
reaction rate (i.e. as for the palladacycle {Pd[P(o-
2
2
.2. Catalytic acti6ity studies
.2.1. Influence of steric hindrance on the
Pd-complexes in zeolite cages
We chose the reaction of bromofluorobenzene with
styrene as the model reaction (Scheme 1). This reaction
gives high yields of Heck product with homogeneous
catalysts, thereby allowing the observation of small
changes in activity of the catalysts.
The conversion and selectivity obtained for the Pd-
complex loaded zeolites (0.1 mol% Pd/aryl bromide 1)
are reported in Table 1 and compared to the results
obtained with the corresponding Pd-complexes in bulk
solution (i.e. homogeneous catalysis).
In accordance with the molecular modelling results
described in Fig. 1 and with the homogeneous catalysis
of the same complexes (see below), the results indicate
that the catalytic activity of the Pd-loaded zeolites is
dependent on the steric hindrance of the complexes.
+
C H CH ) (C H CH )]} ). A comparison of the activity
6
5
3 2
6
5
2
of the free and immobilised [Pd(OAc)2] and
Pd(C H )Cl] at 100°C show that higher activity for the
[
3
5
2
immobilised complexes is obtained. This is probably due
to better stabilisation of the active Pd-species by the
zeolite framework.
Table 1
Results obtained for the Heck reaction (Scheme 1) with different
a
Pd-complex loaded NaY
Pd-complex
T (°C)
140
3 (%)
4 (%)
5 (%)
2
+
[
[
[
Pd(NH ) ]
93.0 [89.1]
1.0 {1.0}
0.7 {1.0}
0.9 {0.9}
0.4 {0.3}
1.0 {0.9}
8.8 {8.8}
6.7 {8.4}
7.2 {7.6}
3.9 {3.9}
8.3 {8.4}
3
4
{
94.0}
Thus,
a
higher activity was observed for the
1
00
140
00
Pd(C H )Cl] 140
94.5 [89.5]
{92.2}
79.2 [68.7]
{83.8}
2
+
[Pd(NH ) ]
complex than for the larger [Pd(OAc)2]
3
4
Pd(OAc)2]
and [Pd(C H )Cl] complexes. Concerning the pallada-
cycle {Pd[P(o-C H CH ) (C H CH )]} entrapped in
3
5
2
+
6
5
3 2
6
5
2
1
57.6 [39.9]
zeolite NaY a very low activity was obtained. This
dependence of activity is probably due to the steric
hindrance created by the complexes in the zeolite super
cage, which limit the diffusion of the educts and/or
products and the accessibility of the metal centre for the
educts. This effect is dramatically enhanced by the large
size of the palladacycle for which a low activity was
observed.
{
47.6}
86.3 [78.2]
{87.1}
3
5
2
1
00
140
00
74.2 {30.9}
3.0 {92.3}
1.0 {90.5}
0.5 {0.2}
0.1 {1.1}
nd {1.0}
5.2 {2.1}
0.2 {7.9}
nd {8.1}
Palladacycle
1
a
GLC yields [isolated yields] are given and compared with {GLC
yields for homogeneous catalysis} (Drel= 910%).

L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
19
Table 2
These results can be compared with results we have
previously reported for Pd(II) or Pd(0) modified zeolites
Results obtained for the Heck reaction (Scheme 1) with recovered
Pd-complex-loaded NaY a
[9]. These were highly active and recoverable heteroge-
neous catalysts toward the Heck reaction at 140°C. We
postulate that the active Pd-species generated for all of
them at 140°C have probably the same structure, prob-
ably Pd(0)-particles bound into the zeolite cage.
Pd-complex
T (°C)
3 (%)
First run
Recycled
2
+
[
[
[
Pd(NH ) ]
140
93.0
94.5
79.2
57.6
86.3
74.2
80.9
1.0
81.2
17.8
72.0
1.4
3
4
1
00
140
00
140
00
2.2.3. Examination of the presence of catalytically
Pd(OAc)2]
Pd(C H )Cl]
2
1
acti6e species in homogeneous bulk solution (‘leaching’)
The leaching of active species from heterogeneous
catalysts used in solution is a crucial question in order
to identify whether the active centres are immobilised in
the zeolite framework or dissolved palladium complexes
in the homogeneous solution. Thus, we have studied
this phenomenon in more detail.
3
5
1
a
GLC yields are given and compared with the results of the first
run of the catalysts (Drel= 910%).
2.2.2. Reco6ery of the Pd-complex-loaded zeolites
The leaching was studied as follows: the organic
phase of a first run was separated from the solid (Pd
modified zeolite). New reagents (p-BrFC H 1, styrene
After separation and washing, the heterogeneous cat-
alysts were used for the same reactions under the same
reaction conditions as for the initial run of the catalyst
without any regeneration. The conversion and selectiv-
ity are reported in Table 2.
Table 2 shows that all the complexes used in a first
run at 140°C still show a high activity that is compara-
ble to the activity observed for the first run. In contrast,
the catalysts used in a first run at 100°C show a lower
activity in these experiments (at 100°C).
In order to explain these results, we tried to regener-
ate the catalysts used in a first run at 100°C. This was
done by treating the recovered catalysts at 140°C for 1
h under the reaction conditions.
The results reported in Table 3 show that the regen-
erated catalysts then show a high activity that is com-
parable to the activity observed for the recovered
catalysts used in a first run at 140°C.
6
4
2
and NaOAc) were added to the clear filtrate, and the
composition of the reaction mixture was determined by
GLC, the amount of pBrFC H 1 being set to 100%.
6
4
This homogeneous reaction mixture was treated as a
standard catalytic experiment (100 or 140°C, 20 h).
After 20 h reaction time, the composition was deter-
mined once more by GLC. The differences observed
between the two GLC determinations give qualitative
information about the leaching phenomenon.
While this method does not allow an absolute quan-
tification of the palladium species (active and/or inac-
tive) dissolved in the bulk solution by the leaching
phenomenon, it gave information concerning the pres-
ence of active species in homogeneous solution. This
method also gives an idea of the relative influence of
these species on the observed activity of the heteroge-
neous catalysts for the Heck reaction.
These observations can be rationalised by the genera-
tion of different active species at the two temperatures,
the species generated at 100°C being irreversibly de-
stroyed during the extraction condition in air while the
species generated at 140°C are stable.
The results reported in Fig. 2 show that the leaching
phenomenon depends on the temperature of the reac-
tion and on the structure of the Pd-complex immo-
bilised in the zeolite cages.
While the difference between the two determinations
stays in the limit error of the analytical GLC, for the
Table 3
2+
[
Pd(NH ) ]
a higher leaching is observed at 140°C as
3
4
Results obtained for the Heck reaction (Scheme 1 at T=100°C) with
recovered Pd-complex loaded NaY before and after the regeneration
a
at 100°C, for which it was not detected.
For the palladacycle {Pd[P(o-C H CH ) (C H -
6
5
3 2
6
5
+
Pd-complex
Reactivation
3 (%)
CH )]} no leaching was detected at all, which is in
2
good agreement with a strong immobilisation of the
complex in zeolite framework, the homogeneous com-
plex being highly active (cf. Table 1).
Concerning the other complexes, the results are quite
different and could not be explained on the basis of
temperature and complex structures only. Thus, we did
First run
Recycled
2
+
[
[
[
Pd(NH ) ]
No
Yes
No
Yes
No
Yes
94.5
94.5
57.6
57.6
74.2
74.2
1.0
75.2
17.8
79.8
1.4
3
4
Pd(OAc)2]
Pd(C H )Cl]
2
3
5
not observe leaching for the immobilised [Pd(OAc) ] at
2
75.3
both reaction temperatures, while we observed a rela-
tively high leaching for the [Pd(C H )Cl] at 140°C. At
a
GLC yields are given and compared with the results of the first
run of the catalysts (Drel= 910%).
3
5
2
least the phenomenon could be strongly reduced by the

20
L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
Fig. 3. Kinetic investigations: Conversion of bromofluorobenzene 1
2
+
and product percentage (3, 4 and 5) obtained with the [Pd(NH ) ]
3
4
loaded NaY for the Heck reaction (Scheme 1).
The results reported in Fig. 4 show that the
2
+
[
Pd(NH ) ] -loaded NaY is much more active as the
3 4
Fig. 2. Leaching of active Pd-species into the bulk solution using the
Pd-loaded zeolites as catalysts in solution for the Heck reaction.
Reaction conditions: 10 mmol bromofluorobenzene, 15 mmol styrene,
other Pd-modified Y zeolites. Thus the Heck reaction of
p-BrFC H 1 with styrene 2 is completed after ca. 20
6
4
2
+
min using [Pd(NH ) ] -loaded NaY where it required
3
4
15 mmol NaOAc, 0.1 mol% [Pd]zeo., 8 ml DMAc, 140 or 100°C, 20 h.
ca. 1450 min with Pd(II) modified zeolites. In addition,
an induction period of ca. 2 min is observed for the
use of a lower reaction temperature (i.e. 100°C). To
explain these observations, we should consider the rela-
tive stability of both complexes under the same reaction
conditions. Thus the [Pd(C H )Cl] will decompose
2
+
[
Pd(NH ) ] -loaded NaY, as it was for the Pd(II)
3 4
modified zeolite (ca. 4 min) [8]. This period is not
observed for the Pd(0)-modified zeolite (Pd(0)-species
being Heck active) and corresponds to the delay re-
quired for the transformation of the Pd(II) species to
Heck active Pd(0) species.
3
5
2
quickly at 140°C (Tdec. =120°C) [17] whereas the
[
[
Pd(OAc) ] will decompose more slowly (Tdec. =205°C)
17] at this temperature, probably leading to higher
2
stabilised Pd-species in the zeolite cages. This is corre-
lated with the results observed at 100°C, where the
2.4. Variation of the educts
[
Pd(C H )Cl] should be slower decomposed to give
3 5 2
2
.4.1. Reaction between aryl bromides and olefins
In order to generalise the results obtained for
more stable Pd-species in the zeolite cage. These argu-
ments also apply to the [Pd(NH ) ]
composes in two steps: Tdec.1 =158°C, and
Tdec.2 =285°C) [18] entrapped into the NaY, which can
explain the slightly better stability of the catalyst ob-
tained at 100°C.
2
+
complex (it de-
3
4
the heterogeneously catalysed Heck reaction with the
2.3. Kinetic studies
Kinetics of the reaction between p-BrFC H 1 and
6
4
the styrene 2 were studied under standard reaction
conditions (Scheme 1: T=140°C). Fig. 3 shows that the
2
+
[
Pd(NH ) ]
-loaded NaY is a highly active catalyst
3
4
for this reaction after a short activation period (ca. 2
min). The selectivity and the activity observed is com-
parable to the one obtained using standard homoge-
neous catalysts [1,2].
Of interest was to compare the kinetics obtained with
Fig. 4. Kinetic investigations: comparison of the activity of different
2
+
the [Pd(NH ) ] -loaded NaY to the results obtained
3
4
Pd-species loaded into the zeolites. Percentage of the product 3
for the same reaction under the same conditions with
Pd(II)- and Pd(0)-loaded Y zeolites [8].
2+
obtained with the [Pd(NH ) ] -, Pd(II)- and Pd(0)-loaded NaY for
3
4
the Heck reaction (Scheme 1).

L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
21
Scheme 2. Reaction conditions: 10 mmol aryl halide 1, 15 mmol olefin 2, 15 mmol NaOAc, 0.1 mol% [Pd]zeo., 8 ml DMAc, 140°C, 20 h.
model reaction (Scheme 2), experiments have been
carried out using both the aryl halides and olefins as
complex to active Pd(0) species. These results are
comparable to the one reported for the homogeneously
catalysed Heck reaction [2].
2
+
substrates using the [Pd(NH ) ] -loaded NaY catalyst
3
4
(
Scheme 2).
The results reported in Table 4 show that the
2.4.2. Acti6ation of aryl chlorides
heterogeneous catalyst has a high activity toward the
activation of aryl bromides also for the non-activated
Another important issue concerning the heteroge-
neously catalysed Heck reaction is the possibility to
activate the industrially more interesting aryl chlorides.
1
3
bromo-anisole (R =CH O, R =Br). Concerning the
3
variation of the olefins, as expected, the electron-poor
In order to examine this possibility, we studied the
2
1
olefins (R =Ph, n-BuO–CO) gave better yields than
reaction between chloroacetophenone (R =CH CO,
3
2
3
2
the electron rich olefins (R =n-BuO). The gradation of
R =Cl; see Scheme 2) and styrene (R =Ph) using the
[Pd(NH ) ] -loaded NaY as heterogeneous catalyst.
2
+
the aryl bromide conversion is similar to homogeneous
catalysis and depends on the activation of the aryl
bromides by a para-substituent and the electron density
of the olefins in the p-bond.
3 4
As expected, the results reported in Table 5 show
that the activation of aryl chlorides required a higher
temperature as the bromo-derivatives. This is connected
with the halogen–carbon bond energy, which was cal-
culated for the bromoacetophenone (ca. 80.5 kcal
In addition to these experiments, kinetic studies have
been performed for the reaction of aryl bromides
1
3
2
−1
(
R =F, H and OCH , R =Br) and styrene (R =Ph)
mol ) [19] and the chloroacetophenone (ca. 96 kcal
3
+
2
−1
using the [Pd(NH ) ] -loaded NaY (Fig. 5).
mol ) [19]. As it was already mentioned in the litera-
3
4
The results reported in Fig. 5 show that the
ture, an interesting salt effect was observed with t-
2
+
[
Pd(NH ) ] -loaded NaY is
a
highly active
Bu NBr [20], which gave a benefit for this reaction, but
3
4
4
heterogeneous catalyst for the Heck reaction, including
the non-activated bromobenzene and the ‘deactivated’
bromo-anisole. As reported above, an initiation period
of ca. 2–5 min is observed before the reaction starts,
corresponding to the transformation of the immobilised
is accompanied in this experiment by a parallel dehalo-
genation reaction leading to the formation of acetophe-
none in 8.3% yield.
3
. Conclusions
Table 4
Palladium-complex-loaded zeolites exhibit a high ac-
Results obtained for the Heck reaction (Scheme 2, T=140°C) with
2
+
tivity towards the Heck reaction of aryl bromides with
olefins, using standard reactions conditions. Remark-
able are the small amounts of palladium (0.1 mol%)
required to perform the heterogeneous catalysed Heck
reaction. As for the homogeneous Heck reaction, the
electronic nature of the aryl bromides and the olefins
has a dominating effect on the reaction yield and
selectivity. The activated aryl bromides and the electron
different aryl halides and olefins using the [Pd(NH ) ]
loaded
3
4
NaY a
R1
R2
3 (%)
4 (%)
5 (%)
H
F
NO2
Ph
Ph
Ph
Ph
84.9 [77.8]
93.0 [89.1]
94.8 [89.8]
81.2 [75.8]
25.7
0.7
1.0
1.1
9.5
20.4
0.5
6.5
8.8
4.1
9.5
CH O
3
H
H
F
n-BuO
CH O–CO
CH O–CO
12.4
0.4
2
poor olefins (R =Ph, n-BuO–CO) almost react quan-
91.0 [69.4]
93.2 [72.6]
3
titatively. The catalysts can be easily separated from the
reaction mixture (filtration) and reused after washing.
Depending of the reaction temperature, some catalysts
0.7
0.6
3
a
GLC yields and [isolated yields] are given (Drel= 910%).

22
L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
port a direct activation of aryl chlorides using a het-
erogeneous catalyst, under standard reaction
conditions. No undesired reactions (i.e. strong dehalo-
genation of the aryl chlorides, biphenyl derivatives or
reacting solvent) were observed.
Whereas the reaction mechanism of the homoge-
neously catalysed Heck reaction is today generally ac-
cepted [1,2], the mechanism of the heterogeneously
catalysed Heck reaction remains unclear. The experi-
ments indicate that Pd(0) species are the active species
(
as for the homogeneous systems [21]). The results
observed are consistent with an oxidative addition of
the aryl bromide to the Pd(0) centre via a S Ar
N
route. In addition, the identical product distributions
(
selectivity) in homogeneous and heterogeneous reac-
Fig. 5. Kinetic investigations: comparison of the activity of the
tions indicate the same reaction mechanism for both.
At this stage of the studies, we propose that a homo-
geneous mechanism take place in the zeolite cage for
the heterogeneously catalysed Heck reaction using Pd-
complexes entrapped in zeolites. The active palladium
species could be palladium complexes formed in situ
by decomposition of the entrapped Pd-complexes,
these complexes co-exist probably with adsorbed enti-
ties by a dissolution–adsorption equilibrium of Pd(0/
II) species but are retained in the zeolite cages.
Current investigations focus on the kinetic and the
homogeneous/heterogeneous mechanistic aspect.
2
+
1
[
Pd(NH ) ] -loaded NaY for the Heck reaction (Scheme 2, R =H,
3 4
2
3
F or CH O, R =Ph and R =Br). Percentages of the product 3 are
3
reported.
were deactivated during the extraction. However, ac-
tive species could be regenerated using an easy pro-
cedure. Except for one catalyst (the [Pd(C H )Cl] -
3
5
2
loaded NaY used at 140°C), no remarkable leaching
was observed for the Pd-complex-loaded zeolites. The
investigations concerning the leaching phenomenon
showed that a more stable catalyst could be obtained
by generating the Pd-active species at a lower temper-
ature in the case of the [Pd(C H )Cl] complex. The
3
5
2
4
. Experimental
stability of the Pd-active species against leaching
seems to be correlated to the temperature of decom-
position (in the starting period) of the immobilised
Pd-complexes in the zeolite cages, a too quick decom-
position leading to stronger leaching.
All preparations, manipulations and reactions were
carried out under argon, including the transfer of the
catalyst to the reaction vessel. All glassware was base-
and acid-washed and oven dried. THF and DMF was
freshly distilled under argon over sodium from purple
benzophenone ketyl before use. The zeolite NaY was
Except for large complexes (i.e. the palladacycle
+
{
Pd[P(o-C H CH ) (C H CH )]} ), no limitation to
6 5 3 2 6 5 2
the diffusion of educts in the zeolite cages was ob-
served for the reaction. The kinetics performed with
purchased from Sigma-Aldrich Chemical (LZ-Z-52)
and dried under 5×10
−2
mmHg at 120°C for 48 h
2
+
the [Pd(NH ) ] -loaded NaY show that the hetero-
3
4
before use in the synthesis of catalysts 2–4. All other
chemicals (organic reagents and solvent) were deaerated
by an argon flow before they were used. The Pd-loaded
zeolites were stored after drying under Ar atmosphere.
Solution NMR spectra of the organic products were
geneous catalysts have a comparable activity to the
homogeneous catalysts, the complete conversion of
the aryl bromides being of ca. 20 min, including the
non activated derivatives. For the first time, we re-
recorded with
a
Bruker AM 400 spectrometer
referenced to the residual
1
(
H-NMR
were
13
Table 5
protio-solvent: CDCl , l=7.25 ppm; C-NMR were
3
Results obtained for the Heck reaction (Scheme 2) with chloroace-
referenced to the C-signal of the deutero solvent:
2
+
tophenone and styrene using the [Pd(NH ) ] -loaded NaY in the
3
4
a
CDCl , l=77 ppm).
3
presence and absence of a co-catalyst
1
13
31
Solid-state H-, C-, and P-MAS-NMR spectra of
the exchanged zeolites were recorded on a Bruker MSL
T (°C)
Co-catalyst
3 (%)
4 (%)
5 (%)
3
00 spectrometer operating at a field strength of 7.05 T.
1
1
1
40
70
70
No
No
Yes
0.3
44.1 [31.2]
59.2 [38.1]
nd
1.2
1.0
0.3
3.3
2.0
1
13
For the H- and C-MAS-NMR spectra, ca. 300 mg of
the sample were packed into 4 mm ZrO Bruker rotors
with Kel-F caps. For the P spectrum a 7 mm ZrO
Bruker rotor with ca. 1 g of sample was used. H- and
2
31
2
a
1
GLC yields and [isolated yields] are given (Drel= 910%).

L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
23
1
3
C-NMR shifts are referenced to an external sample of
adamantane; the proton signal is set to 2 ppm and the
4.1.3. Preparation of the [Pd(C H )Cl] -modified zeolite
3 5 2
[cat. 3]
A solution of the [Pd(C H )Cl] (88.4 mg, 0.47 mmol,
13
low-frequency signal of the C spectrum to 29.472 ppm
relative to TMS. The P-MAS-NMR spectrum is refer-
enced externally to solid NH H PO with a signal shift
3
5
2
3
1
ca. 1% Pd in the final catalyst) in THF was added drop
wise to a suspension of the zeolite NaY (4.95 g) in THF
4
2
4
1
−1
of 1.11 ppm relative to 85% H PO . H-MAS-NMR
(20 ml g
zeolite). The mixture was stirred for 3 days
3
4
spectra were recorded at a sample spinning speed of 15
at r.t. and the [Pd(C H )Cl] -loaded zeolite was filtered
3 5 2
1
3
kHz. C-CP-MAS-NMR spectrum was recorded at a
spinning speed of 8 kHz, using high-power proton
decoupling, with a recycle time of 8 s and a contact
and washed with THF until no trace of non-immo-
bilised complex was detected in the filtrate. Then the
zeolite was allowed to dry at r.t. to give the entrapped
[Pd(C H )Cl] zeolite as a yellow material. The AAS
3
1
time of 5 ms. The P-MAS-NMR spectrum was ob-
tained at a sample spinning speed of 7 kHz.
3
5
2
gave 0.590.1% Pd (w/w).
1
1
2
Gas–liquid chromatograms were performed on an
HP 6890 series chromatograph equipped with a FID
detector and a HP-1 column (cross-linked methylsilox-
ane, 30 m×0.25 mm×0.25 mm film thickness).
The absolute palladium content of the catalysts was
determined by AAS for the Pd-complex-loaded zeolites
after drying and calcination (in order to remove all
organic material) from a solution obtained by treat-
ment with a mixture of HBF , HNO and HCl in a
MAS-NMR: H-NMR, ppm: 2.14 (CH H CH-
1
2
1
2
1
2
1
2
CH H ); 4.03 (CH H CHCH H ); 7.15 (CH H -
1
2
CHCH H ).
13
MAS-NMR: C-NMR, ppm: 107.15 (CH CHCH );
2
2
69.07 (CH CHCH ).
2
2
4
.1.4. Preparation of the {Pd[P(o-C H CH ) -
6 5 3 2
+
(C H CH )]} -modified zeolite [cat. 4]
6 5 2
+
4
3
A solution of the {Pd[P(o-C H CH ) (C H CH )]} ,
6
5
3 2
6
5
2
Teflon reactor at 180°C.
−
NO3 (222.8 mg, 0.47 mmol, ca. 1% Pd in the final
catalyst) in DMF was added drop wise to a suspension
−
1
of the zeolite NaY (4.95 g) in THF (20 ml g
zeolite).
4
4
.1. Preparation of the catalysts
The mixture was stirred for 9 days at r.t. and the
+
{
Pd[P(o-C H CH ) (C H CH )]} -loaded zeolite was
6 5 3 2 6 5 2
.1.1. Preparation of the [Pd(NH ) ]-modified zeolite
filtered and washed with a mixture DMF/THF (2:3)
until no trace of non-immobilised complex was detected
in the filtrate. Then the zeolite was washed with THF
and allowed to dry at r.t. under vacuum to give the
3
4
(cat. 1)
A 0.1 M ammoniac solution of [Pd(NH ) ]Cl pre-
3
4
2
pared from PdCl and a commercial ammoniac solution
2
−
1
+
(
0.95 ml g zeolite, ca. 1% Pd in the final catalyst) was
entrapped {Pd[P(o-C H CH ) (C H CH )]} zeolite as
6
5
3 2
6
5
2
added drop wise to a suspension of the zeolite NaY in
a yellow material. The AAS gave 0.390.1% Pd (w/w).
MAS-NMR: H-NMR, ppm: 33.9 (P).
−
1
31
bidistilled water (100 ml g
zeolite). The mixture was
stirred for 24 h at room temperature (r.t.) and the
exchanged zeolite was filtered and washed until no trace
4.2. Catalytic acti6ity
of chloride was detected in the filtrate (AgNO test).
3
Then the zeolite was allowed to dry at r.t. to give the
Catalytic reactions were carried out in pressure tubes
2
+
entrapped [Pd(NH ) ]
zeolite as a slightly yellow
3
4
under argon. The qualitative and quantitative analysis
of the reactants and the products was made by GLC.
Conversion and selectivity are represented by product
distribution (=relative area of GLC signals), and GLC
yields are in parentheses (=relative area of GLC sig-
nals referred to an internal standard calibrated to the
material. The AAS gave 1.090.2% Pd (w/w).
4
.1.2. Preparation of the [Pd(OAc) ]-modified zeolite
2
[cat. 2]
A solution of the Pd(OAc) (110.2 mg, 0.47 mmol,
2
corresponding pure compound, D = 910%). Where
rel
ca. 1% Pd in the final catalyst) in THF was added drop
wise to a suspension of the zeolite NaY (4.95 g) in THF
available, yields in isolated products are given in brack-
ets. The catalysts were transferred under Ar.
−
1
(
20 ml g
zeolite). The mixture was stirred for 3 days
at r.t. and the [Pd(OAc) ]-loaded zeolite was filtered
2
and washed with THF until no trace of non-immo-
bilised complex was detected in the filtrate. Then the
zeolite was allowed to dry at r.t. to give the entrapped
4.2.1. General procedure for the first run and the
recycling of the catalysts
A total of 10 mmol of aryl halide, 15 mmol of olefin,
15 mmol of NaOAc and 0.1 mol% of Pd (as heteroge-
neous catalysts, the amount in g of catalysts depending
on the palladium concentration) was introduced in a
pressure tube under argon. A total of 8 ml of solvent
(DMAc p.a. previously deaerated) was added and the
[
Pd(OAc) ] zeolite as a slightly brown material. The
2
AAS gave 0.790.2% Pd (w/w).
1
MAS-NMR: H-NMR, ppm: 2.10 (CH CO).
3
1
3
MAS-NMR:
C-NMR, ppm: 186.84 (CH CO);
3
2
3.04 (CH CO).
3

24
L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
mixture was deaerated by an argon flow for 5 min.
The reactor was then placed in a pre-heated oil bath
at 100 or 140°C for 20 h with vigorous stirring and
then cooled to r.t. before the reaction mixture was
analysed by GLC.
4.2.4.1. Data for R=H. M.p.: 118°C, white plates.
1
3
H-NMR, CDCl , 400.13 MHz: 7.64 (d, J=7.0
3
3
Hz, 4H, ortho-vinyl, C H ); 7.48 (pseudo-t, J=7.5
Hz, 4H, meta-vinyl, C H ); 7.39 (pseudo-t, J=7.0
6
5
3
6
5
Hz, 2H, para-vinyl, C H ); 7.24 (s, 2H, CH-vinyl).
6
5
13
For the recycling studies, a catalyst issue from a
first run was used. After separation of the reaction
mixture, it was washed with CH Cl in order to re-
C-NMR, CDCl , 400.13 MHz: 137.28 (C-vinyl-
3
C H ); 128.63 (meta-vinyl-C H ); 128.48 (CH-vinyllic);
6
5
6
5
127.56 (para-vinyl-C H ); 126.48 (ortho-vinyl-C H ).
2
2
6
5
6
5
move adsorbed organic substrates and dried at r.t.
C H , Anal. [Found (Calc.)]: C 92.01 (93.29), H
14 12
6
.64 (6.71).
4
.2.2. General procedure for the examination of the
4
.2.4.2. Data for R=F. M.p.: 122°C, white plates.
presence of catalytically acti6e species in
homogeneous/bulk solution (leaching)
1
H-NMR, CDCl , 400.13 MHz: 7.35–7.40 (m, 4H,
3
ortho-vinyl, C H F and ortho-vinyl, C H ); 7.26
6
4
6
5
A clear filtrate of the first run of the catalyst (ob-
tained by filtration using a microglass Whatman filter
in order to remove the fine particles) was used as
solvent basis for these experiments. The filtrate (free
of catalyst) was placed in a pressure tube and deaer-
ated by an Ar flow for 5 min. Then new organic
reactants were added under argon atmosphere: 10
mmol of aryl halide, 15 mmol of olefin, 15 mmol of
NaOAc and well homogenised. GLC analysis of the
composition of the mixture was carried out before the
reactor was placed in a pre-heated oil bath at 100 or
(
pseudo-t, 2H, meta-vinyl, C H ); 7.16(m, 1H, para-
6 5
vinyl, C H ); 6.92-6.96 (m, 4H, ortho-F, C H F and
CH-vinyl).
6
5
6
4
13
C-NMR, CDCl , 400.13 MHz: 161.06 (C–F,
3
C H F); 137.13 (C-vinyl-C H ); 133.50 (C-vinyl-
C H F); 128.67 (meta-vinyl-C H ); 128.45 and 128.46
6
4
6
5
6
4
6
5
(
ortho-vinyl-C H F); 127.99 and 127.92 (CH-vinyllic);
6 4
1
27.44 (para-vinyl-C H ); 126.42 (ortho-vinyl-C H );
6
5
6
5
1
15.68 and 115.47 (meta-vinyl-C H F).
6
4
+
C H F, mol. wt.: 198.24, MS: m/z (%): [M ]
14
11
1
1
98.2 (100), isotopic distribution: 198.2 (100) and
99.2 (11.6). Anal. [Found (Calc.)]: C 84.65 (84.82),
140°C for 20 h with vigorous stirring. After the reac-
tion, the reactor was cooled to r.t. and a second
GLC analysis was performed. The comparison be-
tween the two GLC analyses gave a qualitative mea-
sure for the presence of active species in
homogeneous/bulk solution.
H 5.68 (5.59).
4.2.4.3. Data for R=NO . M.p.: 107°C, yellow plates.
2
1
3
H-NMR, CDCl , 400.13 MHz: 8.15 (d, J=9.0
3
3
Hz, 2H, ortho-NO , C H NO ); 7.57 (d, J=9.0 Hz,
2
ortho-vinyl, C H ); 7.35 (pseudo-t, J=7.0 Hz, 2H,
meta-vinyl, C H ); 7.16 (pseudo-q, J=7.0 Hz, 1H,
para-vinyl, C H ); 7.21 (d, J=16.6 Hz, 1H, CH-
vinyl); 7.08 (d, J=16.6 Hz, 1H, CH-vinyl).
2
6
4
2
3
H, ortho-vinyl, C H NO ); 7.50 (d, J=7.5 Hz, 2H,
6 4 2
4.2.3. GLC analysis
3
6
5
A homogeneous 3 ml sample of the reaction mix-
3
6
5
ture was sampled and quenched with 3 ml of water in
a test tube. The mixture was extracted with 2 ml of
CH Cl and the organic layer was filtered through a
MgSO pad. The resulting dry organic layer was then
analysed by GLC. GLC-rate program: AS 100-con-
stant pressure (130 kPa): 2 min at 100°C, heating at
3
6
5
3
13
2
2
C-NMR, CDCl , 400.13 MHz: 146.47 (C–NO ,
3
2
4
C H NO ); 143.64 (C-vinyl-C H NO ); 135.93 (C-
vinyl-C H ); 133.07 (meta-vinyl-C H ); 128.65 (ortho-
vinyl-C H NO ); 128.60 (para-vinyl-C H ); 126.80 and
1
6
4
2
6
4
2
6
5
6
5
6
4
2
6
5
−
1
1
1
3
5°C min
5°C min
up to 170°C, 2 min at 170°C, heating at
up to 240°C, 10 min at 240°C, heating
26.64 (CH-vinyllic);126.00 (ortho-vinyl-C H ); 123.85
6 5
−
(
meta-vinyl-C H NO ).
6 4 2
−
1
at 50°C min
up to 270°C and 2 min at 270°C.
+
C H NO , mol. wt.: 225.24, MS: m/z (%): [M ]
1
4
11
2
+
+
2 2
2
25.3 (100) [M −NO ] 179.2 (21) [M −NO −H]
4
.2.4. Purification of the trans-stilbene products 3
After the separation of the heterogeneous catalyst,
178.2 (56), isotopic distribution: 225.3 (100) and 226.3
(13.4) 227.3 (1.8). Anal. [Found (Calc.)]: C 73.89
(74.65), H 5.02 (4.92), N 6.32 (6.22).
the reaction phase was added to CH Cl (50 ml) and
2
2
washed with H O (3×15 ml). The organic layer was
2
separated and dried over MgSO and evaporated. The
4.2.4.4. Data for R=CH O. M.p.: 130°C, light yellow
4
3
residue was dissolved in 15 ml of CH Cl and pen-
tane (8 ml) was added. The solution was placed at
plates.
2
2
1
3
H-NMR, CDCl , 400.13 MHz: 7.55 (d, J=7.04
Hz, 2H, ortho-vinyl, C H ); 7.51 (d, J=9.04 Hz, 2H,
ortho-vinyl, C H OCH ); 7.41 (pseudo-t, J=7.04 Hz,
2H, meta-vinyl, C H ); 7.32 (pseudo-t, J=7.00 Hz,
1H, para-vinyl, C H ); 7.15 (d, J=16.6 Hz, 1H, CH-
vinyl); 7.06 (d, J=16.6 Hz, 1H, CH-vinyl); 6.95 (d,
3
3
18°C to give the trans-stilbene 3, which was collected
6
5
3
by filtration. The mother liquor was concentrated,
treated as the original liquor to give additional
product. The purity of the product was estimated by
GLC to be ]99.8%.
6 4 3
3
6
5
3
6
5
3

L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
25
3
+
J=8.9 Hz, 2H, ortho-OCH , C H OCH ); 3.86 (s, 3H,
C H O , mol. wt.: 204.26, MS: m/z (%): [M ] 204.3
13 16 2
3
6
4
3
+
+
4 9 4 9
CH O).
(100) [M −C H O] 131.3 (22) [M −C H O−CO]
103.3 (42) Anal. [Found (Calc.)]: C 75.96 (76.44), H
7.81 (7.90).
3
3
1
C-NMR, CDCl , 400.13 MHz: 159.20 (C–OCH ,
3
3
C H OCH ); 137.43 (C-vinyl-C H ); 130.00 (C-vinyl-
6
4
3
6
5
C H OCH ); 128.53 (meta-vinyl-C H ); 128.10 (CH-
6
4
3
6
5
vinyllic-C H ); 127.62 (ortho-vinyl-C H OCH ); 127.09
6
5
6
4
3
(
1
CH-vinyllic-, C H OCH ); 126.48 (para-vinyl-C H );
6 4 3 6 5
Acknowledgements
26.16
(ortho-vinyl-C H );
114.02
(meta-vinyl-
6
5
C H OCH ); 55.15 (CH O).
6
4
3
3
We are grateful to the European Community (Marie-
Curie, TMR program) for a research grant to L.
Djakovitch, the Technische Universit a¨ t of M u¨ nchen
and the Fonds der Chemischen Industrie (Germany) for
support. We acknowledge Degussa AG (Germany) for
the donation of PdCl2.
+
C H O, mol. wt.: 210.27, MS: m/z (%): [M ] 210.3
1
5
14
+
+
3 3
(
100) [M −CH ] 195.2 (19) [M −CH O] 179.2 (10).
Anal. [Found (Calc.)]: C 85.42 (85.68), H 6.75 (6.71).
4.2.5. Purification of the styrene deri6ati6e products 3
After the separation of the heterogeneous catalyst,
the reaction phase was added to CH Cl (50 ml) and
2
2
washed with H O (3×15 ml). The organic layer was
2
References
separated and dried over MgSO and evaporated. The
4
−
2
residue was distillate under 5×10
mmHg to give the
[
1] (a) R.F. Heck, J.P. Nolley Jr., J. Org. Chem. 37 (1972) 2320. (b)
R.F. Heck, Palladium Reagents in Organic Synthesis, Academic
Press, New York, 1985. (c) B.M. Trost, I. Flemming (Eds.),
Comprehensive Organic Synthesis, vol. 4, pp. 585, 833 and Refs.
cited therein.
styrene derivative 3 as a colourless oil. The purity of the
product was estimated by GLC to be ]99.8%.
4
.2.5.1. Data for R=F, Eb0.05. M.p.: 128°C
[
2] (a) A. Meijiere, F.E. Meyer, Angew. Chem. 106 (1994) 2473 and
Refs. cited therein. (b) M. Beller, K. K u¨ hlein, Synlett (1995) 441.
(c) W.A. Herrmann, in: B. Cornils, W.A. Herrmann (Eds.),
Applied Homogeneous Catalysis, VCH, Weinheim, 1996, p. 712
and Refs. cited therein. (d) T. Jeffery, Tetrahedron 32 (1996)
10113. (e) M.T. Reetz, G. Lohmer, Chem. Commun. (1996)
1
3
H-NMR, CDCl , 400.13 MHz: 7.91 (d, J=14.6
3
3
Hz, 1H, CH-vinyl, C H F); 7.33 (d, J=8.05 Hz, 2H,
ortho-vinyl, C H F); 6.98 (d, J=8.05 Hz, 2H, meta-
vinyl, C H F); 6.23 (d, J=14.6 Hz, 1H, CH-vinyl,
CO); 4.39 (t, J=7.0 Hz, 2H, CH O); 2.49 (pseudo-
quintet, J=7.0 Hz, 2H, CH CH O); 1.23 (pseudo-sex-
tet, J=7.0 Hz and J=6.65 Hz, 2H, CH CH ); 0.72
6
4
3
6
4
3
6
4
3
2
1
921. (f) W.A. Herrmann, C. Broßmer, C.-P. Reisinger, T.H.
3
2
2
8
Riermeier, K. Ofele, M. Beller, Chem. Eur. J. (1997) 1357. (g) S.
Br a¨ se, A. de Meijere, in: F. Diederich, P.J. Stang (Eds.), Metal-
Catalyzed Cross-Coupling Reactions, Wiley–VCH, Weinheim,
3
3
2
3
3
(
t, J=6.65 Hz, 3H, CH CH ).
2 3
1
3
1
998 and Refs. cited therein. (h) M. Beller, T.H. Riermeier, G.
C-NMR, CDCl , 400.13 MHz: 174.52 (CO); 160.18
3
Stark, in: M. Beller, C. Bolm (Eds.), Transition Metals for
Organic Synthesis, Wiley–VCH, Weinheim, 1998, p. 208 and
Refs. cited therein. (i) M.T. Reetz, G. Lohmer, R. Schwickardi,
Angew. Chem. Int. Ed. Engl. 37 (1998) 481 and Refs. cited
therein.
(
(
(
(
(
C–F, C H F); 141.68 (CH-vinyllic-, C H F); 131.76
6 4 6 4
C-vinyl-C H F); 127.62 (ortho-vinyl-C H F); 127.33
6
4
6
4
CH-vinyllic-CO); 112.49 (meta-vinyl-C H F); 68.15
6
4
CH O); 32.28 (CH CH O); 19.31 (CH CH ); 13.71
2
2
2
2
3
[
[
3] (a) W.A. Herrmann, C. Broßmer, K. Ofele, M. Beller, H.
8
CH CH ).
2
3
+
Fischer, J. Mol Catal. A: Chem. 103 (1995) 133. (b) M. Beller,
H. Fischer, K. K u¨ hlein, C.-P. Reisinger, W.A. Herrmann, J.
Organomet. Chem. 520 (1996) 257.
C H FO , mol. wt.: 222.26, MS: m/z (%): [M ]
1
3
15
2
+
+
4 9 4 9
2
22.3 (100) [M −C H O] 149.3 (22) [M −C H O−
CO] 121.3 (42) Anal. [Found (Calc.)]: C 69.82 (70.25),
H 6.75 (6.80).
4] (a) F.R. Hartley, Supported Metal Complexes: a New Genera-
tion of Catalysts, Riedel, Dordrecht, 1985. (b) Y.I. Ermakov,
L.N. Aezamaskova, Stud. Surf. Sci. Catal. 27 (1986) 459. (c) A.
Eisenstadt, in: F.E. Herkes (Ed.), Catalysis of Organic Reac-
tions, Marcel Dekker, Basel, 1998, pp. 415–427.
4
.2.5.2. Data for R=H, Eb0.05. M.p.: 116°C
1
3
H-NMR, CDCl , 400.13 MHz: 7.91 (d, J=14.6
3
[
5] (a) C.-M. Anderson, K. Karabelas, A. Hallberg, J. Org. Chem.
50 (1985) 3891. (b) P. Yi, Z. Zhuangyu, H. Hongwen, J. Mol.
Catal. 62 (1990) 297. (c) P.W. Wang, M.A. Fox, J. Org. Chem.
Hz, 1H, CH-vinyl, C H ); 7.27 (m, 5H, C H ); 6.23 (d,
6
5
6
5
3
3
J=14.6 Hz, 1H, CH-vinyl, CO); 4.39 (t, J=7.0 Hz,
H, CH O); 2.49 (pseudo-quintet, J=7.0 Hz, 2H,
CH CH O); 1.23 (pseudo-sextet, J=7.0 Hz and J=
3
5
9 (1994) 5358. (d) J. Kiviaho, T. Hanaoka, Y. Kubota, Y. Sugi,
2
2
3
3
J. Mol. Catal. A: Chem. 101 (1995) 25. (e) D. Villemin, P.A.
Jaffres, B. Nechab, F. Courivaud, Tetrahedron Lett. 38 (1997)
2
2
3
6
.65 Hz, 2H, CH CH ); 0.72 (t, J=6.65 Hz, 3H,
2 3
6
581.
CH CH ).
2
3
3
[6] (a) C.-M. Anderson, A. Hallberg, J. Org. Chem. 83 (1988) 235.
(b) B.M. Choudary, R.M. Sarma, K.K. Rao, Tetrahedron 48
1
C-NMR, CDCl , 400.13 MHz: 174.52 (CO); 141.68
3
(
1992) 719. (c) R.L. Augustine, S.T. O’Leary, J. Mol. Catal. A:
(
(
(
(
(
CH-vinyllic-, C H ); 131.52 (C-vinyl-C H ); 127.33
6 5 6 5
Chem. 95 (1995) 277 and Refs. cited therein. (d) M. Beller, K.
K u¨ hlein, Synlett (1995) 441. (e) M. Hiroshige, J.R. Hauske, P.
Zhou, Tetrahedron Lett. 36 (1995) 4567. (f) A. Wali, S.
Muthukumaru, V.K. Kausshik, S. Satish, Appl. Catal. A: Gen.
135 (1996) 83. (g) M.T. Reetz, R. Breinbauer, K. Wanninger,
CH-vinyllic-CO); 126.71 (ortho-vinyl-C H ); 126.72
6
5
meta-vinyl-C H ); 125.12 (para-vinyl-C6H5); 68.15
6
5
CH O); 32.28 (CH CH O); 19.31 (CH CH ); 13.71
2
2
2
2
3
CH CH ).
2
3

26
L. Djako6itch et al. / Journal of Organometallic Chemistry 584 (1999) 16–26
Tetrahedron Lett. 37 (1996) 4499. (f) A.F. Shmidt, T.A.
Vladimirova, E.Y. Shmidt, Kinet. Katal. 38 (1997) 245 and Refs.
cited therein.
[14] Structures of the zeolite Y were obtained from: W.M. Meier,
D.H. Olson, Ch. Baerlocher (Eds.), Atlas of Zeolite Structure
Type, Elsevier, New York, 1996, p. 104.
[
[
[
7] C.P. Mehnert, J.Y. Ying, Chem. Commun. (1997) 2215.
8] L. Djakovitch, K. K o¨ hler, J. Mol. Cat. A: Chem. 142 (1999) 275.
9] (a) M. Julia, M. Dutheil, Bull. Soc. Chim. Fr. (1973) 2790. (b)
M. Julia, M. Duteil, C. Grard, E. Kuntz, Bull. Soc. Chim. Fr.
[15] N.L. Allinger, J. Am. Chem. Soc. 99 (1977) 8127.
[
[
[
[
16] Version 4.0 for PowerMacintosh was used for the molecular
modelling using the parameters provided by E. Chamot.
17] D.R. Lide (Ed.), CRC Handbook of Chemistry and Physics,
(
(
(
1973) 2791. (c) J.J. Bozell, C.E. Vogt, J. Am. Chem. Soc. 110
1988) 2655. (d) W.J. Scott, J. Chem. Soc. Chem. Commun.
1987) 1755. (e) K. Kaneda, M. Higuchi, T. Imanaka, J. Mol.
7
6th ed., CRC Press, Boca Raton, FL, 1995–1996.
18] The temperature of decomposition of the [Pd(NH ) ]2+·2Cl−
3
4
was measured by thermogravimetry.
19] A. Streitwieser, C.H. Heathcock, E.M. Kosower, in: Introduc-
tion to Organic Chemistry, 4th ed., MacMilan Press, London,
Catal. 63 (1990) L33.
[
10] (a) G. Schmid, Chem. Rev. 92 (1992) 1709 and Refs. cited
therein. (b) W.M.H. Sachtler, Acc. Chem. Res. 26 (1993) 383
and Refs. cited therein. (c) W.M.H. Sachtler, Z. Zhang, Adv.
Catal. 39 (1993) 129 and Refs. cited therein. (d) A. Corma,
Chem. Rev. 97 (1997) 2373 and Refs. cited therein.
11] B.C. Gates, Catalytic Chemistry, Wiley, New York, 1992, p. 259
and Refs. cited therein.
12] (a) J. Michalik, M. Narayana, L. Kevan, J. Phys. Chem. 89
1
992.
[
[
20] (a) T. Jeffery, Tetrahedron Lett. 35 (1994) 3051. (b) T. Jeffery,
Tetrahedron Lett. 35 (1994) 4103. (c) T. Jeffery, Tetrahedron 32
(
1996) 10113.
[
[
21] (a) C. Amatore, E. Carre, A. Jutand, M.A. M’Barki, G. Meyer,
Organometallics 14 (1995) 5605. (b) A.F. Shmidt, A. Khalaika,
L.O. Nindakova, E.Y. Shmidt, Kinet. Katal. 39 (1998) 200. (c)
K. Albert, P. Gisdakis, N. R o¨ sch, Organometallics 17 (1998)
1608.
(1985) 4553. (b) W.M.H. Sachtler, F.A.P. Cavalcanti, Catal.
Lett. 9 (1991) 261.
[
13] C.-P. Reisinger, PhD Thesis, University of Munich, 1997.
.