Dendritic effects in catalysis by Pd complexes of bidentate phosphines on a dendronized support: Heck vs. carbonylation reactions

Article abstract of DOI:10.1039/b809715a

Bidentate phosphine ligands have been prepared on polystyrene beads modified with polyether dendron spacers. When complexed to Pd⁰, these systems exhibited a negative dendritic effect on Heck catalysis (contrary to the analogous monodentate phosphine systems), but mostly a positive influence on carbonylation. This opposite influence of the dendronization falls into line with other differences in the optimal ligand structure for the two reactions. The negative effect on the Heck catalysis with bidentate phosphines may indicate that dendrimer-induced reduction in the cross-linking upon Pd complexation is responsible for the positive effect in the corresponding monodentate phosphine systems.

Full text of DOI:10.1039/b809715a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Dendritic effects in catalysis by Pd complexes of bidentate phosphines on a
dendronized support: Heck vs. carbonylation reactions†‡
Amal Mansour, Tzofit Kehat and Moshe Portnoy*
Received 9th June 2008, Accepted 25th June 2008
First published as an Advance Article on the web 25th July 2008
DOI: 10.1039/b809715a
Bidentate phosphine ligands have been prepared on polystyrene beads modified with polyether dendron
spacers. When complexed to Pd0, these systems exhibited a negative dendritic effect on Heck catalysis
(contrary to the analogous monodentate phosphine systems), but mostly a positive influence on
carbonylation. This opposite influence of the dendronization falls into line with other differences in the
optimal ligand structure for the two reactions. The negative effect on the Heck catalysis with bidentate
phosphines may indicate that dendrimer-induced reduction in the cross-linking upon Pd complexation
is responsible for the positive effect in the corresponding monodentate phosphine systems.
dendronization results in a positive net outcome. During these
Introduction
experiments it was found that one of the most important factors,
In the past few years a few research groups have explored the issue
contributing to the positive effect, was reduced cross-linking of
of dendritic catalysts immobilized on solid supports.1,2 In most
cases positive dendritic effects, i.e. improved catalytic output of the
dendritic catalysts as compared to their non-dendritic analogues,
the polymer upon complexation, a property directly derived from
the dendronization-induced proximity of the ligating groups. In
this communication we demonstrate that for bidentate terminal
were observed.3 Thus, we reported polymer-supported catalysts
ligands the contribution of this factor to the net result is of
based on dendrons decorated with monodentate phosphines (e.g.
structure 1), revealing positive dendritic effects on the activity
and chemoselectivity of Co complexes in the Pauson–Khand
substantially reduced magnitude (if it exists at all), and accordingly
the effect of dendronization can become negative.
Herein we report the preparation of dendrons decorated
reaction and Pd complexes in the Heck and Suzuki reactions.3a–c
with chelating bidentate phosphine ligands on a support, their
complexation with a Pd0 precursor and catalysis of Heck and
amidocarbonylation reactions with the obtained complexes.4–6
While dendronization of the polystyrene matrix must influence the
catalyst performance through a variety of different factors, only
some of them will lead to an improvement in the catalyst action,
while others are likely to negatively affect the catalysts.
Results and discussion
In the past, our group communicated the preparation of polyether
dendrons on polystyrene, and their use as spacers for the afore-
mentioned catalytic systems.3a,b,7 Although these dendrons were
prepared in a highly efficient and selective way via a Mitsunobu–
reduction reaction sequence (Scheme 1), the Mitsunobu reaction
required the two-step synthesis of a special coupling agent 3, which
must be prepared from expensive starting materials and has a
limited shelf-life.
In order to remove this obstacle, we reworked the synthesis,
changing the monomer coupling step to the nucleophilic sub-
stitution of benzyl halide by the phenolate of 2, and adding
an additional monomer activation step – chlorodehydroxylation
(Scheme 2).
Clean and efficient substitution reactions yielding Gn(CO2Me)
were carried out using LiH. The reduction step conditions are
the same as those previously reported for the two-step assembly
sequence, while excellent results were obtained for chlorodehy-
droxylation with a PPh3/C2Cl6 reagent mixture, yielding Gn(Cl)
quantitatively. All steps exhibited quantitative conversion. The pu-
rity of each step is excellent as confirmed by the cleavage analysis.
Morever, gel phase 13C NMR did not detect any impurities on the
support. Thus, replacing the two-step sequence with the three-step
one, we were able to reach approximately the same yield and purity
of the dendrons, using only readily available commercial reagents.
The aforementioned experiments demonstrated that, in the
case of the monodentate phosphine systems in the tested re-
actions, the superposition of all the factors affected by the
School of Chemistry, Raymond and Beverly Sackler Faculty of Exact
Sciences, Tel Aviv University, Tel Aviv, 69978, Israel. E-mail: portnoy@
post.tau.ac.il; Fax: +972 3640 9293; Tel: +972 3640 6517
† This paper is dedicated to Prof. D. Milstein on the occasion of his 60th
birthday.
‡ Electronic supplementary information (ESI) available: General experi-
mental conditions and characterization data of intermediates Gn(CO2Me),
Gn(serinol-OH) and Gn(serinol-Cl). See DOI: 10.1039/b809715a
3382 | Org. Biomol. Chem., 2008, 6, 3382–3387
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Scheme 1 The reported synthesis of polyether dendrons.6
Scheme 2 The modified synthesis of polyether dendrons.
This alternative sequence also directly provided the chloromethyl-
terminated dendrons Gn(Cl) that were used as starting materials
for the bidentate ligand synthesis.
The synthesis of the diphosphine ligand followed the route pre-
viously reported for its assembly on the Wang Bromo polystyrene,
and was initiated from immobilization of serinol on the Gn(Cl) via
a nucleophilic substitution reaction (Scheme 3).8 Although for the
synthesis of the first two generation ligands, serinol itself is used in
this step as an HCl-sequestering base, diisopropylethylamine must
be added for the immobilization of serinol on G3(Cl), to prevent
dendron cleavage. It is possible that a higher local concentration
of HCl is temporarily generated in this reaction with G3(Cl) and is
responsible for the cleavage. The next step, chlorodehydroxylation,
followed under the conditions previously described. The synthesis
was accomplished via phosphinodechlorination with KPPh2.
For most steps en route from Gn(Cl) to Gn(serinol-PPh2) the
conversion and yield were close to quantitative, and only for
the last two steps of the G3(serinol-PPh2) synthesis was a lower
yield observed. The synthesis of the ligands was monitored by
the cleavage analysis (first two steps) and gel-phase NMR (the
last step). The yield of the phosphination in the last step was
determined using a 31P NMR-based quantification experiment
with the commercial supported diphenylphosphino polystyrene
reference.9 The cleavage analysis and quantification, being in a
very good agreement, demonstrated that the overall yield of the
G1(serinol-PPh2) is 71%. For the higher generations, the 9-step and
12-step reaction sequences yielded 45% and 11% of the second-
and third-generation structures respectively.
The ligands Gn(serinol-PPh2) were converted to catalytic resins,
following their complexation with Pd(dba)2 (Scheme 4).10 Two
series of catalysts were prepared. For the first series, the incubation
procedure was based on a 1 : 1 ratio between the bisphosphine
ligand and the soluble Pd precursor, thus leading to systems where
each ligating site is complexed by the Pd(dba) fragment (Gn-I).
Scheme 3 Synthesis of the bidentate phosphines.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3382–3387 | 3383
©

View Article Online
Table 2 The amidocarbonylation reaction of bromobenzene with
diethylaminea
Scheme 4 The complexation with Pd0.
Entry
Catalyst
Conversion (%)
Yield (%)b
The second series of the catalysts was generated using a 2 : 1 ratio
between the bidentate ligand and the Pd precursor, thus leading
to systems with only half of the ligating sites being occupied
by Pd (Gn-II). In the synthesis of the second series, complete
discoloration of the brown Pd(dba)2 solution occurred during the
incubation, while in the first series synthesis the discoloration was
almost complete. This observation provides evidence for complete
incorporation of the calculated amount of Pd into the ligating sites
on the resin. Additional evidence for the generation of the planned
catalytic series was obtained from gel-phase 31P NMR. While the
phosphorus signal of the free ligand appears at −23 ppm, that of
the ligand complexed with Pd(dba) appears at +13 ppm. For the
first series of dendritic catalysts the complete disappearance of the
broad signal at −23 ppm was accompanied by the appearance of
a new broad signal at +13 ppm. For the second series two signals
in ca. 1 : 1 ratio appeared at −23 and +13 ppm.
1
2
3
4
5
6
7
G0-I
74
43
21
100
80
97
100
100
44
72
42
20
99
80
97
99
99
40
78
G1-I
G2-I
G3-I
G0-II
G1-II
G2-II
G3-II
G1-II
G2-II
8
9c
10c
79
a Reagents and conditions: 0.5 mmol bromobenzene, 2.0 mmol diethy-
lamine, 0.7 mmol triethylamine and 1% catalyst (0.005 mmol Pd) in 2 ml
NMP under 70 psi CO at 150 ◦C, 17 h. b HPLC yield. c Reaction time 6 h.
This is not the case with the strongly chelating bidentate ligands.
Therefore, the dendronized resins do not gain a strong cross-
linking related advantage, as compared to the non-dendronized
one, when decorated with the bidentate ligands. Most likely other
factors associated with the dendronization of the support, such as
reduced pore volume and diminished swelling,2a are responsible for
the differences between the dendritic and non-dendritic supported
catalysts. These properties can definitely lead to the restricted
access of reagents to the pores and, consequently, the decreased
activity in the higher generations.
In the case of the amidocarbonylation reaction (Table 2),
the zeroth generation ligand demonstrated an even stronger
dependence on the degree of complexation than in the Heck
reaction.10,11 In this case we decided to examine both series of
the catalyst, fully complexed Gn-I and partially complexed Gn-
II. In the series Gn-I a negative dendritic effect was observed
up to the second-generation G2-I catalyst (Table 2, entries 1–3).
This trend was drastically altered for the G3-I catalyst, which
led to quantitative conversion (entry 4). A substantially better
performance was demonstrated by the second series of catalysts
(entries 5–8). In this case a positive effect was observed, with
the quantitative conversion for the second- and third-generation
catalysts. At shorter reaction times the dendritic influence is even
more dramatic, as a comparison between G1-II and G2-II in the
6 hour reaction demonstrated (entries 9 and 10).
Usually, Pd leaching, revealed by black Pd particle precipitation,
is observed in both catalytic processes. However, this phenomenon
was minimal for G3-II in the carbonylation reaction and, upon
recycling, the second run yielded 97% of the product. It is even
plausible that high ligand density of the third-generation dendrons
prevented complete precipitation of Pd from the G3-I system,
converting it into a G3-II-like catalyst and, thus, eventually leading
to its high activity in the amidocarbonylation reaction.12 On the
other hand, G0-II recycled from the Heck reaction (the most active
catalyst in the first run) led to substantially lower conversion and
yield in the second run (ca. 10%).
The preliminary studies with the zeroth-generation ligand
(G0(serinol-PPh2)) prepared on the Wang resin through a sequence
similar to that in Scheme 3 demonstrated that partial complexation
with Pd (versus full complexation) somewhat improves the perfor-
mance of the catalytic system in the Heck reaction. Therefore, the
second series (Gn-II) was examined in the Heck arylation of methyl
acrylate with bromobenzene (Table 1). The reaction was carried
out at 110 ◦C for 14 h with the amount of catalytic resin normalized
with respect to Pd equivalents, under conditions similar (though
not equivalent) to those used in the dendritic effect study of the
monodentate phosphine-based catalysts.3b,c The results in Table 1
demonstrate a negative effect of the dendronization on the catalytic
performance of the complexes of the bisphosphine ligand. There
is a notable decrease in the activity of G1-II, G2-II and G3-II
compared with that of G0-II. The activity of the catalysis decreases
steadily with the increase of dendron generation.
The striking difference with the monodentate phosphine system
may point to the source of the dendritic effect in the monodentate
ligand case. We have demonstrated that the catalytic systems are
based on the (phosphine)2Pd complex fragments.3c Clearly in the
monodentate phosphine case each formation of such fragment is a
cross-linking event, unless the two phosphines are chemically con-
nected to the same attachment point (e.g. dendronized support).
Table 1 The Heck reaction of bromobenzene with methyl acrylatea
Entry
Catalyst
Conversion (%)
Yield (%)b
1
2
3
4
G0-II
G1-II
G2-II
G3-II
72
56
48
16
68
52
43
12
The considerable differences between the Heck and carbony-
lation reactions, in regard to the dendritic influence, fall into
line with the differences in the preferred ligand architecture that
we observed for the two reactions catalyzed by the supported
a Reagents and conditions: 0.53 mmol bromobenzene, 0.64 mmol methyl
acrylate, 0.7 mmol triethylamine and 3.75% catalyst (0.02 mmol Pd) in
1 ml NMP at 110 ◦C, 14 h. b HPLC yield.
3384 | Org. Biomol. Chem., 2008, 6, 3382–3387
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 3 Differences in the performance of bisphosphine ligands in the Heck and amidocarbonylation reactionsa
Heck reaction
Carbonylation
Entry
Ligand
Conversion (%)
Yield (%)b
Conversion (%)
Yield (%)b
1
2
3
4
G0-I
65
91
41
61
72
37
96
74
18
74
56
72
16
73
54
4
5
6
100
a Reagents and conditions: Heck reaction: 0.53 mmol bromobenzene, 0.64 mmol methyl acrylate, 0.7 mmol triethylamine and 3.75% catalyst (0.02 mmol
Pd, Pd:P ratio 1 : 2) in 1 ml NMP at 110 ◦C, 14 h. Carbonylation reaction: 0.5 mmol bromobenzene, 2.0 mmol diethylamine, 0.7 mmol triethylamine and
1% catalyst (0.005 mmol Pd, Pd:P ratio 1 : 2) in 2 ml NMP under 70 psi CO at 150 ◦C, 17 h. b HPLC yield.
non-dendritic phosphine–Pd complexes. Ligands 4–6 were pre-
pared on Wang support, complexed with Pd(dba)2, and used
Experimental
General synthesis of Gn(CO2Me)
in catalysis of the two reactions. The results (Table 3) reveal
dramatic differences between the influences of different structural
features of the ligand on the outcome of the two reactions. Thus,
carbonylation catalysis is strongly favored by complexes of six-
membered chelates, while the Heck reaction can also proceed
effectively with ligands forming larger chelate rings (if at all)
(entries 1 and 4). Substituents on the bridge of the six-membered
chelates improve the Heck catalysis, while negatively affecting
the carbonylation reaction (entries 1 and 2). The replacement of
diarylalkylphosphines by strongly electron-donating trialkylphos-
phines substantially improves the carbonylation catalysis, while
it is disadvantageous in the case of the Heck reaction (entries 3
and 4).
A mixture of dimethyl 5-hydroxyisophthalate (10 equiv.) and
LiH (5 equiv.) was stirred in the minimal volume of dry DMF
at room temperature for 5 minutes, and added to a suspension
of the appropriate halomethyl-functionalized resin (1 equiv. of
halomethyl group.) in the minimal volume of dry DMF. TBAI
(3 equiv.) was added and the mixture was stirred at 60 ◦C overnight.
The resin was filtered off and washed with DMF, DMF–water,
THF, THF–water, THF and DCM and dried in vacuum.
Synthesis of Gn(OH)
G1(OH), G2(OH) and G3(OH) were prepared via the reported
procedures.6
General synthesis of Gn(Cl)
Hexachloroethane (5.5 equiv.) and triphenylphosphine (5.5 equiv.)
were added to a suspension of the resin-bound alcohol (1 equiv.
of OH groups) in dry THF (10 ml g−1 resin). The suspension was
mixed at room temperature overnight. The resin was washed with
THF (× 3) and dichloromethane and dried under vacuum.
The common factor of these findings is that strong chelates/high
electron density on Pd are “good” for carbonylation, while they are
not contributing as much, or are even disadvantageous, for Heck
reaction. Plausibly, Pd with a single coordinated phosphine (likely
PdII) is a necessary intermediate in the Heck reaction, though the
ability to easily form the (phosphine)2Pd fragment is also crucial
(probably for the stabilization of Pd0). On the other hand, the
stable bisphosphine chelate–Pd fragment can propagate without
dissociation through the intermediates of the carbonylation cycle.
Though it is difficult to supply direct experimental evidence for
this hypothesis in supported catalysis, the mechanistic study of
the related systems in solution revealed a similar explanation for
the differences in the chelating strength requirement from the
optimal ligand in the Heck and alkoxycarbonylation catalysis of
chloroarenes.13 It is possible that at some stage of the carbonylation
reaction even three phosphine donors are required for stabilizing
the catalytic intermediate (frequently, phosphine excess is used in
homogeneous Pd-catalyzed carbonylation processes14).
G1(Cl). Prepared from G1(OH) (0.54 mmol g−1). Yield 96%,
purity >99%, loading 0.50 mmol g−1. Gel-phase 13C NMR
(100.8 MHz, C6D6): d 158.5, 144.6, 138.7, 120.8, 114.5, 69.2, 45.0.
1
Following TFA-induced cleavage: H NMR (200 MHz, CDCl3–
TFA 1 : 1): d 7.11 (s, 1H), 6.94 (s, 2H), 4.56 (s, 4H). 13C NMR
(100.8 MHz, CDCl3–TFA 1 : 1): d 154.1, 140.0 122.1, 115.4, 45.0.
G2(Cl). Prepared from G2(OH) (0.31 mmol g−1). Yield >99%,
purity >95%, loading 0.30 mmol g−1. Following TFA-induced
cleavage: 1H NMR (200 MHz, CDCl3–TFA 1 : 1): d 7.01–6.82 (m,
9H), 5.28 (s, 4H), 4.56 (s, 8H). 13C NMR (100.8 MHz, CDCl3–TFA
1 : 1): d 160.9, 139.7, 138.7, 122.0, 121.9, 115.3, 114.4, 70.2, 54.3.
G3(Cl). Prepared from G3(OH) (0.28 mmol g−1). Yield 95%,
purity >90%, loading 0.26 mmol g−1. Gel-phase 13C NMR
(100.8 MHz, C6D6): d 158.7, 144.9, 120.8, 118.5, 114.4, 113.0,
69.1, 45.0. Following TFA-induced cleavage: 1H NMR (200 MHz,
CDCl3–TFA 1 : 1): d 7.07–6.91 (m, 21H), 5.13 (broad s, 12H), 4.55
(s, 16H). 13C NMR (100.8 MHz, CDCl3–TFA 1 : 1): d 158.4, 158.5,
139.6, 138.9, 138.5, 121.8, 119.9, 115.3, 114.3, 114.2, 70.2, 70.1,
45.4.
For all these reasons we believe that, in the case of bidentate
phosphine ligands, the high local density of phosphines dictated
by the dendritic architecture is an advantage for the carbonylation
process, while it does not contribute to (or even inhibits) the Heck
catalytic reaction.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3382–3387 | 3385
©

View Article Online
1
General synthesis of Gn(serinol-OH)
d −10.1. Following TFA-induced cleavage: H NMR (200 MHz,
CDCl3–TFA 1 : 1): d 8.03 (br, 1H), 7.53 (s, 2H), 5.70 (d, JHP
=
Serinol (10 equiv.) was added to a suspension of the Gn(Cl) resin
(1 equiv. of Cl groups) in a minimal volume of dry DMF. The
suspension was stirred for 17 hours at 50 ◦C, filtered, the resin
washed with ethanol (× 2) and DCM (× 2) and dried in vacuo.
460 Hz, 4H), 4.39 (t, J = 6.5 Hz, 2H), 4.02 (m, 2H), 3.90 (m, 4H),
2.55 (m, 4H), 1.6 (m, 98H). 13C NMR (100.8 MHz, CDCl3–TFA
1 : 1): d 170.0, 159.7, 134.2, 118.6, 117.5, 69.3, 69.0, 42.8, 29.2,
29.0, 28.8, 28.7, 27.8, 25.7, 25.2. 31P NMR (162 MHz, C6D6):
d 22.5.
General synthesis of Gn(serinol-Cl)
Synthesis of 6
The synthesis followed the procedure for preparation of Gn(Cl).
The general procedure for synthesis of Gn(serinol-PPh2)
was applied on Wang PS-bound 11-(4-(2,6-bis(2-chloroethyl)-
carboxamide)phenyloxy)undec-1-yl (0.49 mmol g−1).15 Yield 90%,
purity >95%, loading 0.38 mmol g−1. Partial gel-phase 13C NMR
(100.8 MHz, C6D6): d 138.2, 136.4, 132.5, 116.0, 115.1, 69.9, 29.4,
25.9. Gel-phase 31P NMR (162 MHz, C6D6): d −23.4.
General synthesis of Gn(serinol-PPh2)
N,N-Diisopropylethylamine (20 equiv.) was added to a suspension
of the Gn(serinol-Cl) resin (1 equiv. of NH groups) in the minimal
volume of dry THF. After 1 hour of stirring of the suspension at
room temperature, the resin was washed with THF. For better
results the procedure was performed twice. The resin was re-
suspended in a minimal volume of THF, in a glove box, and
potassium diphenylphosphide (0.5 M in THF, 20 equiv.) was
added to the suspension of resin. The suspension was stirred for
24 hours. The resin was filtered off and washed with water, acetone,
chloroform and ether and dried in vacuum.
The complexation with Pd(dba)2 and the carbonylation reaction
procedures were reported elsewhere.10
General procedure for the Heck reaction
Bromobenzene (0.056 ml, 0.53 mmol, 1 equiv.), methyl acrylate
(0.057 ml, 0.64 mmol, 1.2 equiv.) and triethylamine (0.1 ml,
0.73 mmol, 1.3 equiv.) were added to the suspension of the
supported complex (0.02 mmol Pd) in 1 ml NMP. The mixture
was stirred for 14 h at 110 ◦C. The suspension was filtered and
washed with acetonitrile. The filtrate was diluted with acetonitrile
and analyzed by HPLC.
G1(serinol-PPh2). Prepared from G1(serinol-Cl) (0.44 mmol
g−1). Yield >99%, purity >99%, loading 0.36 mmol g−1. Partial
gel-phase 13C NMR (100.8 MHz, C6D6): d 139.2, 133.0, 114.2,
69.6, 52.6, 50.8, 36.2. Gel-phase 31P NMR (162 MHz, C6D6):
d −23.3.
G2(serinol-PPh2). Prepared from G2(serinol-Cl) (0.23 mmol
g−1). Yield >99%, purity >95%, loading 0.19 mmol g−1. Gel-phase
31P NMR (162 MHz, C6D6): d −23.4.
Conclusions
In conclusion, we have demonstrated synthesis of and catalysis
with a new type of supported dendritic ligand. The dendritic
influence in this case reflects its multifactor nature. The delicate
balance between the factors depends on the chosen catalytic
process and even on the conditions of the particular reaction.
For the carbonylation reaction it was demonstrated that an
overall negative dendritic effect could be turned into a positive
one. The Heck reaction catalysis with the bidentate ligands
displayed an overall negative dendritic effect. This finding supports
the hypothesis of the reduced cross-linking as the main factor
responsible for the positive dendritic effect, observed previously
for supported dendritic monodentate phosphine-based catalysts.
G3(serinol-PPh2). Prepared
from
G3(serinol-PPh2)
(0.11 mmol g−1). Yield 50%, loading 0.043 mmol g−1. Gel-
phase 31P NMR (162 MHz, C6D6): d −23.4. TFA-induced
cleavage: 1H NMR (200 MHz, CDCl3–TFA 1 : 1): d 7.87–7.67 (m,
160H, Ph), 7.00–6.70 (m, 28H), 4.90–3.90 (m, 68H).
Synthesis of 4
The synthesis followed the route reported earlier for G0(serinol-
PPh2).8 Yield >99%. Partial gel-phase 13C NMR (100.8 MHz,
C6D6): d 139.8, 132.7, 56.2, 27.6, 26.0, 16.1. Gel-phase 31P NMR
(162 MHz, C6D6): d −24.5.
Acknowledgements
Synthesis of 5
This research was supported by the Israel Science Foundation. The
fellowship of A. M. by the Planning & Budgeting Committee of the
Israel Council for Higher Education is gratefully acknowledged.
In a glove box, a solution of tert-butyllithium (10% in pentane,
3.2 ml, 4.9 mmol, 10 equiv.), was added slowly to a solution of
dicyclohexylphosphine (7.4 mmol, 15 equiv.) in dry THF. The
mixture was stirred for 3 hours, the solvent was evaporated and the
resulting salt was dissolved in a minimal amount of dry THF. This
solution was added to the suspension of Wang PS-bound 11-(4-
(2,6-bis(2-chloroethyl)carboxamide)phenyloxy)undec-1-yl (1.0 g,
0.49 mmol g−1, 0.98 mmol of Cl groups, 4 equiv.) in 2 ml THF and
stirred overnight.15 The resin was filtered off (in the hood), washed
with water, acetone, chloroform and ether, and dried in vacuum.
Yield 95%, purity >95%, loading 0.40 mmol g−1. Partial gel-phase
13C NMR (100.8 MHz, C6D6): d 167.3, 114.6, 69.8, 56.8, 40.6, 33.2,
29.4, 28.9, 27.2 26.4, 22.2. Gel-phase 31P NMR (162 MHz, C6D6):
References and notes
1 For general reviews on dendritic catalysis, see: (a) D. Astruc and F.
Chardac, Chem. Rev., 2001, 101, 2991; (b) R. van Heerbeek, P. C. J.
Kamer, P. W. N. M. van Leeuwen and J. N. H. Reek, Chem. Rev., 2002,
102, 3717; (c) R. Kreiter, A. W. Kleij, R. J. M. Klein Gebbink and G.
van Koten, Top. Curr. Chem., 2001, 217, 163.
2 For reviews on dendritic catalysis on solid supports, see: (a) A. Dahan
and M. Portnoy, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 235;
(b) T. Kehat, K. Goren and M. Portnoy, New J. Chem., 2007, 31, 1218.
3 (a) A. Dahan and M. Portnoy, Chem. Commun., 2002, 2700; (b) A.
Dahan and M. Portnoy, Org. Lett., 2003, 5, 1197; (c) A. Dahan and
3386 | Org. Biomol. Chem., 2008, 6, 3382–3387
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
M. Portnoy, J. Am. Chem. Soc., 2007, 129, 5860; (d) H. Alper, P. Arya,
S. C. Bourque, G. R. Jefferson and L. E. Manzer, J. Am. Chem. Soc.,
2001, 123, 2889; (e) J. Bu, Z. M. A. Judeh, C. B. Ching and S. Kawi,
Catal. Lett., 2003, 85, 183; (f) Y.-M. Chung and H.-K. Rhee, Chem.
Commun., 2002, 238; (g) T. Kehat and M. Portnoy, Chem. Commun.,
2007, 2823.
4 For general reviews on the Heck reaction, see: (a) I. P. Beletskaya and
A. V. Cheprakov, Chem. Rev., 2000, 100, 3009; (b) N. J. Whitcombe,
K. K. Hii and S. E. Gibson, Tetrahedron, 2001, 57, 7449; (c) A. M.
Trzeciak and J. J. Ziolkowski, Coord. Chem. Rev., 2007, 251, 1281.
5 For general reviews on the carbonylation reaction, see: (a) Y. V.
Gulevich, N. A. Bumagin and I. P. Beletskaya, Russ. Chem. Rev., 1988,
57, 299; (b) R. Skoda-Foldes and L. Kollar, Curr. Org. Chem., 2002, 6,
1097.
7 A. Dahan and M. Portnoy, Macromolecules, 2003, 36, 1034.
8 A. Mansour and M. Portnoy, Tetrahedron Lett., 2003, 44, 2195.
9 B. Ben-Aroya Bar-Nir and M. Portnoy, Tetrahedron, 2002, 58,
5147.
10 A. Mansour and M. Portnoy, J. Mol. Catal. A: Chem., 2006, 250,
40.
11 With the monodentate phosphine dendritic systems, only traces of the
product were observed under similar conditions.
12 Precipitation of Pd in the fully complexed systems is much more severe
than in the partially complexed ones; see ref. 10.
13 (a) Y. Ben-David, M. Portnoy and D. Milstein, J. Am. Chem. Soc.,
1989, 111, 8742; (b) Y. Ben-David, M. Portnoy, M. Gozin and D.
Milstein, Organometallics, 1992, 11, 1995; (c) M. Portnoy, Y. Ben-
David and D. Milstein, Organometallics, 1993, 12, 4734; (d) M. Portnoy,
Y. Ben-David, I. Rousso and D. Milstein, Organometallics, 1994, 13,
3465.
6 For other supported dendritic catalysts for the Heck and carbonylation
reactions, see: (a) H. Alper, P. Arya, S. C. Bourque, G. R. Jefferson and
L. E. Manzer, Can. J. Chem., 2000, 78, 920; (b) S. Antebi, P. Arya, L. E.
Manzer and H. Alper, J. Org. Chem., 2002, 67, 6623; (c) S. M. Lu and
H. Alper, J. Am. Chem. Soc., 2005, 127, 14776.
14 W. Magerlein, A. F. Indolese and M. Beller, J. Organomet. Chem., 2002,
641, 30.
15 A. Weissberg and M. Portnoy, Chem. Commun., 2003, 1538.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3382–3387 | 3387
©