SCS-pincer palladium-catalyzed auto-tandem catalysis using dendritic catalysts in semi-permeable compartments

Article abstract of DOI:10.1039/c1dt10502g

Novel monometallic and dendritic SCS-pincer palladium complexes 2, 3 and 4 have been synthesized in good yields (60-89%) and high purity (palladium loading >97%). These complexes were successfully used as catalysts in the stannylation of cinnamyl chloride

Full text of DOI:10.1039/c1dt10502g

View Article Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
Pincers and other hemilabile ligands
Guest Editors Dr Bert Klein Gebbink and Gerard van Koten
Published in issue 35, 2011 of Dalton Transactions
Image reproduced with permission of Jun-Fang Gong
Articles in the issue include:
PERSPECTIVE:
Cleavage of unreactive bonds with pincer metal complexes
Martin Albrecht and Monika M. Lindner
Dalton Trans., 2011, DOI: 10.1039/C1DT10339C
ARTICLES:
Pincer Ru and Os complexes as efficient catalysts for racemization and deuteration of alcohols
Gianluca Bossi, Elisabetta Putignano, Pierluigi Rigo and Walter Baratta
Dalton Trans., 2011, DOI: 10.1039/C1DT10498E
Mechanistic analysis of trans C–N reductive elimination from a square-planar macrocyclic aryl-
copper(III) complex
Lauren M. Huffman and Shannon S. Stahl
Dalton Trans., 2011, DOI: 10.1039/C1DT10463B
CSC-pincer versus pseudo-pincer complexes of palladium(II): a comparative study on
complexation and catalytic activities of NHC complexes
Dan Yuan, Haoyun Tang, Linfei Xiao and Han Vinh Huynh
Dalton Trans., 2011, DOI: 10.1039/C1DT10269A
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 8896
www.rsc.org/dalton
PAPER
SCS-pincer palladium-catalyzed auto-tandem catalysis using dendritic
catalysts in semi-permeable compartments
Niels J. M. Pijnenburg, Harm P. Dijkstra, Gerard van Koten and Robertus J. M. Klein Gebbink*
Received 24th March 2011, Accepted 3rd June 2011
DOI: 10.1039/c1dt10502g
Novel monometallic and dendritic SCS-pincer palladium complexes 2, 3 and 4 have been synthesized in
good yields (60–89%) and high purity (palladium loading >97%). These complexes were successfully
used as catalysts in the stannylation of cinnamyl chloride with hexamethylditin and in the catalytic
auto-tandem reaction consisting of this stannylation followed by an electrophilic addition with
4-nitrobenzaldehyde, showing similar reaction rates and selectivities for all complexes. Dendritic
complex 4 has furthermore been used in the compartmentalized catalysis of single and auto-tandem
reactions, allowing catalyst reuse for four consecutive runs.
catalytic activity is linked to location. Recent publications by
Introduction
Rothenberg and Vogt report on dendritic catalysts in combination
Immobilization of homogeneous catalysts on a solid or macro-
with solution-phase compartmentalized catalysis in tailor-made
molecular soluble support1–7 allows for the separation of catalysts
from reaction mixtures by (membrane) filtration techniques and
enables catalyst recycling and continuous catalytic processes, e.g. in
membrane reactors. The use of dendrimers as support permits con-
trol over the number of catalyst units on each single macromolecu-
lar entity. Dendrimers have excellent solubility properties and tend
to dissolve better in organic solvents than their linear and cross-
linked polymeric analogues.8 The controlled, molecular synthesis
of dendrimers provides access to monodisperse materials, which
further enhances their solubility and separation properties. The
first example of a functionalized dendrimer containing multiple
peripheral metal catalysts was published in 1994 by Van Koten
and Van Leeuwen9 and since then the field of dendrimer catalysis
has further developed. Several comprehensive reviews on dendritic
catalysts, their use in synthesis and their separation and reuse have
been reported.1,2,4,10–18
The possibility of performing sequential or tandem catalysis, i.e.
a sequence of chemical reactions in which every transformation
is catalyzed, with a single or several dendritic catalysts has been
pointed out in an early stage,15 but no experimental reports on this
topic have been published so far. We are interested in such systems,
because the ability to carry out multiple sequential catalyzed
reaction steps in combination with the possibility of separating and
recycling the costly catalysts creates an interesting reaction setup
in view of process intensification. The various aspects of tandem
catalysis are nicely compiled in a review by Fogg and Dos Santos.19
One approach towards tandem catalysis using multiple catalysts
is the use of compartmentalized catalysts that prohibit disadvan-
tageous effects between different types of catalysts and in which
membrane reactors.20–22 Also in heterogeneous23 and enzyme
catalysis24 similar systems have been described. We have chosen
to investigate a reaction setup that does not require sophisticated
membrane reactors. We therefore set out to investigate the use
of closed membrane dialysis bags filled with dendritic catalysts.
Catalytic (tandem) reactions can be performed using such semi-
permeable compartments by inserting one or several of these
compartments in a single reaction mixture. After reaction the
dialysis bag(s) can be easily removed from the reaction mixture
and in principle be placed into a fresh reaction solution. Proofs
of concept for such a ‘teabag’ approach have been reported earlier
by us25,26 and more recently by Gade.27
A prerequisite for the separation and recycling of dendritic
catalysts is that the active catalytic centers should be tightly bound
to the dendritic support to prevent catalyst leaching. Because of
its robust metal-carbon bond, dendritic ECE-pincer complexes are
good catalyst candidates in this respect. Many studies on ECE-
pincer metal complexes (ECE = C6H3(CH2E)2-2,6]-; E = NR2,
PR2, SR, etc., see Fig. 1) in which the metal ions are bound to the
pincer ligand via a covalent M–C bond and two coordinative M–
E bonds have been reported.28–31 Amongst others, we have shown
that dendritic ECE-pincer complexes can indeed be separated by
membrane separation techniques and reused without significant
metal leaching for several runs.25,32
Fig. 1 A pincer metal complex. M = metal (Ni, Pd, Pt, etc.), X = co-ligand
(e.g. Cl, Br, I, NCMe, OH2), E = electron donating groups (NMe2, PPh2,
SPh, etc.).
Organic Chemistry & Catalysis, Debye Institute for Nanomaterials Science,
Utrecht University, Universiteitsweg 99, 3584, CG Utrecht, The Netherlands.
E-mail: r.j.m.kleingebbink@uu.nl; Fax: +31-30-252-3615
8896 | Dalton Trans., 2011, 40, 8896–8905
This journal is
The Royal Society of Chemistry 2011
©

In a recent collaboration with the group of Szabo´, we have
found that two independent reactions that are catalyzed by two
different pincer metal complexes33,34 (Fig. 2, reaction 1 and 2),
can in fact be catalyzed in a consecutive, one pot manner by
a single, different pincer metal complex35,36 (Fig. 2, reaction
3). The stannylation of cinnamyl chloride with hexamethylditin
to form cinnamyl trimethylstannane (reaction 1) is typically
catalyzed by NCN-pincer Pd-complexes and the homoallylation
of 4-nitrobenzaldehyde by cinnamyl trimethylstannane to form
1-(4-nitrophenyl)-2-phenyl-3-buten-1-ol (reaction 2) is typically
catalyzed by PCP-pincer Pd-complexes. In contrast, monometallic
SCS- and PCS-pincer palladium-complexes are able to catalyze
both of those mechanistically different reactions. In this catalytic
auto-tandem reaction (reaction 3) all three starting materials are
present in the reaction mixture from the offset of the reaction and
the reaction is carried out in one pot. In the presence of 5 mol%
SCS-pincer palladium-catalyst, the homoallylic alcohol products
◦
are formed in 74% yield after 16 h at 40 C.35 A diastereomeric
product ratio of 9 : 1 favoring the anti diastereomers was observed.
Here, we present our investigations on a dendrimer supported
version of the SCS-pincer Pd auto-tandem catalyst and have
explored its use in a compartmentalized set-up. The objectives of
our investigations were twofold. First, a first generation dendritic
catalyst was to be tested in solution and its activity compared
to those of its monometallic and its zeroes generation dendritic
analogues. For this purpose a series of four catalysts was prepared
(Fig. 3). This series consists of two monometallic analogues
(1 and 2) and two dendritic SCS-pincer palladium-complexes
Fig. 2 The NCN-pincer palladium-catalyzed stannylation of cinnamyl chloride to cinnamyl trimethylstannane (reaction 1) and the PCP-pincer
palladium-catalyzed electrophilic cross-coupling of cinnamyl trimethylstannane and 4-nitrobenzaldehyde to 1-(4-nitrophenyl)-2-phenyl-3-buten-1-ol
(reaction 2) can be combined to a SCS- or PCS-pincer palladium-catalyzed auto-tandem reaction where all substrates are present from the beginning of
the reaction (reaction 3).
Fig. 3 Monomeric and dendritic SCS-pincer palladium-complexes 1–4 used as catalyst in the tandem reaction.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 8896–8905 | 8897
©

(3 and 4). Compound 2 contains a para-functionalized trimethylsi-
lyl group that resembles the connectivity of the pincer moieties to
the dendritic scaffold. As a second objective, dendritic catalyst 4
was tested in a compartmentalized set-up and reused several times
to evaluate its stability and recyclability.
ated 5, chloro-terminated carbosilane dendrimers 7 and 8 were
used (Fig. 5). These dendritic scaffolds where synthesized ac-
cording to a standard literature procedure via a repetitive
Grignard/hydrosilylation reaction sequence starting from tetra-
chlorosilane and allylbromide.40
Synthesis of compound 1–4
Parent SCS-pincer palladium complex 1 has been reported by
Sillanpa¨a¨ and co-workers and was synthesized as described.37 The
trimethylsilyl-functionalized complex 2 was synthesized starting
from the para-bromo functionalized pincer ligand precursor 5 in
a two-step synthetic protocol (Fig. 4). Ligand precursor 5 was
lithiated via lith◦ium-halogen exchange by addition of 2 equiv.
of tBuLi at -80 C and subsequent treatment with trimethylsilyl
chloride. After workup this procedure gave pincer ligand precursor
6 in 96% isolated yield. Palladation of 6 was achieved via direct
C–H-activation with [Pd(MeCN)4](BF4)2 in refluxing acetonitrile.
The resulting cationic acetonitrile complex was treated with
a saturated aqueous sodium chloride solution to replace the
acetonitrile ligand by the stronger chloride ligand. Although
cationic SCS-pincer Pd(MeCN)-complexes have shown to be very
reactive in the stannylation reaction33 and in the tandem reaction,35
our catalytic studies have been performed with the neutral SCS-
pincer Pd-Cl-complexes, because of the poor solubility properties
of the polycationic SCS-pincer Pd(MeCN) dendrimers and the
improved storage properties of the Pd-Cl dendrimers.
Fig. 5 Synthesis of carbosilane-based dendritic SCS-pincer palladium–
complexes 3 and 4.
An excess of lithiated 5 (1.2 equivalents per Si–Cl moiety) was
used to ensure a complete conversion of all chlorosilyl groups
in order to yield monodisperse dendritic materials. The excess of
pincer ligand precursor (and eventual other low molecular weight
contaminants) was removed by passive dialysis. After three dialysis
cycles, the dendritic ligand precursors were isolated in good yields
(76% for dendrimer 9, and 61% for dendrimer 10) and high purity
(vide infra).
Dendrimers 9 and 10 were palladated by using an excess
of [Pd(MeCN)4](BF4)2 (1.2 equivalents per dendritic arm) in
a refluxing mixture of acetonitrile and toluene. This solvent
combination was used to improve the solubility of the dendritic
ligands. A treatment with NaCl resulted in the chloride complexes.
After passive dialysis and PVPy treatment, orange–yellow powders
were obtained in reasonable to good yields (83% for G0 dendrimer
3, 60% for G1 dendrimer 4).
Fig. 4 Synthesis of trimethylsilyl functionalized SCS-pincer palladium
complex 2.
The integrity and purity of palladium complexes 1–4 was
1
verified by means of a combination of analytical techniques. H
The resulting SCS-pincer Pd-complex was purified by col-
umn chromatography, and further treated with 100 equiv. of
poly(vinylpyridine) (PVPy) in dichloromethane for 3 h. PVPy is
known to intercept free palladium particles in solution and is
regularly used in catalyst poisoning experiments.38,39 By doing so,
minute amounts of potentially catalytically active Pd(0) species
were excluded from palladium complexes that were subsequently
used in catalysis experiments. After this workup, 2 was obtained
as an air and moisture stable orange–yellow solid in 89% yield.
Dendritic complexes 3 and 4 were synthesized via a sim-
ilar protocol as trimethylsilyl-pincer complex 2. Instead of
using trimethylsilyl chloride as the quenching agent for lithi-
NMR analysis of the complexes showed a single characteristic
signal for the benzylic protons around 4.6 ppm. The introduction
of the palladium center caused a large downfield shift for this
signal compared to the corresponding pre-ligands, which show
this resonance at 4.0 ppm. Besides the chemical shift of this signal,
also its line width differs significantly for the palladium complexes
compared to the pre-ligands. A sharp singlet was observed for the
ligand precursors 6, 9 and 10, whereas a broad to very broad singlet
was seen for the corresponding SCS-pincer palladium-complexes
2–4. This observation is in agreement with observations in the
literature, and is caused by the flexible structure of SCS-pincer Pd-
complexes due to a combination of inversion of the conformation
8898 | Dalton Trans., 2011, 40, 8896–8905
This journal is
The Royal Society of Chemistry 2011
©

of the sulfur atoms, and puckering of the S–Pd–C chelate.41,42
A
Table 2 Formation of cinnamyl trimethylstannane (%) in the stannyla-
tion of cinnamyl chloride as catalyzed by complexes 1–4a
similar shift of the benzylic signals was observed in the 13C NMR
spectra; for all reported species the signal corresponding to the
benzylic carbon shifts from 39 ppm for the ligand precursor to
52 ppm for the palladium complex. 1H as well as 13C NMR spectra
of complexes 7–10 did not show any residual ligand precursor
signals. The dendritic ligands and complexes show broad signals
for all observed protons, in agreement with their macromolecular
nature.
Catalyst
1 h, CH2Cl2
5 h, CH2Cl2
1 h, THF
5 h, THF
1
2
3
4
58
71
69
55
99
83
82
83
82
100
100
100
100
100
100
99
a Conditions: 0.80 mmol cinnamyl chloride, 0.80 mmol hexamethylditin
and 2 mol% Pd catalyst in 6 mL CH2Cl2 or THF, ambient temperature,
N2 atmosphere.
G0 ligand precursor 9 and complex 3 were successfully analyzed
by MALDI TOF MS. For 9, a parent peak at m/z = 1714.69 was
observed (theoretical value for [M+H]+ is m/z = 1714.94). Analysis
of complex 3 showed a distinct signal at m/z = 2478.73 for a tetra-
palladium species (theoretical value for [M+Na]+ is 2479.19). No
signals corresponding to species of lower palladium content or
to the free ligand were observed. For the G1 ligand 10 and G1
complex 4 the conditions for a good mass analysis were not found.
Changing to ESI-MS also did not result in a proper analysis.
Besides the absorption around 250 nm that is typical of
compounds that contain aromatic groups, SCS-pincer palladium
complexes show a characteristic UV absorption at 330 nm. The
corresponding pre-ligands do not show an absorption in this re-
gion (e330 < 0.001). This specific optical feature was therefore used
to determine the Pd content of the isolated dendritic materials.
Analytically pure trimethylsilyl functionalized complex 2 was used
as a calibrant. Separate dilution series of monometallic complex
2 and dendritic complexes 3 and 4 in CH2Cl2 were prepared by
using an equal concentration of palladium centers (theoretical
value). The absorption at 330 nm for these dilution series was
plotted against their theoretical Pd-concentration. The ratio of
the slopes of these straights for 3 and 4 to the slope of the straight
of 2 were taken as a measure of the percentage of palladium
centers in the dendritic complexes. These experiments showed an
average of 3.9 and 11.9 metalated arms per molecule, respectively,
thus showing a full palladium loading of both 3 and 4 (Table 1).
Metalations of identical dendritic compounds bearing NCN-type
pincer ligands were performed via stepwise lithiation of the pincer
ligands with t-BuLi followed by transmetalation with a metal(II)
precursor.32 Although these reactions appeared to be high yielding
for momomeric pincer complexes, for dendritic complexes no
complete metal loading was achieved due to partial hydrolysis of
the extremely sensitive lithio-intermediate. Metalation percentages
around 80–90% were generally observed for these NCN-pincer
dendrimers. To the best of our knowledge, the present SCS-pincer
dendrimers are the first to show full metalation for these types of
structures.
Catalysis with 1–4 in solution
Dendrimers 3 and 4 and their mononuclear counterparts 1 and
2 were tested as catalysts in the stannylation reaction (reaction
1) and in the two-step tandem reaction (reaction 3). In both
reactions, the substrates were combined in either THF or CH2Cl2
in equimolar amounts. These solutions were found to be stable for
a prolonged time without catalyst: no blank reactions took place.
The catalysts were added at 2 mol% palladium (i.e. 2 mol% of 1 or
2, 0.5 mol% of 3 or 0.167 mol% of 4). A nitrogen atmosphere
was necessary because of the sensitivity towards hydrolysis of
hexamethylditin, and due to the lability of the intermediate
cinnamyl trimethylstannane.
Both the monometallic and the dendritic complexes were found
to be excellent catalysts for the stannylation of cinnamyl chloride
(Table 2). Complete conversion was obtained in all cases in less
than 5 h, showing very similar reaction kinetics. In CH2Cl2 the
reactions were complete in 5 h, with a conversion of 55–71% in the
first hour. In THF a similar trend was observed, although after
1 h the conversion is significantly higher (~83%).
In the tandem reaction (reaction 3) in THF, it was found that the
overall activity after 24 and 72 h was very similar for compounds
1–4 (Table 3). Initially, a fast decrease of cinnamyl chloride was
observed within 5 h (Fig. 6), next to a fast increase of cinnamyl
trimethylstannane. The second reaction step was much slower
and reached full conversion only after several days. The dendritic
catalysts 3 and 4 proved to be more effective than the monomeric
catalysts in the first reaction step. For the overall tandem reaction,
the ‘head start’ of the dendritic catalysts was averaged out due
to the much longer reaction time that was required for the
second reaction step. Overall, monomeric catalyst 1, which lacks
a trimethylsilyl group on the para-position with respect to the
palladium center, appeared to be slightly slower than the other
catalysts used. The ratio between anti and syn products at the end
of the reaction showed a slight dependence on the catalyst, with the
dendritic catalysts being somewhat more selective for the anti 1-(4-
nitrophenyl)-2-phenyl-3-buten-1-ol products. The anti/syn ratio
varies from 5 for the monometallic species to 6 for the dendritic
catalysts.
Table 1 Determination of the palladium content of complexes 3 and 4
based on absorption intensities (330 nm) of dilution seriesa
Slope of
dilution
Catalyst series
R2 value of
dilution
series
% of palla- Calc. nr. of
dated arms palladated arms
Compartmentalized catalysis with dendritic complex 4
2
3
4
0.0074700
0.0072627
0.0074227
0.9972
0.9988
0.9991
100
97.22
99.37
1
3.9
11.9
Next, the catalytic performance of first generation metalloden-
dritic catalyst 4 was investigated in a compartmentalized setting.
For this purpose, catalyst 4 was placed inside a semi-permeable
compartment and tested in the stannylation reaction 1 and
in tandem reaction 3. As the semi-permeable compartment a
a The dilution curve of complex 2 was used as a standard and the Pd-
content of 3 and 4 were calibrated accordingly.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 8896–8905 | 8899
©

Table 3 Substrate conversion, intermediate built-up, and product formation in tandem reaction 3 catalyzed by 1–4 in THFa
Cinnamyl chloride (%) Cinnamyl trimethylstannane (%) Homo-allyl alcohol (%)
1 h
24 h
72 h
1 h
24 h
72 h
1 h
24 h
72 h
Anti/Syn ratio
1
2
3
4
37
44
29
15
1
1
0
0
0
0
0
0
38
36
39
53
61
50
58
56
21
6
12
6
25
20
32
32
38
49
42
44
79
94
88
94
5.0
5.2
5.7
6.0
a Conditions: 0.80 mmol cinnamyl chloride, 0.80 mmol hexamethylditin, 0.80 mmol 4-nitrobenzaldehyde and 2 mol% Pd catalyst in 6 mL THF at ambient
temperatures, N2 atmosphere.
Stannylation of cinnamyl chloride with hexamethylditin
(reaction 1)
Initial studies with compartmentalized dendritic catalyst 4 were
performed in the palladium-catalyzed stannylation reaction of
cinnamyl chloride. To this end, a membrane dialysis bag con-
taining a solution of dendritic catalyst 4 in THF was placed
into a solution containing the substrates. After 24 h of reaction,
the membrane dialysis bag was removed and placed into a
second vessel containing a new batch of substrates in THF. The
presence of the semi-permeable barrier had a large effect on the
reaction rate. Not only was a significant lag phase in substrate
conversion observed, the reaction also took considerably longer
Fig. 6 Conversion of cinnamyl chloride to cinnamyl trimethylstannane
catalyzed by Pd-catalysts 1–4 in the tandem setup.
to go to completion (24 h) compared to the reaction under
standard homogeneous conditions (5 h). Diffusion of substrates
into the membrane bag and of products out of the membrane bag
driven by osmotic pressure seems an obvious reason why a lower
membrane dialysis bag (benzoylated regenerated cellulose dialysis
reaction rate was observed under compartmentalized conditions,
tubing) was used. The mass weight cut-off (MWCO) of this
as product sampling was carried out from the outer membrane
membrane is 1000 Da, whereas the molecular weight of the
phase.
dendritic catalyst is 7593 Da. As shown by UV/Vis analysis,
The same catalyst batch was used in four consecutive runs by
no detectable dendrimer leaching took place in CH2Cl2 or THF
removal of the membrane dialysis bag from the reaction solution
for a week when the dendritic catalyst was placed inside such a
after 24 h and placement of the bag into a new batch of substrates.
membrane dialysis bag. For these studies a reaction vessel was
The compartmentalized catalyst remained active over the four
used that is equipped with a glass raster at approximately 1 cm
runs, although the reaction rate steadily decreased in every run.
above the bottom of the reactor in order to protect the membrane
Substrate conversions of 100%, 95%, 72%, and 61%, respectively,
dialysis bag from being damaged by the rotating stirring bar at the
were observed for the four runs. The fourth run was prolonged to
bottom of the reactor (Fig. 7).
72 h of reaction time and a conversion of 99% was observed at this
time, indicating the endured activity of the dendritic catalyst.
The palladium contents of the outer membrane solution of
these four runs were analyzed via ICP mass spectroscopy in
order to probe possible palladium leaching from the membrane
bag. Various amounts of palladium were found in all four runs
(Table 4). In the first run 13 ppm Pd was detected, corresponding to
5.2% of the total starting amount of palladium. In the second, third
and fourth runs lower amounts of palladium were found. In these
Table 4 Pd ICP MS analysis of the outer membrane solution in the
consecutive formation of cinnamyl trimethylstannane using compartmen-
talized catalyst 4
Palladium (ppm)
Leaching (%)
Run 1
Run 2
Run 3
Run 4
13
10
5
5.2
4.0
2.0
1.2
Fig. 7 Reactor used for compartmentalized (tandem) catalysis experi-
ments. A dialysis membrane bag closed by two white clamps contains the
catalyst solution.
3
8900 | Dalton Trans., 2011, 40, 8896–8905
This journal is
The Royal Society of Chemistry 2011
©

Table 5 Substrate conversion, intermediate built-up, and product formation in tandem reaction 3 catalyzed by dendritic catalyst 4 inside a membrane
dialysis bag in THFa
Cinnamyl chloride (%)
Cinnamyl trimethylstannane (%)
Homo-allyl alcohol (%)
1 h
24 h
168 h
1 h
24 h
168 h
1 h
24 h
168 h
Anti/Syn ratio
Run 1
Run 2
Run 3
Run 4
48
47
88
83
0
0
14
44
0
0
4
45
39
36
5
56
54
59
33
36
49
43
23
13
17
7
44
46
29
23
64
51
53
32
4.2
3.8
1.9
1.9
11
6
a Conditions: 8.0 mmol cinnamyl chloride, 8.0 mmol hexamethylditin, 8.0 mmol 4-nitrobenzaldehyde and 2 mol% Pd catalyst inside a membrane dialysis
bag in 60 mL THF at ambient temperatures and N2 environment.
four runs, a total of 12.4% of the starting amount of palladium
had leached through the membrane into the outer solution.
After the fourth run, complex 4 was regained from the dialysis
bag in 89% yield. NMR analysis showed that 4 was not regained
unmodified. Analysis of the peaks corresponding to the benzylic
protons showed that a significant amount of the SCS-pincer
moieties (approximately 30%) no longer contained a palladium
center and had been transformed in a preligand form. Apparently,
significant amounts of palladium were released from the pincer
ligand manifold during catalysis via overall protonolysis. The
difference between the amount of Pd leaching as determined by
ICP MS and by NMR analysis (12.4% and 30% respectively) might
be caused by agglomeration of released palladium into larger
clusters inside the membrane dialysis bag or by adsorption of
Pd by the membrane itself.
Upon further use of the same compartmentalized catalyst in
three additional consecutive runs by replacing the contents of
the reaction vessel by a fresh batch of substrates after 1 week,
it was found that the catalyst was active in all of these runs.
This might be somewhat surprising taking into account that the
electrophilic addition of the first run did not reach complete
conversion. Actually, a very similar reaction profile was found
for the second run as for the first run. The rate of conversion and
of product formation were almost identical in the first two runs
(Table 5). The amount of tandem products found in the second
run was even slightly higher compared to the first run, which
is most probably caused by the remaining amount of product
in the dialysis bag present after the first run. The anti/syn ratio
of the products in run 1 (4.2 : 1) and 2 (3.8 : 1) was found to be
similar.
Possible leaching of dendritic catalyst 4 itself from the mem-
brane bag was tested in the presence of each individual substrate
in THF under catalytic conditions (i.e. 2 mol% Pd per substrate
and identical concentration). In none of these experiments any
palladium leaching was observed ([Pd] < 0.1 ppm). This leads to
the conclusion that only under true catalytic conditions, i.e. in the
presence of both substrates, palladium leaching took place.
In the third run, the first step of the tandem reaction was
considerably slower than in the first two runs. Consequently, the
second step showed a reduced reaction rate in the initial stage.
However, after one week the same amount of tandem product
had formed as compared to the second run. In the fourth run a
considerable decrease in reaction rate for both reaction steps was
observed: half of the starting amount of cinnamyl chloride was still
present after one week, and tandem product formation proceeded
to only 37%. The anti/syn product ratio in run 3 and run 4 (1.9 : 1)
differed significantly from that in the previous runs.
Tandem stannylation and electrophilic addition (reaction 3)
ICP MS analysis was again carried out to probe the palladium
concentration in the outer solutions after 1 week of reaction
(Table 6). The amount of palladium observed in these cases was
higher than for the single step stannylation reaction, which might
be caused by the longer reaction times that are necessary for the
tandem catalysis reaction. In the first run a considerable 35.8 ppm
of palladium leached out of the membrane bag, which corresponds
to 14.3% of the total amount of palladium centers. Pd-leaching in
the three subsequent runs decreased to 1.6 ppm in run 4.
After the single step stannylation reaction, the compartmen-
talized dendritic catalyst 4 was investigated in the tandem
stannylation/homo-allylation reaction. The compartmentalized
tandem reaction was found to progress more slowly than the stan-
nylation reaction: NMR-monitoring of the reaction was carried
out for one week whereupon the dialysis bag was transferred into
a new batch of substrates.
In the first run, the stannylation reaction took place readily
and went to completion within 24 h (Table 5). The second
step of the tandem reaction, i.e. the electrophilic substitution of
nitrobenzaldehyde and cinnamyl trimethylstannane was found to
progress at a much lower rate: noticeable amounts of cinnamyl
trimethylstannane were still present after 2 days of reaction. In
fact, also after a prolonged reaction time of one week the reaction
had not gone to completion. At that point a steady state mixture
of tandem products (60%) and cinnamyl trimethylstannane (40%)
was observed. An anti/syn product ratio of 4.2 : 1 was found
in this mixture, which is a somewhat lower ratio than for the
homogeneous reaction (6.0 : 1).
Table 6 Pd-analysis (ICP MS analysis) of the outer membrane solution
for four consecutive compartmentalized runs of reaction 3 using complex
4
Palladium (ppm)
Leaching (%)
Run 1
Run 2
Run 3
Run 4
35.8
23.1
5.7
14.3
9.3
2.3
1.6
4.1
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 8896–8905 | 8901
©

Table 7 Formation of secondary product and Pd ICP MS analysis of
the outer membrane solution for the compartmentalized tandem catalysis
experiment after 2 h under optmised conditionsa
ol (reaction 3). This particular auto-tandem reaction was studied
in a standard batch manner as well as in a compartmentalized
manner. We found that compartmentalized tandem catalysis is
indeed possible using simple commercially available dialysis bags
in a tea-bag approach. These studies provide the first example for
compartmentalized auto-tandem catalysis using a semi-permeable
compartment in which a molecularly enlarged catalyst is retained.
Upon reuse of a catalyst-loaded membrane bag, the catalytic
reaction was found to take place with a similar reaction profile.
However, upon further reuse, i.e. in a third and fourth catalytic
run, the reaction rate slowed down, while still significant amounts
of products were formed. On the other hand, complex 4 was used
for one month in this recycling experiment, which shows an overall
stability of the organometallic complex.
Product
formation (%)
Anti/Syn
ratio
Palladium
(ppm)
Observed
leaching (%)
Run 1
Run 2
Run 3
Run 4
89
96
96
92
3.9
4.0
3.8
3.7
0
0
3
8
0
0
1
3
a Conditions are equal to those used in Table 5, except that 3 equiv. of
cinnamyl chloride (24 mmol) and 3 equiv. of hexamethylditin (24 mmol)
have been used.
In addition, these studies showed an anticipated drawback of
the membrane bag setup. When relying on passive diffusion for
substrate and product transport, overall reaction rates tend to
be low. The use of excess amounts of only two of the reaction
substrates, however, was found to considerably speed up the
compartmentalized tandem reaction.
Optimization of reaction conditions
Preliminary results from mechanistic studies aimed at optimiz-
ing the conditions to perform compartmentalized auto-tandem
catalysis showed a very fast tandem reaction takes place when
three (or more) equivalents of cinnamyl chloride and hexam-
ethylditin are used compared to 4-nitrobenzaldehyde.43 Using
these reagent excess conditions, dendritic SCS-pincer palladium
complex 4 (2 mol% Pd) was investigated in a compartmental-
ized setup to explore the possibility of performing recyclable
auto-tandem catalysis. This modification led to a dramatic
improvement in reaction time: an almost full conversion to
1-(4-nitrophenyl)-2-phenyl-3-buten-1-ol was obtained in only
two hours (Table 7), whereas only 50% of product was synthesized
in a week by using just one equivalent of cinnamyl chloride and
hexamethylditin.
The subsequent second, third and fourth runs also yielded al-
most quantitative amounts of 1-(4-nitrophenyl)-2-phenyl-3-buten-
1-ol in the same amount of time. When the contents of the
(outer) reaction solutions were analyzed for palladium leaching
by ICP/MS analysis (Table 7), palladium leaching was found to
be significantly diminished. No detectable amounts of palladium
leaching were observed in the first two runs, whereas starting in the
third run palladium leaching was observed to a very minor extent.
Furthermore a constant anti/syn product ratio was observed for
four consecutive runs. These optimized reaction conditions and
their mechanistic implications will be the subject of a forthcoming
manuscript.
For the particular palladium-catalyzed tandem reaction stud-
ied here, palladium leaching through the membrane was also
observed. Obviously, this possesses a serious drawback in view
of recyclability and reproducibility of the system. For further
research more robust catalytic systems and/or different reaction
conditions would have to lead to an improved recyclability. We will,
furthermore, investigate other dendritic catalyst systems in order
to further develop the concept of compartmentalized catalysis
and compartmentalized tandem reactions. At the same time, we
are looking into the SCS-pincer palladium-catalysts discussed
here. These complexes have earlier been found to be instable at
high temperatures (>120 ◦C) in the presence of strong bases
and are used in these conditions among others as precursor
for Pd(0)-catalyzed Heck reactions.44–46 In the current reactions,
however, neither high temperatures nor strong bases were used,
so palladium leaching from the pincer manifold was rather
unexpected. To the best of our knowledge, this is the first example
in which it has been found that SCS-pincer palladium-complexes
are instable under mild conditions, i.e. ambient temperature and
non-acidic or non-basic conditions.
We are currently carrying out further mechanistic studies on
this particular catalytic tandem reaction in which palladium is
combined with ditin and tin halide species. These studies have al-
ready led to an optimized and very promising compartmentalized
auto-tandem system that performs fast catalysis for this tandem
reaction with very minor Pd(0) leaching.
Conclusions
In this paper, the synthesis of novel SCS-pincer palladium
dendrimers is described. For the first time, full metal loadings
were obtained for this type of dendritic carbosilane compound.
Since the palladation of the SCS-pincer ligands in the present
study was performed by a direct C–H activation in absence of
reactive intermediates, a full metal loading for both the G0 and
the G1 dendritic complexes could be achieved in a synthetic route
that also contains less reaction steps than in the earlier report on
NCN-pincer Pd-dendrimers.
The new dendritic SCS-pincer palladium complexes appear
to be efficient catalysts for the stannylation of allyl chlo-
rides by hexamethylditin (reaction 1) and for the auto-tandem
reaction between cinnamyl chloride, hexamethylditin and 4-
nitrobenzaldehyde to form 1-(4-nitrophenyl)-2-phenyl-3-buten-1-
Experimental
General
All reactions were carried out using standard Schlenk techniques
under an inert dinitrogen atmosphere unless stated otherwise.
All solvents were carefully dried and distilled prior to use. All
standard reagents were purchased commercially and used without
further purification. Carbosilane dendrimers 7 and 8,40 SCS-pincer
ligand 547,48 and [Pd(MeCN)4](BF4)2 were prepared according
49
to literature procedures. All other reagents were purchased from
Acros Organics and Sigma–Aldrich Chemical Co. Inc. and used as
8902 | Dalton Trans., 2011, 40, 8896–8905
This journal is
The Royal Society of Chemistry 2011
©

received. 1H (300 MHz), 13C (100 MHz) and 29Si (60 MHz) NMR
spectra were recorded on a Varian 400 MHz spectrometer at 25 ◦C,
chemical shifts are given in ppm referenced to residual solvent
resonances. UV-vis spectra were recorded on a Cary 50 Scan
UV-visible spectrophotometer. MALDI-TOF MS spectra were
acquired using a Voyager-DE Bio-Spectrometry Workstation mass
spectrometer equipped with a nitrogen laser emitting at 337 nm.
High resolution mass spectroscopy (HRMS) has been performed
on a Waters LCT Premier XE Micromass instrument using the
electrospray ionization (ESI) technique. Elemental analyses and
ICP-MS analyses were carried out by Kolbe Mikroanalytisches
Laboratorium (Mu¨lheim a.d. Ruhr, Germany).
G0 dendritic SCS-pincer ligand 9
A solution of 5 (4.4 eq., 1.87 mmol, 0.75 g) in Et2O (40 mL)
was cooled to -80 ◦C and a 1.6 M tBuLi solution in pentane
(8.6 eq., 3.65 mmol, 2.28 mL) was added dropwise in 5 min. The
solution immediately turned dark red. The reaction was stirred
for another 5 min at -80 ◦C. Next, carbosilane dendrimer 7
(1 eq., 0.425 mmol, 0.242 g) was added in one portion. The
cooling bath was removed and the reaction mixture was allowed
to reach ambient temperature. The color turned to red/orange.
Subsequently, MeOH (5 mL) was added to quench the excess
lithio-pincer. Immediately, the solution turned pale yellow. After
30 min, all volatiles were evaporated and dichloromethane was
added to the residue. This solution was washed with water (2 ¥ 50
mL) and brine (2 ¥ 50 mL), dried over MgSO4 and concentrated.
The resulting orange syrup was dissolved in 5 mL of a CH2Cl2:
MeOH mixture (1 : 1, v/v) and placed into a dialysis bag. This
bag was placed into a beaker containing a mixture of CH2Cl2:
MeOH (500 mL; 1 : 1 v:v) and dialyzed for 2 h. This procedure
was repeated twice. Finally, the contents of the dialysis bag were
evaporated, yielding 9 as a pale orange syrup in 0.57 g (76%).
1H NMR (CDCl3, 300 MHz): d 7.30–7.16 (m, 48H, CHarom),
4.06 (s, 16H, SCH2), 1.31 (m, 8H, SiCH2CH2), 0.74 (t, 8H, J =
6.2 Hz, SiMe2CH2), 0.56 (t, 8H, J = 6.2 Hz, SicoreCH2), 0.17 (s,
24H, SiMe2). 13C NMR (CDCl3, 75 MHz): d 140.5, 137.1, 136.5,
133.1, 130.4, 130.1, 129.0, 126.7, 39.4, 20.7, 18.8, 17.7, -2.7. 29Si
NMR (CDCl3, 60 MHz): d 0.73, -3.82. MALDI-TOF MS for
C100H116S8Si5 (m/z): [M+H]+ 1714.69 (calc. 1714.94).
3,5-bis(phenylthiomethyl)phenyl)trimethylsilane 6 (TMS-SCS-H)
A solution of 5 (2.49 mmol, 1.00 g) in Et2O (30 mL) was cooled
to -80 ◦C whereupon a 1.6 M tBuLi solution in pentane (2.0 eq.,
4.98 mmol, 3.11 mL) was added slowly in 5 min. The solution
immediately turned dark red. After stirring for another 5 min.
at -80 ◦C trimethylsilyl chloride (1.1 eq., 2.74 mmol, 348 mL)
was added in one portion. The cooling bath was removed and
the reaction mixture was allowed to reach ambient temperature.
The solution turned pale yellow. After 30 min, all volatiles were
evaporated and dichloromethane was added to the residue. This
solution was washed with water (2 ¥ 50 mL) and a saturated NaCl
solution (2 ¥ 50 mL), dried over MgSO4 and concentrated, yielding
6 as a colorless syrup in 0.90 g (92%).
1H NMR (CDCl3, 300 MHz): d 7.32–7.21 (m, 13H, CHarom), 4.10
(s, 4H, CH2), 0.21 (s, 9H, SiMe3). 13C NMR (CDCl3, 75 MHz): d
141.1, 137.3, 136.5, 133.0, 130.7, 130.3, 129.1, 126.7, 39.5, -0.8.
29Si NMR (CDCl3, 60 MHz): d -3.76. ESI-HRMS for C23H26S2Si
(m/z): [M+Na]+ 417.1143 (calc. 417.1108).
G0 dendritic SCS-pincer Pd-complex 3
To a solution of compound 9 (0.32 mmol, 550 mg) in
an acetonitrile/toluene mixture (20 mL, 1 : 1 v:v) was added
[Pd(MeCN)4](BF4)2 (5 eq., 1.6 mmol, 740 mg). The reaction mix-
ture was stirred for 16 h at reflux temperature. After removal of the
solvent in vacuo a biphasic system consisting of dichloromethane
(10 mL) and brine was added. The reaction mixture stirred for
1 h. Subsequently, the organic phase was separated and the
aqueous phase was washed with dichloromethane (2 ¥ 20 mL).
The combined organic fractions were washed with water, dried
over MgSO4 and evaporated to dryness. The product was purified
with column chromatography (CH2Cl2:EtOAc 4 : 1 v:v). Finally,
the product is dissolved in dichloromethane and a large excess
(~100 eq.) of PVPy was added. After the solution has stirred for
2 h, the solution was filtered over Celite. After evaporation of
the dichloromethane, the product was obtained as yellow powder.
Yield: 650 mg (83%).
1H NMR (CD2Cl2, 300 MHz): d 7.83 (d, 16H, SPhortho), 7.39 (m,
24H, SPhmeta+para), 7.07 (s, 8H, CHpincer), 4.62 (br. s, 16H, SCH2),
1.31 (m, 8H, SiCH2CH2), 0.74 (t, 8H, J = 6.3 Hz, CH2SiMe2),
0.53 (t, 8H, J = 6.3 Hz, SicoreCH2), 0.18 (s, 24H, SiMe2). 13C NMR
(CDCl3, 75 MHz): d 149.2, 136.3, 132.8, 132.4, 131.4, 130.4,
127.9, 125,9, 52.3, 20.7, 18.8, 17.6, -2.9. 29Si NMR (CDCl3, 60
MHz): d 0.62, -3.82. UV-VIS: 330.1 nm. MALDI-TOF MS for
C100H112Br4Pd4S8Si5 (m/z): [M + Na]+ 2478.73 (calc. 2479.19).
para-TMS SCS-pincer Pd-complex 2
To a solution of 6 (1.27 mmol, 500 mg) in acetonitrile (30 mL)
was added [Pd(MeCN)4](BF4)2 (1.2 eq., 1.52 mmol, 674 mg).
The reaction mixture was stirred for 16 h at reflux temperature
followed by filtration over Celite and evaporation of the solvent.
A biphasic solution consisting of dichloromethane (10 mL) and
a saturated aqueous solution of NaCl (10 mL) was added. The
resulting mixture was stirred for 1 h. Subsequently, the organic
phase was separated and the aqueous phase was washed with
dichloromethane (2 ¥ 20 mL). The combined organic fractions
were washed with water, dried over MgSO4 and evaporated to
dryness. Column chromatography (CH2Cl2) was used for further
purification of the product. Finally, the product is dissolved in
dichloromethane and a large excess (~100 eq.) of PVPy was added.
After the solution has stirred for 2 h, the solution was filtered over
Celite. After evaporation of the dichloromethane, the product was
obtained as yellow powder. Yield: 605 mg (89%).
1H NMR (CD2Cl2, 300 MHz): d 7.86 (m, 4H, C SPhortho), 7.41
(m, 6H, SPhmeta+para), 7.13 (s, 2H, CHarom.pincer), 4.64 (br.s, 4H, SCH2),
0.24 (s, 9H, SiMe3). 13C NMR (CDCl3, 75 MHz): d 149.7, 136.9,
132.8, 131.7, 130.0, 129.8, 127.2, 125,1, 52.1, -1.1. 29Si NMR
(CDCl3, 60 MHz): d -3.67. UV/Vis (CH2Cl2): lmax = 331.0 nm.
ESI-HRMS for C23H25ClPdS2Si (m/z): [M-Cl]+ 499.0239 (calc.
499.0210).
G1 dendritic SCS-pincer ligand 10
The used procedure of the synthesis of dendritic ligand 10 was
similar to the one described for the compound 9. For 10 14 equiv.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 8896–8905 | 8903
©

1
of 5 and 27 equiv. of tBuLi have been used. An orange-brown
syrup was obtained in 61% yield (0.94 gram).
samples of the outer solution were taken and analyzed by H
NMR.
1H NMR (CDCl3, 300 MHz): d 7.29–7.17 (m, 144H, CHarom).,
4.03 (s, 48H, SCH2), 1.33 (m, 32H, SiCH2CH2 (inner and outer)),
0.72 (bt, 24H, CH2SiMe2), 0.58 (m, 40H, SiCH2), 0.13 (s, 72H,
SiMe2). 13C NMR (CDCl3, 75 MHz): d 140.5, 137.2, 136.7, 132.7,
130.8, 129.7, 128.7, 126.3, 39.4, 20.5, 20.2, 19.5, 18.9, 18.4, 17.8,
-2.7. 29Si NMR (CDCl3, 59.6 MHz): d 0.80 (core), 0.57 (middle),
-3.75 (periphery).
General protocol for the compartmentalized tandem coupling
reaction with dendritic catalyst 4 present inside a membrane
dialysis bag
In a tailor-made reaction vessel, which is equipped with a stirring
bar, a NS50 joint and a nitrogen inlet, dry THF (60 mL)
was added. To the solvent were subsequently added cinnamyl
chloride (8.0 mmol, 1.22 g, 1.13 mL), hexamethylditin (8.0 mmol,
2.75 g, 1.74 mL), 4-nitrobenzaldehyde (8.0 mmol, 1.21 g) and
hexamethylbenzene (internal standard, 0.89 mmol, 144 mg). A
dialysis bag (Aldrich, benzoylated cellulose membranes, MWCO =
1000 Da.) with 2.5 mL THF and 0.0133 mmol (2% Pd) 4 that was
closed by plastic clamps was added to this solution. In regular
intervals, samples of the outer solution were taken and analyzed
by 1H NMR.
G1 dendritic SCS-pincer Pd-complex 4
The used procedure for the synthesis of 4 was similar to the
one described for the synthesis of 3, but now 15 equivalents
of [Pd(MeCN)4](BF4)2 were used. The product appeared as an
orange–yellow foam. Yield: 110 mg (60%).
1H NMR (CD2Cl2, 300 MHz): d. 7.86 (m, 48H, SPhortho), 7.45
(m, 72H, SPhmeta+para), 7.12 (bs, 24H, CHarom,pincer), 4.61 (br. s, 16H,
CH2S), 1.37 (m, 32H, SiCH2CH2), 0.79 (m, 24H, CH2SiMe2),
0.60 (m, 40H, SiCH2), 0.21 (s, 72H, SiMe2). 13C NMR (CDCl3,
75 MHz): d. 149.2, 136.0, 132.8, 131.9, 131.8, 130.1, 129.7, 127.0,
52.3, 20.7, 20.0, 19.7, 18.8, 18.4, 17.6, -2.8. 29Si NMR (CDCl3,
59.6 MHz): d 0.77 (core), 0.57 (middle), -3.64 (periphery). UV-
VIS: 330.1 nm
After the reaction has finished, the dialysis bag was directly
placed reused in a fresh batch of substrates to start a new catalytic
run.
Notes and references
1 G. E. Oosterom, J. N. H. Reek, P. C. J. Kamer and P. W. N. M. van
Leeuwen, Angew. Chem., Int. Ed., 2001, 40, 1828–1849.
2 R. van Heerbeek, P. C. J. Kamer, P. W. N. M. van Leeuwen and J. N. H.
Reek, Chem. Rev., 2002, 102, 3717–3756.
3 H. P. Dijkstra, G. P. M. van Klink and G. van Koten, Acc. Chem. Res.,
2002, 35, 798–810.
4 P. A. Chase, R. J. M. Klein Gebbink and G. van Koten, J. Organomet.
Chem., 2004, 689, 4016–4054.
5 C. Mu¨ller, M. G. Nijkamp and D. Vogt, Eur. J. Inorg. Chem., 2005,
4011–4021.
6 I. F. J. Vankelecom, Chem. Rev., 2002, 102, 3779–3810.
7 J. T. Scarpello, D. Nair, L. M. F. dos Santos, L. S. White and A. G.
Livingston, J. Membr. Sci., 2002, 203, 71–85.
8 A. W. Bosman, H. M. Janssen and E. W. Meijer, Chem. Rev., 1999, 99,
1665–1688.
General protocol for the stannylation reaction with the catalyst
present in solution
In a representative experiment, the appropriate catalyst (2 mol%
[Pd], 0.016 mmol), was added to a solution of cinnamyl chloride
(0.80 mmol, 122.1 mg, 113 mL), hexamethylditin (1.05 eq.,
0.84 mmol, 275 mg, 174 mL) and hexamethylbenzene (internal
standard, 0.088 mmol, 14.4 mg) in 6 mL dry THF or CH2Cl2. The
reaction stirred at room temperature in a nitrogen environment.
Aliquots of 50 mL for NMR/GC analysis were regularly taken
with an airtight syringe.
9 J. W. J. Knapen, A. W. van der Made, J. C. de Wilde, P. W. N. M. van
Leeuwen, P. Wijkens, D. M. Grove and G. van Koten, Nature, 1994,
372, 659–663.
10 D. Astruc and F. Chardac, Chem. Rev., 2001, 101, 2991–3023.
11 R. Kreiter, A. W. Kleij, R. J. M. Klein Gebbink and G. van Koten, Top.
Curr. Chem., 2001, 217, 163–199.
General protocol for the tandem coupling reaction with the catalyst
present in solution
In a representative experiment, the appropriate catalyst (2 mol%
[Pd], 0.016 mmol), was added to a solution of cinnamyl chloride
(0.80 mmol, 122.1 mg, 113 mL), hexamethylditin (1.05 eq.,
0.84 mmol, 275 mg, 174 mL), 4-nitrobenzaldehyde (1.05 eq.,
0.84 mmol, 126.9 mg) and hexamethylbenzene (internal standard,
0.088 mmol, 14.4 mg) in 6 mL dry THF. The reaction stirred at
room temperature in a nitrogen environment. Aliquots of 50 mL for
NMR/GC analysis were regularly taken with an airtight syringe.
12 L. J. Twyman, A. S. H. King and I. K. Martin, Chem. Soc. Rev., 2002,
31, 69–82.
13 D. Astruc, E. Boisselier and C. Ornelas, Chem. Rev., 2010, 110, 1857–
1959.
14 E. de Jesu´s and J. C. Flores, Ind. Eng. Chem. Res., 2008, 47, 7968–7981.
15 A. Berger, R. J. M. Klein Gebbink and G. van Koten, Top. Organomet.
Chem., 2006, 20(Dendrimer Catal.), 1–38.
16 D. Mery and D. Astruc, Coord. Chem. Rev., 2006, 250, 1965–1979.
17 V. A. Yazerski, R. J. M. Klein Gebbink, Catal. Metal Complexes, 33
(Heterogenized Homogeneous Catalysts for Fine Chemicals Produc-
tion), 2010, 171–201.
18 N. J. M. Pijnenburg, T. J. Korstanje, G. van Koten and R. J. M. Klein
Gebbink, Palladacycles on Dendrimers and Star-Shaped Molecules,
Wiley-VCH Verlag GmbH & Co. KGaA, 2008, 361–398.
19 D. E. Fogg and E. N. dos Santos, Coord. Chem. Rev., 2004, 248, 2365–
2379.
General protocol for the compartmentalized stannylation reaction
with dendritic catalyst 4 present inside a membrane dialysis bag
In a tailor-made reaction vessel, which is equipped with a stirring
bar, a NS50 joint and a nitrogen inlet, dry THF (60 mL) was
added. To the solvent were subsequently added cinnamyl chloride
(8.0 mmol, 1.22 g, 1.13 mL), hexamethylditin (8.0 mmol, 2.75 g,
1.74 mL), and hexamethylbenzene (internal standard, 0.89 mmol,
144 mg). A dialysis bag (Aldrich, benzoylated cellulose mem-
branes, MWCO = 1000 Da.) with 2.5 mL THF and 0.0133 mmol
(2 mol% Pd) 4 was added to this solution. In regular intervals,
20 A. V. Gaikwad, V. Boffa, J. E. ten Elshof and G. Rothenberg, Angew.
Chem., Int. Ed., 2008, 47, 5407–5410.
21 M. Janssen, C. Mu¨ller and D. Vogt, Adv. Synth. Catal., 2009, 351,
313–318.
22 M. Janssen, J. Wilting, C. Mu¨ller and D. Vogt, Angew. Chem., Int. Ed.,
2010, 49, 7738–7741.
23 R. A. Sheldon and H. van Bekkum, Fine Chemicals Through Heterege-
neous Catalysis, 2001, 1st ed., Wiley, New York.
8904 | Dalton Trans., 2011, 40, 8896–8905
This journal is
The Royal Society of Chemistry 2011
©

24 M. T. Grimes and D. G. Drueckhammer, J. Org. Chem., 1993, 58,
6148–6150.
37 N. Lucena, J. Casabo, L. Escriche, G. Sanchez-Castello, F. Teixidor, R.
Kivekas and R. Sillanpa¨a¨, Polyhedron, 1996, 15, 3009–3018.
38 M. Weck and C. W. Jones, Inorg. Chem., 2007, 46, 1865–1875.
39 J. S. Chen, A. N. Vasiliev, A. P. Panarello and J. G. Khinast, Appl.
Catal., A, 2007, 325, 76–86.
40 A. W. van der Made and P. W. N. M. van Leeuwen, J. Chem. Soc.,
Chem. Commun., 1992, 1400–1401.
25 M. Albrecht, N. J. Hovestad, J. Boersma and G. van Koten, Chem.–
Eur. J., 2001, 7, 1289–1294.
26 A. M. Arink, R. van de Coevering, B. Wieczorek, J. Firet, J. T. B. H.
Jastrzebski, R. J. M. Klein Gebbink and G. van Koten, J. Organomet.
Chem., 2004, 689, 3813–3819.
27 M. Gaab, S. Bellemin-Laponnaz and L. H. Gade, Chem.–Eur. J., 2009,
15, 5450–5462.
41 J. Dupont, N. Beydoun and M. Pfeffer, J. Chem. Soc., Dalton Trans.,
1989, 1715–1720.
42 C. A. Kruithof, H. P. Dijkstra, M. Lutz, A. L. Spek, R. J. M. Klein
Gebbink and G. van Koten, Organometallics, 2008, 27, 4928–4937.
43 N. J. M. Pijnenburg, Y. H. M. Cabon, G. van Koten and R. J. M. Klein
Gebbink, manuscript in preparation.
28 M. Albrecht and G. van Koten, Angew. Chem., Int. Ed., 2001, 40,
3750–3781.
29 J. T. Singleton, Tetrahedron, 2003, 59, 1837–1857.
30 N. Selander and K. J. Szabo´, Chem. Rev., 2011, 111, 2048–2076.
31 C. M. Jensen and D. Morales-Morales, The Chemistry of Pincer
Compounds, 2007, Elsevier Science Ltd.
44 D. E. Bergbreiter, P. L. Osburn and J. D. Frels, Adv. Synth. Catal., 2005,
347, 172–184.
32 A. W. Kleij, R. A. Gossage, R. J. M. Klein Gebbink, N. Brinkmann,
E. J. Reijerse, U. Kragl, M. Lutz, A. L. Spek and G. van Koten, J. Am.
Chem. Soc., 2000, 122, 12112–12124.
33 O. A. Wallner and K. J. Szabo´, Org. Lett., 2004, 6, 1829–1831.
34 N. Solin, J. Kjellgren and K. J. Szabo´, J. Am. Chem. Soc., 2004, 126,
7026–7033.
35 M. Gagliardo, N. Selander, N. C. Mehendale, G. van Koten, R. J. M.
Klein Gebbink and K. J. Szabo´, Chem.–Eur. J., 2008, 14, 4800–4809.
36 J. Li, M. Siegler, M. Lutz, A. L. Spek, R. J. M. Klein Gebbink and G.
van Koten, Adv. Synth. Catal., 2010, 352, 2474–2488.
45 K. Q. Yu, W. Sommer, J. M. Richardson, M. Weck and C. W. Jones,
Adv. Synth. Catal., 2005, 347, 161–171.
46 W. J. Sommer, K. Q. Yu, J. S. Sears, Y. Y. Ji, X. L. Zheng, R. J. Davis,
C. D. Sherrill, C. W. Jones and M. Weck, Organometallics, 2005, 24,
4351–4361.
47 M. D. Meijer, B. Mulder, G. P. M. van Klink and G. van Koten, Inorg.
Chim. Acta, 2003, 352, 247–252.
48 P. Steenwinkel, S. L. James, D. M. Grove, N. Veldman, A. L. Spek and
G. van Koten, Chem.–Eur. J., 1996, 2, 1440–1445.
49 T. W. Lai and A. Sen, Organometallics, 1984, 3, 866.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 8896–8905 | 8905
©