New mono- and tricyclopalladated dendritic systems with encapsulated catalytic sites

Article abstract of DOI:10.1002/1521-3765(20020104)8:1<45::AID-CHEM45>3.0.CO;2-P

The preparation of a series of new macrocyclic carbodiazasilane molecules functionalized with the monoanionic [2,6-(CH₂NMe₂)₂C₆H₃] ^- ≡ N,C,N-pincer ligand has been accomplished. Palladation

Full text of DOI:10.1002/1521-3765(20020104)8:1<45::AID-CHEM45>3.0.CO;2-P

G. van Koten et al.
Dendritic Systems as Active Lewis
Acid Catalysts
For more information
see the following pages.
Chem. Eur. J. 2002, 8, No. 1
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0045 $ 17.50+.50/0
45

FULL PAPER
New Mono- and Tricyclopalladated Dendritic Systems with Encapsulated
Catalytic Sites
Gema Rodríguez,[a] Martin Lutz,[b] Anthony L. Spek,[b] and Gerard van Koten*[a]
Abstract: The preparation of a series
of new macrocyclic carbodiazasilane
molecules functionalized with the
monoanionic [2,6-(CH2NMe2)2C6H3]
ꢀ N,C,N-pincer ligand has been accom-
plished. Palladation of these systems was
possible through oxidative addition with
[Pd(dba)2] affording exclusive formation
of the meso diastereoisomer. The X-ray
crystal structures of these novel ligands
and of the palladium(ii) complex 10 were
determined and confirmed the stereo-
chemistry of the organopalladium cage.
Attachment of the para-OH functional-
ized carbodiazasilane macrocycle 16 to a
central core led to the formation of the
dendritic structure 18 which was palla-
dated to afford the novel multimetallic
dendritic system with encapsulated cat-
alytic sites 1. This cyclopalladated car-
bosilane dendrimer (1) as well as the
mononuclear organopalladium cage 10
can be conveniently converted into ac-
tive Lewis acid catalysts for the aldol
condensation reaction. The catalytic da-
ta showed higher reaction rates for the
dendritic structure than for the corre-
sponding mononuclear systems.
Keywords: cage compounds ´ cata-
lysts ´ dendrimers ´ macrocycles ´
macrocyclic ligands
dendrimers instead of polymers for anchoring catalytic sites
leads to well-defined nanosized species in which the number
and location of the catalytic sites can be controlled.
Introduction
In the past few years the chemistry of dendrimers has
experienced spectacular developments and, very recently,
functionalized dendrimers have received substantial atten-
tion.[1, 2] In this regard, attractive new materials with interest-
ing chemical, physical and catalytic properties have been
prepared consisting of dendrimers or dendritic wedges which
contain organometallic functional groups.[3, 4]
One of the most interesting applications of these metal-
lodendrimers is their use in catalysis. Dendrimers having
nanoscopic dimensions can be molecularly dissolved. Thus,
soluble dendrimers carrying a defined number of catalytic
sites can be removed from homogeneous reaction mixtures by
simple nanofiltration techniques. The combination of these
properties makes them suitable to bridge the gap between
homo- and heterogeneous catalysts. Moreover, the use of
In 1994, our group reported on the first carbosilane
dendrimer;[5] its periphery is functionalized with catalytic
sites based on the monoanionic pincer ligand [2,6-
(CH2NMe2)2C6H3] (N,C,N).[6] This metallodendritic system
was successfully applied as a homogeneous catalyst in organic
synthesis and turned out to be suitable for separation by
nanomembrane filtration techniques. This multimetallic car-
bosilane dendrimer was the starting point of our studies in the
synthesis and applications of new dendrimers with reactive
sites based on N,C,N-pincer and [2-(CH2NMe2)C6H4] (C,N)
ligands.[7]
Over the last five years, numerous metallodendritic cata-
lysts have been reported which contain the catalytically active
sites on the outer surface[4, 5, 7, 8] or at the core[9] of the
molecule. However, such structures can also have their
limitations. Recently, it has been reported that only low
generations of peripherally functionalized dendrimers can be
suitable carriers of catalytically active sites in atom transfer
radical addition reactions.[7a, b] When the catalytic units are
placed in a densely packed surface (of a higher generation
dendrimer) they can interfere with each other resulting in
decreased activity. On the other hand, catalytic sites residing
at the core (focal point) of the dendrimer (or dendritic wedge)
can be used to change, for example, the solubility properties of
the catalyst[10] and it can also result in beneficial interactions
between the substrate and the dendritic branches around the
catalyst.[11] However, when the catalytic site is located inside a
[a] Prof. Dr. G. van Koten, Dr. G. Rodríguez
Debye Institute, Department of Metal-Mediated Synthesis
Utrecht University, Padualaan 8
3584 CH Utrecht (The Netherlands)
Fax: (‡31)30-2523615
E-mail: g.vankoten@chem.uu.nl
[b] Dr. M. Lutz, Prof. Dr. A. L. Spek[‡]
Bijvoet Center for Biomolecular Research
Department of Crystal and Structural Chemistry
Utrecht University, Padualaan 8
3584 CH Utrecht (The Netherlands)
Fax: (‡31)30-2533940
E-mail: a.l.spek@chem.uu.nl
‡
[ ] Correspondence pertaining crystallographic studies.
46
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0046 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Dendritic Systems
45±57
higher generation dendrimer, the branches can prevent access
of the substrate to the reactive center.[12] To date, the
introduction of regio- or stereocontrol in a chemical reaction
by using dendrimers with an interior isolated catalytic site has
not been straightforward. The molecular network of the
dendrimers studied have been too flexible and thus unable to
impose distinct spatial constraints on the course of the
reaction. We therefore considered that progress in this field
seems to require specific combinations of dendrimers and
encapsulated catalytic sites to encourage regio- and stereo-
control. Following this approach, we set out to develop new
multimetallic dendritic systems with catalytically active tran-
sition metal complexes placed neither at the periphery nor at
the core but encapsulated in a dendritic branch.[13] As a first
model, we designed the molecule 1 shown in Figure 1.
In this model, the N,C,N-pincer moiety is linked on one side
to the core (branching point) and on the other side it is used to
coordinate the active metal atom as well as to provide the next
branching point (Si) for further extension of the dendritic
structure. An interesting aspect of this structure is that 1 has
an estimated size of approximately 1.5 nm. Recently we
showed that even species of this size could be removed after
catalysis from the product-containing solution for reuse by
nanomembrane filtration techniques.[7a]
In this paper we report the synthesis of 1 following a
convergent procedure compromising first the synthesis of the
new macrocyclic carbodiazasilane ligands 8 (see Scheme 2)
and the para-OH functionalized 16 (see Scheme 5) as the
cages, then their attachment to a central core and the
subsequent formation of the corresponding palladium(ii)
complexes by oxidative addition.
Ph
Ph
We also describe the preliminary results obtained in the use
of the aqua complexes of 1 and 10 as homogeneous catalysts
using the aldol condensation reaction of benzaldehyde and
methyl isocyanoacetate.[14]
Si
Br
N
N
O
Pd
Results and Discussion
O
Synthesis of the macrocyclic carbodiazasilane ligands 8a and
8b: Several approaches to the synthesis of macrocycles have
been reported.[15] One of these, described by Kellogg et al.,[16]
uses Cs2CO3 as a template in an aprotic solvent such as DMF.
Depending on the nature of the starting materials, sometimes
the use of other salts such as Na2CO3 or K2CO3 or other
solvents such as acetonitrile gave better results. We therefore
pursued this strategy to obtain the desired carbodiazasilanes 8
(Scheme 2), utilising the bisbenzylic bromide 5 and the
diamines 7a and 7b, as precursors (Scheme 1).
O
O
N
N
O
O
Pd
Pd
Br
Br
N
N
Ph
Ph
Si
Ph
Si
Ph
1
Figure 1. Multimetallic dendritic system with encapsulated catalytic sites 1.
Br
Ph2Si
Ph2Si
Abstract in Dutch: Dit onderzoek beschrijft de synthese van
een reeks nieuwe, macrocyclische carbodiazasilaan moleculen,
die het monoanionisch drievoudig-gecoördineerde tangligand
[2,6-(CH2NMe2)2C6H3] ꢀ N,C,N bevatten. Deze systemen
kunnen door oxidatieve additie met [Pd(dba)2] gepalladeerd
worden, waarbij selectief de meso-diastereoisomeren gevormd
worden. De structuren van de nieuwe liganden en van het
palladium(ii)complex 10 zijn kristallografisch onderzocht.
Deze kristalstructuuranalyse bevestigt de meso-stereochemie
van complex 10. Het gebruik van de para-OH gefunctionali-
seerd macrocyclische carbodiazasilaan verbinding 16 in een
convergente synthese-route leidde tot de vorming van de
dendritische structuur 18. Het multi-metallo-dendritische sys-
teem 1 met ingekapselde katalytische centra werd gesyntheti-
seerd door palladering van verbinding 18. Het cyclopalla-
deerde carbosilanedendrimeer 1 en de mononucleaire orga-
nopalladium-kooi 10 kunnen eenvoudig omgezet worden in
kationische (Lewis zure) centra en zijn getest als katalysatoren
in de aldol-condensatiereactie. De katalytische activiteit van de
dendritische structuur blijk hoger te zijn dan die van de
mononucleaire systemen.
iii
i
Br
OR
2
2
OTBDMS
3, R= TBDMS
4, R= H
ii
2
5
NH
R
HN
Br
R
Br
iv
6a R= Br
6b R= H
7a R= Br
7b R= H
Scheme 1. Synthesis of the bisbenzylic bromide 5 and diamines 7a and 7b.
i) tBuLi, Et2O, 788C, then Ph2SiCl2, 12 h; ii) AcOH/THF/H2O 3:1:1,
508C, 3 h; iii) PBr3, C6H6, RT, 3 h; iv) MeNH2, Et2O, 08C, 1.5 h.
Treatment of a solution of 3-bromobenzyl tert-butyldi-
methylsilyl ether (2) in diethyl ether with tBuLi (2.0 equiv) at
788C followed by addition of dichlorodiphenylsilane
(0.45 equiv) afforded after work-up the protected benzylic
alcohol 3 in good yield (Scheme 1). Deprotection of 3 with a
Chem. Eur. J. 2002, 8, No. 1
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0047 $ 17.50+.50/0
47

G. van Koten et al.
FULL PAPER
Table 1. Yields of monomer 8 and dimer 9 using different alkali metal
mixture of AcOH/THF/H2O 3:1:1 gave rise to the formation
of the bisbenzylic alcohol 4 which was easily transformed to
the corresponding bisbenzylic bromide 5 by treatment with
PBr3 (0.7 equiv) in benzene at room temperature (81% over
two steps). The meta-bis(bromomethyl)aryl bromide 6a and
the meta-bis(bromomethyl)arene 6b were prepared according
to literature procedures,[17] and subsequently treated with
methylamine which afforded the diamines 7a and 7b. The
building blocks for the synthesis of 8 were then used in the
[2 ‡ 2]-macrocyclization reactions shown in Scheme 2. Thus,
deprotonation of the bisamino compounds 7a and 7b by an
alkali metal carbonate followed by reaction of the bisamine
with 5 afforded the desired macrocyclic carbodiazasilanes 8a
and 8b, respectively, by a sequence of two nucleophilic
substitutions. Interestingly, formation of dimer 9 was only
observed for the reaction of the bisaminoaryl 7a with 5
(Scheme 2).
carbonates.[a]
Yield [%][b]
Entry
7
Base
Solvent
8
9
overall
1
2
3
4
5
7a
7a
7a
7a
7b
Cs2CO3
K2CO3
Na2CO3
Na2CO3
Na2CO3
DMF
DMF
DMF
CH3CN
DMF
21
26
30
8
12
23
16
41
±
33
49
46
49
63
63
[a] Reaction conditions: Dropwise addition of a solution of 5 in DMF to a
suspension of the metal carbonate and 7 also in DMF, at 508C over a period
of 2 h. [b] Isolated yield after column chromatography.
(entry 3).[18] The preferred initial concentration of 7 appeared
to be 1.3 Â 10 2 m. More concentrated solutions gave lower
yields of the reaction products. The use of CH3CN as solvent
(entry 4) inverts the ratio monomer (8)/dimer (9) providing a
suitable synthetic route to the dimer 9 in a 41% yield. It is
important to note that the absence of bromide, that is the use
of the bisaminoarene 7b as the starting material, leads to the
selective formation of the monomer 8b (entry 5) in a much
higher yield than in the corresponding other cases (entries 1 ±
4) which most probably is due to steric interference of the
bromine atom with the cyclization process.
Several experimental procedures involving different alkali
metal carbonates and solvents were tested in order to
establish the best reaction conditions (see Table 1). The best
protocol discovered is the dropwise addition of a solution of 5
in DMF to a suspension of 7 and Na2CO3 over a period of 2 h
Ph
Ph
Si
NH R2 HN
Synthesis of the palladium(ii) complex 10: Reaction of the aryl
bromide 8a with [Pd(dba)2][19] in refluxing benzene for 12 h
gave rise to the formation of the desired macrocyclic PdII
complex 10, which was isolated as a yellow solid in 47% yield
(Scheme 3). However, alternative synthesis from macrocycle
8b, following electrophilic palladation procedures developed
for related hydrocarbons[20] were unsuccessful.
+
Br
Br
R1
5
7a R1= H, R2= Br
7b R1= H, R2= H
13 R1= OTBDMS, R2= Br
i
Ph
Ph
Si
Si
Si
i
N
R2
N
Br
N
Br
N
N
Pd
N
8a R1= H, R2= Br
8b R1= H, R2= H
14 R1= OTBDMS, R2= Br
R1
+
10
8a
Ph
Scheme 3. Transformation of the new carbodiazasilane ligand 8a into the
corresponding PdII complex 10 by oxidative addition. i) [Pd(dba)2], C6H6,
D, 12 h.
Ph
Si
The structure of 10 was identified on the basis of elemental
N
N
1
analysis, MALDI-TOF and H, 13C{1H} and two-dimensional
R1
Br
N
R1
Br
NMR spectroscopic data. It is important to note that in 10 the
nitrogen atoms are stereogenic centers. As a consequence,
four stereoisomers could be formed (RR, SS, RS, SR); the RS
and SR forms are identical, when ring flipping of the two fused
five membered chelate rings is fast on the NMR time scale,
they represent the meso compound with an apparent internal
mirror plane. The RR and SS isomers are enantiomers and
therefore indistinguishable by 1H NMR. Therefore, if the
palladation reaction of 8a would give rise to the four possible
N
9 R1= H
15 R1= OTBDMS
Si
Ph
Ph
Scheme 2. Synthesis of carbodiazasilane ligands 8a, 8b and 14 by a [2‡2]-
macrocyclization reaction of the benzylic dibromide 5 and diamines 7a, 7b
and 13. i) Na2CO3, DMF, 508C, 14 h.
48
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0048 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Dendritic Systems
45±57
stereoisomers of 10, two sets of signals should be observed,
one attributable to the meso compound and one correspond-
ing to the RR/SS diastereoisomer. Interestingly, the 1H NMR
spectrum of complex 10 shows only one group of resonances,
indicating the formation of only one diastereomeric form.
Definitive proof for the structure of 10 in the solid state was
obtained from a single crystal X-ray diffraction study (see
below, Figure 3).
This disposition may be caused by a repulsive effect between
the bromine and the two nitrogen atoms. The fixed config-
uration of the nitrogen centers (R and S for the N1 and N2,
respectively, for the molecule shown[21]) in the solid state (the
nitrogen inversion process typical for tertiary amines in
solution was not observed) forces the bromine atom to tilt out
of the macrocyclic ring and situates the C Br bond in an
assumed suitable disposition to react with [Pd(dba)2].
A different situation is found in the case of macrocyclic
arene 8b. Although the respective configurations of the
nitrogen centers in this molecule are the same as those for
these centers in 8a, the absence of the bromine atom gives
more flexibility to the macrocycle. The benzylic rings bonded
to the Si atom now exhibit conformations leading to a
different orientation of the protons at C2 and C22 (distance
2.54 Š). Likewise, the torsion angles C25-Si-C1-C2 and C25-
Si-C21-C22 differ significantly ( 98.90(10) and 55.64(11)8,
respectively).
Molecular structures of 8a, 8b and 10 in the solid state: To
obtain more structural information concerning the conforma-
tional preferences of the macrocyclic ring in the arylpalladium
compound 10, the structures of both 10 and the ligands 8a and
8b were studied by single crystal X-ray techniques. The
molecular geometries of 8a, 8b and 10 are shown in Figures 2,
3, respectively, while in Table 2 some pertinent bond lengths
and bond and torsion angles have been listed.
The molecular structure of the macrocyclic arylpalladium
compound 10 (Figure 3) shows the palladium atom bound to
the two nitrogen atoms, to C14 (i.e., Cipso of the monoanionic
h3-N,C,N bonded moiety) and to the bromine atom trans to
Cipso. The square-planar coordination geometry is only slightly
distorted, in particular the N1-Pd-N2 angle of 161.80(8)8,
which is a result of the intrinsically small N-Pd-C14 bite angles
of the two neighboring five-membered chelate rings, 81.02(9)
and 80.81(9)8, respectively. As a result of the coupled
puckering of the two five-membered rings, the N and methyl
C atoms are at opposite sides of the plane containing the aryl-
Pd portion of the molecule. This puckering conformation is
similar to that found for, namely the simple complex
[PdBr(2,6-{CH2NMe2}2C6H3)].[22]
The two five-membered chelate rings are puckered in such
a way that the rest of the macrocycle is orientated perpen-
dicular to the coordination plane. Consequently, several
structural features can be observed: i) the bromine atom in
10 is tilted out of the Pd-coordination plane, Br-Pd-C14
174.62(7)8, whereas in other cases such as 1,3,5-C6H3-
[4'-(PdBr)(2',6'-{CH2NMe2}2C6H3)]3 this atom is in the coor-
dination plane,[23] ii) one of the methyl groups is in an axial
position while the other methyl group is placed equatorially,
which is a feature that the structure of 10 has in common with
other organoplatinum macrocyclic compounds reported ear-
lier, that is, [PtI{CH2NMe(CH2)10MeNCH2C6H3}],[24] iii) the
aromatic rings from the macrocycle which do not belong to
the N,C,N moiety are now orientated in opposite directions.
This is in contrast to what is observed for the molecular
structure of 8a in which these aromatic rings are orientated
towards the same focal point. However, comparison of
conformations of the nitrogen centers in 8a and 10 show
various striking similarities which indicate that these con-
formations are largely determined by the macrocyclic ring.
Figure 2. Displacement ellipsoid plots (50% probability) and numbering
schemes of the molecular structures of the macrocyclic ligands a) 8a and
b) 8b. Hydrogen atoms have been omitted for clarity.
The molecular structure of the macrocyclic aryl bromide 8a
(Figure 2a) shows that in the solid state the phenyl groups
forming part of the cavity have similar orientations. Thus,
protons bound at carbons C2 and C22 which are only 2.38 Š
apart from each other and the bromine atom are orientated to
the same focal point. The torsion angles for C25-Si-C1-C2
NMR Spectroscopy of ligands 8a, 8b and PdII complex 10:
Ligands 8a and 8b and arylpalladium complex 10 have been
characterized in solution by 1H, 13C{1H} and two-dimensional
NMR techniques. The NMR data reveal a high degree of
symmetry for these compounds in solution, due to an
apparent molecular symmetry plane which contains the
[
56.46(18)8] and C25-Si-C21-C22 [59.52(17)8] indicate sim-
ilar orientations of the benzylic rings attached to the Si atom.
Chem. Eur. J. 2002, 8, No. 1
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0049 $ 17.50+.50/0
49

G. van Koten et al.
FULL PAPER
Table 2. Selected bond lengths [Š] and bond and torsion angles [8] for the ligands 8a, 8b, 14, 16 and the PdII
complex 10.
and SR pair of enantiomers are
present in the unit cell). For-
mation of the RR/SS pair is less
likely because the aromatic
rings of the macrocycle are
constrained to one side of the
palladium coordination plane,
and consequently the methyl
groups are positioned at the
opposite side. Furthermore,
the puckering of the two fused
five-membered chelate rings
represent another chiral ele-
ment, which makes the RS and
SR enantiomers unique stereo-
isomers. However, this is only
the case when the chelate ring-
flip process is slow on the NMR
time scale. When this process
becomes fast on the NMR time
scale the RS and SR forms
become identical and they then
represent the meso compound
having an internal mirror plane.
The 1H NMR spectrum of the
8a
8b
10
14
16
bond lengths
Br C14
N1 C7
N1 C8
N1 C23
C3 C7
C8 C9
C9 C10
C9 C14
Pd Br
Pd N1
Pd N2
Pd C14
1.905(2)
1.459(3)
1.468(3)
1.459(3)
1.513(3)
1.514(3)
1.390(3)
1.399(3)
1.9110(6)
1.458(2)
1.457(2)
1.460(2)
1.513(2)
1.509(2)
1.390(2)
1.400(2)
1.9081(17)
1.457(2)
1.461(2)
1.455(2)
1.517(2)
1.509(2)
1.391(2)
1.401(3)
1.4670(15)
1.4620(15)
1.4605(15)
1.5060(16)
1.5113(16)
1.3958(16)
1.3947(16)
1.506(3)
1.510(3)
1.481(3)
1.511(3)
1.501(4)
1.387(4)
1.400(3)
2.5480(3)
2.127(2)
2.132(2)
1.919(2)
C11 O
1.366(2)
1.357(2)
Si2 O
1.6673(12)
bond angles
Br-Pd-N1
Br-Pd-N2
Br-Pd-C14
N1-Pd-N2
N1-Pd-C14
N2-Pd-C14
Pd-N1-C7
Pd-N1-C8
C7-N1-C8
Pd-N2-C15
Pd-N2-C16
C15-N2-C16
O-C11-C10
Si2-O-C11
torsion angles
C25-Si-C21-C22
C25-Si-C1-C2
C31-Si-C21-C22
C31-Si-C1-C2
100.04(6)
97.97(5)
174.62(7)
161.80(8)
81.02(9)
80.81(9)
109.12(15)
107.68(15)
105.67(19)
109.04(14)
117.93(15)
110.1(2)
112.04(16)
112.49(17)
109.61(9)
112.53(13)
113.55(14)
aryl
bromide
ligand
8a
111.55(10)
112.78(12)
117.52(15)
128.32(11)
112.26(13)
117.47(16)
(300 MHz, CD2Cl2, 298 K)
shows a singlet for the NMe
protons at d ˆ 2.30 and a very
broad signal for the benzylic
protons in the range d ˆ 3.20 ±
3.80. The aromatic protons
were all found as multiplets
between d ˆ 7.12 and 7.52 ex-
59.52(17)
56.46(18)
176.27(15)
174.12(16)
55.64(11)
98.90(10)
172.64(9)
144.49(10)
91.3(2)
159.6(2)
144.6(2)
78.6(2)
94.81(15)
54.45(14)
143.89(14)
176.68(12)
83.23(16)
48.30(15)
156.17(15)
169.92(13)
cept for those bound to C2 and C22 (cf. Figure 2a) which
appear as an apparent singlet at d ˆ 7.80.
1
On the other hand, in the H NMR (300 MHz, CD2Cl2,
298 K) of the arene ligand 8b the resonance for the NMe
protons is a singlet at d ˆ 2.24 while the benzylic protons also
appear as a singlet at d ˆ 3.53. The aromatic region is again
characterized by the presence of an apparent singlet at d ˆ
8.07 corresponding to the aromatic protons bound to C2 and
C22 (cf. Figure 2b) and several multiplets between d ˆ 7.06
and 7.59.
These NMR data indicate that in the case of 8b, pyramidal
inversion at the nitrogen centers is a fast process on the NMR
time scale. Accordingly, the potentially diastereotopic ben-
zylic protons appear as a singlet. The decoalescence of the
benzylic protons could not be noted, even at 183 K where only
very broad resonances at d ˆ 3.04, 3.43 and 3.53 for such
benzylic protons were observed. This indicates that the
inversion of configuration at the nitrogen centers has a very
small activation barrier and this process reaches the inter-
mediate exchange rate on the NMR time scale.
A different situation was found for macrocycle 8a, in which
at room temperature broad signals are observed for the
protons of the diastereotopic benzylic protons. At low
temperatures (203 K) the inversion process at the nitrogen
centers is reaching the slow exchange limit shown by the
Figure 3. Displacement ellipsoid plot (50% probability) and numbering
scheme of the molecular structure of the PdII complex 10. Hydrogen atoms
have been omitted for clarity.
C14 ± C11 axis and the Si atom and it is perpendicular to the
C9 ± C13 axis (see Figure 2 and Figure 3).
1H NMR spectra: As mentioned above, from the four possible
stereoisomers that could be formed in 10 (RR, SS, RS, SR)
only the RS/SR pair was observed in the solid state (both RS
50
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0050 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Dendritic Systems
45±57
observation of two AB patterns at d ˆ 2.14 and 3.80 and 3.22
and 3.63, respectively, corresponding to the diastereotopic
benzylic protons. The presence of the bromine atom in 8a
apparently slows down the inversion process at the nitrogen
centers, as this process involves also a movement of the
macrocyclic ring.
Evidence for a rigid nitrogen ± palladium interaction in the
arylpalladium complex 10 on the NMR time scale comes from
the observation of diastereotopic resonance patterns of the
prochiral methylene protons (1H NMR at 300 MHz, C6D6,
298 K). The benzylic protons appear as two AB patterns,
protons bound to C7 and C16, at d ˆ 2.87 and 4.39 while
protons bound to C8 and C15 appear at d ˆ 3.03 and 3.92,
respectively (cf. Figure 3). These patterns establish that the
nitrogen atoms are stereogenic centers with a stable config-
uration. This stability arises from the strong Pd ± N coordina-
tion, which efficiently blocks the inversion process at the
nitrogen centers.
Synthesis and characterization of the multimetallic dendritic
system 1: Having optimized the conditions for the synthesis of
the metallic cage 10, we pursued the synthesis of the desired
dendrimer 1 following a convergent procedure. For the
coupling of this cage to the core benzenetricarbonyl trichlor-
ide 17 we prepared the new functionalized cage 16 with a
phenol group as a binding site (Scheme 5). For the synthesis of
the macrocycle 16 the same synthetic procedure was used as
for the synthesis of ligands 8, that is, a [2 ‡ 2]-macrocyclization
reaction between the bisbenzylic bromide 5 and the bisami-
noaryl bromide 13 (see Scheme 2). Compound 13 was
prepared following the route shown in Scheme 4. Treatment
of a solution of 11[7g] in methanol with PyH ´ Br3 (1 equiv) in
the presence of iron powder afforded the brominated
bisbenzylic alcohol 12.[25] Reaction of 12 with mesyl chloride
using Et3N as a base followed by nucleophilic attack of
methylamine on the resulting mesylate led to the bisamine 13
(88% yield, two steps).
Notably 10 exists as stable RS and SR enantiomers in the
solid state. In these stereoisomers, the NMe groups take
different positions, that is, one axial and one equatorial, while
also the respective protons bound to carbons C2 and C22 have
non-equivalent chemical environments. If these distinct
orientations were to be maintained in solution, one would
expect the 1H NMR spectrum of 10, for example, to show two
resonances for the Me groups and another two for the
respective protons bound to carbons C2 and C22. Thus, the
chelate ring-flip process involving inversion of the five-
membered chelate ring conformations by wagging of the aryl
plane about the C11-C14-Pd axis has a low activation barrier
and is fast on the NMR time scale.[24b] When low temperature
1H NMR spectra of 10 (300 MHz, CD2Cl2, 188 K) were
performed, the slow exchange limit for the ring-flip process
could not be reached. At 188 K three broad signals for the
benzylic protons between d ˆ 3.30 and 5.00 and a broad
singlet at 3.14 for the protons of the methyl group were
observed. This spectrum corresponds to the situation in which
ring flipping is at an intermediate exchange rate on the NMR
time scale.
HO
OH
HO
Br
OH
NH
Br HN
i
ii, iii
OTBDMS
OTBDMS
OTBDMS
11
12
13
Scheme 4. Synthesis of diamine 13. i) PyH ´ Br3, Fe powder, CH2Cl2, RT,
1 h; ii) MsCl, Et3N, CH2Cl2, 788C, 1.5 h; iii) MeNH2, Et2O, 788C, 1 h.
The [2 ‡ 2]-macrocyclization reaction between compounds
5 and 13 was carried out following the best protocol found for
the preparation of ligands 8a and 8b (see above). Thus,
dropwise addition of a solution of 5 in DMF to a suspension of
13 and Na2CO3 also in DMF over a period of 2 h afforded the
desired carbodiazasilane ligand 14 and the dimer 15 in a 30%
and 4% yield, respectively (Scheme 2). When this reaction
was performed using K2CO3 as a base in addition to monomer
14 and dimer 15, formation of the deprotected phenol 16 was
also observed. Further deprotection of macrocycle 14 to yield
the para-functionalized carbodiazasilane ligand 16 took place
by treatment of a solution of 14 in THF with tetrabutylam-
monium fluoride (1.3 equiv) for 1 h (Scheme 5).
13C NMR spectra: APT, DEPTand 13C{H1} NMR experiments
were carried out for solutions of the ligands 8a and 8b in C6D6
and CD2Cl2, respectively, while a 1H,13C COSY spectrum was
also obtained from a solution of complex 10 in C6D6, in order
to assign all signals. Although for macrocycles 8a and 8b the
13C NMR data reveal high symmetry, a different result is
observed for compound 10. The 13C NMR of the arylpalla-
dium macrocycle 10 shows six signals for tertiary carbons and
another two for quaternary carbons corresponding to the
phenyl rings bound to the silicon atom which do not belong to
the macrocycle. This result is not only further evidence for the
strong Pd ± N coordination, which blocks the inversion
process at the nitrogen centers, but also reveals that part of
the 12-membered macrocycle, which does not include the
N,C,N moiety, is now orientated perpendicular to the
coordination plane (cf. Figure 3). As a result, the two phenyl
rings of the Ph2Si moiety have also now become diastereo-
topic, that is, the 12-membered macrocycle has a distinct
puckering at this temperature.
Molecular structures of 14 and 16: Unequivocal confirmation
of the proposed connectivity was obtained from single crystal
structure determinations of 14 and 16 (see Figure 4 and
Table 2). The overall structural features of 14 were expected
to be similar to those found for macrocycle 8a. However, the
configuration of one of the nitrogen centers in 14 is different
than in 8a thus, the two nitrogen centers in 14 have the same
configuration S and therefore the benzylic rings in this
molecule have a different orientation than in 8a (torsion
angles for C25-Si-C1-C2 and C25-Si-C21-C22 are 54.45(14)
and 94.81(15)8, respectively).
In the solid state, ligand 16 forms dimers resulting from the
formation of intermolecular hydrogen bonds between the
phenolic hydrogen atom and one of the nitrogen atoms of the
adjacent molecule. This hydrogen bonding brings about a
Chem. Eur. J. 2002, 8, No. 1
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0051 $ 17.50+.50/0
51

G. van Koten et al.
FULL PAPER
change in the configuration of the nitrogen centers relative to
those in the bromine ligand 8a (N1 and N2 have R and S
configuration in 8a, respectively, while the configuration in 16
for N1 is R and for N2 is S). Therefore, the disposition of the
benzylic rings attached to the Si atom in 14 is again not
comparable to the one in 8a (torsion angles for C25-Si-C1-C2
and C25-Si-C21-C22 are 48.30(15) and 83.23(16)8, respec-
tively).
Ph
Ph
Si
Cl
O
3
N
N
Br
+
Cl
Cl
O
O
OR
We attempted the palladation of ligands 14 and 16 by
oxidative addition to [Pd(dba)2]. Unfortunately, when the
respective solutions of the carbodiazasilanes 14 and 16 in
benzene were heated under reflux overnight in the presence
of [Pd(dba)2] the desired palladated species were not formed
and only starting material could be recovered. Attempts to
increase the reactivity of these compounds by using other
solvents such as toluene or 1,2-dichlorobenzene failed. The
low reactivity of macrocycles 14 and 16 compared with that of
ligand 8a is probably due to the presence of an electron
donating group in the N,C,N moiety. Therefore, we decided to
prepare the dendritic compound 18 (Scheme 5) in order to
obtain cages functionalized with more electron withdrawing
groups, such as ester moieties, expecting to have a more
reactive system towards palladation. Treatment of a solution
of freshly recrystallized 1,3,5-benzenetricarbonyl chloride
(15) in THF with the phenolic ligand 16 (3.5 equiv), in the
presence of Et3N afforded the dendritic molecule 18 in an
17
14 R= TBDMS
16 R= H
i
ii
Ph
Ph
Si
N
N
O
Br
O
iii
1
O
O
N
N
O
O
Br
Br
1
88% yield. The H NMR spectrum of 18 shows a similar
N
N
Ph
Ph
Ph
Si
Ph
Si
pattern to that observed for compound 16 with the most
remarkable difference being the singlet at d ˆ 9.50 corre-
sponding to the three aromatic protons of the triester core.
The final step in the synthesis of 1 was the palladation
reaction of 18 which indeed took place readily by oxidative
addition to [Pd(dba)2]. Thus, refluxing of a solution of the
multicage dendritic system 18 in toluene overnight in the
presence of [Pd(dba)2] gave rise to 1 as a yellow solid in a 60%
yield. A notable feature of 1 is the two diagnostic downfield
18
Scheme 5. Synthesis of the multimetallic dendritic system 1. i) Bu4NF,
THF, RT, 1 h; ii) Et3N, THF, RT, 12 h; iii) Pd(dba)2, toluene, D, 12 h.
1
singlets in the H NMR spectrum which are assigned to the
three protons of the central aromatic triester ring (d ˆ 9.44)
and to the hydrogens bound to C2 and C22 (d ˆ 10.00, cf.
Figure 3), respectively.
Molecular mechanics calculations[26] suggest that the overall
geometry of dendrimer 1 is best described as a helix (see
Figure 5). The average distance between two of the three
palladium centers of 1 is ꢁ1.5 nm. Although this molecule has
a fairly low molecular weight of 2334 Da its helical structure
gives it true nanoparticle size dimensions and thus, appro-
priate properties for retainment by (nano)membrane filtra-
tion materials.[27]
Catalysis: The macrocyclic PdII complex 10 and dendrimer 1
were used as catalyst precursors in the aldol condensation
reaction of benzaldehyde and methyl isocyanoacetate to form
oxazolines (Scheme 6). Their catalytic preformance was
compared with the activity of the corresponding mononuclear
and model compounds [PdBr(2,6-{CH2NMe2}2C6H3)] (19),
[PdI(4-CO2Me-2,6-{CH2NMe2}2C6H2)] (20) and [PdBr(2,6-
{CH2NMeBn}2C6H3)] (21).
Figure 4. Displacement ellipsoid plots (50% probability) and numbering
schemes of the structures of the macrocyclic ligands a) 14 and b) the
hydrogen bonded dimer of 16. Hydrogen atoms have been omitted for
clarity, except H1O in 16.
Consequently, 1, 10, 19, 20 and 21 were converted in their
corresponding (poly)cationic analogues by abstracting the
52
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0052 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Dendritic Systems
45±57
crease in the reaction rate is not observed for the case of the
cyclic analogue 10 (entry 1) demonstrating the positive
influence of the cage. Unfortunately, the steric environment
around the Pd site caused by the presence of the cavity in 10
does not have an influence on the diastereoselectivity of the
reaction.
Conclusion
In the present investigation, we have developed the synthesis
of two new macrocyclic carbodiazasilane molecules contain-
ing the monoanionic N,C,N-ligand 8a and 8b. Macrocycle 8a
was successfully transformed to the organometallic PdII
complex 10 by oxidative addition to [Pd(dba)2]. A remarkable
aspect in the synthesis of complex 10 is the formation of only
the meso compound from all the possible stereoisomers.
Modification of the macrocycles 8a by introduction of a
phenol group at the para-position of the N,C,N moiety
afforded the ligands 14 and 16. The attachment of cage 16
to a central core gave rise to the novel multicage dendritic
structure 18 which was palladated in order to obtain the novel
multimetallic dendrimer 1. The cationic derivative of 1
obtained by halide abstraction with AgBF4 was successfully
applied as homogeneous catalyst showing higher reaction
rates than the mononuclear analogues.
Figure 5. Molecular modelling structure of 1 (MMFF94).
PdII complex/AgBF4
Ph
CO2Me
Ph
CO2Me
O
OMe
iPr2EtN
+
CN
+
O
N
Ph
H
O
N
O
CH2Cl2, RT
cis
trans
Scheme 6. Aldol condensation reaction of benzaldehyde and methyl
isocyanoacetate yielding oxazolines.
halide anion using AgBF4 in wet acetone.[28] After appropiate
work-up involving a thorough filtration of the solutions
containing the cationic complexes through a path of Celite
to remove the insoluble silver halide salts[29] active catalytic
species were isolated and used without further purification in
the aldol condensation reaction.
Experimental Section
General: All sensitive manipulations were performed under a dry and
deoxygenated dinitrogen atmosphere using standard Schlenk techniques
unless otherwise stated. All solvents were carefully dried and distilled prior
to use. All standard chemicals were purchased from Acros Chimica or
Aldrich and used without further purification. 1,3,5-Benzenetricarboxylic
acid chloride was recrystallized from hot hexanes prior to use. Flash
chromatography was performed using 230 ± 400 mesh silica (Merck). The
starting materials 1,3-bis(hydroxymethylbenzene) tert-butyldimethylsilyl
ether,[30] [Pd2(dba)2],[19b] 19[19, 24a] and 20[31] were synthesized according to
literature procedures. 1H (200 or 300 MHz), 13C (50 or 75 MHz) and 29Si
(75 MHz) NMR spectra were recorded on a Varian Inova spectrometer.
Chemical shifts are given in ppm using TMS as an external standard.
Elemental analyses were performed by Dornis and Kolbe, Mikroanaly-
tisches Laboratorium (Mülheim a.d. Ruhr, Germany). MALDI-TOF-MS
spectra were acquired using a Voyager-DE BioSpectrometry Workstation
(PerSeptive Biosystems Inc., Framingham, MA) mass spectrometer
equipped with a nitrogen laser emitting at 337 nm. The instrument was
operated in the linear mode at an accelerating voltage in the range 22000 V.
External calibration was performed using C60/C70, and detection was
performed by means of a linear detector and digitizing oscilloscope
These preliminary experiments were carried out with
iPr2EtN (Huenigꢁs base, 10 mol%) as a base, using 1 mol%
of the catalyst, methyl isocyanoacetate (100 mol%) and
benzaldehyde (100 mol%) in dichloromethane at room
temperature. The results obtained with different PdII com-
plexes are shown in Table 3.
The polycationic dendritic system derived from 1 turned
out to have a higher catalytic activity than the other
mononuclear model compounds tested (entry 2). The pres-
ence of an ester moiety at the para-position of the catalytic
site seems to play an important role in the activity of these
systems (entries 3 and 4). An interesting difference on the
reaction rate was noted between compounds 19 and 21
(entries 3 and 5). Apparently the alkyl substituents at the
nitrogen centers seem to influence the reactivity of these
catalysts for the acyclic compounds. Interestingly, this de-
1
operating at 500 MHz. Sample solutions with ꢁ10 mgmL in THF were
used, and the matrix was 3,5-dihydroxybenzoic acid in THF (10 mgmL 1).
A solution of silver(i) trifluoroacetate in THF was added to the sample in
order to improve the peak resolution. The sample solution (0.2 mL) and the
matrix solution (0.2 mL) were combined and placed on a gold MALDI
target and analyzed after evaporation of the solvents.
Table 3. Aldol condensation reaction of methylisocyanate and benzaldehyde.[a, b]
Entry PdII complex
Time Conver-
[h]
trans/
sion[b] [%] cis[b]
1
2
3
4
5
10 (Pd1 species)
1 (Pd3 species)
[PdBr(2,6-{CH2NMe2}2C6H3)] (19)
[PdI(4-CO2Me-2,6-{CH2NMe2}2C6H2)] (20)
[PdBr(2,6-{CH2NMeBn}2C6H3)](21)
7
4
7
7
7
89
> 99
83
> 99
40
71/29
61/39
62/38
62/38
63/37
Ph2Si(C6H4CH2OSiMe2tBu)2 (3): A sample of 3-bromobenzyl tert-butyldi-
methylsilyl ether (2) (20.15 g, 66.87 mmol) was dissolved in Et2O (250 mL)
and the solution cooled to 788C. tert-Butyllithium (85 mL of a 1.5m
solution in pentane, 127.06 mmol) was added dropwise and the mixture
was stirred for 30 min, followed by the addition of dichlorodiphenylsilane
(7.62 g, 6.33 mL, 30.09 mmol). The yellow suspension was allowed to warm
slowly to room temperature and then stirred overnight. To the resulting
white suspension was added an extra amount of tBuLi (4.25 mL). After
stirring for 10 min, the reaction mixture was quenched with a saturated
[a] Reaction carried out in CH2Cl2 (5 mL) at RT with ca. 10 mol% Hüningꢁs base
[Et(iPr)2N]. [b] In all catalytic rounds the amount of palladium was kept constant
(i.e., ca. 1 mol%). [c] Conversion and trans/cis ratio calculated using specific signal
1
integration in the H NMR spectra.
Chem. Eur. J. 2002, 8, No. 1
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0053 $ 17.50+.50/0
53

G. van Koten et al.
FULL PAPER
aqueous solution of NH4Cl until a clear two-phase system was obtained.
The aqueous layer was separated and washed with Et2O (2 Â 100 mL). The
combined organic layers were washed with H2O (2 Â 50 mL) and brine
(50 mL), and then dried over MgSO4. This solution was filtered and
reduced under vacuum to a crude yellow oil, which was purified by flash
column chromatography (EtOAc/hexanes 1:2). The product was obtained
as a colorless oil (16.36 g, 87%). 1H NMR (300 MHz, C6D6, 258C): d ˆ
0.03 (s, 12H; SiMe2), 0.89 (s, 18H; SitBu), 4.52 (s, 4H; ArCH2), 7.13 ± 7.24
(m, 8H; ArH), 7.39 (d, 3J(H,H) ˆ 7.8 Hz, 2H; ArH), 7.62 (d, 3J(H,H) ˆ
7.2 Hz, 2H; ArH), 7.70 ± 7.74 (m, 4H; ArH), 7.78 (s, 2H; ArH); 13C NMR
(50 MHz, C6D6, 258C): d ˆ 5.10 (SiMe2), 18.5 [SiC(CH3)3], 26.18
[SiC(CH3)3], 65.15 (ArCH2), 127.81, 128.28, 128.32, 129.88, 134.42, 134.71
(C), 134.92 (C), 135.62, 136.92, 141.36 (C); 29Si NMR (75 MHz, C6D6,
258C): d ˆ 13.3 (Ar2Si), 20.0 (SiO); MS (MALDI-TOF): m/z: calcd
column chromatography (EtOAc/hexanes 1:4) to afford two products: the
desired macrocycle 8a (0.14 g, 30%) and the dimer 9 (20 mg, 4%).
Macrocycle 8a: 1H NMR (300 MHz, CD2Cl2, 258C): d ˆ 2.30 (s, 6H;
NCH3), 3.20 ± 3.80 (brm, 8H; ArCH2), 7.12 ± 7.23 (m, 6H; ArH), 7.29 ± 7.52
(m, 13H; ArH), 7.80 (s, 2H; ArH); 1H NMR (300 MHz, C6D6, 258C): d ˆ
2.20 (s, 6H; NCH3), 3.10 ± 3.70 (brm, 8H; ArCH2), 6.89 ± 6.94 (m, 1H;
ArH), 7.02 ± 7.21 (m, 12H; ArH), 7.57 (d, 3J(H,H) ˆ 6.9 Hz, 2H; ArH),
7.66 ± 7.73 (m, 4H; ArH), 8.09 (s, 2H; ArH); 13C NMR (75 MHz, C6D6,
258C): d ˆ 42.48 (NCH3), 58.36 (ArCH2), 61.75 (ArCH2), 126.05, 127.44,
128.10 (C), 128.88 (C), 129.72, 130.07, 130.42, 135.04, 135.34 (C), 137.21 (2 Â
ArH), 137.56, 138.78 (C), 140.24 (C); 29Si NMR (75 MHz, C6D6, 258C): d ˆ
‡
14.47; MS (MALDI-TOF): m/z: calcd for: 603.7; found: 603.3 [M] , 523.7
‡
[M Br] ; elemental analysis calcd (%) for C36H35BrN2Si (603.67): C 71.63,
H 5.84, N 4.64, Si 4.65; found C 71.80, H 5.76, N 4.54, Si 4.55.
‡
for: 732.9; found: 732.8 [M‡Ag] ; elemental analysis calcd (%) for
Dimer 9: 1H NMR (300 MHz, C6D6, 258C): d ˆ 2.00 (s, 12H; NCH3), 3.36
(s, 8H; ArCH2), 3.55 (s, 8H; ArCH2), 6.99 (t, 3J(H,H) ˆ 7.5 Hz, 2H; ArH),
7.16 ± 7.22 (m, 16H; ArH), 7.31 (d, 3J(H,H) ˆ 7.8 Hz, 4H; ArH), 7.39 (d,
3J(H,H) ˆ 7.5 Hz, 4H; ArH), 7.65 (d, 3J(H,H) ˆ 7.2 Hz, 4H; ArH), 7.75 ±
7.79 (m, 8H; ArH), 7.93 (s, 4H; ArH); 13C NMR (75 MHz, C6D6, 258C):
d ˆ 42.18 (NCH3), 61.81 (ArCH2), 62.41 (ArCH2), 126.72 (C), 127.16,
127.91, 128.34, 129.19, 129.90, 130.60, 134.84 (C), 135.05 (C), 135.66, 136.96,
137.40, 139.29 (C), 139.37 (C); 29Si NMR (75 MHz, C6D6, 258C): d ˆ
C38H52O2Si3 (625.07): C 73.02, H 8.39, Si 13.48; found C 73.10, H 8.49, Si
13.57.
Ph2Si(C6H4CH2OH)2 (4): A solution of 3 (3.53 g, 5.65 mmol) in a mixture of
AcOH/THF/H2O (3:1:1, 30 mL) was warmed to 508C and stirred for 3 h.
After this time, the reaction mixture was cooled to room temperature and
all volatiles were removed in vacuo. The residue obtained was dissolved in
Et2O (30 mL) and an aqueous solution of NaOH (1m) was added until
neutral pH was reached. The organic layer was separated, washed with H2O
(2 Â 10 mL) and brine (1 Â 10 mL), and then dried over MgSO4. This
solution was filtered and the volatiles were removed in vacuo to yield a
crude yellow oil, which was purified by flash column chromatography
(EtOAc/hexanes 2:1). The product was obtained as a colorless oil (1.99 g;
‡
13.46; MS (MALDI-TOF): m/z: calcd for 1207.3; found: 1207.6 [M] ,
‡
1128.0 [M Br] ; elemental analysis calcd (%) for C72H70Br2N4Si2 (1207.3):
C 71.63, H 5.84, N 4.64, Si 4.65; found C 71.41, H 5.98, N 4.39, Si 4.48.
Synthesis of macrocycle 8b: This compound was prepared as described for
8a, starting from 7b in 63% yield as a white solid. 1H NMR (300 MHz,
CD2Cl2, 258C): d ˆ 2.24 (s, 6H; NCH3), 3.53 (s, 8H; ArCH2), 7.07 (d,
3J(H,H) ˆ 7.5 Hz, 2H; ArH), 7.18 ± 7.23 (m, 1H; ArH), 7.30 ± 7.49 (m, 12H;
ArH), 7.56 ± 7.59 (m, 5H; ArH), 8.07 (s, 2H; ArH); 1H NMR (200 MHz,
C6D6, 258C): d ˆ 2.05 (s, 6H; NCH3), 3.32 (s, 4H; ArCH2), 3.41 (s, 4H;
ArCH2), 6.96 (d, 3J(H,H) ˆ 7.4 Hz, 4H; ArH), 7.10 ± 7.20 (m, 9H; ArH),
7.62 ± 7.65 (m, 2H; ArH), 7.72 ± 7.76 (m, 4H; ArH), 7.96 (s, 1H; ArH), 8.42
(s, 2H; ArH); 13C NMR (75 MHz, CD2Cl2, 258C): d ˆ 43.12 (NCH3), 60.72
(ArCH2), 61.77 (ArCH2), 128.00, 128.04, 128.16, 128.44, 129.01 (C), 130.14,
131.24, 134.91 (C), 135.23 (C), 135.68, 136.98, 137.12, 139.72 (C); MS
1
89%). H NMR (200 MHz, C6D6, 258C): d ˆ 3.39 (brs, 2H; OH), 4.20 (s,
4H; ArCH2), 7.04 ± 7.18 (m, 10H; ArH), 7.57 (d, 3J(H,H) ˆ 6.2 Hz, 2H;
ArH), 7.66 ± 7.69 (m, 4H; ArH), 7.76 (s, 2H; ArH); 13C NMR (50 MHz,
C6D6, 258C): d ˆ 64.83 (ArCH2), 128.37, 128.43, 128.90, 129.99, 134.75 (C),
‡
134.82 (C), 135.24, 135.90, 136.89, 141.33 (C); FAB-MS: m/z: 396.1 [M] ;
elemental analysis calcd (%) for C26H24O2Si (396.55): C 78.75, H 6.10, Si
7.08; found C 78.64, H 6.14, Si 7.02.
Ph2Si(C6H4CH2Br)2 (5): A solution of PBr3 (0.94 g, 3.45 mmol) in benzene
(10 mL) was slowly added (15 min) to a solution of 4 (1.99 g, 4.97 mmol) in
benzene (20 mL) at room temperature. After stirring the reaction mixture
for 3 h, the solvent was removed in vacuo. The residue obtained was
purified by flash column chromatography (EtOAc/hexanes 0.25:9.75) to
afford the desired bisbenzyl bromide 7 as a colorless oil (2.13 g, 91%).
1H NMR (300 MHz, C6D6, 258C): d ˆ 3.87 (s, 4H; ArCH2), 7.02 (t,
3J(H,H) ˆ 7.2 Hz, 2H; ArH), 7.10 ± 7.22 (m, 8H; ArH), 7.47 ± 7.50 (m, 2H;
ArH), 7.60 ± 7.63 (m, 6H; ArH); 13C NMR (50 MHz, C6D6, 258C): d ˆ 33.41
(ArCH2), 128.45, 128.75, 130.21, 130.90, 134.00 (C), 135.23 (C), 136.71,
136.80, 137.00, 138.10 (C); 29Si NMR (75 MHz, C6D6, 258C): d ˆ 13.59;
‡
(MALDI-TOF): m/z: calcd for: 524.8; found: 524.0 [M] ; elemental
analysis calcd (%) for C36H36N2Si (603.67): C 82.40, H 6.91, N 5.34, Si 5.35;
found C 82.46, H 6.85, N 5.28, Si 5.29.
Synthesis of the PdII complex 10: [Pd(dba)2] (66 mg, 0.11 mmol) was added
to a solution of 8a (63 mg, 0.10 mmol) in benzene (5 mL). The resulting
solution was refluxed overnight, during which time the color changed from
deep purple to yellow. The reaction mixture was filtered through Celite and
the solvent removed under reduced pressure. The solid residue was
dissolved in wet acetone (8 mL) and AgBF4 (21 mg, 0.11 mmol) was added,
the suspension was stirred for 1 h. After this time, the resulting cloudy
suspension was filtered through Celite, and the solvent was concentrated to
ca. 2 mL. Et2O was added to this solution in order to precipitate the
product. The precipitate was purified by washing several times with Et2O.
The precipitate was then dissolved in CH2Cl2 and an excess of LiBr was
added to the solution, which was stirred for 2 h. Filtration of the suspension
through Celite and revomal of the solvent in vacuo afforded compound 10
which was isolated as an air and temperature-stable yellow crystalline solid
(35 mg, 47%). 1H NMR (300 MHz, CD2Cl2, 258C): d ˆ 3.21 (s, 6H; NCH3),
3.40 (d, 3J(H,H) ˆ 11.8 Hz, 2H; ArCH2), 3.63 (d, 3J(H,H) ˆ 14.8 Hz, 2H;
ArCH2), 4.23 (d, 3J(H,H) ˆ 14.8 Hz, 2H; ArCH2), 4.50 (d, 3J(H,H) ˆ
‡
FAB-MS: m/z: 521/523/525 [M‡H] ; elemental analysis calcd (%) for
C26H22Br2Si (522.35): C 59.78, H 4.25, Si 5.38; found C 59.72, H 4.18, Si 5.41.
1-Bromo-2,6-bis[(methylamino)methyl]benzene (7a): A solution of 2-bro-
mo-1,3-bis(bromomethyl)benzene (6a) (2.07 g, 6.04 mmol) in Et2O
(20 mL) was cooled to 08C. MeNH2 was bubbled through this solution
for a period of 4 min. The reaction mixture was then stirred for 20 min and
MeNH2 gas was again bubbled through the reaction for 4 min. Formation of
a white precipitate was observed. The suspension was allowed to warm to
room temperature and stirred for 1 h. After this time, water was added and
the organic layer was separated. The obtained organic layer was washed
with brine (10 mL), dried over MgSO4 and concentrated to afford the
desired diamine 7a as a yellow oil (1.00 g, 68%). 1H NMR (200 MHz, C6D6,
258C): d ˆ 0.92 (brs, 2H; NH), 2.21 (brs, 6H; NCH3), 3.74 (brs, 4H;
ArCH2), 7.05 (t, 3J(H,H) ˆ 5.0 Hz, 2H; ArH), 7.23 (d, 3J(H,H) ˆ 5.0 Hz,
1H; ArH); 13C NMR (75 MHz, CDCl3, 258C): d ˆ 35.73 (NCH3), 56.11
(ArCH2), 125.57 (C), 126.94, 128.95, 139.46 (C).
3
11.8 Hz, 2H; ArCH2), 6.70 (d, J(H,H) ˆ 7.5 Hz, 2H; ArH), 6.88 (m, 1H;
1
ArH), 7.28 ± 7.59 (m, 16H; ArH), 9.58 (s, 2H; ArH); H NMR (300 MHz,
C6D6, 258C): d ˆ 2.87 (d, 3J(H,H) ˆ 12.0 Hz, 2H; ArCH2), 2.91 (s, 6H;
NCH3), 3.03 (d, 3J(H,H) ˆ 14.6 Hz, 2H; ArCH2), 3.92 (d, 3J(H,H) ˆ
14.6 Hz, 2H; ArCH2), 4.39 (d, 3J(H,H) ˆ 12.0 Hz, 2H; ArCH2), 6.46 (d,
3
Synthesis of ligands 8a and 9: The following is an example of the
experimental conditions used for the performance of the [2‡2]-macro-
cyclization reaction. A suspension of 2-bromo-1,3-bis[(methylamino)me-
thyl]benzene (0.19 g, 0.80 mmol) and Na2CO3 (0.21 g, 2.01 mmol) in DMF
(60 mL) was stirred for 15 min at 508C. A solution of 5 (0.42, 0.80 mmol) in
DMF (20 mL) was then added dropwise over 2 h and the reaction mixture
was stirred overnight at this temperature. DMF was removed in vacuo and
the crude product was dissolved in CH2Cl2 (80 mL). The organic layer was
washed with H2O (2 Â 40 mL) and brine (2 Â 40 mL), dried over MgSO4,
filtered and concentrated. The residue obtained was purified by flash
3J(H,H) ˆ 7.5 Hz, 2H; ArH), 6.80 (t, J(H,H) ˆ 7.4 Hz, 1H; ArH), 6.87 (d,
3J(H,H) ˆ 7.5 Hz, 2H; ArH), 7.06 ± 7.25 (m, 8H; ArH), 7.69 (d, 3J(H,H) ˆ
7.2 Hz, 4H; ArH), 7.87 ± 7.90 (m, 2H; ArH), 10.09 (s, 2H; ArH); 13C NMR
(50 MHz, C6D6, 258C): d ˆ 52.49 (NCH3), 66.97 (ArCH2), 69.71 (ArCH2),
119.79, 124.35, 127.60, 128.34, 128.44, 129.90, 129.99, 133.24, 134.01 (C),
134.77 (C), 134.94 (C), 135.63 (C), 136.78, 137.05, 137.31, 140.64, 145.46 (C),
157.87 (C); MS (MALDI-TOF): m/z: calcd for 630.2; found: 630.5 [M
‡
‡
Br] , 524.7 [M PdBr] ; elemental analysis calcd (%) for C36H35BrN2PdSi
(710.09): C 60.89, H 4.97, N 3.95, Si 3.96; found C 61.00, H 4.90, N 3.99,
Si 3.87.
54
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0054 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Dendritic Systems
45±57
1-Bromo-2,6-bis(hydroxymethyl)benzene tert-butyldimethylsilyl ether
(12): A solution of PyH ´ Br3 (3.66 g, 11.44 mmol) in MeOH (30 mL) was
added dropwise at room temperature to a suspension of 3,5-bis(hydroxy-
methyl)benzene tert-butyldimethylsilyl ether (11) (3.07 g, 11.44 mmol) and
iron powder in CH2Cl2 (30 mL). After 1 h, the reaction mixture was
filtered, washed with water (2 Â 15 mL) and brine (2 Â 15 mL) and dried
over MgSO4. The solvent was concentrated to afford the desired dialcohol
Synthesis of dendrimer 18: Et3N (0.20 mL, 1.44 mmol) was added to a
solution of 16 (67 mg, 0.11 mmol) in THF (5 mL) under nitrogen at room
temperature. The reaction mixture was stirred for 15 min and then 1,3,5-
benzenetricarboxylic acid chloride (91 mg, 0.03 mmol) was added. The
resulting solution was allowed to stir overnight. After this time, the solvent
was removed in vacuo, the residue was dissolved in CH2Cl2 (10 mL) and
washed with water (5 mL) and brine (5 mL). The organic layer was then
dried with MgSO4 and evaporated to afford 18 as a white solid (61 mg,
88%). 1H NMR (200 MHz, C6D6, 258C): d ˆ 2.20 (s, 18H; NCH3), 2.80 ±
3.90 (brm, 24H; ArCH2), 7.04 ± 7.22 (m, 30H; ArH), 7.28 (s, 6H, ArH), 7.55
(d, 3J(H,H) ˆ 6.8 Hz, 6H; ArH), 7.70 (brm, 12H; ArH), 7.93 (s, 6H; ArH),
9.50 (s, 3H, ArH); 13C NMR (75 MHz, C6D6, 258C): d ˆ 42.87 (NCH3),
58.36 (ArCH2), 60.73 (ArCH2), 123.16, 124.82 (C), 126.67 (C), 126.98 (C),
127.22 (C), 129.77, 130.43, 131.85 (C), 135.14, 135.27, 136.12 (C), 137.19,
1
12 as a white solid (2.88 g, 73%). H NMR (200 MHz, CDCl3, 258C): d ˆ
0.20 (s, 6H; SiMe2), 0.98 (s, 9H; SitBu), 2.19 (s, 2H; OH), 4.67 (s, 4H;
ArCH2), 6.92 (s, 2H; ArH); 13C NMR (50 MHz, CDCl3, 258C): d ˆ 4.38
(SiMe2), 18.17 [SiC(CH3)3], 25.64 [SiC(CH3)3], 64.91 (ArCH2), 119.31 (C),
141.33, 155.57 (C), 164.63 (C); elemental analysis calcd (%) for C14H23BrO3-
Si (347.34): C 48.41, H 6.67, Si 8.09; found C 48.44, H 6.59, Si 8.27.
1-Bromo-2,6-bis((methylamino)methyl)benzene
tert-butyldimethylsilyl
ˆ
137.69, 137.96, 141.77, 149.46 (C), 163.09 (C O); MS (MALDI-TOF): m/z:
ether (13): Et3N (2.80 mL, 20.15 mmol) was added to a solution of 12
(1.00 g, 2.88 mmol) in CH2Cl2 (20 mL) under nitrogen at 788C and
subsequently mesyl chloride (1.65 g, 1.11 mL, 14.40 mmol). The reaction
was stirred for 1.5 h at this temperature and then MeNH2 was bubbled
through the reaction for a period of 5 min. Formation of a white precipitate
was observed. The suspension was allowed to warm up to room temper-
ature and stirred for 1 h. After this time, water was added (10 mL) and the
organic layer was separated. The obtained organic layer was washed with
brine (10 mL), dried over MgSO4 and concentrated to afford the desired
diamine 13 as a yellow oil (0.95 g, 88%). 1H NMR (200 MHz, C6D6, 258C):
d ˆ 0.17 (s, 6H; SiMe2), 0.99 (s, 9H; SitBu), 2.22 (brs, 6H; NCH3), 3.71
(brm, 4H; ArCH2), 7.07 (s, 2H; ArH); 13C NMR (50 MHz, C6D6, 258C):
d ˆ 4.18 (SiMe2), 18.49 [SiC(CH3)3], 26.10 [SiC(CH3)3], 35.98 (NCH3),
56.13 (ArCH2), 116.78 (C), 120.24, 141.51 (C), 155.37 (C).
‡
‡
‡
calcd for 2015.1; found: 2123.9 [M‡Ag] , 2014.3 [M] , 1934.6 [M Br] ;
elemental analysis calcd (%) for C117H105Br3N6O6Si3 ´ 3ꢂ2CH2Cl2 (2142.6): C
66.43, H 5.08, N 3.92; found C 66.28, H 5.19, N 3.56.
Synthesis of dendrimer 1: [Pd(dba)2] (66 mg, 0.11 mmol) was added to a
stirred solution of 18 (59 mg, 0.03 mmol) in toluene (10 mL). The resulting
solution was heated under reflux overnight, during which time the color
changed from deep purple to yellow. The reaction mixture was diluted with
CH2Cl2, filtered through Celite and the solvent removed under reduced
pressure. The solid residue was dissolved in CH2Cl2 (5 mL) and hexanes
were added in order to precipitate the product. The precipitate was purified
by washing first with hexane (2 Â 20 mL) and then with Et2O (20 mL) to
afford 1 as yellow powder (40 mg; 60%). 1H NMR (300 MHz, C6D6, 258C):
d ˆ 2.75 ± 2.84 (m, 6H; ArCH2 and 18H; NCH3), 2.92 (d, 3J(H,H) ˆ
15.0 Hz, 6H; ArCH2), 3.81 (d, 3J(H,H) ˆ 15.0 Hz, 6H; ArCH2), 4.32 (d,
3J(H,H) ˆ 12.6 Hz, 6H; ArCH2), 6.40 (d, 3J(H,H) ˆ 6.3 Hz, 6H; ArH),
6.80 ± 6.82 (m, 6H; ArH), 7.03 ± 7.16 (m, 24H; ArH), 7.38 (s, 6H; ArH), 7.68
(m, 12H; ArH), 7.84 ± 7.87 (m, 6H; ArH), 9.44 (s, 3H; ArH), 10.02 (s, 6H;
ArH); 13C NMR (50 MHz, C6D6, 258C): d ˆ 52.60 (CH3), 67.05 (ArCH2),
69.60 (ArCH2), 113.26, 123.15, 127.91, 128.59, 129.82 (C), 130.02, 131.83 (C),
132.19 (C), 133.22, 133.87 (C), 134.60, 134.85 (C), 135.72 (C), 136.76, 137.15,
Synthesis of macrocycles 14 and 15: Macrocycles 14 and 15 were prepared
as described for 8a, starting from 5 and 13. Yields: 30% for 14 and 4% for
15. Macrocycle 14: 1H NMR (200 MHz, C6D6, 258C): d ˆ 0.11 (s, 6H;
SiMe2), 0.95 (s, 9H; SitBu), 2.18 (s, 6H; NCH3), 2.80 ± 3.90 (brs, 8H;
3
ArCH2), 6.96 ± 7.18 (m, 12H; ArH), 7.53 (d, J(H,H) ˆ 7.0 Hz, 2H; ArH),
7.64 ± 7.67 (brm, 4H; ArH), 8.01 (s, 2H; ArH); 13C NMR (75 MHz, C6D6,
258C): d ˆ 4.25 (SiMe2), 18.43 [SiC(CH3)3], 25.87 [SiC(CH3)3], 42.66
(NCH3), 58.10 (ArCH2), 61.26 (ArCH2), 119.88 (C), 122.12, 127.45, 127.74
(C), 128.38, 129.71, 130.24, 135.12, 135.23 (C), 137.21, 137.73, 138.31 (C),
141.22 (C), 154.39 (C); 29Si NMR (75 MHz, C6D6, 258C): d ˆ 14.28
(Ar4Si), 21.30 (SiO); MS (MALDI-TOF): m/z: calcd for: 732.2; found:
ˆ
137.29, 140.66, 146.13, 148.48 (C), 154.65 (C), 163.31 (C O); MS (MALDI-
‡
TOF): m/z: calcd for 2254.5; found: 2254.9 [M Br] , 2148.8
‡
‡
[M PdBr] , 1961.6 [M 2(PdBr)] ; elemental analysis calcd (%) for
117H105Br3N6O6Pd3Si3 ´ 3C6H6 (2568.8): C 63.12, H 4.83, N 3.27; found C
C
63.05, H 4.12, N 3.31.
‡
‡
730.9 [M] , 649.8 [M Br] ; elemental analysis calcd (%) for
C42H49BrN2OSi2 (733.96): C 68.73, H 6.73, N 3.82, Si 7.65; found C 68.49,
H 6.85, N 3.67, Si 7.64.
Procedure for the PdII-catalyzed aldol reaction: The following is an
example of the experimental conditions used for the PdII catalyzed aldol
condensation reaction. A suspension of 10 (10 mg, 0.016 mmol) and AgBF4
(4 mg, 0.020 mmol) in CH2Cl2 (5 mL) was stirred for ca. 30 min at room
temperature. The resulting cloudy solution was filtered through Celite.
Solvent was removed under reduced pressure to give the active catalyst
which was dissolved in CH2Cl2 (5 mL). To this solution were sequentially
added benzaldehyde (173 mg, 1.63 mmol), diisopropylethylamine (21 mg,
0.16 mmol) and methyl a-isocyanoacetate (161 mg, 1.62 mmol). The
mixture was stirred at room temperature for 24 h. From the clear and
completely homogeneous, stirred reaction mixture samples (0.1 mL) were
taken after regular time intervals and analyzed by 1H NMR spectroscopy in
CDCl3 after careful removal of the solvent. 1H NMR (300 MHz, CDCl3,
258C): d ˆ 3.19 (s, 3H, CO2CH3, cis isomer), 3.83 (s, 3H, CO2CH3, trans
Dimer 15: 1H NMR (200 MHz, C6D6, 258C): d ˆ 0.09 (s, 12H; SiMe2), 0.92
(s, 18H; SitBu), 2.01 (s, 12H; NCH3), 3.35 (s, 8H; ArCH2), 3.51 (s, 8H;
ArCH2), 7.11 ± 7.24 (m, 20H; ArH), 7.49 (d, 3J(H,H) ˆ 7.8 Hz, 4H; ArH),
7.59 (d, 3J(H,H) ˆ 7.4 Hz, 4H; ArH), 7.69 ± 7.74 (m, 8H; ArH), 7.78 (s, 4H;
ArH); 13C NMR (75 MHz, C6D6, 258C): d ˆ 4.20 (SiCH3), 18.51
[SiC(CH3)3], 25.96 [SiC(CH3)3], 42.26 (NCH3), 61.65 (ArCH2), 62.44
(ArCH2), 118.18 (C), 121.06, 128.34, 129.90 130.58, 134.65 (C), 135.10
(C), 135.77, 136.96 (2 Â ArH), 137.38, 139.19 (C), 140.55 (C), 155.28 (C); 29Si
NMR (75 MHz, C6D6, 258C): d ˆ 13.55 (Ar4Si), 21.61 (SiO); MS
‡
(MALDI-TOF): m/z: calcd for: 1467.9; found: 1468.0 [M] , 1387.7
‡
[M Br] .
3
4
Synthesis of ligand 16: Tetrabutylammonium fluoride (1m, 0.04 mL,
0.04 mmol) was added to a solution of 14 (0.20 g, 0.03 mmol) in THF
(5 mL). The reaction was stirred at room temperature for 1 h. After this
time, the solvent was removed in vacuo and the residue was dissolved in
CH2Cl2 (10 mL), washed with a saturated aqueous solution of NH4Cl
(5 mL) and brine (5 mL), dried over MgSO4 and evaporated to dryness.
The phenol 16 was obtained as a white powder (160 mg, 85%) which was
crystallized by diffusion of pentane into a concentrated solution of 16 in
isomer), 4.62 (dd, J(H,H) ˆ 7.8, J(H,H) ˆ 2.2 Hz, 1H; -CHCO2Me, trans
isomer), 5.08 (dd, 3J(H,H) ˆ 11.1, 4J(H,H) ˆ 1.8 Hz, 1H; -CHCO2Me, cis
isomer), 5.68 (d, 3J(H,H) ˆ 7.8 Hz, 1H; ArC(H)O), 5.73 (d, 3J(H,H) ˆ
11.1 Hz, 1H; ArC(H)O), 7.11 (d, 3J(H,H) ˆ 2.2 Hz, 1H; HC ˆ N trans
isomer, other signal not visible due to overlap), 7.22 ± 7.43 (m, 11H; ArH).
Crystal structure determinations: X-ray intensities were measured on a
Nonius KappaCCD diffractometer with rotating anode (l ˆ 0.71073 Š) at a
temperature of 150(2) K. The structures were solved with the automated
Patterson program DIRDIF[32] (compound 10) and the direct methods
programs SIR97[33] (compound 16) and SHELXS97[34] (compounds 8a, 8b,
and 14). The structures were refined with SHELXL97[35] against F2 of all
reflections. Non-hydrogen atoms were refined with anisotropic displace-
ment parameters, hydrogen atoms were refined as rigid groups. Structure
calculations, checking for higher symmetry and preparations of molecular
plots were performed with the PLATON[36] package. Further experimental
details are given in Table 4.
1
CH2Cl2. H NMR (200 MHz, C6D6, 258C): d ˆ 2.21 (s, 6H; NCH3), 2.80 ±
3.90 (brm, 8H; ArCH2), 6.56 (s, 2H; ArH), 7.01 ± 7.22 (m, 10H; ArH), 7.58
(d, 3J(H,H) ˆ 6.8 Hz, 2H; ArH), 7.70 (brm, 4H; ArH), 8.04 (s, 2H; ArH);
13C NMR (75 MHz, CDCl3, 258C): d ˆ 42.76 (NCH3), 57.63 (ArCH2), 60.65
(ArCH2), 117.02, 117.82 (C), 126.94, 127.62, 129.34, 129.98, 134.60, 136.32
(C), 136.37 (C), 136.67, 137.23, 137.46 (C), 140.39 (C), 154.13 (C); MS
‡
‡
(MALDI-TOF): m/z: calcd for: 619.7; found: 619.3 [M] , 539.2 [M Br] ;
elemental analysis calcd (%) for C36H35BrN2OSi (619.67): C 69.78, H 5.69,
N 4.52, Si 4.53; found C 69.86, H 5.63, N 4.40, Si 4.68.
Chem. Eur. J. 2002, 8, No. 1
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0055 $ 17.50+.50/0
55

G. van Koten et al.
FULL PAPER
Table 4. Crystallographic data for the compounds 8a, 8b, 10, 14, 16.
8a
8b
10
14
16
formula
MW
C36H35BrN2Si
603.66
C36H36N2Si
524.76
C36H35BrN2PdSi
710.06
C42H49BrN2OSi2
733.92
C36H35BrN2OSi[37]
619.66
crystal size [mm3]
crystal colour
crystal system
space group
a [Š]
b [Š]
c [Š]
a [8]
b [8]
0.50 Â 0.25 Â 0.09
pale yellow
triclinic
0.45 Â 0.29 Â 0.21
colourless
triclinic
0.38 Â 0.38 Â 0.25
0.35 Â 0.18 Â 0.11
colourless
monoclinic
P21/c (No. 14)
11.3853(1)
11.8896(1)
29.2089(2)
90
95.5501(4)
90
3935.38(5)
4
0.30 Â 0.30 Â 0.18
colourless
monoclinic
C2/c (No. 15)
27.7378(3)
18.5891(1)
15.7505(1)
90
yellow
monoclinic
P21/c (No. 14)
11.3370(4)
13.7887(5)
19.4084(7)
90
92.3084(9)
90
3031.51(19)
4
Å
Å
P1 (No. 2)
P1 (No. 2)
9.9087(4)
11.6169(5)
14.4548(5)
67.583(2)
88.335(2)
79.043(2)
1508.36(10)
2
9.9722(1)
12.1487(2)
13.2232(1)
82.8613(7)
71.5698(8)
74.0779(6)
1460.27(3)
2
120.0925(4)
90
g [8]
V [Š3]
7026.67(10)
8
Z
3
1 [gcm
m [mm
]
1.329
1.430
1.193
0.108
1.556
1.998
1.239
1.139
1.172[37]
1
]
1.232[37]
abs. correction
PLATON[36]
(MULABS)
0.78 ± 0.87
13623
PLATON[36]
(MULABS)
0.94 ± 0.99
30139
PLATON[36]
(MULABS)
0.60 ± 0.72
21404
none
none
transmission
measured refl.
unique refl.
±
±
98696
9028
76492
8051
6887
6512
6924
param./restraints
R1/wR2 (all refl.)
R1/wR2 (obs. refl.)
GoF
363/0
354/0
372/0
440/0
374/0
0.0523/0.0864
0.0375/0.0805
1.031
0.0427/0.0973
0.0367/0.0933
1.026
0.0390/0.0746
0.0307/0.0717
1.085
0.0437/0.0854
0.0332/0.0804
1.021
0.0453/0.0975
0.0405/0.0947
1.093
3
1 (max/min) [e Š
]
0.43/0.35
0.27/0.28
0.59/0.51
0.40/0.30
0.51/0.50
Â
Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication no. CCDC 162098 (8a),
162099 (8b), 162100 (10), 162101 (14), and 162102 (16). Copies of the data
can be obtained free of charge on application to CCDC, 12 Union Road,
Cambridge CB21EZ, UK (fax: (‡44)1223-336-033; e-mail: deposit@ccdc.
cam.ac.uk).
120, 5480 ± 5487; d) I. Cuadrado, C. M. Casado, B. Alonso, M. Moran,
J. Losada, V. Belsky, J. Am. Chem. Soc. 1997, 119, 7613 ± 7614.
[4] See for example: a) C. Köllner, B. Pugin, A. Togni, J. Am. Chem. Soc.
1998, 120, 10274 ± 10275; b) M. T. Reetz, G. Lohmer, R. Schwickardi,
Angew. Chem. 1997, 109, 1559 ± 1562; Angew. Chem. Int. Ed. Engl.
1997, 36, 1526 ± 1529; c) T. Marquardt, U. Lüning, Chem. Commun.
1997, 1681 ± 1682; d) J.-J. Lee, W. T. Ford, J. A. Moore, Y. Li, Macro-
molecules 1994, 27, 4632 ± 4634.
[5] J. W. J. Knapen, A. W. van der Made, J. C. De Wilde, P. W. M. N.
van Leeuwen, P. Wijkens, D. M. Grove, G. van Koten, Nature 1994,
372, 659 ± 663.
[6] a) M. Albrecht, G. van Koten, Angew. Chem. 2001, 113, 3866 ± 3898;
Angew. Chem. Int. Ed. 2001, 40, 3750 ± 3781; b) M. P. H. Rietveld,
D. M. Grove, G. van Koten, New J. Chem. 1997, 21, 751 ± 771, and
references therein; c) G. van Koten, Pure Appl. Chem. 1989, 61,
1681 ± 1694.
Acknowledgements
We wish to thank Prof. A. Canty and Dr. S. Williams for helpful discussions
during the preparation of this manuscript. G. Rodríguez thanks the
Â
Ministerio de Educacion y Cultura (Spain) for a postdoctoral fellowship
and the European Commission for a TMR Grant (Contract No. HPMF-CT-
1999-00236). This work was supported by the Council for Chemical
Sciences of the Netherlands Organization for Scientific Research (CW-
NWO).
[7] a) A. W. Kleij, R. A. Gossage, R. J. M. Klein Gebbink, N. Brinkman,
E. J. Reyerse, U. Kragl, M. Lutz, A. L. Spek, G. van Koten, J. Am.
Chem. Soc. 2000, 122, 12112 ± 12124; b) A. W. Kleij, R. A. Gossage,
J. T. B. H. Jastrzebski, J. Boersma, G. van Koten, Angew. Chem. 2000,
112, 176 ± 178; Angew. Chem. Int. Ed. 2000, 39, 176 ± 178; c) R. A.
Gossage, J. T. B. H. Jastrzebski, J. van Ameijde, S. J. E. Mulders, A. J.
Brouwer, R. M. J. Liskamp, G. van Koten, Tetrahedron Lett. 1999, 40,
1413 ± 1416; d) A. W. Kleij, H. Kleijn, J. T. B. H. Jastrzebski, W. J. J.
Smeets, A. L. Spek, G. van Koten, Organometallics 1999, 18, 268 ± 276;
e) M. Albrecht, G. van Koten, Adv. Mater. 1999, 11, 171 ± 174; f) M.
Albrecht, R. A. Gossage, A. L. Spek, G. van Koten, Chem. Commun.
1998, 1003 ± 1004; g) P. J. Davies, D. M. Grove, G. van Koten, Organo-
metallics 1997, 16, 800 ± 802.
[8] See for example: a) R. Breinbauer, E. N. Jacobsen, Angew. Chem.
2000, 112, 3750 ± 3753; Angew. Chem. Int. Ed. 2000, 39, 3604 ± 3607;
b) D. de Groot, E. B. Eggeling, J. C. de Wilde, H. Kooijman, R. J.
van Haaren, A. W. van der Made, A. L. Spek, D. Vogt, J. N. H. Reek,
P. C. J. Kamer, P. W. N. M. van Leeuwen, Chem. Commun. 1999,
1623 ± 1624; c) K. Vassilev, W. T. Ford, Polym. Prepr. 1998, 39, 322 ±
323; d) T. Suzuki, Y. Hirokawa, K. Ohtake, T. Shibata, K. Soai,
Tetrahedron: Asymmetry 1997, 8, 4033 ± 4040.
[1] G. Newkome, C. N. Moorefield, F. Vögtle, Dendritic Molecules-
Concepts, Syntheses, Perspective, VCH, Weinheim, 1996.
[2] For recent reviews on dendritic molecules, see: a) R. Kreiter, A. W.
Kleij, R. J. M. Klein Gebbink, G. van Koten, Top. Curr. Chem. 2001,
217, 163 ± 199; b) G. Newkome, E. He, C. N. Moorefield, Chem. Rev.
1999, 99, 1689 ± 1746; c) A. W. Bosman, H. M. Janssen, E. W. Meijer,
Chem. Rev. 1999, 99, 1665 ± 1688; d) J.-P. Majoral, A.-M. Caminade,
Chem. Rev. 1999, 99, 845 ± 880; e) H. Frey, C. Lach, K. Lorenz, Adv.
Mater. 1998, 10, 279 ± 293; f) C. Gorman, Adv. Mater. 1998, 10, 295 ±
309; g) A. Archut, F. Vögtle, Chem. Soc. Rev. 1998, 27, 233 ± 240; h) F.
Zeng, S. Zimmerman, Chem. Rev. 1997, 97, 1681 ± 1712; i) D. Gudat,
Angew. Chem. 1997, 109, 2039 ± 2043; Angew. Chem. Int. Ed. Engl.
1997, 36, 1951 ± 1955.
[3] See for example: a) M. A. Hearshaw, J. R. Moss, Chem. Commun.
1999, 1 ± 8; b) C. M. Cardona, A. E. Kaifer, J. Am. Chem. Soc. 1998,
120, 4023 ± 4024; c) P. Ceroni, F. Paolucci, C. Parasisi, A. Juris, S.
Rofia, S. Serroni, S. Campagna, A. J. Bard, J. Am. Chem. Soc. 1998,
[9] See for example: a) P. B. Rheiner, H. Sellner, D. Seebach, Helv. Chim.
Acta 1997, 80, 2027 ± 2032; b) C. C. Mak, H.-F. Chow, Macromolecules
56
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0056 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Dendritic Systems
45±57
1997, 30, 1228 ± 1230; c) D. Seebach, R. E. Marti, T. Hintermann, Helv.
Chim. Acta 1996, 79, 1710 ± 1740; d) P. Bhyrappa, J. K. Young, J. S.
Moore, K. S. Suslick, J. Am. Chem. Soc. 1996, 118, 5708 ± 5711.
[10] P. B. Rheiner, D. Seebach, Chem. Eur. J. 1999, 5, 3221 ± 3236.
[11] H. Brunner, J. Organomet. Chem. 1995, 500, 39 ± 46.
[12] H.-F. Chow, C. C. Mak, J. Org. Chem. 1997, 62, 5116 ± 5127.
[13] We are also interested in the linkage between different branches of the
dendrimer with the concomitant formation of cages where the
reactions would take place, and therefore, the size and structure of
the cavity would determine the reaction products.
[14] a) J. M. Longmire, X. Zhang, M. Shang, Organometallics 1998, 17,
4374 ± 4379; b) M. A. Stark, C. J. Richards, Tetrahedron Lett. 1997, 38,
5881 ± 5884; c) R. Nesper, P. Pregosin, K. Püntener, M. Wörle, A.
Albinati, J. Organomet. Chem. 1996, 507, 85 ± 101; d) F. Gorla, A.
Togni, L. M. Venanzi, A. Albinati, F. Lianza, Organometallics 1994,
13, 1607 ± 1616; e) R. Nesper, P. S. Pregosin, K. Püntener, M. Wörle,
Helv. Chim. Acta 1993, 76, 2239 ± 2249; f) T. Hayashi, M. Sawamura,
Y. Ito, Tetrahedron 1992, 48, 1999 ± 2012.
[15] See for example: a) S. Höger, A.-D. Meckenstock, S. Müller, Chem.
Eur. J. 1998, 4, 2423 ± 2434; b) V. Hensel, K. Lützow, J. Jacob, K.
Gessler, W. Saenger, A.-D. Schüter, Angew. Chem. 1997, 109, 2768 ±
2770; Angew. Chem. Int. Ed. Engl. 1997, 36, 2654 ± 2655; c) Y.
Nagasaki, E. Honzawa, M. Kato, K. Kataoka, T. Tsuruta, Chem. Lett.
[22] a) M.-C. Lagunas, R. A. Gossage, A. L. Spek, G. van Koten, Organo-
metallics 1998, 17, 731 ± 741; b) J. Terheijden, G. van Koten, F. Muller,
D. M. Grove, K. Vrieze, J. Organomet. Chem. 1986, 315, 401 ± 417.
[23] H. Dijkstra, M. D. Meijer, J. Patel, R. Kreiter, G. P. M. van Klink, M.
Lutz, A. L. Spek, A. Canty, G. van Koten, Organometallics 2001, 20,
3159 ± 3168.
[24] a) J. A. M. van Beek, G. van Koten, G. P. C. M. Dekker, E. Wissing,
M. C. Zoutberg, C. H. Stam, J. Organomet. Chem. 1990, 394, 659 ± 678;
b) J. Terheijden, G. van Koten, J. A. M. van Beek, B. K. Vriesema,
R. M. Kellogg, M. C. Zoutberg, C. H. Stam, Organometallics 1987, 6,
89 ± 93.
[25] A. M. Clark, C. E. F. Richard, W. R. Roper, L. J. Wright, Organo-
metallics 1998, 17, 4535 ± 4537.
[26] Molecular modelling was performed using MMFF94, Spartan 5.1,
SGI.
[27] U. Kragl, C. Dreisbach, Angew. Chem. 1996, 108, 684 ± 685; Angew.
Chem. Int. Ed. Engl. 1996, 35, 642 ± 664.
[28] D. M. Grove, G. van Koten, J. N. Louwen, J. G. Noltes, A. L. Spek,
H. J. C. Ubbels, J. Am. Chem. Soc. 1982, 104, 6609 ± 6616.
[29] This is important to note as the precipitated silver salt can also act as a
Lewis acid catalyst.
[30] J. Brunner, F. M. Richards, J. Biol. Chem. 1980, 255, 3319 ± 3329.
[31] Amination of 12 with Me2NH gave the dimethylamino analogues of
13. Further deprotection, acylation with AcCl and oxidative addition
of [Pd(dba)2] afforded compound 20 in 44% overall yield.
[32] P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, S. Garcia-
Granda, R. O. Gould, J. M. M. Smits, C. Smykalla, The DIRDIF97
program system, Technical Report of the Crystallography Laboratory,
University of Nijmegen (The Netherlands), 1997.
[33] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giaco-
vazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna J.
Appl. Crystallogr. 1999, 32, 115 ± 119.
[34] G. M. Sheldrick, SHELXS97, Program for crystal structure solution,
University of Göttingen (Germany), 1997.
Â
1993, 1825 ± 1828; d) M. A. Perez, J. M. Bermejo, J. Org. Chem. 1993,
58, 2628 ± 2630; e) N. A. Porter, B. Lacher, V. H.-T. Chang, D. R.
Magnin, B. T. Wright, J. Am. Chem. Soc. 1988, 110, 3554 ± 3560; f) S.
Warwel, H. Kätker, C. Rauenbusch, Angew. Chem. 1987, 99, 714 ± 715;
Angew. Chem. Int. Ed. Engl. 1987, 26, 702 ± 703.
[16] a) B. K. Vriesema, J. Buter, R. M. Kellogg, J. Org. Chem. 1984, 49,
110 ± 113; b) W. H. Kruizinga, R. M. Kellogg, J. Am. Chem. Soc. 1981,
103, 5183 ± 5189.
[17] L. A. van de Kuil, H. Luitjes, D. M. Grove, J. W. Zwikker, J. G. M.
van der Linden, A. M. Roelofsen, L. W. Jenneskens, W. Drenth, G.
van Koten, Organometallics 1994, 13, 468 ± 477.
[18] Prolonged addition did not afford higher yields.
[35] G. M. Sheldrick, SHELXL97, Program for crystal structure refine-
ment, University of Göttingen (Germany), 1997.
[19] a) P. L. Alsters, P. J. Baesjou, M. D. Janssen, H. Kooijman, A. Sicherer-
Roetman, A. L. Spek, G. van Koten, Organometallics 1992, 11, 4124 ±
4135; b) S. Komiya, Synthesis of Organometallic Compounds, Wiley,
Chichester, 1997.
[36] A. L. Spek, PLATON, A multipurpose crystallographic tool. Utrecht
University (The Netherlands), 2001.
3
[37] The crystal structure of compound 16 contains large voids (1325.9 Š
[20] a) H. P. Dijkstra, P. Steenwinkel, D. M. Grove, M. Lutz, A. L. Spek, G.
van Koten, Angew. Chem. 1999, 111, 2322 ± 2324; Angew. Chem. Int.
Ed. 1999, 38, 2185 ± 2288; b) P. Steenwinkel, R. A. Gossage, T.
Maunula, D. M. Grove, G. van Koten, Chem. Eur. J. 1998, 4, 763 ± 768.
[21] In the centrosymmetric unit cell, two isomers are observed. The RS
isomer shown in Figure 2a, and its SR enantiomer.
per unit cell) filled with disordered solvent molecules. Their contri-
bution to the structure factors was secured by back-Fourier trans-
formation using the routine CALC SQUEEZE of the program
PLATON[35] (144 electrons/unit cell). Derived values (formula weight,
density, absorption coefficient) do not contain the contribution of the
disordered solvent.
Received: May 21, 2001 [F3272]
Chem. Eur. J. 2002, 8, No. 1
ꢀ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0057 $ 17.50+.50/0
57