Polyphenylene dendrimers via Diels-Alder reactions: The convergent approach

Article abstract of DOI:10.1039/a907339f

Polyphenylene dendrimers and dendrons can be obtained using [2 + 4]cycloaddition and Knoevenagel cyclocondensation in a convergent approach.

Full text of DOI:10.1039/a907339f

Polyphenylene dendrimers via Diels–Alder reactions: the convergent approach
Uwe-Martin Wiesler and Klaus Müllen*
Max-Planck-Institut für Polymerforschung, Ackermannweg 10, 55128 Mainz, Germany.
E-mail: muellen@mpip-mainz.mpg.de
Received (in Cambridge, UK) 10th September 1999, Accepted 4th October 1999
Polyphenylene dendrimers and dendrons can be obtained
using [2 + 4]cycloaddition and Knoevenagel cyclocondensa-
tion in a convergent approach.
nucleophilic attack of 5 and subsequent formation of the
cyclopentadienone ring (see Scheme 1).
The four-fold cycloaddition of an excess of cyclopentadie-
none dendron 4 to the tetraethynylbiphenyl 7 and the tetra-
kis(ethynylphenyl)methane 8 in Ph2O at 200 °C leads to
dendrimers 9 and 10, respectively (see Scheme 2). The products
are isolated by precipitation from the reaction mixture using
EtOH (89 and 85% isolated yield, respectively). Molecules 9
and 10 correspond to the second generation polyphenylene
dendrimer previously made by the divergent method.6
Whereas in the synthesis of dendrimer 9 the reactants have to
be heated for two days, the reaction time in the case of
dendrimer 10 is about one week. The MALDI-TOF mass
spectra taken at various stages during the reaction reveal that
three ethynyl groups of both cores react rapidly with dendron 4,
Due to their outstanding properties, such as molecular shape-
and size-persistence and narrow molecular weight distribution,
dendrimers are the object of increasing interest.1 In this context
we developed a divergent method to synthesize polyphenylene
dendrimers based on the selective Diels–Alder cycloaddition
between tetraphenylcyclopentadienones and ethynyls.2 We are
thus able to synthesize dendrimers up to the fourth generation,
which exist as shape-persistent nanoparticles.3
While the divergent approach builds dendrimers layer-by-
layer from a central core to the periphery, the convergent
strategy builds the dendrimer in the opposite manner, from the
periphery toward the central core. Thus, the dendrimer
branches, also called dendrons, are synthesized first and then
added to a central core. This approach not only allows the
symmetric functionalization of the dendrimer surface as in the
divergent approach, but also the controlled functionalization of
the dendrimer by adding differently functionalized dendrons to
selectively protected cores.4
Here, we present a convergent approach for the synthesis of
polyphenylene dendrons and dendrimers based on two alternat-
ing orthogonal reactions: (i) the Diels–Alder cycloaddition of
tetraphenylcyclopentadienones to phenylene-substituted ethy-
nyls, and (ii) the Knoevenagel condensation of benzils with
1,3-diphenylacetones to give cyclopentadienones.
The starting point of the synthesis is 4,4A-diethynylbenzil 1
which contains two ethynylic dienophile units and one ethane-
dione function which can react in a Knoevenagel condensation
(Scheme 1). The starting material 1 is prepared via the
Sonogashira5 coupling of triisopropylsilylacetylene and 4,4A-
dibromobenzil and subsequent cleavage of the TIPS groups
with KF in DMF (70%, yellow crystals).
The two-fold Diels–Alder cycloaddition of an excess of
tetraphenylcyclopentadienone 2, which has to be regarded as a
first-generation dendron, to 1 in refluxing xylene leads to the
second generation 3 in the form of a benzilic dendron (91%,
pale yellow amorphous powder). The Knoevenagel condensa-
tion of 3 to the corresponding cyclopentadienone dendron 4
with 1,3-diphenylacetone 5 is achieved in dioxane and in the
presence of Bu4NOH as base. Like 3, the cyclopentadienone
dendron 4 can be easily precipitated in EtOH as a pale red solid
in 85% isolated yield.
The next generation benzil 6 is obtained, similar to the second
generation, by addition of an excess of dendron 4 to 1 in
refluxing xylene. Isolation after precipitation in EtOH affords
the benzil 6 as an amorphous yellow powder in 89% yield.
Further base-catalyzed condensation of 6 with 1,3-diph-
enylacetone 5 provides a complex mixture of products,
determined by mass spectrometry, consisting of small amounts
of third generation cyclopentadienone dendron and further side-
products of the polycondensation of 5 with 6. This can be
explained by the high steric hindrance of the polyphenylenic
branches of the benzil 6, which will prefer a conformation
where the dendrons (and thus the vicinal carbonyls) are oriented
in opposite directions. This arrangement prevents the favour-
able orientation of the benzilic carbonyls required for the
Scheme 1 Reagents and conditions: i, xylene, 14 h, 91%; ii, 5, 1,4-dioxane,
Bu4NOH, 100 °C, 1 h, 85%; iii, 1, xylene, reflux, 14 h, 89%.
Chem. Commun., 1999, 2293–2294
This journal is © The Royal Society of Chemistry 1999
2293

Compared to the divergent approach, the convergent method
at first glance yields the same monodisperse products with
similarly high yields. However, though the convergent method
can be used to synthesize dendrimers only up to the second
generation, this method opens the way to dendritic polymers
carrying substituents of more than just one kind. In contrast, the
divergent approach may be successfully used up to the fourth
generation, but it only allows the ordered attachment of one kind
of functional group. Both methods have in common the fact that
they yield monodiperse polyphenylene dendrimers in high
yield.
In conclusion, we can say that: (i) using a convergent
approach, polyphenylene dendrons and dendrimers with more
than 60 benzene rings can be obtained via [2 + 4] cycloadditions
and Knoevenagel condensations; (ii) by attachment of the
polyphenylene dendrons to different cores, dendrimers with
different shapes are obtained; (iii) the dendrimers show high
thermal and chemical stability.
Our current investigations involve the convergent synthesis
of dendrimers with various substitution patterns as possible
carriers in catalysis.
Scheme 2 Reagents and conditions: i, 7, Ph2O, 200 °C, 2 days, 89%; ii, 8,
Ph2O, 200 °C, 7 days, 85%.
We thank the Foundation of the Chemical Industries (Fonds
der Chemischen Industrie) and the Bundesministerium für
Bildung und Forschung for financial support.
while the reactivity of the last group depends on the flexibility
of the core. In this view the tetraethynylbiphenyl core 7 allows
rotation about the phenyl–phenyl bond of the core, so that the
fourth ethynyl group can be brought into a favorable position for
the sterically demanding cycloaddition of dendron 6. In contrast
the fourth ethynyl groups of the tetrakis(ethynylphenyl)me-
thane core 8 does not have such mobility.
The disparity in reaction times required for obtaining 9 and
dendrimer 10 can also be understood examining ball-and-stick
models, which are generated using the MM2 (85) force field
with the CERIUS 2 program package and applying the
Conjugate Gradient 200 algorithm (compare Scheme 2).7
Whereas dendrimer 9 possesses a dumb-bell like structure, with
easily accessible holes close to the phenyl–phenyl bond of the
core, the methane center of dendrimer 10 is densely surrounded
by phenylene groups which increase the stiffness of the whole
molecule and render the core difficult to access.
Notes and references
† Selected data for 4: FD-MS: m/z 1145.47, calc. for C89H60O: 1145.45;
dH(500 MHz, C2D2Cl4, 393 K) 7.49 (s, 2H), 7.25–7.13 (m, 20H), 6.97 (d,
3J 8.25, 4H), 6.90–6.72 (m, 30H), 6.67 (d, 3J 8.25, 4H); dC(125 MHz,
C2D2Cl4, 303 K) 199.0 (CNO), 154.6, 142.5, 142.0, 141.9, 140.9, 140.4,
140.3, 140.1, 140.0, 139.8, 139.4, 131.8, 131.8, 131.7, 131.1, 131.0, 130.5,
130.4, 130.3, 129.7, 129.2, 128.3, 127.9, 127.7, 127.1, 127.0, 126.8, 126.5,
126.0, 125.7, 125.5, 125.2. For 9: MALD-TOF MS: m/z 4720, calc. for
C372H250: 4720.07; dH(500 MHz, C2D2Cl4, 393 K) 7.41 (s, 4H), 7.34 (s,
4H), 7.28 (m, 6H), 7.18–6.25 (m, 236H); dC(125 MHz, C2D2Cl4, 303 K)
199.0 (CNO), 142.3, 142.1, 141.4, 140.8, 140.7, 140.5, 139.5, 139.4, 139.0,
138.7, 138.3, 132.7, 132.0, 131.9, 131.6, 131.3, 130.4, 130.2, 128.9, 128.6,
127.7, 127.2, 127.0, 126.7, 126.3, 125.9, 125.6, 125.4.
Neverthless, both the dendrons and the dendrimers are very
soluble in organic solvents such as toluene and CH2Cl2.
Characterization was carried out by field-desorption mass
spectrometry for the dendrons with molecular masses up to
2000 g mol21 and matrix-assisted laser desorption ionization
time of flight mass spectrometry (MALDI-TOF MS) for the
higher molecular mass dendrons and the dendrimers. The purity
and monodispersity of the analyzed molecules was confirmed
by the perfect agreement between calculated and experimen-
tally determined m/z ratios.† A characteristic feature of the
synthesized dendrimers 9 and 10 is their high thermal and
chemical stability. They decompose under air only at tem-
peratures higher than 550 °C according to thermogravimetric
analysis. Their chemical stability is also noteworthy. For
example boiling in concentrated HCl or in 30% KOH for seven
days fails to produce any decomposition. The dendrimers react
only with strong electrophiles, like H2SO4, via aromatic
electrophilic substitution, but without change in the poly-
phenylene framework. This opens a way to chemically
functionalize dendrimers at their surface.
1 G. R. Newkome, C. N. Moorefield and F. Vögtle, Dendritic Molecules,
VCH, Weinheim, 1996; A. W. Bosman, H. M. Janssen and E. W. Meijer,
Chem. Rev., 1999, 99, 1665.
2 F. Morgenroth, E. Reuter and K. Müllen, Angew. Chem., 1997, 109, 647;
Angew. Chem., Int. Ed. Engl., 1997, 36, 631; F. Morgenroth A. J.
Berresheim, M. Wagner and K. Müllen, Chem. Commun., 1998, 1139.
3 F. Morgenroth, C. Kübel and K. Müllen, J. Mater. Chem., 1997, 7,
1207.
4 C. J. Hawker and J. M. J. Fréchet, Macromolecules, 1990, 23, 4726; K. L.
Wooley, C. J. Hawker and J. M. J. Fréchet, J. Chem. Soc., Perkin Trans.
1, 1991, 1059; K. L. Wooley, C. J. Hawker and J. M. J. Fréchet, J. Am.
Chem. Soc., 1993, 115, 11496.
5 S. Takahashi, Y. Kuroyama, K. Sonogashira and N. Hagihara, Synthesis,
1980, 627.
6 F. Morgenroth and K. Müllen, Tetrahedron, 1997, 53, 15 349.
7 CERIUS2, Molecular Simulations Inc., Waltham, MA, USA. For more
details see ref. 3. Additional data are to be published.
Communication 9/07339F
2294
Chem. Commun., 1999, 2293–2294