Dendronized polymer organogels from click chemistry: A remarkable gelation property owing to synergistic functional-group binding and dendritic size effects

Article abstract of DOI:10.1002/anie.200801870

(Figure Presented) Click together and gel: Non-gelating AB-type heterobifunctional macromonomers click together to form polytriazole-containing dendronized polymers. These polymers form interchain hydrogen bonds, the extent of which is influenced by the size of the dendritic appendage.

Full text of DOI:10.1002/anie.200801870

Communications
DOI: 10.1002/anie.200801870
Click chemistry
Dendronized Polymer Organogels from Click Chemistry: A
Remarkable Gelation Property Owing to Synergistic Functional-Group
Binding and Dendritic Size Effects**
Kwun-Ngai Lau, Hak-Fun Chow,* Man-Chor Chan, and Ka-Wai Wong
The use of 1,3-dipolar cycloaddition reactions between azides
and alkynes (click chemistry) has been extremely successful
as a versatile synthetic tool to construct novel polymeric
systems.[1] Whereas the main thrust has been focused on
building up highly elaborate polymeric architectures, such as
block copolymers, star polymers, dendrimers, and hyper-
branched polymers, it is also noted that the physical proper-
ties of such poly(triazole)-based materials are little studied.[1c]
Additionally, although there are many examples of the
synthesis of dendrimers using click chemistry, only a few
concern the preparation of dendronized polymers.[2] These
are polymers incorporating multiple dendron segments stem-
ming from a linear polymer backbone and are commonly
prepared by graft-to, graft-from, or macromonomer polymer-
ization approaches.[3] The major challenges for these
approaches are the difficulty in ensuring complete dendron
coverage in the graft-to and graft-from strategies, and the
sometimes poor polymerization efficiency in the macromo-
nomer strategy. To improve the synthetic efficacy, it is
necessary to make use of reactions that offer perfect
conversion efficiency (such as click chemistry). Herein we
wish to report a) the successful click synthesis of two different
series of dendronized polymers (DPs), AmDP1–AmDP3 and
EsDP1–EsDP3, from heterobifunctional amide-linked mac-
romonomers (AmM1–AmM3) and ester-linked macromono-
mers (EsM1--EsM3), respectively, b) the novel and unique
organogelation property of one such poly(triazole)-based
dendronized polymer AmDP2, c) the remarkable functional-
group synergistic effect on polymer interchain H-bonding,
owing to the placing of many amide functionalities in close
proximity along the polymer chain, and most importantly
d) that the macromolecular interactions among the dendron-
ized polymer chains are strongly influenced by the size of
dendritic appendages and the nature of the linker function-
ality. To our knowledge, synthesis of dendronized polymers by
AB-type macromonomer polymerization has not been
reported before. Moreover, although physical organogels
based on dendrimers[4] and linear polymers[5] are known,
those based on click poly(triazole) polymers[6] and dendron-
ized polymers[7] are extremely rare.
The click macromonomer polymerization is basically a
step-growth polymerization. Therefore, to achieve a higher
degree of polymerization (DP), it must be ensured that the
AB-hetero-bifunctional monomers AmM1–AmM3 are of
perfect purity and structural homogeneity. For this purpose,
we made use of the symmetrical aliphatic hydrocarbon-based
Meldrumꢀs acids 1–3[8] as our starting materials. Simple
functional-group transformations then led to the target
amide-linked macromonomers AmM1–AmM3 in good
yields and high purities (see Supporting Information). The
macromonomers AmM1–AmM3 were then polymerized in
the presence of sodium ascorbate and CuSO4 in a 1:1:1 solvent
mixture of THF, DMF, and water at 258C for 4 days. To
counteract the poor solubility of the products, DMF was
added to maintain a homogenous reaction mixture, at least
during the initial stage of the polymerizations. The dendron-
ized polymers AmDP1 (70% yield) and AmDP2 (62% yield)
[*] K.-N. Lau, Prof. Dr. H.-F. Chow, Prof. Dr. M.-C. Chan
Department of Chemistry and The Center of Novel Functional
Molecules
The Chinese University of Hong Kong
Shatin, NT, Hong Kong SAR (China)
Fax: (+852)2603-5057
E-mail: hfchow@cuhk.edu.hk
Prof. Dr. K.-W. Wong
Department of Physics
The Chinese University of Hong Kong
Shatin, NT, Hong Kong SAR (China)
[**] We thank the Physical Science Panel, CUHK for the financial support
(Project Code: 2060322).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200801870.
6912
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6912 –6916

Angewandte
Chemie
were isolated by precipitation from the reaction mixture,
whereas AmDP3 (81% yield) was obtained as a solid by
solvent extraction. For comparison studies, the corresponding
ester-linked macromonomers EsM1–EsM3 were also pre-
pared and polymerized to give the corresponding click
dendronized polymers EsDP1–EsDP3 (see Supporting Infor-
mation). Both the ester-linked EsM1–EsM3 and, to a lesser
extent, the amide-linked AmM1–AmM3 were shown to
undergo thermal self-clicking when stored as neat oils at
258C. The presence of the amide and triazole repeating units
rendered the dendronized polymers quite polar. For AmDP1,
solubility was so poor that its characterization could only be
conducted in DMF or DMSO solution. However, because of
the presence of larger, nonpolar hydrocarbon sectors, better
solubilities were observed for AmDP2 and AmDP3.
gelating in [D8]THF, indicating that the amide hydrogen
atoms must be involved in a stronger H-bonding environment
in AmDP2 than those in AmDP3. Surprisingly, the SEC
chromatographs of AmDP2 and AmDP3 showed the pres-
ence of 2–4 wt% of oligomeric species (Figure 2). In both
The structure of the amide-linked dendronized polymers
were characterized by 1H and 13C NMR spectroscopy, IR
spectroscopy, and size-exclusion chromatography (SEC). The
1H NMR spectra revealed the complete disappearance of the
acetylenic proton signal at approximately d = 2.2 ppm and the
appearance of the triazole proton signal at approximately d =
7.9 ppm (in [D6]DMSO) or 7.4–7.7 ppm (in [D8]THF),
indicating that all monomer was consumed during the
polymerization (Figure 1). However, there was a notable
Figure 2. Stacked SEC chromatographs of a) AmDP2 and AmDP3 and
b) EsDP1-EsDP3. The wt% of the oligomers are exaggerated in these
plots as the x-axis (elution time) is on a logarithm scale of polymer
molecular weight.
cases, there was even a peak with a longer retention time than
that of the monomer (see Supporting Information). In
addition, subjecting the initial click dendronized polymer
EsDP2 to reaction in the presence of CuSO4 for an additional
48 h did not change the appearance of the SEC chromato-
graph, indicating that these oligomeric species were most
likely cyclic. The formation of cyclic oligomers using click
chemistry has been reported before.[9] Disregarding the
signals of the lower molecular weight oligomers up to the
hexamer, the SEC data of the dendronized polymers were
tabulated (Table 1). The polydispersity values (PDI) were
close to 2, a typical value expected from a step-growth
polymerization. The degree of polymerization (DP) for the
amide-linked compounds was generally higher than for the
ester-linked series of the same generation (Gn), owing mainly
to the lower tendency of amide-linked monomers to form
cyclic oligomers. As cyclization was an inevitable side
reaction, the DP values were generally low (30–60). However,
as many previous studies revealed that SEC measurements
tend to underestimate the actual Mw values by a factor of 2–
8,[3d] the actual DP values could be between 50 and 500, and
therefore click chemistry could still be considered an efficient
polymerization process. The DP values dropped with increas-
ing dendrimer generation, suggesting that steric hindrance
Figure 1. 1H NMR spectra of a) third generation (G3) macromonomer
AmM3 (in CDCl3) and b) dendronized polymer AmDP3 (15% w/v in
[D8]THF at 508C). Peaks marked withan asterisk are signals due to
residual solvent.
difference between the 1H NMR spectra (at 508C) of AmDP2
À
and AmDP3 (see Supporting Information). While all N H
signals from AmDP3 (15% w/v in [D8]THF) resonated
between d = 7.7 and d = 8.5 ppm, two additional, but unusu-
À
ally downfield N H signals were found at approximately d =
10.6 ppm for AmDP2 when the sample concentration was
> 3% w/v in [D8]THF. Additionally, AmDP2 was also found
to form a partial gel when the concentration was > 3% in
À
[D8]THF. It was interesting to note that such downfield N H
signals were not found for AmDP2 when the concentration
was < 3%, at which concentration the sample was non-
Angew. Chem. Int. Ed. 2008, 47, 6912 –6916
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6913

Communications
Table 1: SEC data of dendronized polymers AmDP1–AmDP3 and
EsDP1–EsDP3.[a]
Polymer
Mw
PDI
DP
Weight% of
LMW fraction[b]
AmDP1[c]
AmDP2
AmDP3
EsDP1
EsDP2
EsDP3
–
–
–
40
37
66
36
12
–
2
4
2
4
27300
43700
28000
24300
14000
1.84
2.21
2.19
1.71
1.11
49
[a] All experiments were conducted in THF using polystyrenes as
calibration standards. [b] Low MW fraction up to hexamer. See Support-
ing Information for details of calculation. [c] Not determined because of
poor solubility in THF.
Figure 3. Two regions of the FT-IR spectra of a) AmM2 (5% solution
in toluene); b) AmDP2 (1% gel in toluene) and c) AmDP2 (solid in
KBr disc).
was a dominant retardation factor for the polymerization
process. In fact, oligomers (49% by weight) were the major
products from the ester-linked G3 monomer EsM3.
Among the three amide-linked dendronized polymers,
only AmDP2 was found to form strong organogels in organic
solvents, with minimum gelation concentration (MGC) down
to 5 mgmLÀ1 (Table 2), whereas AmDP1 and AMDP3 did
Table 3: FT-IR data [cmÀ1] of macromonomers and dendronized poly-
mers in different physical states.
À
N H
=
C O
Sample State
ꢀCH N3
AmM1 5% solution in PhMe
AmM2 5% solution in PhMe
AmM3 20% solution in PhMe
AmDP1 solid
3430
3436
3431
3338
3307 2098 1676
3310 2099 1679
3314 2098 1679
–
–
–
–
–
–
–
–
–
–
–
–
1668
1649
1653
1652
1653
1676
AmDP2 1% gel in PhMe
3378, 3285
Table 2: Minimum gel concentration (MGC) values [mgmLÀ1
dendronized polymer AmDP2 in organic solvents.[a]
]
of
1% freeze dried gel in PhMe 3378, 3287
solid in KBr disc
AmDP3 solid
3377, 3279
3350
3418, 3328
Solvent
MGC
Solvent
MGC
40% solution in PhMe
n-hexane
CHCl3
EtOAc
acetone
EtOH
I
toluene
o-xylene
m-xylene
p-xylene
o-DiClbenzene
anisole
5 (CG)
5 (CG)
5 (CG)
5 (CG)
5 (CG)
10 (CG)
10 (TG)
S[b]
10 (OG)
10 (OG)
20 (OG)
50 (OG)
10 (CG)
À
=
N H and C O absorption peaks match very well to those of
non-hydrogen-bonded secondary amides,[11] indicating that
the monomers did not form H-bond aggregates in toluene.
THF
benzene
ꢀ
After click polymerization, the peaks corresponding to the
nitrobenzene
À
=
C-H and N3 absorptions disappeared, and the N H and C O
[a] CG=transparent gel; TG=translucent gel; OG=opaque gel; S=
absorptions were significantly red-shifted (50–150 cmÀ1 for
N H, 10–30 cm for C O) in the solid samples of AmDP1–
AmDP3, indicating the presence of interchain H-bonds in the
solid state. For the organogel and xerogel samples obtained
soluble; I=insoluble. [b] Solubility ꢁ50 mgmLÀ1
.
À1
À
=
not exhibit any gelation in any of the solvents tested.
Interestingly, stirring a sample of AmDP2 in a CHCl3/
aqueous ethylenediaminetetraacetic acid (EDTA) biphasic
solution for 24 h at 258C did not reduce its gelation power,
indicating that gelation was not due to cross-linking of the
dendronized polymer chains through Cu complexation to the
triazole moieties.[6] Moreover, the three ester-linked dendron-
ized polymers EsDP1–EsDP3 were also devoid of gelation
properties. Therefore the key structural element that was
responsible for gelation must be the amide-linker function-
ality. It was also noted that weak, opaque gels were formed
with non-aromatic solvents, such as CHCl3 or EtOH, while
strong transparent gels were formed with aromatic solvents,
suggesting that the p–p stacking effect between the benzene/
triazole units and aromatic solvents may also contribute to the
gelation mechanism.[10]
À
from 1% AmDP2 in toluene, the N H absorption was split
into two peaks. Both peaks were red-shifted (60 and 140 cmÀ1
)
=
relative to the monomer, as was the C O absorption
(30 cmÀ1). In contrast, the corresponding red-shift values of
a 40% solution sample of AmDP3 in toluene were much
smaller (ca. 10 and 100 cmÀ1 for N H, and 3 cm for C O),
suggesting that the extent of interchain H-bonding was
significantly weaker than that for AmDP2. This finding was
À1
À
=
1
consistent with those obtained from the H NMR spectro-
scopic data. Unfortunately, we were unable to record the FT-
IR spectrum of AmDP1 in solution in toluene, as the
compound is insoluble in nonpolar solvents.
The morphology of the supramolecular dendronized
polymer gel AmDP2 was visualized by scanning electron
microscopy (SEM) (Figure 4). The xerogels obtained from
freeze drying of a 1% sample in p-xylene showed the
presence of three dimensional networks formed by the
entanglement of fibers with length of tens of micrometers
and diameter ranging from 50–70 nm. The diameter of such
nanofibres was much larger than expected for a single
dendronized polymer chain, and suggested that the xerogels
FT-IR studies were carried out on the monomers and the
amide-linked polymers (Figure 3 and Table 3). 5–20% sol-
utions of the monomers in toluene showed four characteristic
IR absorptions at approximately 3430, 3310, 2100, and
À1
À
ꢀ À
1680 cm , which could be attributed to N H, C H, N3,
=
and C O stretching vibrations, respectively. In particular, the
6914 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6912 –6916

Angewandte
Chemie
microenvironment that further enhanced the strength of H-
bonding.[12] In practice, the G3 dendrons were too bulky, and
prevented sufficiently close contacts among the polymer
chains to form a stable H-bond network structure in the
solution state, as confirmed by the smaller red shift values of
À
=
N H and C O stretchings in the FT-IR spectrum of AmDP3
in toluene solution. With regard to the gelation-specificity of
the AmDP2 dendronized polymer, the G2 branching hydro-
carbon residue seemingly possessed the optimized size to
allow close contacts of the dendronized polymer chains, and
hence the formation of a stronger H–bond-mediated network
structure in the solution state, as confirmed by FT-IR study of
AmDP2 in toluene solution. At the same time, the internal
voids created within the network were of the appropriate size
to accommodate solvent molecules and therefore physical
gels were formed. For AmDP1, with the smallest hydro-
carbon side chains that did not hinder chain associations,
interchain H-bonding was very strong and hence network
packing was much denser. Hence AmDP1 is a highly
insoluble compound and the internal voids within it are too
small to accommodate any solvent molecules. In addition, the
role of solubility should also play a role in controlling
gelation. The different hydrophobic side chains could mod-
ulate the solubility of the resulting dendronized polymers.
Hence, AmDP1 does not have enough hydrophobic function-
ality to dissolve and therefore is a solid, whereas AmDP3 has
too much hydrophobicity, interacts too effectively with the
solvent, and simply dissolves.
Figure 4. SEM Images of a) freeze-dried 1% AmDP2 gel in p-xylene at
ꢀ35000 magnification and b) the same sample at ꢀ121000 magnifi-
cation.
were super-bundles formed from the intertwining of many
polymer chains. However, it is not certain whether these were
present in the native gel or formed during the sample drying
stage.
Based on the facts that only AmDP2 formed strong
organogels and that the monomers were all non-aggregating
in the solution state, a gelation model was postulated
(Figure 5). Since the monomers contain only two H-bonding
amide units that are not pre-organized by a rigid spacer, there
is little cooperative binding effect. Hence AmM1–AmM3
showed little self-association in the solution state, as con-
firmed by FT-IR study. After click polymerization, the
dendronized polymers contain a large number of H-bonding
amide units packed at regular intervals along the polymer
chain, producing a zip templating effect, as initial interchain
H-bonding between a few amide units would facilitate
bindings of those located further down the polymer chain,
provided that the steric size of the dendrons did not interfere
with the binding process. In addition, aliphatic hydrocarbon
dendrons also provided a highly nonpolar hydrophobic
In summary, we have reported herein the first efficient
synthesis of G1–G3 dendronized polymers starting from AB-
type heterobifunctional macromonomers using click poly-
merization. A significant functional-group synergistic effect
was noted on the interchain hydrogen-bonding capability of
the many amide functionalities in the resulting dendronized
polymers. Despite their structural similarities, the strength of
the hydrogen-bond networks, and hence the phys-
ical properties of the three dendronized polymers,
were different and were controlled by the size of
the dendritic appendage. It is of interest to note
that our findings are reminiscent to those made for
the hydrogelating properties of poly(N-alkyl)acry-
lamides, wherein their gelating properties are
À
dependent on the amide N H hydrogen bonding
and the nature of the alkyl side chains.[13] Notably,
the G2 dendronized polymer AmDP2 was found
to have very strong organogelating properties, with
MGC values down to 5 mgmLÀ1. We believe the
new findings can provide valuable insights into the
self-assembly of dendronized polymers, and offer
new understandings of the intricate mechanism of
polymer–polymer hydrogen-bonding interactions.
Received: April 22, 2008
Published online: July 24, 2008
Keywords: click chemistry · dendrimers · gels ·
.
Figure 5. Proposed gelation model of AmDP2. For clarity, some dendrons are
omitted in some of the structures.
hydrogen bonding · polymerization
Angew. Chem. Int. Ed. 2008, 47, 6912 –6916
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6915

Communications
Tetrahedron 2005, 61, 11279 – 11290; g) A. R. Hirst, D. K. Smith,
Top. Curr. Chem. 2005, 256, 237 – 273; h) D. K. Smith, Chem.
Commun. 2006, 34 – 44.
[1] For recent reviews: a) W. H. Binder, C. Kluger, Curr. Org. Chem.
2006, 10, 1791 – 1815; b) W. H. Binder, R. Sachsenhofer, Macro-
mol. Rapid Commun. 2007, 28, 15 – 54; c) J.-L. Lutz, Angew.
Chem. 2007, 119, 1036 – 1043; Angew. Chem. Int. Ed. 2007, 46,
1018 – 1025; d) B. Voit, New J. Chem. 2007, 31, 1139 – 1151; e) H.
Nandivada, X. Jiang, J. Lahann, Adv. Mater. 2007, 19, 2197 –
2208.
[5] A. Vintiloiu, J.-C. Leroux, J. Controlled Release 2008, 125, 179 –
192.
[6] D. D. Dꢁaz, J. J. M. Tellado, D. G. Velꢂzquez, A. G. Ravelo,
Tetrahedron Lett. 2008, 49, 1340 – 1343.
[7] a) M. Yoshida, Z. M. Fresco, S. Ohnishi, J. M. J. FrØchet, Macro-
molecules 2005, 38, 334 – 344; b) C. Park, K. S. Choi, Y. Song, H.-
J. Jeon, H. H. Song, J. Y. Chang, C. Kim, Langmuir 2006, 22,
3812 – 3817.
[8] H.-F. Chow, K.-F. Ng, Z.-Y. Wang, C.-H. Wong, T. Luk, C.-M. Lo,
Y.-Y. Yang, Org. Lett. 2006, 8, 471 – 474.
[9] a) N. V. Tsarevsky, B. S. Sumerlin, K. Matyjaszewski, Macro-
molecules 2005, 38, 3558 – 3561; b) B. A. Laurent, S. M. Grayson,
J. Am. Chem. Soc. 2006, 128, 4238 – 4239.
[10] H.-F. Chow, J. Zhang, C.-M. Lo, S.-Y. Cheung, K.-W. Wong,
Tetrahedron 2007, 63, 363 – 373.
[11] R. M. Silverstein, F. X. Webster in Spectrometric Identification of
Organic Compounds, 6th ed., Wiley, New York, 1998, pp. 99 –
102.
[12] a) H. Sun, A. E. Kaifer, Org. Lett. 2005, 7, 3845 – 3848; b) C.-H.
Wong, H.-F. Chow, S.-K. Hui, K.-H. Sze, Org. Lett. 2006, 8, 1811 –
1814.
[2] a) B. Helms, J. L. Mynar, C. J. Hawker, J. M. J. FrØchet, J. Am.
Chem. Soc. 2004, 126, 15020 – 15021; b) J. L. Mynar, T.-L. Choi,
M. Yoshida, V. Kim, C. J. Hawker, J. M. J. FrØchet, Chem.
Commun. 2005, 5169 – 5171.
[3] For recent reviews: a) A. Zhang, L. Shu, Z. Bo, A. D. Schlüter,
Macromol. Chem. Phys. 2003, 204, 328 – 339; b) A. D. Schlüter,
C. R. Chim. 2003, 6, 843 – 851; c) A. D. Schlüter, Top. Curr.
Chem. 2005, 245, 151 – 191; d) H. Frauenrath, Prog. Polym. Sci.
2005, 30, 325 – 384.
[4] a) W.-D. Jang, D.-L. Jiang, T. Aida, J. Am. Chem. Soc. 2000, 122,
3232 – 3233; b) C. Kim, K. T. Kim, Y. Chang, H. H. Song, T. Y.
Cho, H. J. Jeon, J. Am. Chem. Soc. 2001, 123, 5586 – 5587;
c) A. R. Hirst, D. K. Smith, M. C. Feiters, H. P. M. Geurts, A. C.
Wright, J. Am. Chem. Soc. 2003, 125, 9010 – 9011; d) Y. Ji, Y.-F.
Luo, X.-R. Jia, E.-Q. Chen, Y. Huang, C. Ye, B.-B. Wang, Q.-F.
Zhou, Y. Wei, Angew. Chem. 2005, 117, 6179 – 6183; Angew.
Chem. Int. Ed. 2005, 44, 6025 – 6029; e) H.-F. Chow, J. Zhang,
Chem. Eur. J. 2005, 11, 5817 – 5831; f) H.-F. Chow, J. Zhang,
[13] H. G. Schild, Prog. Polym. Sci. 1992, 17, 163 – 249.
6916 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6912 –6916