A novel hyperbranched polyether by melt transetherification

Article abstract of DOI:10.1039/b005552m

Melt self-condensation of 1-(2-hydroxyethoxy)-3,5-bis-(methoxymethyl)-2,4,6-trimethylbenzene in the presence of an acid catalyst via a transetherification process yielded a soluble high molecular weight hyperbranched polyether, whose structure was establi

Full text of DOI:10.1039/b005552m

A novel hyperbranched polyether by melt transetherification
M. Jayakannan and S. Ramakrishnan*
Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore 560012, India.
E-mail: raman@ipc.iisc.ernet.in; Fax: 91-80-360-1552
Received (in Cambridge, UK) 10th July 2000, Accepted 24th August 2000
First published as an Advance Article on the web
Melt self-condensation of 1-(2-hydroxyethoxy)-3,5-bis-
(methoxymethyl)-2,4,6-trimethylbenzene in the presence of
an acid catalyst via a transetherification process yielded a
soluble high molecular weight hyperbranched polyether,
whose structure was established by NMR spectroscopy.
was then treated with 2-chloroethanol to give the monomer 3.
This monomer was readily polymerized at 145–150 °C in the
presence of toulene-p-sulfonic acid (PTSA) as catalyst. Typi-
cally, the polymerization is carried out under N2 purge for 30
min and then under reduced pressure (0.01 mm of Hg) for an
additional 15 min to attain high molecular weight. To start with,
the contents of the reaction vessel formed a clear melt but as the
condensation proceeded it transformed into a solid mass. The
solid polymer was dissolved in THF, filtered to remove any
unwanted insoluble material, and then precipitated into metha-
nol. The precipitate was isolated and dried to give the polymer
in 82% yield. The expected structure of the resulting hyper-
branched polyether is shown in Scheme 1.
Highly branched macromolecules, such as dendrimers and
hyperbranched polymers, adopt compact conformations and
possess a large number of terminal functional groups. This
drastically alters both their solution and bulk properties. Several
reviews have appeared in the recent literature, that describe the
synthesis, structural properties and applications of dendrimers
and hyperbranched polymers.1–4 Despite the presence of
significant levels of structural imperfections, hyperbranched
polymers present one major advantage over their structurally
perfect cousins and that is the ease of synthesis, which is most
often a one-pot self-condensation of an AB2-type monomer.
More importantly, several of the interesting dendrimeric
properties, such as low melt-viscosity and high functionality,
are also exhibited by hyperbranched polymers, thereby making
them excellent alternatives to the synthetically cumbersome
dendrimers for certain kinds of applications. A large variety of
hyperbranched polymers, namely polyesters, polyurethanes,
polyamides, polyphenylenes, polysiloxanes, polycarbonates
and polyethers have been reported.3,4 One important feature of
the polyether class is their high solubility in common organic
solvents, which make them more readily amenable for the study
and exploitation of their properties in solution. Hyperbranched
polyethers of several types have also been reported in the
literature.5–8 Most of these procedures yielded moderate to high
molecular weight polymers, however, they lack adaptability
toward structural variation, in that they always require a
phenolic group and an aryl/benzyl halide.
Recently, we reported a melt transetherification methodology
for the synthesis of linear polyethers.9 Fully substituted
bisbenzyl methyl ethers readily underwent melt-condensation
with diols in the presence of an acid catalyst to give polyethers
of moderate molecular weights. In this report the transether-
ification methodology is extended for the preparation of a
hyperbranched polyether under melt condition. For this purpose
an AB2 monomer having a hydroxy group (A) and two benzyl
methyl ether groups (B) was targeted. An additional require-
ment for the transetherification approach to function effectively
without crosslinking is the preclusion electrophilic aromatic
substitution, which is fulfilled by complete substitution of all
the aromatic sites in the monomer.9 The simplest monomer that
meets all these criteria is 1-(2-hydroxyethoxy)-3,5-bis(meth-
oxymethyl)-2,4,6-trimethylbenzene (3), which is readily pre-
pared from commercially available mesitol, as shown in
Scheme 1. Mesitol was bis-chloromethylated and reacted with
NaOCH3–methanol to give the intermediate 2 in good yield. It
The 1H-NMR spectra of the monomer 3 and the polymer are
shown in Fig. 1, along with the assignments of the various
peaks. Comparison of the spectrum of the monomer with that of
the polymer reveals several interesting features. Firstly, as
expected for high conversions, there is a 50% decrease in the
relative intensity of the peak d, corresponding to Ar-CH2OCH3
protons, and a complete disappearance of the peak b, corre-
sponding to the –CH2OH protons, in the spectrum of the
polymer (Fig. 1-B). Furthermore, two types of benzylic protons
(a1 and a2) of equal intensity—one corresponding to the
unreacted benzyl methyl ether (Ar-CH2OCH3) and the other to
the reacted one (Ar-CH2OR–), are seen in the polymer
spectrum. These observations confirm that the transetherifica-
tion process has indeed occurred to very high conversions to
yield a polymer with the expected structure. One other
interesting feature is the transformation of the peaks corre-
sponding to the aromatic methyl groups from two singlets (e and
f) in the monomer to a cluster of at least five well-resolved
peaks (six inclusive of a shoulder), suggesting the presence of
several distinct chemical environments in which they are
present.
In general, hyperbranched polymers are expected to have
different types of subunits, such as dendritic (D), linear (L) and
terminal (T) units. Most often, the presence of these various
units is established and quantified based on the 1H-NMR signals
corresponding to the aromatic protons belonging to them.
Interestingly, in the present polyether a significantly more
intense Ar-CH3 signal appears to reveal the presence of these
subunits. A total of six aromatic methyl signals (in the region
between 2.3–2.5 ppm) are seen, as opposed to a total of seven
that might be expected if no coincidental overlaps are present—
two each corresponding to dendritic and terminal units and three
to the linear unit. As expected, the sum total of the intensities of
these Ar-CH3 peaks is in the expected ratio with respect to the
other peaks, c and d. In order to calculate the degree of
branching (DB), it is essential to identify and quantify the mole
fraction of the various subunits D, L and T, in such
hyperbranched structures. Based on topological considerations
Scheme 1
DOI: 10.1039/b005552m
Chem. Commun., 2000, 1967–1968
This journal is © The Royal Society of Chemistry 2000
1967

absence of the dendritic model compound,13 we rely on the ca.
1+2 intensity ratio of the shoulder to peak (e+f) in the polymer
spectrum to confirm the absence of peaks due to other subunits
within this one. Typical values for DB quoted in the literature
range from 0.4–0.8.14 Special efforts using a gradual addition of
the AB2 monomer to a polyfunctional B-type core monomer
(Bf), are often needed to further increase the degree of
branching.15
The molecular weight of the polymer was determined by
GPC using polystyrene standards and it showed a broad
distribution with MW of 103 000 and a polydispersity of 5.8. The
high polydispersity is typical of such hyperbranched structures.
The DSC thermogram of the polymer was recorded under a dry
N2 purge, and the sample was first heated to 150 °C and then
quenched to 2100 °C at 30 °C min21. It was then reheated from
2100 to 150 °C at 10 °C min21 and cooled at the same rate to
record the thermograms. No crystallization and melting transi-
tions were observed either during the heating or cooling runs,
which suggests that the samples were completely amorphous. A
glass transition temperature (Tg) was, however, clearly visible at
28 °C.
In conclusion, we have shown that the melt transetherifica-
tion approach can be readily adapted for the preparation of
hyperbranched polyethers. More importantly, the intermediate
2 is readily amenable to a variety of structural variation by
incorporation of various types of alkyl and oligoethyleneoxy
spacers, to generate a wide range of AB2 monomers. Such an
inclusion of spacer segments into hyperbranched polyethers is
not readily possible using most other previous methods, which
makes this approach especially useful for preparing a range of
hyperbranched structures starting from a single intermediate. A
particularly interesting series of hyperbranched polyethers
would be the one that incorporates oligoethyleneoxy segments.
This would lead essentially to branched polyethylene oxides,
which could serve as potential candidates for solid polymer
electrolyte applications.16,17 Work along these lines is currently
in progress and will be reported shortly.
1
Fig. 1 500 MHz H-NMR spectra of the monomer (A), polymer (b) and
We would like to thank Department of Science and
Technology, New Delhi, for financial support, and Dr Shiv
Venkataramani, 3M-Center, St. Paul, for GPC analysis.
model compound (C), in CDCl3.
it has been shown that hyperbranched structures will always
possess equal number of D and T units (at high conversions),
and the DB has been calculated for several polymers using a
simplified expression (eqn. 1), from the mole fraction of the
terminal units alone.8,10–13
Notes and references
1 A. W. Bosman, H. M. Janssen and E. W. Meijer, Chem. Rev., 1999, 99,
1665.
2 C. J. Hawker, Adv. Polym. Sci., 1999, 147, 144.
3 Y. H. Kim, J. Polym. Sci., Polym. Chem. Ed., 1998, 36, 1685.
4 A. Hult, M. Johansson and E. Malmstrom, Adv. Polym. Sci., 1999, 143,
1.
5 V. Percec and M. Kawasumi, Macromolecules, 1992, 25, 3843.
6 K. E. Uhrich, C. J. Hawker and J. M. J. Frechet, Macromolecules, 1992,
25, 4583.
[D]+ [T]
2[T]
DB =
=
(1)
[D]+ [L]+ [T] [D]+ [L]+ [T]
In order to identify the methyl peaks belonging to the
terminal units, a model compound 4 (see Fig. 1-C) was prepared
by reacting 2 with 2-benzyloxy-1-chloroethane in presence of a
1
7 T. M. Miller, T. X. Neenan, E. W. Kwock and S. M. Stein, J. Am. Chem.
Soc., 1993, 115, 356.
base, and its H-NMR spectrum is shown in Fig. 1. Upon
comparison of the spectrum of the model compound with that of
the polymer, it is clear that the peaks between 2.34–2.39 ppm in
the polymer corresponds to the Ar-CH3 protons of the terminal
unit. The sensitivity of the chemical shifts of these methyl
protons to its chemical environment is apparent when one
compares the spectrum of model compound 4 with that of the
monomer; the transformation of a remote hydroxy group to a
benzyl ether causes a significant downfield shift of the two
ortho-methyl protons peak, f. Thus, a non-exact match between
the chemical shifts of model compound 4 and the polymer is not
surprising; the ortho-methyl signal f appears slightly further
downfield in the polymer. Thus, the DB was calculated from the
relative intensities of the two peaks assigned to the terminal unit
(between 2.34–2.39 ppm) with respect to the total intensity of
all the methyl signals (between 2.3–2.49 ppm), as per the above
formula. This number works out to be 0.78 suggesting that a
fairly high degree of branching has been achieved. In the
8 C. J. Hawker and F. Chu, Macromolecules, 1996, 29, 4370.
9 M. Jayakannan and S. Ramakrishnan, Macromol. Chem. Phys., 2000,
201, 759.
10 K. L. Wooley, C. J. Hawker, R. Lee and J. M. J. Frechet, Polym. J.,
1994, 26, 187.
11 D. Yan, A. H. E. Muller and K. Matyjaszewski, Macromolecules, 1997,
30, 7024.
12 H. R. Kricheldorf, O. Stober and D. Lubbers, Macomolecules, 1995, 28,
2118.
13 Attempts to prepare the dendritic model compound using standard
approaches did not yield the desired product, and hence the unequivocal
confirmation of the absence of coincidental overlap is yet to be
established.
14 D. Holter, A. Burgath and H. Frey, Acta Polym., 1997, 48, 30.
15 D. Holter and H. Frey, Acta Polym., 1997, 48, 298.
16 C. J. Hawker, F. Chu, P. J. Pomery and D. J. T. Hill, Macromolecules,
1996, 29, 3831.
17 A. Kumar and S. Ramakrishnan, Macromolecules, 1996, 29, 2524.
1968
Chem. Commun., 2000, 1967–1968