Synthesis and characterization of double-hydrophilic model networks based on cross-linked star polymers of poly(ethylene glycol) methacrylate and methacrylic acid

Article abstract of DOI:10.1021/ma011822z

A series of model networks based on cross-linked star polymers of poly(ethylene glycol) methacrylate (PEGMA) and tetrahydropyranyl methacrylate (THPMA), a protected form of methacrylic acid (MAA), were synthesized by group transfer polymerization (GTP), covering various compositions (homopolymers and copolymers), arm architectures (homopolymer, block and statistical copolymers), and star architectures (heteroarm and star block). The synthesis took place in tetrahydrofuran (THF) using tetrabutylammonium bibenzoate (TBABB) as the catalyst and 1-methoxy-1-(trimethytsiloxy)-2-methyl-propene (MTS) es the initiator, and comprised a four- or six-step procedure: first, the formation of linear polymer chains upon the GTP of the monomer(s); next, the interconnection of the linear segments to star polymers ("arm-first" stars) by the in situ polymerization of the ethylene glycol dimethacrylate (EGDMA) cross-linker, followed by the addition of monomer(s) for a second time to grow new chains from the cores of the stars outward ("in-out" stars); finally, the cross-linking of the "in-out" stars to a network with the addition of more EGDMA. The molecular weights (MWs) and molecular weight distributions (MWDs) of all the linear and the star precursors to the networks were determined using gel permeation chromatography (GPC). The networks which contained THPMA were hydrolyzed under acidic conditions to convert the THPMA units to MAA units and thus to obtain polyelectrolytic, double-hydrophilic model networks. The aqueous degrees of swelling of all the networks were measured as a function of pH. At low pH, where MAA is protonated, low degrees of swelling were obtained, which increased with the network content in PEGMA. An increase of the degrees of swelling was observed at high pH. These high pH degrees of swelling were found to be dependent on both the composition and the structure of the network.

Full text of DOI:10.1021/ma011822z

Macromolecules 2002, 35, 4903-4911
4903
Synthesis and Characterization of Double-Hydrophilic Model Networks
Based on Cross-linked Star Polymers of Poly(ethylene glycol)
Methacrylate and Methacrylic Acid
Sa vva s N. Geor gia d es, Ma r ia Va m va k a k i, a n d Costa s S. P a tr ick ios*
Department of Chemistry, University of Cyprus, P.O. Box 20537, 1678 Nicosia, Cyprus
Received October 19, 2001; Revised Manuscript Received March 20, 2002
ABSTRACT: A series of model networks based on cross-linked star polymers of poly(ethylene glycol)
methacrylate (PEGMA) and tetrahydropyranyl methacrylate (THPMA), a protected form of methacrylic
acid (MAA), were synthesized by group transfer polymerization (GTP), covering various compositions
(homopolymers and copolymers), arm architectures (homopolymer, block and statistical copolymers), and
star architectures (heteroarm and star block). The synthesis took place in tetrahydrofuran (THF) using
tetrabutylammonium bibenzoate (TBABB) as the catalyst and 1-methoxy-1-(trimethylsiloxy)-2-methyl-
propene (MTS) as the initiator, and comprised a four- or six-step procedure: first, the formation of linear
polymer chains upon the GTP of the monomer(s); next, the interconnection of the linear segments to star
polymers (“arm-first” stars) by the in situ polymerization of the ethylene glycol dimethacrylate (EGDMA)
cross-linker, followed by the addition of monomer(s) for a second time to grow new chains from the cores
of the stars outward (“in-out” stars); finally, the cross-linking of the “in-out” stars to a network with
the addition of more EGDMA. The molecular weights (MWs) and molecular weight distributions (MWDs)
of all the linear and the star precursors to the networks were determined using gel permeation
chromatography (GPC). The networks which contained THPMA were hydrolyzed under acidic conditions
to convert the THPMA units to MAA units and thus to obtain polyelectrolytic, double-hydrophilic model
networks. The aqueous degrees of swelling of all the networks were measured as a function of pH. At low
pH, where MAA is protonated, low degrees of swelling were obtained, which increased with the network
content in PEGMA. An increase of the degrees of swelling was observed at high pH. These high pH
degrees of swelling were found to be dependent on both the composition and the structure of the network.
In tr od u ction
route),23 to “living” ring-opening cationic,24 oxyanionic,25
group transfer,26 “living” cationic,27,28 and atom transfer
radical polymerization.29 Potential uses of these materi-
als as stabilizers,23,30,31 as dispersants,32 and as gene
delivery vehicles33 have been suggested. The synthesis
of both block and statistical copolymers of methacrylate-
based PEO macromonomers containing acidic residues
was reported using group transfer polymerization.26,34
The block copolymer analogues exhibited interesting
aqueous solution properties and reversible micellization
as a function of both pH and potassium carbonate
concentration.
Poly(ethylene oxide), PEO, is a widely used nonionic,
water-soluble, biocompatible polymer. Because of its
unique properties, PEO has been incorporated in poly-
meric materials in many different forms including
macromonomers,1 macroinitiators,2 and graft3 and lin-
ear polymers.4 Amphiphilic (i.e., hydrophilic-hydropho-
bic) block copolymers of PEO have been proposed as
efficient emulsifiers,5,6 stabilizers,7-9 and drug-delivery
matrices.10,11 Among them, triblock copolymers of PEO
with poly(propylene oxide), known as Pluronics or
Synperonics,12-16 are extensively studied polymeric
surfactants with various industrial applications. DuPont
has also filed patents on the group transfer polymeri-
zation synthesis of ABC triblock copolymers containing
at least one hydrophilic and one hydrophobic block and
low molecular weight PEO monomethacrylate as the
third comonomer. The use of these copolymers for
pigment dispersion has been described.17,18
However, it is only during the past decade that PEO-
based double-hydrophilic (i.e., hydrophilic-hydrophilic)
block copolymers have attracted great scientific interest.
Among the first examples are studies of the ionic
interactions between mixtures of PEO-b-polyanion co-
polymers with either cationic polyelectrolytes19 or PEO-
b-polycations.20 Such systems were shown to lead to the
formation of complex mixed micelles with potential drug
delivery applications.21,22 In subsequent studies, various
PEO-based hydrophilic-hydrophilic block copolymers
were synthesized using different synthetic techniques,
from conventional free radical (via a macroinitiator
Apart from the linear polymers, another class of
interesting synthetic materials with many applications
are polymer networks, in which several linear polymer
chains are interconnected at both chain ends.35 Hydro-
gels (water-swellable networks)36,37 are used as super-
absorbents,38,39 tissue engineering scaffolds,40 actuators,
molecular valves, sensors, chemical memories, molec-
ular separation systems, drug delivery systems, artifi-
cial muscles for robotics, and other biomedical materi-
als.41,42 These applications are based on the ability of
these networks to respond to changes of the conditions
of the external environment. In particular, pH-sensitive
hydrogels have been proposed as promising candidates
for the controlled release of peptides43 and drugs.44
Among them, copolymer networks of poly(methacrylic
acid) (polyMAA) grafted with poly(ethylene glycol)
monomethacrylate have attracted attention because
they combine the pH response of the MAA units and
the biocompatibility of the ethylene glycol units. The
synthesis of such networks was first reported by Klier
et al.45 using conventional free-radical polymerization.
* To whom correspondence should be addressed.
10.1021/ma011822z CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/16/2002

4904 Georgiades et al.
Macromolecules, Vol. 35, No. 13, 2002
F igu r e 1. Chemical formulas of the main reagents used for the network synthesis. Monomers: methoxy poly(ethylene glycol)
methacrylate (PEGMA), tetrahydropyranyl methacrylate (THPMA), and methacrylic acid (MAA). Cross-linker: ethylene glycol
dimethacrylate (EGDMA). Initiator: 1-methoxy-1-(trimethylsiloxy)-2-methylpropene (MTS).
proton nuclear magnetic resonance (1H NMR) spectrum (typi-
The swelling behavior and adsorption properties of these
networks have been studied extensively.43,46,47 However,
cal purity by gas chromatography 99%). Both THPMA and
EGDMA were stirred overnight over calcium hydride to
due to limitations of the synthetic technique employed,
remove the last traces of moisture and protonic impurities in
these networks exhibited a random distribution of the
the presence of an added free radical inhibitor, 2,2-diphenyl-
two comonomers along the network elastic chains and
1-picrylhydrazyl hydrate (DPPH, 95%), to avoid thermal
polymerization. MTS, THPMA, and EGDMA were all freshly
distilled prior to the polymerization. PEGMA was used as a
also variable lengths of the segments between cross-
links.
In the present work, anionic, double-hydrophilic
model48 networks (accurate size and composition of their
components) containing MAA units and methoxy poly-
(ethylene glycol) methacrylate (PEGMA) (six ethylene
glycol units in the side-chain) and based on both block
and statistical copolymers were prepared. The network
structure was more complicated than that of conven-
tional model networks. Thus, the present networks, in
addition to the elastic chains, contained an equal
number of dangling chains.49 The model networks were
synthesized in a one-pot preparation by group transfer
polymerization (GTP).50-53 Both the architecture (block
and statistical) and the size of the dangling and elastic
chains of the networks were controlled due to the ‘living’
character of GTP. All networks prepared were charac-
terized in terms of their swelling behavior in water as
a function of pH. The degree of swelling was shown to
be affected by both the solution pH and the network
structure.
50% v/v solution in freshly distilled THF because it is a highly
viscous liquid and rather difficult to handle without dilution.
The solution was passed twice through a basic alumina column
to remove inhibitors and acidic impurities. It was stirred over
calcium hydride for 2 h and kept in the refrigerator until use.
Finally, it was syringed through a 0.45 µm filter (to keep any
suspended calcium hydride) directly into the polymerization
flask. All glassware was dried overnight in an oven at 120 °C
and assembled hot under dynamic vacuum. The polymeriza-
tions were carried out in 100 mL round-bottom flasks. Cata-
lytic amounts of TBABB (∼10 mg) were transferred to each
flask, which was fitted with a rubber septum and purged with
dry nitrogen. Freshly distilled THF was transferred from the
still to the flask via a glass syringe. The initiator was added
next, followed by the addition of the monomers and cross-
linker.
Netw or k Syn th esis. All polymerization reactions were
carried out at ambient temperature without thermostating the
polymerization flasks. The polymerization exotherm was
monitored by a digital thermometer and was used to follow
the progress of the reaction. The synthetic sequence for all
cross-linked star structures prepared is illustrated in Figure
2. Homopolymer cross-linked stars were prepared when only
one monomer was added before each addition of cross-linker,
and the monomer added second was the same as that used
first. When the monomer added after the first cross-linking
was different from that added before, heteroarm copolymer
star-based networks were synthesized. The sequential or
simultaneous addition of the two different monomers before
each addition of the EGDMA cross-linker resulted in the
preparation of arms of block or statistical architecture, and it
ultimately led to the formation of block or statistical copolymer
cross-linked stars, respectively. A typical polymerization pro-
cedure is detailed below which describes the synthesis of a
block copolymer cross-linked star having primary and second-
ary arms both comprising linear diblock copolymers with 20
PEGMA units and 5 THPMA units. Freshly distilled THF (50
mL) and 0.2 mL of MTS initiator (0.98 mmol) were syringed
in this order to a 100 mL round-bottom flask containing a
small amount (∼10 mg) of TBABB. PEGMA (13.1 mL of 50
v/v% solution in THF, 19.6 mmol) monomer was slowly added
under stirring. The polymerization exotherm (25.8-30.4 °C)
abated within 5 min, samples for GPC were extracted (full
monomer conversion, polymer Mn ) 8400 g mol-1 and Mw/Mn
) 1.08), and 0.84 mL of THPMA (4.9 mmol) was added slowly,
giving an exotherm (30.3-31.7 °C). After sampling (full
monomer conversion, polymer Mn ) 8700 g mol-1 and Mw/Mn
Exp er im en ta l Section
Ma ter ia ls a n d Meth od s. All reagents used were purchased
from Aldrich, with the exception of the PEGMA monomer
(purity 99% w/w, 0.4% water), which was kindly donated (as
Bisomer MPEG350MA) by Laporte Performance Chemicals.
Figure 1 shows the chemical formulas and names of the
monomers, the cross-linker and the initiator used for the
network synthesis. Tetrahydropyranyl methacrylate (THPMA)
is not commercially available and was in-house synthesized
by the sulfuric acid-catalyzed reaction of 3,4-dihydro-2H-pyran
(DHP, 97%) with MAA (99%), following a modification of the
method reported by Hertler.54 The polymerization catalyst,
tetrabutylammonium bibenzoate (TBABB), was in-house syn-
thesized by the method described by Dicker et al.52 The catalyst
powder was stored under vacuum until use. Tetrahydrofuran
(THF, 99.8%; Labscan) served as the polymerization solvent.
It was dried by refluxing it over a potassium/sodium alloy for
3 days and was freshly distilled prior to the polymerization.
1-Methoxy-1-(trimethylsiloxy)-2-methylpropene (MTS, 95%)
was used as the polymerization initiator. The THPMA mono-
mer and the cross-linker, ethylene glycol dimethacrylate
(EGDMA, 98%), were passed through basic alumina columns
to remove inhibitors and acidic impurities. THPMA was passed
more than once (typically 3-4 times) through the alumina
columns until no MAA olefinic protons were detected in its

Macromolecules, Vol. 35, No. 13, 2002
Double-Hydrophilic Cross-linked Stars 4905
F igu r e 2. Synthetic sequences employed for the preparation of the various cross-linked star architectures of this study: PEGMA,
methoxy poly(ethylene glycol) methacrylate; THPMA, tetrahydropyranyl methacrylate; EGDMA, ethylene glycol dimethacrylate;
MTS, 1-methoxy-1-(trimethylsiloxy)-2-methylpropene; TBABB, tetrabutylammonium bibenzoate.
) 1.09), 0.75 mL of EGDMA (3.94 mmol) was added which
produced an exotherm (31.4-33.7 °C). Samples were with-
drawn again (full monomer conversion, polymer Mn ) 16 400
g mol-1 and Mw/Mn ) 1.47) before 0.84 mL of THPMA (4.9
mmol) was added giving an exotherm from 33.1 to 34.0 °C.
Samples for GPC were withdrawn again (full monomer
conversion, polymer Mn ) 19 300 g mol-1 and Mw/Mn ) 1.46),
followed by the addition of 13.1 mL PEGMA (50 v/v% solution
in THF, 19.6 mmol) with an exotherm from 31.2 to 32.7 °C
(full monomer conversion, polymer Mn ) 21 000 g mol-1 and
Mw/Mn ) 1.40). In the final stage, 0.75 mL of EGDMA (3.94
mmol) was added, which promoted gelation within seconds.
Ch a r a cter iza tion of Netw or k P r ecu r sor s by Gel P er -
m ea tion Ch r om a togr a p h y. Molecular weights (MWs) and
molecular weight distributions (MWDs) of the linear, “arm-
first” and “in-out” star polymer precursors to the networks
were obtained by gel permeation chromatography (GPC). A
Polymer Laboratories system was used, equipped with a PL-
LC1120 isocratic pump, an ERC-7515A refractive index detec-
tor, and a nonthermostated (room temperature ∼25 °C) PL-
mixed “D” GPC column (bead size ) 5 µm; pore sizes ) 100,
500, 103, and 104 Å). The mobile phase was THF, delivered at
a flow rate of 1 mL min-1. The calibration curve was based on
seven narrow MW (630, 4250, 13000, 28900, 50000, 128000,
and 260000) linear poly(methyl methacrylate) standards,
which provided relatively accurate MW calculations for the
linear segments but only qualitative estimates for the MWs
of the star polymers.
mL of THF) and left for 2 weeks. Fourier transform infrared
(FTIR) spectroscopy using a VECTOR 22 Bruker FTIR spec-
trometer was employed to qualitatively confirm the hydrolysis
of the THPMA units, by comparing the spectra taken from a
vacuum-dried sample of the heteroarm copolymer star network
(50 mol % THPMA) before and after the addition (and washing)
of HCl.
Ch a r a cter iza tion of th e Degr ee of Sw ellin g in Wa ter .
After hydrolysis the gels were washed in 500 mL of distilled
water, which was changed every 2-3 days for 2 weeks to
remove all THF and the excess HCl. Measurements of the pH
and the total gel mass took place before each water change.
When the pH and mass values stabilized, the water was
replaced once more and the gels were allowed to equilibrate
for 1 additional week. Two cubic samples of size ∼1 cm were
then cut from each network. The mass of the water swollen
cubes was measured gravimetrically before placing them in a
vacuum oven to dry for 48 h at room temperature. The dry
gel mass was determined and the degrees of swelling in the
aqueous phase were calculated as the ratio of wet network
mass divided by the dry gel mass.
Ten cubic samples from each network of size ∼1 cm were
taken, and each was placed in a separate vial. Then, 10 mL of
distilled water were added to each vial. The pH of the gel
samples in water was varied between 2 and 12 by the addition
of the appropriate volume of 0.5 M HCl or 0.1 M NaOH
solutions. The samples were allowed to equilibrate for 3 weeks
and the solution pH and gel mass were measured. The degree
of swelling for each gel sample was calculated. All degrees of
swelling were determined in triplicate and the averages of the
measurements are presented.
Ca lcu la tion of th e Degr ee of Ion iza tion a n d th e p K.
The degree of ionization of each sample was calculated as the
number of NaOH equivalents added divided by the number of
MAA equivalents (calculated from network wet mass, known
degree of swelling, and network composition) in the sample.
The effective pK of each network was estimated as the pH (of
the supernatant solution) at 50% ionization.
Deter m in a tion of th e Sol F r a ction . The gels were taken
out of the polymerization flasks by cutting the flasks with a
diamond knife and were subsequently washed in 200 mL of
THF for 1 week to remove the sol fraction. The solvent was
replaced twice during that time. The THF solution was
recovered by filtration each time. The solvent was evaporated
and the extractables were dried under vacuum for 48 h at room
temperature. The sol fraction was calculated as the ratio of
the mass of the extracted polymer to the theoretical mass of
the gel.
THP MA Hyd r olysis. After the sol fraction was removed
and while the gels were in THF, 100% mol excess HCl 2 M
with respect to the theoretical number of THPMA equivalents
in the gels was added to the networks containing THPMA. For
example, 10 mL of a 2 M HCl solution were added to the
network [(P20-b-T5)-star-(T5-b-P20)]-network (still swollen in 200
Resu lts a n d Discu ssion
Syn th esis a n d Str u ctu r e of Cr oss-Lin k ed Sta r
P olym er s. Eight polymer networks were synthesized
by GTP, using MTS as the initiator, TBABB as the

4906 Georgiades et al.
Macromolecules, Vol. 35, No. 13, 2002
F igu r e 3. Schematic representation of the synthetic sequence followed for the preparation of the block copolymer-based network
[(P20-b-T5)-star-(T5-b-P20)]-network (D). The white color indicates methoxy poly(ethylene glycol) methacrylate (PEGMA or P), while
the black indicates tetrahydropyranyl methacrylate (THPMA or T). Key: EGDMA, ethylene glycol dimethacrylate; MTS, 1-methoxy-
1-(trimethylsiloxy)-2-methylpropene; TBABB, tetrabutylammonium bibenzoate. The number of arms at the cross-links is not 2,
4, or 8 as shown in the figure but around 30-60.
catalyst, and PEGMA and THPMA as the monomers.
THPMA is the protected form of MAA which cannot be
polymerized directly by GTP because it contains labile
protons which cause termination of the polymerization.
The THPMA units were readily converted to MAA units
after the polymerization by acid hydrolysis. EGDMA
served as the cross-linker for the formation of star
polymers and cross-linked star polymer networks due
to its ability to be added repeatedly at the core, bringing
together a large number of chains.55 The hydrogels
obtained were model networks, having accurate MWs
of the dangling chains and of the segments between
cross-links, due to the “livingness” of GTP, which allows
good control over MWs and MWDs.
four, as shown in the figure, but much higher, between
30 and 60, as determined by static light scattering on
similar star polymers.49,55 Further addition of THPMA
formed secondary arms from the core of the stars
outward, giving a structure known as “in-out” stars.
Chain extension of the secondary arms was accom-
plished by addition of PEGMA leading to secondary
arms of the same composition and architecture as that
of the primary arms. The active sites at that point were
at the ends of the secondary arms. The “livingness” of
the polymerization implies that the number of second-
ary arms is equal to the number of primary arms.49 The
final addition of cross-linker induced the inter-cross-
linking of the “in-out” stars via their secondary arms
to form the network.
Figure 3 is a schematic representation of the synthetic
procedure used for the preparation of one of the net-
works, network D, which is based on diblock copolymer
linear chains. The PEGMA units are shown in white,
while the THPMA/MAA units are shown in black. The
large black dots indicate the secondary EGDMA cores,
and the active sites of the polymerization are shown by
the arrows. For this particular network, the procedure
included six steps, four additions of monomer and two
additions of cross-linker. All steps took place in the same
reactor. After the catalyst, the solvent and the initiator,
the two monomers were added sequentially to the
reaction flask. First, the addition of PEGMA gave linear
homopolymer chains followed by the addition of THP-
MA, which induced chain growth and formation of
diblock copolymer chains. The subsequent in situ po-
lymerization of EGDMA in 4-fold molar excess with
respect to the initiator led to the formation of “arm-first”
star polymers, which maintained the active sites in the
core of the star. These cross-linked chains constitute the
primary arms of the stars and the dangling chains of
the networks. The 4:1 ratio of cross-linker to initiator
used was found in previous studies to give satisfactory
incorporation of chains to stars, without using large
quantities of cross-linker.56,57 The number of arms is not
Figure 4 shows all network structures prepared in
this study. These comprise the two homopolymer cross-
linked stars, A and B, one equimolar heteroarm copoly-
mer cross-linked star, C, and five isomeric copolymer
cross-linked stars with 80% mol PEGMA and 20 mol %
MAA, D, E, F, G, and H. The sequence of reagent
addition for the preparation of each of the networks was
presented in Figure 2. Table 1 shows the theoretical
chemical formulas of all networks precursors and their
MW characterization results.
Molecu la r Weigh t An a lysis. Figure 5 shows the
GPC chromatograms of the five precursors to network
D. The MWDs of the linear PEGMA homopolymer, P20
and the linear PEGMA-THPMA diblock copolymer, P20
,
-
b-T5, are narrow and unimodal. The MWD of the “arm-
first” star, (P20-b-T5)-star, is bimodal, containing a small
amount of unattached linear chains in addition to the
stars. Similar are the chromatograms of the “in-out”
stars, (P20-b-T5)-star-(T5) and (P20-b-T5)-star-(T5-b-P20),
which are multimodal, with a narrow MWD of the main
peak corresponding to the “in-out” star. The MW of the
main peak is increasing from the initial homopolymer
to the final “in-out” star precursor, as expected. The
relatively small difference between the GPC retention

Macromolecules, Vol. 35, No. 13, 2002
Double-Hydrophilic Cross-linked Stars 4907
successive step. It should be noted though that both the
Mns and the Mps determined by GPC for the star
polymers are only apparent values and much lower than
the actual ones due to the compact nature of the star
structure compared to the linear PMMA calibration
standards. Recent characterization of similar systems
by static light scattering suggested that the absolute
weight-average MW, Mw, of the star polymers is about
10 times higher than the Mn determined by GPC.49,59
For all the isomeric co-networks D-H the increase of
the apparent Mn from the linear to the “arm-first” star
as well as the increase from the “arm-first” star to the
final “in-out” star is about the same, as expected.
THP MA Hyd r olysis. Using acidic conditions, the
gels containing THPMA were transformed into MAA
gels.60 The THPMA hydrolysis reaction is exactly the
reverse to its synthesis reaction. The deprotection of
THPMA was confirmed qualitatively by FTIR, using
samples from gel C, which are representative for all the
other gels containing THPMA, given that excess HCl
was used and extensive time was allowed for hydrolysis
in all gels. The appearance of a double characteristic
band at 2940-2860 cm-1, due to the O-H stretching
vibration of the carboxylic group of MAA, provided
evidence for successful hydrolysis of THPMA to MAA.61
The hydrogen ion titrations of the hydrolyzed network
samples, which indicated the presence of a concentration
of carboxylic acid groups close to that stoichiometrically
expected, were used for the quantitative confirmation
of the hydrolysis. The final hydrogels are either hydro-
philic (homopolymer networks A and B) or double-
hydrophilic (heteroarm copolymer star network C and
statistical and block copolymer star networks D-H)
gels, containing one or two types of hydrophilic units,
respectively.
F igu r e 4. Schematic representation of the architectures of
all the model networks prepared in this study. The methoxy
poly(ethylene glycol) methacrylate (PEGMA) units are depicted
in white, while in black are shown the tetrahydropyranyl
methacrylate (THPMA) units, which are to be converted
subsequently to methacrylic acid (MAA) units. The number
of arms at the cross-links is not 2 or 8 as shown in the figure
but around 30-60.
Sol F r a ction of th e Netw or k s. Table 2 shows the
percent extractables for each network. Network B, a
THPMA homopolymer network, has the highest per-
centage of extractables (15.1%) while all the other
networks exhibit relatively low sol fractions (<10%). The
low values of these sol fractions support findings of our
previous work56,57 where a 4-fold molar excess of EGD-
MA cross-linker with respect to the initiator has been
determined as the optimal ratio for sufficiently high
conversion to star polymer during “arm-first” star
polymer synthesis. Similar results were also obtained
for studies on both the anionic and “living” cationic
“arm-first” star polymer synthesis62,63 in which high star
polymer yields (>85% for anionic) were obtained.
Sw ellin g Beh a vior of t h e Hyd r ogels in Wa t er .
Figure 6 shows the pH dependence of the degrees of
swelling (DSs) and the degrees of ionization for the eight
hydrolyzed model networks. Network A does not contain
any MAA ionizable units and, therefore, its degree of
ionization is zero within the whole pH range. As
expected, its DS remains practically constant and unaf-
fected by the pH. All the other networks contain MAA
ionizable units and are therefore sensitive to the pH,
exhibiting similar pH dependence of their DSs. Thus,
at low pH values, MAA is protonated and electrically
neutral, leading to low DSs. An increase in the DS is
observed due to the ionization of the MAA units at
higher pH (>4). This ionization destroys the hydrogen
bonds between the MAA units, it builds up repulsive
forces between the now negatively charged chains of the
network, and it causes an increase in the osmotic
pressure due to accumulation of counterions within the
volumes of the “arm-first” and “in-out” star polymers
is attributed to the small difference of the hydrodynamic
volume of these two structures which is common in star
polymers with numerous arms and with primary and
secondary arms of similar length. This was also ob-
served in similar work on the synthesis of “in-out” star
polymers by anionic polymerization.58
Table 1 summarizes the GPC results for the precur-
sors to all eight networks. These include the apparent
number-average MWs, Mns, the polydispersity indices,
Mw/Mns, and the peak MWs, Mps. The theoretical MWs
of the linear precursors are also shown. All linear
polymers exhibit narrow MWDs (Mw/Mn < 1.3), and
their apparent Mns are reasonably close to their theo-
retical MWs. For the star polymers the MWDs are
broader, mainly due to the incomplete incorporation of
linear chains into stars and the inclusion of the unat-
tached chain population in the calculation due to the
overlap of the two peaks. Incomplete incorporation can
be attributed to high solution viscosity and low chain
mobility toward the later stages of the polymerization
procedure.56 An additional reason in our case is the large
size of the PEGMA units, which causes considerable
steric hindrance during the polymerization.
An increase in the apparent experimental MWs (Mns
and Mps) of the network precursors in all stages of the
synthetic procedure was determined by GPC, suggesting
the growth of the whole distribution of the stars at each

4908 Georgiades et al.
Macromolecules, Vol. 35, No. 13, 2002
Ta ble 1. Gel P er m ea tion Ch r om a togr a p h y (GP C) Ch a r a cter iza tion of th e Cr oss-Lin k ed Sta r Netw or k P r ecu r sor s
theoretical structure
of precursora
theor
MW
apparent
Mn by GPC
Mw/Mn
by GPC
apparent peak
MW (Mp) by GPC
network
A
P25
25-star
25-star-P25
T25
25-star
25-star-T25
T25
25-star
25-star-P25
P20
8925
4425
4425
12 800
25 800
47 900
4900
23 400
96 300
5800
25 600
104 500
8400
1.08
1.65
1.56
1.10
2.49
1.79
1.09
2.26
1.25
1.08
1.09
1.47
1.46
1.40
1.28
1.07
1.33
1.29
1.38
1.08
1.34
1.36
1.08
1.36
1.36
1.07
1.37
1.38
13 200
59 800
79 100
5300
68 800
96 100
6100
68 800
116 900
9100
P
P
B
C
D
T
T
T
T
7175
8025
P
20-b-T5
8700
9600
(P20-b-T5)-star
(P20-b-T5)-star-T5
(P20-b-T5)-star-(T5-b-P20
T5
T5-b-P20
(T5-b-P20)-star
(T5-b-P20)-star-P20
(T5-b-P20)-star-(P20-b-T5)
20-co-T5
(P20-co-T5)-star
(P20-co-T5)-star-(P20-co-T5)
P25
16 400
19 300
21 000
900
28 800
33 000
33 000
900
)
E
1025
8025
10 800
17 200
21 500
25 000
10 500
19 100
25 700
10 400
17 200
21 700
11 200
21 400
24 200
11 500
25 900
28 800
36 800
11 200
29 600
36 800
11 200
28 100
32 100
11 900
33 900
32 100
F
P
8025
8925
7125
G
H
P
P
P
25-star
25-star-(P15-co-T10
15-co-T10
)
(P15-co-T10)-star
(P15-co-T10)-star-P25
a
P and T are (further) abbreviations for methoxy poly(ethylene glycol) methacrylate (PEGMA) and tetrahydropyranyl methacrylate
(THPMA), respectively.
decrease in the DS at high pH values is due to the high
ionic strength of the solution at this pH, resulting from
the relatively high sodium hydroxide concentration,
which screens the electrostatic interactions.
The degrees of ionization also increase systematically
with increasing pH. The effective pKs (taken as the pH
at 50% ionization) were determined for the MAA
containing networks from their degree of ionization
curves. pKs of around 6.8 were found which are higher
than the corresponding values for linear MAA polymers
of around 5.4.67 This is attributed to the fact that the
pH was measured in the surrounding solution where
proton concentration is lower due to counterion parti-
tioning into the gel phase.66
In agreement with previous results for homopolymer
based cross-linked stars of ionizable methacrylate mono-
mers,49 we have observed that homopolymer star net-
work B, heteroarm copolymer star network C, and
statistical copolymer star network G break into small
pieces of size ∼1 mm above pH 9. This has also been
reported by Flory68 on ionized poly(MAA) networks. It
can be attributed to intense electrostatic repulsion in
the gels, which causes the rupture of the carbon-carbon
bonds of the backbone. The breakage of the gels can lead
to an overestimation of the DSs, caused by water
entrapment between gel pieces.
Figure 7 shows the DSs for each network at low and
high pH values. The DSs at low pH are composition-
dependent but architecture-independent. A linear de-
crease of the DSs is observed as a function of the
network content in MAA. This decrease is attributed
to the greater hydrophilicity of PEGMA compared to
protonated MAA and the formation of hydrogen bonds
between the protonated MAA units, which act as ad-
ditional physical cross-links and cause the shrinkage
of the networks.69 The DSs of the isomeric copolymer
networks D-H are all around 10 and independent of
F igu r e 5. Gel permeation chromatograms of the five precur-
sors to network [(P20-b-T5)-star-(T5-b-P20)]-network (D): P,
methoxy poly(ethylene glycol) methacrylate (PEGMA); T,
tetrahydropyranyl methacrylate (THPMA). P20
: linear ho-
mopolymer of PEGMA; P20-b-T5, linear diblock copolymer of
PEGMA and THPMA; (P20-b-T5)-star, “arm-first” star-block;
(P20-b-T5)-star-(T5), first “in-out” star; (P20-b-T5)-star-(T5-b-P20),
final “in-out” star.
Ta ble 2. Sol Extr a cta bles fr om th e Cr oss-Lin k ed Sta r
Netw or k s
% extractables
network
theoretical structurea
[P25-star-P25]-network
[T25-star-T25]-network
[T25-star-P25]-network
[(P20-b-T5)-star-(T5-b-P20)]-network
[(T5-b-P20)-star-(P20-b-T5)]-network
[(P20-co-T5)-star-(P20-co-T5)]-network
[P25-star-(P15-co-T10)]-network
[(P15-co-T10)-star-P25]-network
(w/w)
A
B
C
D
E
F
G
H
9.0
15.1
10.1
8.2
9.1
7.6
10.2
8.5
a
P and T are (further) abbreviations for methoxy poly(ethylene
glycol) methacrylate (PEGMA) and tetrahydropyranyl methacry-
late (THPMA), respectively.
gels,64-66 thus leading to the expansion of the network.
The DS curves reach a maximum (which is dependent
on the structure of the networks as discussed in a later
section) and then decrease with increasing pH. This

Macromolecules, Vol. 35, No. 13, 2002
Double-Hydrophilic Cross-linked Stars 4909
F igu r e 6. pH dependence of the aqueous degrees of swelling and the degrees of ionization of all the cross-linked star networks:
P, methoxy poly(ethylene glycol) methacrylate (PEGMA); A, methacrylic acid (MAA).
their structure, suggesting that the location of the MAA
units in the network structure does not affect their
swelling properties at low pH.
In contrast, the DSs at high pH increase with the
MAA content of the network and are strongly architec-
ture-dependent. Thus, an increase of the high pH DS is
observed following the order A < (D, H) ≈ C ≈ B
corresponding to networks which contain 0, 20, 50, and
100% MAA, respectively. Under these conditions, the
main contribution to swelling comes from the ionized
MAA units since the high pH DSs are at least two times
greater than the corresponding low pH DSs (2-3.5-fold
increase from low to high pH for the heteroarm star
network and the statistical and block copolymer star
networks and 6-fold increase for the MAA homopolymer
network) and much higher than the DS of the PEGMA
homopolymer network. However, at high pH, apart from
the composition, there is also a strong dependence of

4910 Georgiades et al.
Macromolecules, Vol. 35, No. 13, 2002
static repulsions developed between the charged MAA
blocks, located at the end of the dangling chains of
network E, which cause its expansion and ultimately
result in statistically the same DS for the two networks.
It is noteworthy that the effective pK of network F is
slightly lower than that of network E.
Con clu sion s
We have used GTP to synthesize complex model
network structures based on cross-linked “in-out” star
polymers of PEGMA and MAA. These networks are new
materials exhibiting interesting swelling behavior. Their
degree of swelling was found to be affected by the
solution pH, network composition and star architecture.
Therefore, network design can be critical in controlling
their swelling properties. Future work will focus on
further investigation of the effects of polymer composi-
tion and molecular weight on the swelling behavior.
Possible applications based on both the ionic and double-
hydrophilic character of these networks will be also
explored.
F igu r e 7. Low pH and high pH degrees of swelling of the
networks as a function of the methacrylic acid (MAA) mol
fraction. The curves are drawn to guide the eye.
the DS on the star architecture. This is also shown in
Figure 7 where the five isomeric copolymer networks
(D-H) exhibit a significant variation of their high pH
DSs, which was not observed for the low pH DSs. The
factors that affect the DS and result in these differences
in the high pH DSs of the isomeric copolymer networks
are, first, the MAA content of the elastic chains of the
networks (secondary arms of the stars) and, second, the
location of the MAA units within the star structure.
Thus, network G, which has the highest MAA content
in its elastic chains among the copolymer networks,
exhibits the highest DS, while the lowest DS is obtained
for network H, which has no MAA units in the second-
ary arms. This is consistent with the fact that the main
contribution to swelling at high pH comes from the
ionized MAA units, and it suggests that the elastic
chains have a greater effect on the swelling behavior of
the networks compared to the dangling chains, which
has also been observed for homopolymer and am-
phiphilic copolymer-based cross-linked stars.49,59 The
effect of the location of the MAA units in the network
structure on the swelling behavior of the networks is
seen when comparing the high pH DS of the isomeric
networks D, E, and F. Although all three networks have
the same MAA content in both their elastic and dangling
chains, network D exhibits the lowest DS at high pH.
This is attributed to the fact that the ionized MAA units
in network D are all located in blocks around the cores
of the “in-out” stars. This initially high charge density
may lead to significant counterion condensation and
ultimately result in the weakest electrostatic repulsion
and lowest increase in osmotic pressure.70 Comparing
networks D and E, the latter has a higher DS even
though it also has MAA blocks around the secondary
cores of the network (formed upon the inter-cross-
linking of the “in-out” stars to obtain the polymer
network). This seems surprising at a first glance but
can be explained by the presence of MAA units at the
end of the dangling chains of network E, which are not
“locked” around a star core and can extend to a greater
volume, thus reducing counterion condensation. On the
other hand, networks E and F exhibit, within experi-
mental error, the same DSs, consistent with the same
composition of both their dangling and elastic chains.
This result also suggests that the random distribution
of the MAA units in network F, or their location in
blocks around the secondary cores and at the end of the
dangling chains in network E, has the same effect on
the swelling behavior. A more effective counterion
condensation would be expected to take place around
the secondary cores of network E, which would result
in a lower high pH DS compared to network F. However,
this effect is counterbalanced by the stronger electro-
Ack n ow led gm en t. We gratefully acknowledge the
financial support by the University of Cyprus Research
Committee (Grant 2000-2003). We also thank Dr. R.
Budd of Laporte Performance Chemicals, for kindly
donating the PEGMA monomer. Finally, we thank our
colleague Dr. C. Sergides of Medochemie Ltd., for
obtaining the FTIR spectra.
Refer en ces a n d Notes
(1) Nomura, E.; Ito, K.; Kamachi, M. Macromolecules 1997, 30,
2811-2817.
(2) Angot, S.; Taton, D.; Gnanou, Y. Macromolecules 2000, 33,
5418-5426.
(3) Rempp, P.; Lutz, P. J . Graft Copolymers In Encyclopedia of
Advanced Materials; Pergamon: Oxford, U.K., 1994; p 934.
(4) Mortensen, K. Colloids Surf. A: Physicochem. Eng. Aspects
2001, 183-185, 277-292.
(5) Barker, M. C.; Vincent, B. Colloids Surf. 1984, 8, 289-295.
(6) Xu, Z.; Chen, Y.; Zhang, G.; Cheng, S.; Feng, L. J . Polym.
Sci., Part A: Polym. Chem. 1999, 37, 2719-2725.
(7) Winzor, C. L.; Mra´zek, Z.; Winnik, M. A.; Croucher, M. D.;
Riess, G. Eur. Polym. J . 1994, 30, 121-128.
(8) Liang, W.; Bognolo, G.; Tadros, Th. F. Langmuir 1995, 11,
2899-2904.
(9) Bourgeat-Lami, E.; Guyot, A. Polym. Bull. (Berlin) 1995, 35,
691-696.
(10) Eisenberg, A. Bioconjugate Chem. 1998, 9, 564-572.
(11) Allen, C.; Eisenberg, A.; Mrsic, J .; Maysinger, D. Drug
Delivery 2000, 7, 139-145.
(12) Alexandridis, P.; Holzwarth, J . F.; Hatton, T. A. Macromol-
ecules 1994, 27, 2414-2425.
(13) Michels, B.; Waton, G.; Zana, R. Langmuir 1997, 13, 3111-
3118.
(14) Linse, P. J . Phys. Chem. 1993, 97, 13896-13902.
(15) Malmsten, M.; Linse, P.; Zhang, K. W. Macromolecules 1993,
26, 2905-2910.
(16) Mortensen, K.; Brown, W. Macromolecules 1993, 26, 4128-
4135.
(17) Dicker, I. B.; Hertler, W. R.; Ma, S.-H. US Pat. 5,219,945,
1993.
(18) Ma, S.-H.; Dicker, I. B.; Hertler, W. R. Eur. Pat. 0 556 649
A1, 1993.
(19) Kabanov, A. V.; Bronich, T. K.; Kabanov, V. A.; Yu, K.;
Eisenberg, A. Macromolecules 1996, 29, 6797-6802.
(20) Harada, A.; Kataoka, K. Macromolecules 1995, 28, 5294-
5299.
(21) Kabanov, A. V.; Kabanov, V. A. Adv. Drug Delivery Rev. 1998,
30 (1-3), 49-60.
(22) Kataoka, K.; Togawa, H.; Harada, A.; Yasugi, K.; Matsumoto,
T.; Katayose, S. Macromolecules 1996, 29, 8556-8557.
(23) Lieske, A.; J aeger, W. Macromol. Chem. Phys. 1998, 199,
255-260.

Macromolecules, Vol. 35, No. 13, 2002
Double-Hydrophilic Cross-linked Stars 4911
(24) Bijsterbosch, H. D.; Cohen-Stuart, M. A.; Fleer, G. J .; van
Caeter, P.; Goethals, E. J . Macromolecules 1998, 31, 7436-
7444.
(48) Hild, G. Prog. Polym. Sci. 1998, 23, 1019-1149.
(49) Vamvakaki, M.; Hadjiyannakou, S. C.; Loizidou, E.; Patrick-
ios, C. S.; Armes, S. P.; Billingham, N. C. Chem. Mater. 2001,
13, 4738-4744.
(25) Vamvakaki, M.; Billingham, N. C.; Armes, S. P. Polymer
1999, 40, 5161-5171.
(50) Webster, O. W.; Hertler, W. R.; Sogah, D. Y.; Farnham, W.
B.; RajanBabu, T. V. J . Am. Chem. Soc. 1983, 105, 5706-
5708.
(26) Bu¨tu¨n, V.; Vamvakaki, M.; Billingham, N. C.; Armes, S. P.
Polymer 2000, 41, 3173-3182.
(27) Forder, C.; Patrickios, C. S.; Armes, S. P.; Billingham, N. C.
Macromolecules 1996, 29, 8160-8169.
(51) Sogah, D. Y.; Hertler, W. R.; Webster, O. W.; Cohen, G. M.
Macromolecules 1987, 20, 1473-1488.
(52) Dicker, I. B.; Cohen, G. M.; Farnham, W. B.; Hertler, W. R.;
Laganis, E. D.; Sogah, D. Y. Macromolecules 1990, 23, 4034-
4041.
(53) Webster, O. W. J . Polym. Sci., Part A: Polym. Chem. 2000,
38, 2855-2860.
(54) Hertler, W. R. U.S. Patent 5,072,029, 1991. Demosthenous,
E.; Hadjiyannakou, S. C.; Vamvakaki, M.; Patrickios, C. S.
Macromolecules 2002, 35, 2252-2260.
(28) Forder, C.; Patrickios, C. S.; Armes, S. P.; Billingham, N. C.
Macromolecules 1997, 30, 5758-5762.
(29) Ashford, E. J .; Naldi, V.; O’Dell, R.; Billingham, N. C.; Armes,
S. P. Chem. Commun. 1999, 1285-1286.
(30) Colfen, H.; Antonietti, M. Langmuir 1998, 14, 582-589.
(31) Spatz, J .; Mossmer, S.; Moller, M.; Kocher, M.; Neher, D.;
Wegner, G. Adv. Mater. 1998, 10, 473-474.
(32) Vamvakaki, M.; Billingham, N. C.; Armes, S. P.; Watts, J .
F.; Greaves, S. J . J . Mater. Chem. 2001, 11, 2437-2444.
(33) Rungsardthong, U.; Deshpande, M.; Bailey, L.; Vamvakaki,
M.; Armes, S. P.; Garnett, M. C.; Stolnik, S. J . Controlled
Release 2001, 73, 359-380.
(55) Simms, J . A. Rubber Chem. Technol. 1991, 64, 139-151.
(56) Simmons, M. R.; Yamasaki, E. N.; Patrickios, C. S. Polymer
2000, 41, 8523-8529.
(57) Simmons, M. R.; Yamasaki, E. N.; Patrickios, C. S. Macro-
molecules 2000, 33, 3176-3179.
(34) Vamvakaki, M.; Armes, S. P.; Billingham, N. C. Macromol-
ecules 1999, 32, 2088-2090.
(35) Dickie, R. A., Labana, S. S., Bauer, R. S., Eds.; Cross-Linked
Polymers: Chemistry, Properties and Applications; ACS
Symposium Series 367; American Chemical Society: Wash-
ington, DC, 1988.
(36) Tanaka, T. Sci. Am. 1981, 244 (1), 124-138.
(37) Osada, Y.; Ross-Murphy, S. B. Sci. Am. 1993, 268 (5), 42-
47.
(38) Buchholz, F. L., Peppas, N. A., Eds.; Superabsorbent Poly-
mers; ACS Symposium Series 573; American Chemical
Society: Washington, DC, 1995.
(39) Buchholz, F. L., Graham, A. T.; Eds.; Modern Superabsorbent
Polymer Technology; Wiley: New York, 1998.
(40) J en, A. C.; Wake, M. C.; Mikos, A. G. Biotechnol. Bioeng.
1996, 50, 357-364.
(41) Dagani, R. Chem. Eng. News 1997, 75 (23), 26-37.
(42) Lee, K. Y.; Mooney, D. J . Chem. Rev. 2001, 101, 1869-1880.
(43) Torres-Lugo, M.; Peppas, N. A. Macromolecules 1999, 32,
6646-6651.
(44) Bae, Y. H.; Knon, I. C. In Biorelated Polymers and Gels;
Okano, T., Ed.; Academic Press: Boston, MA, 1998; pp 93-
134.
(58) Tsitsilianis, C.; Chaumont, P.; Rempp, P. Macromol. Chem.
1990, 191, 2319-2328.
(59) Vamvakaki, M.; Patrickios, C. S. Chem. Mater. 2002, 14,
1630-1638.
(60) Kearns, J . E.; McLean, C. D.; Solomon, D. H. J . Macromol.
Sci., Chem. 1974, A8, 673-685.
(61) Williams, D. H.; Fleming, R. I. Spectroscopic Methods in
Organic Chemistry, 4th ed.; McGraw-Hill: London, 1989; p
49.
(62) Kanaoka, S.; Omura, T.; Sawamoto, M.; Higashimura, T.
Macromolecules 1992, 25, 6407-6413.
(63) Tsitsilianis, C.; Voulgaris, D. Macromol. Chem. Phys. 1997,
198, 997-1007.
(64) Siegel, R. A.; Firestone, B. A. Macromolecules 1988, 21, 3254-
3259.
(65) Shibayama, M.; Tanaka, T. Adv. Polym. Sci. 1993, 109, 1-62.
(66) Philippova, O. E.; Hourdet, D.; Audebert, R.; Klokhlov, A. R.
Macromolecules 1997, 30, 8278-8285.
(67) Merle, Y. J . Phys. Chem. 1987, 91, 3092-9098.
(68) Flory, P. J . Principles of Polymer Chemistry; Cornell Univer-
sity Press: Ithaca, NY, 1953; pp 584-589.
(69) Annaka, M.; Tanaka, T. Nature (London) 1992, 335, 430-
432.
(45) Klier, J .; Scranton, A. B.; Peppas, N. A. Macromolecules 1990,
23, 4944-4949.
(70) Oosawa, F. Polyelectrolytes; Marcel Dekker: New York, 1971;
pp 23-26.
(46) Lowman, A. M.; Peppas, N. A. Macromolecules 1997, 30,
4959-4965.
(47) Lowman, A. M.; Peppas, N. A. Polymer 2000, 41, 73-80.
MA011822Z