Poly(ethylene oxide)-grafted poly(N-isopropylacrylamide) networks: Preparation, characterization and rapid deswelling and reswelling behavior of hydrogels

Article abstract of DOI:10.1016/j.reactfunctpolym.2011.12.006

Poly(ethylene oxide)-grafted poly(N-isopropylacrylamide) networks (PNIPAAm-g-PEO) were prepared via the reversible addition-fragmentation chain transfer polymerization (RAFT) of N-isopropylacrylamide with trithiocarbonate-terminated poly(ethylene oxide) a

Full text of DOI:10.1016/j.reactfunctpolym.2011.12.006

Reactive & Functional Polymers 72 (2012) 176–184
Contents lists available at SciVerse ScienceDirect
Reactive & Functional Polymers
journal homepage: www.elsevier.com/locate/react
Poly(ethylene oxide)-grafted poly(N-isopropylacrylamide) networks: Preparation,
characterization and rapid deswelling and reswelling behavior of hydrogels
⇑
Yaochen Zheng, Sixun Zheng
Department of Polymer Science and Engineering, the State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Poly(ethylene oxide)-grafted poly(N-isopropylacrylamide) networks (PNIPAAm-g-PEO) were prepared
via the reversible addition–fragmentation chain transfer polymerization (RAFT) of N-isopropylacrylamide
with trithiocarbonate-terminated poly(ethylene oxide) and N,N -methylenebisacrylamide as the chain
transfer agent and the crosslinking agent, respectively. It was found that the PNIPAAm-g-PEO copolymer
networks were microphase-separated and that PEO microdomains were dispersed in the PNIPAAm
matrix. The hydrogel behavior of the PNIPAAm-g-PEO networks was investigated using swelling,
deswelling and reswelling tests. The PNIPAAm-g-PEO hydrogels displayed faster responses to external
temperature changes than did the control PNIPAAm hydrogel.
Received 12 September 2011
Received in revised form 5 December 2011
Accepted 6 December 2011
Available online 13 December 2011
0
Keywords:
Poly(N-isopropylacrylamide)
Poly(ethylene oxide)
RAFT polymerization
Behavior of hydrogels
Ó 2011 Elsevier Ltd. All rights reserved.
1
. Introduction
Stimuli-responsive hydrogels have attracted considerable inter-
Recently, an enhancement of temperature-response rate of hydro-
gels has reportedly been achieved through the introduction of
structural inhomogeneity into hydrogel systems [11,29–33]. The
structural inhomogeneity can be either hydrophilic or hydropho-
bic. The basic consideration for the modification of hydrogels with
hydrophilic components is that the hydrophilic components can
act as a tunnel for water molecules across the dense and hydropho-
bic layer at the surface of hydrogels and thus accelerate the diffu-
sion of water [34]. Nonetheless, the incorporation of hydrophobic
blocks (or segments) into PNIPAAm hydrogels has recently been
reported as an alternative method of improving the properties of
hydrogels, although the mechanism of the enhancement of the
thermoresponsive rates has not yet been well elucidated [35–41].
In addition to the incorporation of chemically – different compo-
nents into hydrogels, improvements in the thermoresponsive
properties can be achieved via the formation of comb-type grafted
PNIPAAm networks [30,41–43]. It is proposed that the accelerated
thermoresponsive properties result from the high mobility of the
grafted PNIPAAm chains in the networks [30].
est because of their ability to change volume or shape in response
to environmental alteration. Such environment-responsive proper-
ties have potential applications in the biomedical and biochemical
fields [1]. Poly(N-alkylacrylamide) hydrogels have been among the
most investigated thermoresponsive systems over the past several
decades [1–4]. Among this class of hydrogels, poly(N-isopropylac-
rylamide) (PNIPAAm) hydrogels can exhibit volume-phase transi-
tion (VPT) behavior in response to temperature alterations [4–12].
The VPT properties stem from the lower critical solution tempera-
ture (LCST) behavior of PNIPAAm chains in aqueous solutions
[
13–16]. However, traditional PNIPAAm hydrogels exhibit inherent
drawbacks, such as their low reswelling and deswelling rates that
result from the collective diffusion of water molecules into the
crosslinked networks [17–19]. Therefore, modification of PNIPAAm
hydrogels is necessary to meet the requirements of practical
applications.
It has been proposed that fast deswelling and reswelling can be
achieved by the introduction of porosities into hydrogels [20–28].
A variety of approaches have been employed in the preparation of
the hydrogels to introduce porosity [4]. However, care must be
taken because the mechanical properties of porous hydrogels can
be significantly weaker than those of nonporous hydrogels. There-
fore, measures should be taken to balance the accelerated response
rates with the decreased mechanical strength of porous hydrogels.
PNIPAAm hydrogels modified with poly(ethylene oxide) (PEO)
are among the most investigated systems [30,34,41–43]. Okano
et al. [34] first reported the modification of PNIPAAm hydrogels
through the copolymerization of N-isopropylacrylamide with
a-acryloyl-x-methoxy-poly(ethylene oxide). The PEO-modified
hydrogels displayed fast thermoresponsive properties compared
to control PNIPAAm hydrogels. The hydrophilic component (viz.
PEO) has been proposed to act as a tunnel for water molecules across
the dense and hydrophobic layer at the surface of hydrogels and thus
accelerate the diffusion of water [34]. More recently, PEO has also
been used to access the semi-interpenetrating polymer networks
⇑
Corresponding author. Tel.: +86 21 54743278; fax: +86 21 54741297.
E-mail address: szheng@sjtu.edu.cn (S. Zheng).
1
381-5148/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.reactfunctpolym.2011.12.006

Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
177
(
semi-IPNs) to improve the thermoresponsive properties of the
was added to a solution of potassium hydroxide (5.2 g, 92 mmol)
dissolved in water (50 ml). Carbon disulfide (6.0 ml) was slowly
added to the solution within 30 min with vigorous stirring. The
reaction was conducted for an additional 5 h; the mixture was
heated to 80 °C, and benzyl bromide (7.9 g, 46 mmol) was then
added. The reaction system was maintained at 80 °C for 12 h under
vigorous stirring. Afterwards, the system was cooled to room tem-
perature, and chloroform (30 ml) was added. The reacted mixture
was acidified with hydrochloric acid until the organic layer became
yellow. The aqueous phase was extracted with chloroform
(2 ꢀ 100 ml). The combined organic layers were washed with
10% aqueous solution of sodium carbonate (2 ꢀ 100 ml) and dried
over anhydrous magnesium sulfate. After the solvents were re-
moved via rotary evaporation, the remaining product was recrys-
hydrogels [44]. The PEO chains in the semi-IPNs could be lost after
undergoing several cycles of deswelling and reswelling because they
are physically interpenetrated into PNIPAAm networks. Zeng et al.
[
45] have reported a facile approach to the preparation of fully
interpenetrated polymer networks (full-IPNs) of PNIPAAm using
hepta(3,3,3-trifluropropyl) polyhedral oligomeric silsesquioxane-
capped PEO telechelics. The organic–inorganic PEO telechelics were
physically inter-locked with PNIPAAm networks. In addition to the
above uses of PEO as one of the components of modified PNIPAAm
networks, PEO has also been used as a porogen to obtain hydrogels
with macroporous structures [23,46].
In this work, we report the preparation of the comb-type PEO-
grafted PNIPAAm hydrogels via reversible addition–fragmentation
chain transfer (RAFT) polymerization. Toward this end, a trithiocar-
bonate-terminated PEO was synthesized and used as the macromo-
lecular chain transfer agent (PEO–CTA); the RAFT polymerizations
among NIPAAm, PEO–CTA and N,N-methylenebisacrylamide (BIS)
were then performed to afford the PNIPAAm-g-PEO networks with
tallized from cold dichloromethane to afford a yellow solid
1
(11.30 g) with a yield of 90%. H NMR (CDCl
3
, ppm): 7.33–7.27
), 3.62 (t,
(m, 5H, protons of aromatic ring), 4.61 (s, 2H, CH
2 6 5
C H
2H, CH CH S), 2.84 (t, 2H, CH COOH).
2
2
2
2
,2-azobisisobutyronitrile as the initiator. This approach to prepare
2.3. Synthesis of PEO macromolecular chain transfer agent (PEO–CTA)
PEO-grafted PNIPAAm networks differs significantly from that of
Okano et al. [34], who used a direct copolymerization reaction
between NIPAAm and a-acryloyl-x-methoxy-poly(ethylene oxide).
The prepared BSPA (2.73 g, 10 mmol) was dissolved in 25 ml of
anhydrous dichloromethane, and freshly distilled thionyl chloride
(5 ml) was then slowly dripped into the vigorously stirred mixture
at 0 °C over a period of 2 h. The reaction was conducted for 12 h,
and the unreacted thionyl chloride was removed via distillation
under reduced pressure. The reacted product was slowly dripped
into a flask containing a vigorously stirred mixture of anhydrous
poly(ethylene oxide) monomethyl ether (4.00 g, 2 mmol), triethyl-
amine (1.4 ml, 10 mmol) and dichloromethane (50 ml) within
30 min with vigorous stirring, and the flask was immersed in an
ice-water bath. The reaction was performed at 50 °C for 12 h. After
concentration by rotary evaporation, the reacted mixture was
precipitated into 50 ml of cold diethyl ether. This procedure was
repeated three times to purify the product, and a light-yellow solid
The use of PEO–CTA rather than an acryloyl-terminated PEO macro-
mer facilitates the preparation of PNIPAAm-g-PEO copolymers in
which PEO is randomly distributed into the networks. This approach
also avoids the formation of possible blocky copolymers due to the
difference in reactivity ratios between NIPAAm monomer and acry-
loyl-terminated PEO macromer [47]. In addition, the utilization of a
living radical polymerization approach overcomes some of the
drawbacks of traditional radical random copolymerization in the
formation of PNIPAAm network, such as morphological inhomoge-
neity and a broad distribution of polymer chain lengths between
crosslinking points [11,38,48–51]. To the best of our knowledge,
no previous reports on PEO-grafted PNIPAAm networks via RAFT
polymerization have been published. In this work, the morphology
of the networks was investigated using differential scanning calo-
rimetry (DSC) and small-angle X-ray scattering (SAXS). The thermo-
responsive behavior of hydrogels was addressed based on swelling,
deswelling and reswelling tests.
(3.84 g) was obtained after the product was dried in vacuo at room
1
temperature for 24 h with a yield of 84.6%. H NMR (ppm, CDCl
3
):
7.29 (5H, C
CH –O–CO), 3.50–3.77 (174H, –O–CH
3.36 (3H, CH O–), 2.79 (2H, S–CH –CH
2
6
H
5
–CH
2
–), 4.60 (2H, C
6
H
5
–CH
–CH
–CO).
2
–S–), 4.24 (2H, CH
2
–
2
2
2
–O, S–CH –CH –CO),
2
2
3
2
2
. Experimental
2.4. Preparation of plain PNIPAAm network
2.1. Materials
Typically, NIPAAm (2.00 g) and BIS (27.3 mg, i.e., 1 mol% with
respect to the quantity of NIPAAm) were charged to a glass tube
with a diameter of 10 mm, and anhydrous THF (2.5 ml) was added.
The system was purged with highly pure nitrogen for 30 min, and
Poly(ethylene oxide) monomethyl ether with a quoted molecular
weight of M = 2000 was purchased from Fluka, Germany. Mercap-
n
topropionic acid (97%) was purchased from Alfa Aesar, China. N-
Isopropylacrylamide (NIPAAm) was prepared by following the
method of literature [52]. 2,2-Azobisisobutyronitrile (AIBN) was
used as the initiator and was purchased from Shanghai Reagent, Chi-
na. Prior to use, AIBN was purified by double recrystallization from
2
,2-azobisisobutyronitrile (AIBN) (4.0 mg) was added. The radical
polymerization was initiated and performed at 60 °C for 24 h.
The system gelled as the polymerization proceeded. The gel was
extracted with water for 72 h to remove any unreacted NIPAAm
and then dried in vacuo at 60 °C for 48 h.
0
anhydrous ethanol solution. N,N -methylenebisacrylamide (BIS)
was of chemically pure grade, supplied by Shanghai Reagent, China.
All the solvents were obtained from commercial sources. Prior to
use, tetrahydrofuran and 1,4-dioxane were refluxed over sodium
and then distilled; triethylamine and dichloromethane were re-
2
.5. Synthesis of model PNIPAAm-b-PEO diblock copolymer
The model PEO-b-PNIPAAm diblock copolymer was synthesized
via RAFT polymerization using PEO–CTA as the chain transfer
agent. In a typical procedure, NIPAAm (1.306 g, 12.56 mmol),
PEO–CTA (0.196 g, 0.09 mmol), AIBN (2.4 mg, 0.015 mmol) and
2
fluxed over CaH and then distilled. Unless specially indicated, all
other reagents were used as received.
4
.0 ml of anhydrous 1,4-dioxane were charged to a flask equipped
2
.2. Synthesis of 3-benzylsulfanylthiocarbonylsufanylpropionic acid
with a magnetic stirrer. The trace of oxygen in the mixture was
eliminated via bubbling with highly pure nitrogen for 40 min.
The polymerization was performed at 70 °C for 18 h. The reacted
mixture was dripped into 50 ml of diethyl ether to afford the pre-
cipitate. After drying in a vacuum oven at 30 °C for 12 h, the poly-
mer (1.263 g) was obtained with an NIPAAm conversion of 81.7%.
(
BSPA)
3
-Benzylsulfanylthiocarbonylsufanylpropionic acid (BSPA) was
prepared according to the method reported in the literature
18,53]. In the first step, mercaptopropionic acid (4.0 ml, 46 mmol)
[

1
78
Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
Gel permeation chromatography (GPC): M
= 1.28.
n
= 16,100 with M
w
/
wavelength of k = 1.24 Å. A TH MS600 Linkam hotstage with a pre-
cision of ±0.1 °C was mounted and used for the variable-tempera-
ture experiments. The intensity profiles were output as a plot of
the scattering intensity, I, versus the scattering vector, q, where
M
n
2.6. Preparation of PNIPAAm-g-PEO copolymer networks
q = (4p/k) sin (h/2) (h = scattering angle). To eliminate the effect of
Typically, NIPAAm (1.2000 g, 11.5 mmol), PEO–CTA (0.30 g,
.134 mmol), BIS (17.8 mg, 0.115 mmol), and AIBN (4.9 mg,
.029 mmol) were dissolved in 2.0 ml of anhydrous 1,4-dioxane
PEO crystals on the scattering, the SAXS measurements were
performed 80 °C, which is a temperature greater than the melting
temperature of PEO (i.e., 52 °C).
0
0
and the solution was charged to a glass tube equipped with a mag-
netic stirrer. The reactive system was connected to a standard
Schlenk line and degassed using three freeze–evacuate–thaw cy-
cles. Thereafter, the polymerization was performed in an oil bath
at 70 °C for 48 h. The unreacted NIPAAm monomer were removed
by extraction with deionized water at 20 °C for one week and then
with tetrahydrofuran for 3 d. After the majority of solvent was
evaporated, the copolymer network was dried in vacuo to a con-
stant weight. The conversion of NIPAAm was then calculated. In
this work, the PEO content in the networks was varied from 0 to
2.7.6. Dynamic laser scattering
The solution of polymer (10.0 mg) dissolved in 0.5 ml of tetra-
hydrofuran was charged to 50 ml with ultra-pure water with a
dropping funnel under vigorous stirring. The solutions were dia-
lyzed in ultrapure water for 3 d prior to the measurements. The
laser light scattering experiments were conducted on a Malvern
Nano ZS90 equipped with a He–Ne laser operated at the wave-
length of 633 nm, and the data were collected at a fixed scattering
angle of 90°.
3
0 wt.%.
2
.7.7. Swelling ratios and volume-phase transition behavior
All the specimens for swelling, deswelling and reswelling
2
2
.7. Characterization and measurements
experiments were cut from cylinders of the dried PNIPAAm-g-
PEO networks with a diameter of 10 mm, and the heights of the
specimens were controlled to be ca. 5 mm. Swelling ratios of the
gel samples were gravimetrically measured in the temperature
range from 24 to 48 °C with an interval of 2 °C. All of the specimens
were equilibrated in distilled water for 24 h at each particular tem-
perature. The swelling ratio (SR) at each temperature was defined
as follows:
.7.1. Nuclear magnetic resonance spectroscopy (NMR)
The NMR measurements were conducted on a Varian Mercury
Plus 400 MHz NMR spectrometer at 25 °C using deuterated chloro-
form as the solvent and tetramethylsilane (TMS) as an internal
reference.
2
.7.2. Fourier-transform infrared spectroscopy (FTIR)
FTIR spectroscopy was conducted on a Perkin–Elmer Spectrum
000 Fourier-transform spectrometer at room temperature. The
SR ¼ W
s
=W
d
ð1Þ
1
sample films were prepared by dissolution of the polymers with
ethanol (5 wt.%), and the solutions were cast onto KBr windows.
The residual solvent was removed in a vacuum oven at 60 °C for
where W
s
is the weight of the water absorbed in the swollen hydro-
gels and W is the weight of the dried networks.
d
2
h. The crosslinked samples were granulated, and the powder
2
.7.8. Deswelling kinetics
The deswelling kinetics of the hydrogels was gravimetrically
was mixed with KBr pellets to press into small flakes for measure-
ments. All of the specimens were sufficiently thin to be within a
range where the Beer–Lambert law is obeyed. In all cases, the spec-
measured at 48 °C without free water on the surface. The gel sam-
ples were fully saturated in distilled water at 24 °C for 24 h prior to
the measurements. The weight changes of the gels were recorded
at regular time intervals. Water retention (WR) at regular time
intervals was defined as follows:
ꢁ1
tra were recorded with 64 scans at a resolution of 2 cm
.
2
.7.3. Gel permeation chromatography (GPC)
Molecular weights were measured on a Perkin–Elmer Series
00 gel permeation chromatography system (100 l injection
m PL gel 300 mm ꢀ 7.5 mm mixed B columns) at
0 °C with a refractive index detector. N,N-dimethylformamide
DMF) that contained 0.01 M lithium bromide (LiBr) was used as
2
l
WR ¼ ðW
where W
s
ꢁ W
d
Þ=W
d
ꢀ 100%
ð2Þ
column, 10
7
(
l
t
is the weight of the gel at each particular time. To analyze
the deswelling process quantitatively, a semilogarithmic plot as a
first-order rate analysis was applied to the time dependence of
deswelling as follows:
an eluent at a flow rate of 1.0 ml/min. The molecular weight was
measured relative to polystyrene standards.
ln½ðW
t
ꢁ W
e
Þ=ðW
0
ꢁ W
e
Þꢂ ¼ ꢁkt
ð3Þ
2
.7.4. Differential scanning calorimetry (DSC)
DSC measurements were performed on a TA DSC Q2000 differ-
where W
e
and W
0
are the weights of the hydrogels in the equilib-
rium state under certain conditions and the weights of the hydro-
gels at 24 °C, respectively, k is the rate constant, and t is the
deswelling time.
ential scanning calorimeter under a dry nitrogen atmosphere. Tem-
perature and enthalpy were calibrated with an indium standard.
The quenched samples of approximately 10.0 mg were heated
from ꢁ60 to 160 °C, and a heating rate of 20 °C/min was used in
all cases. The midpoint of the heat capacity change and the maxi-
mum of exothermic transition were defined as the glass-transition
2.7.9. Reswelling kinetics
The reswelling kinetics of the gels equilibrated in distilled water
at 48 °C for 24 h were gravimetrically measured at 24 °C. At regular
time intervals, the weight changes of fully saturated gels were
recorded after water was wiped from the gels using filter paper.
Water uptake (WU) at regular time intervals is defined as follows:
g m
temperature (T ) and the melting temperature (T ), respectively.
2
.7.5. Small angle X-ray scattering (SAXS)
The SAXS measurements were performed on a small-angle X-ray
scattering station (BL16B1) at the Shanghai Synchrotron Radiation
Facility (SSRF), Shanghai, China, which used the third-generation
synchrotron radiation light source. Two-dimensional diffraction
patterns were recorded with an image-intensified CCD detector.
The experiments were performed using X-ray radiation with a
WU ¼ ðW
where W
t
ꢁ W
d
Þ=W
d
ꢀ 100%
ð4Þ
t
is the weight of the gel at each particular time, and the
other symbols are the same as previously defined. To analyze the
reswelling process quantitatively, a semilogarithmic plot as a

Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
179
first-order rate analysis was applied to the time dependence of the
reswelling as follows:
chain, resonance signals at 1.14, 1.64, 2.02 and 4.01 ppm were ob-
served and were assigned to the methyl protons of isopropyl
groups, methylene protons, methine protons of the main chain,
and methine protons of the PNIPAAm main chain, respectively.
The broad resonance signals at 6–7 ppm were assignable to the
ln½1 ꢁ ðW
where W
t
ꢁ W
0
Þ=W
e
ꢂ ¼ ꢁkt
ð5Þ
t
is the weight of hydrogels at each particular time, W
e
and
1
protons of the N–H moiety. The H NMR spectroscopic results indi-
0
W are the weights of the hydrogels in the equilibrium state under
certain conditions and the weights of the hydrogels at 24 °C, respec-
tively, k is the rate constant, and t is the reswelling time.
cate that the product combined the structural features of PEO and
PNIPAAm chains. The PNIPAAm-b-PEO diblock copolymer was sub-
jected to gel permeation chromatography (GPC), and a unimodal
distribution of molecular weight was observed, which suggested
that no unreacted PEO–CTA was detected. The polymer exhibited
3
. Results and discussion
n w
a molecular weight of M = 16,100 and a polydispersity of M /
M = 1.28. The unimodal and fairly narrow distribution of molecu-
n
3.1. Preparation of PNIPAAm-g-PEO copolymer networks
lar weight indicates that the PNIPAAm-b-PEO diblock copolymer
In this work, PNIPAAm-g-PEO copolymer networks were pre-
pared via the reversible addition–fragmentation chain transfer
RAFT) polymerization of N-isopropylacrylamide (NIPAAm) with
was successfully prepared. Based on the molecular weight, the
mass fraction of PEO in the diblock copolymer was calculated to
be 12.4%.
(
The 1H NMR and GPC results indicate that PEO–CTA was an
effective macromolecular chain-transfer agent for the RAFT poly-
merization of NIPAAm, which reminds us of preparing PEO-grafted
PNIPAAm networks via the RAFT polymerization of NIPAAm with
the PEO–CTA as the chain-transfer agent and BIS as the crosslink-
ing agent (see Scheme 1). With this approach, we anticipated that
the PEO chains would participate in the formation of the cross-
linked networks and act as the pendant side chains. As the poly-
merization proceeded, the systems were observed to gradually
gel, which suggested the formation of crosslinked networks. All
of the as-prepared gels were homogenous and transparent, which
implied that no macroscopic phase separation occurred. The cross-
linked networks were extracted with deionized water for 7 d and
with tetrahydrofuran for 3 d, respectively, to remove any unre-
acted monomer. The real compositions of the PNIPAAm-g-PEO
copolymer networks were determined in terms of the conversion
of the monomer and are summarized in Table 1. The PNIPAAm-
g-PEO copolymer networks were obtained with a PEO content of
as much as ca. 30 wt.%. The PNIPAAm-g-PEO copolymer networks
were subjected to Fourier-transform infrared spectroscopy (FTIR);
the FTIR spectra are shown in Fig. 2. Plain PNIPAAm is character-
ized by N–H stretching vibration bands in the range of 3200–
trithiocarbonate-terminated poly(ethylene oxide) monomethyl
0
ether as the chain-transfer agent (i.e., PEO–CTA) and with N,N -
methylenebisacrylamide (BIS) as the crosslinking agent. Toward
this end, we had explored the synthesis of poly(N-isopropylacryl-
amide)-block-poly(ethylene oxide) (PNIPAAm-b-PEO) diblock
copolymer via a similar approach but without the use of BIS. First,
the PEO–CTA was prepared via the reaction between poly(ethylene
oxide) monomethyl ether and 3-benzylsulfanylthiocarbonylsufa-
nylpropionic acid chloride. The 1H NMR spectrum of the product
is shown in Fig. 1. The resonance signals at 4.2, 4.6 and 7.1–
7
.4 ppm are assignable to the methylene and aromatic-ring pro-
tons, respectively, from the trithiocarbonate moiety. The resonance
peaks at 2.8, 3.3 and 3.6 ppm were ascribed to the methyl and
methylene protons, respectively, of PEO. According to the ratio of
the integral intensity of the PEO protons to that of trithiocarbonate
protons, the trithiocarbonate-terminated PEO has been success-
fully synthesized. The trithiocarbonate-terminated PEO was used
as the chain-transfer agent for the RAFT polymerization of NIPAAm
monomer using AIBN as the imitator. The polymerization was per-
formed at 70 °C for 18 h, and the conversion of NIPAAm monomer
was as high as 81.7% under the experimental conditions. The as-
1
prepared polymer was subjected to H NMR spectroscopy and gel
permeation chromatography (GPC). The 1H NMR spectrum is also
shown in Fig. 1. In addition to the signals assignable to the PEO
ꢁ1
ꢁ1
3600 cm as well as by an amide I band at ꢃ1650 cm and an
ꢁ1
amide II band at ꢃ1547 cm . For the PNIPAAm-g-PEO copolymers,
ꢁ1
bands at 1126 cm were discernable, which were assigned to the
stretching vibrations of the ether bond (O–C–O) of PEO. The inten-
sity of the band increased with increasing PEO content. The FTIR
results indicate that the resulting copolymers combined the struc-
tural features from PNIPAAm and PEO.
g
OCH3
j
e
f
h
S
S
O
m
O
c
O
n
S
O
HNm
f+c
k
The PNIPAAm-g-PEO copolymer networks were subjected to dif-
ferential scanning calorimetry (DSC); the DSC curves are presented
in Fig. 3. For comparison, the DSC curve of the model PNIPAAm-b-
PEO diblock copolymer is also included in this figure. The plain PNI-
PAAm network exhibited a glass transition at ca. 142 °C. For the
model diblock copolymer, one exothermic transition appeared
k
l
e+j+h
l
g
m
5
2 °C and one glass transition was detected at 121 °C. The exother-
mic transition was attributed to the melting of the PEO block,
whereas the T at 121 °C was assigned to the PNIPAAm subchain
O
S
b
h
c
O
f
O
OCH3
g
g
S
S
d
e
n
g
of the diblock copolymer. The fact that the melting transition of
the PEO block (i.e., at 52 °C) occurred at a lower temperature than
the glass transition of the PNIPAAm block (i.e., at 120 °C) indicated
that the diblock copolymer was microphase-separated, i.e., the PEO
microdomains were dispersed in the matrix of PNIPAAm in the di-
a
e+f+h
a
b
c
d
block copolymer. Notably, the T
g
of the PNIPAAm block was lower
of PNI-
than the T of plain PNIPAAm (i.e., 142 °C). The depressed T
g
g
8
7
6
5
4
3
2
1
0
PAAm microdomains in the diblock copolymer is responsible for the
partial miscibility of PNIPAAm and PEO [54]. Similar DSC profiles
were displayed for the PNIPAAm-g-PEO copolymer networks. The
remnant exothermic transitions were detected at ca. 50 °C, which
Chemical shift (ppm)
Fig. 1. 1H NMR spectra of trithiocarbonate-terminated PEO and PNIPAAm-b-PEO
diblock copolymer.

1
80
Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
S
O
S
O
SOCl2
S
S
OH
S
S
Cl
S
O
O
O
CH O
OH
3
n
S
S
O
n
OCH
3
PEO-CTA
O
NH
O
AIBN
PEO-CTA
+
+
HN
NH
O
S
S
O
OCH3
O
a
O
b
n
O
O
O
O
S
HN
HN
HN
HN
c
O
d
O
HN
HN
PNIPAAm-g-PEO Networks
Scheme 1. Synthesis of PEO–BSPA and PNIPAAm-g-PEO copolymer networks.
Table 1
Results of RAFT polymerizations for PNIPAAm-g-PEO networks.
Samples
[NIPAAm]:[PEO] (wt)
Wight fraction of hydrogels (wt.%)
PEO (wt.%)
PNIPAAm
100:0
95:5
90:10
85:15
80:20
95.4
94.6
86.7
77.4
68.3
0
5.3
11.5
19.3
29.3
PNIPAAm–PEO5
PNIPAAm–PEO10
PNIPAAm–PEO20
PNIPAAm–PEO30
were assigned to PEO microdomains. The intensity of the melting
peaks for the PEO microdomains was significantly lower than that
in the diblock copolymer, which could result from the restriction
of PEO (i.e., 52 °C). The SAXS profiles of the copolymer networks
that contained 5, 10 and 30 wt.% of PEO are shown in Fig. 4. For
the control PNIPAAm network, no scattering peak was detected,
as expected. However, the virtual increases in scattering intensity
g
of PNIPAAm networks on the PEO chains. The T s of the PNIPAAm
ꢁ1
matrices in the PNIPAAm-g-PEO networks were lower than that
of plain PNIPAAm and decreased with increasing PEO content.
The DSC results suggested that the PNIPAAm-g-PEO copolymer
networks were microphase-separated, i.e., the PEO microdomains
were dispersed in the matrices of crosslinked PNIPAAm networks.
The microphase-separated morphology for the PNIPAAm-g-PEO
copolymer networks is readily confirmed through small-angle
X-ray scattering (SAXS) experiments. To eliminate the effect of
PEO crystals on the scattering, the SAXS measurements were
performed at 80 °C, a temperature higher than the melting point
were displayed in the range of q = 0.1–0.5 nm
for the PNI-
PAAm-g-PEO copolymer networks; the scattering intensity in-
creased with increasing PEO content. The increased scattering
intensity for the PNIPAAm-g-PEO copolymer networks indicates
that the copolymer networks were microphase-separated, i.e.,
PEO microdomains were dispersed in the matrix of PNIPAAm.
The results of the SAXS experiments were in a good agreement
with the DSC results. The fact that no well-defined scattering peaks
were detected for the PEO-grafted PNIPAAm network suggests that
the sizes of the PEO microdomains and the average distance

Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
181
4
10
1
126
3
E
10
D
C
B
PNIPAAm-PEO30
PNIPAAm-PEO10
PNIPAAm-PEO5
PNIPAAm
2
10
A
1
4000
3500
3000
2500
2000
1500
1000
500
10
0
.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
-
1
Wavenumber (cm )
-1
q (nm )
Fig. 2. FTIR spectra of plain PNIPAAm and PNIPAAm-g-PEO networks: (A) PNIPAAm,
B) PNIPAAm–PEO5, (C) PNIPAAm–PEO10, (D) PNIPAAm–PEO20 and (E) PNIPAAm–
PEO30.
Fig. 4. SAXS profiles of PNIPAAm and PNIPAAm-g-PEO networks at 80 °C.
(
3
0 °C), the PNIPAAm-g-PEO hydrogels were homogeneous and
transparent. When heated to higher temperatures (e.g., greater than
0 °C), the hydrogels shrank and became cloudy, which implied the
A
4
occurrence of a volume-phase transition (VPT). After cooling, the
PNIPAAm-g-PEO hydrogels reswelled and again became transpar-
ent. The reversible and temperature-dependent swelling and
deswelling behavior is characteristic of thermo-responsive proper-
ties of PNIPAAm hydrogels. It is of interest to investigate the swell-
ing, deswelling and reswelling kinetics of the PEO-modified
PNIPAAm hydrogels.
B
C
D
E
3
.3. Swelling ratios and volume-phase transition behavior
The swelling behaviors of the PNIPAAm and PNIPAAm-g-PEO
hydrogels were measured. The plots of the stable swelling ratio
as functions of temperature are shown in Fig. 5. In all the cases,
the plots of swelling ratio versus temperature exhibited sigmoid
curves, which is characteristic of the volume-phase transition
behavior of temperature-responsive hydrogels. The PNIPAAm-g-
PEO hydrogels exhibited a volume-phase transition (VPT) at ca.
F
3
2 °C, which was similar to the VPT temperature of plain PNIPAAm
hydrogels. The LCSTs of the PNIPAAm-g-PEO hydrogels were less
affected by the PEO chains that were grafted onto the PNIPAAm
networks. At temperatures below the VPT temperature, the PNI-
PAAm-g-PEO hydrogels exhibited swelling ratios that were signif-
icantly higher than those of the control PNIPAAm hydrogel; the
swelling ratios of the PNIPAAm-g-PEO hydrogels increased with
increasing PEO content. For the PNIPAAm-PEO30 hydrogel, the
swelling ratio was four times greater than that of the control PNI-
PAAm hydrogel. The enhanced swelling ratios for the PNIPAAm–
PEO hydrogels are attributed to the inclusion of the hydrophilic
component (i.e., PEO), which may be important for the application
of thermoresponsive hydrogels in drug delivery. For the control
PNIPAAm hydrogel, the swelling ratio was reduced from 8.8 to
-
40
0
40
80
120
160
o
Temperature ( C)
Fig. 3. DSC curves of PNIPAAm-g-PEO networks and PNIPAAm-b-PEO diblock
copolymer: (A) PNIPAAm, (B) PNIPAAm–PEO5, (C) PNIPAAm–PEO10, (D) PNIPAAm–
PEO20, (E) PNIPAAm–PEO30 and (F) the model PNIPAAm-b-PEO diblock copolymer.
between adjacent microdomains were not uniform because of the
restriction of PNIPAAm networks on the microphase separation of
the PEO chains. The hydrophilic PEO microdomains in the PNI-
PAAm-g-PEO networks are expected to profoundly influence the
properties of the thermoresponsive hydrogels, which can be evalu-
ated in terms of swelling, deswelling and reswelling tests.
1
.9, i.e., 78 wt.% of the absorbed water was released with the occur-
rence of the volume-phase transition in the temperature range of
24–48 °C. For the PNIPAAm–PEO5, PNIPAAm–PEO10, PNIPAAm–
PEO20 and PNIPAAm–PEO30 hydrogels, the amount of absorbed
water that was released was 95% or greater. The increased amount
of water released with the occurrence of VPT relative to that
3
.2. Swelling kinetics of PNIPAAm-g-PEO hydrogels
All the PNIPAAm-g-PEO copolymer networks can be swelled in
water without dissolving. At low temperatures (e.g., less than

1
82
Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
3
2
2
1
1
0
5
0
5
0
5
0
100
PNIPAAm
0
1
2
3
4
PNIPPAm
PNIPAAm-PEO5
PNIPAAm-PEO10
PNIPAAm-PEO20
PNIPAAm-PEO30
PNIPAAm-PEO5
PNIPAAm-PEO10
PNIPPAm-PEO20
PNIPAAm-PEO30
-
-
-
-
80
6
4
2
0
0
0
0
0
10
20
30
40
Time (min)
0
150
300
450
600
750
Time (min)
2
4
28
32
36
40
44
48
o
Fig. 6. Plots of water retention as functions of deswelling time for the control
PNIPAAm and the PNIPAAm-g-PEO hydrogels. The inset shows the plots of
Temperature ( C)
ln[(W
at 48 °C.
t
ꢁ W
e
)/(W
0
e
ꢁ W )] versus time t used to obtain the deswelling rate constants
Fig. 5. Plots of swelling ratios as functions of temperature for control PNIPAAm and
PNIPAAm-g-PEO hydrogels.
could significantly affect the sizes and distribution of hydrophilic
PEO microdomains in the PNIPAAm networks and thus affect the
hydrophobic aggregation of the PNIPAAm backbone.
released from the control hydrogel suggests that the inclusion of a
water-soluble component (i.e., PEO chains) facilitated the release of
the absorbed water.
3.4. Deswelling behavior
3.5. Reswelling behavior
When the hydrogels were heated to temperatures above their
To evaluate the reswelling behavior, all the hydrogels were first
deswelledin water at 48 °C to reach equilibrium. Thereafter, thedes-
welled samples were immersed in water at 25 °C, and the reswelling
ratios were measured gravimetrically at 25 °C. Fig. 7 shows the plots
of water uptake as functions of time for the control PNIPAAm and
PNIPAAm-g-PEO hydrogels. Within the first 160 min, the water
uptake of the control PNIPAAm hydrogel was slightly higher than
that of the PNIPAAm-g-PEO hydrogels. The slightly faster reswelling
behavior for the control PNIPAAm may be associated with the for-
mation of bubbles at the surface due to collective diffusion of water
under the dehydration conditions [29]. These water bubbles blew up
and were finally broken by the increased pressure inside the hydro-
gel. The broken bubbles could constitute the porosity at the surface
of the gel and thus facilitate the occupation of water in the reswell-
ing experiment within this initial stage. However, after the initial
stage, the water uptake of the control hydrogel slowly increased
with increasing reswelling time, which was in marked contrast to
the behaviors of the PNIPAAm-g-PEO hydrogels. It is noted that after
the initial stage the increase in water uptake of the PEO-grafted
PNIPAAm hydrogels was much faster than the control PNIPAAm
hydrogel and significantly exceeded that of the control PNIPAAm
hydrogel. With the identical reswelling times, the PNIPAAm–
PEO10 hydrogel displayed the highest water uptake among all the
PEO-grafted PNIPAAm hydrogels. To analyze the reswelling process
VPTT, thermoresponsive hydrogels shrank and released absorbed
water. The deswelling behavior of the PNIPAAm-g-PEO hydrogels
was evaluated by measurement of the swelling ratios of the hydro-
gels at 48 °C for various times. The deswelling curves are shown in
Fig. 6. The deswelling curves of all the PNIPAAm-g-PEO hydrogels
are below that of the control PNIPAAm hydrogel, which indicates
that, under identical conditions, the water retention of the PNI-
PAAm-g-PEO hydrogels was significantly less than that of the con-
trol PNIPAAm hydrogel, i.e., the PEO-modified hydrogels achieved
stable water retention much faster than the control PNIPAAm
hydrogel. For example, it took c.a. 127 min to attain the stable water
retention of ca. 33 % for the control PNIPAAm hydrogel. In contrast,
the stable water retentions of all the PNIPAAm-g-PEO hydrogels
were less than 25 wt.% within the same time period. The water
retentions of the PNIPAAm-g-PEO hydrogels decreased with
increasing PEO content. To analyze the deswelling process quantita-
tively, ln[(W
t
ꢁ W
e
)/(W
0
ꢁ W
e
)] was plotted as a function of time t
(
see the inset in Fig. 6). The plots were linearly fitted with coeffi-
cients greater than 0.984, which indicated that the deswelling
process can be well described using first-order kinetics (Eq. (3)).
Based on the slopes of these straight lines, deswelling rate constants
ꢁ
3
ꢁ3
ꢁ3
ꢁ3
(
k) of 3.37 ꢀ 10 , 6.01 ꢀ 10 , 6.39 ꢀ 10 and 7.31 ꢀ 10 were
obtained for the control PNIPAAm, PNIPAAm–PEO5, PNIPAAm–
PEO10, PNIPAAm–PEO20 and PNIPAAm–PEO30 hydrogels, respec-
tively. The k values increased with increasing PEO content in the
PEO-grafted hydrogels. Notably, the deswelling kinetics as a func-
tion of PEO content in the present system is slightly different from
that reported by Okano et al. [34], who reported a slower deswelling
rate when the PEO content was ca. 11 wt.% or greater. The difference
in deswelling kinetics may be related to the different molecular
weights of PEO used in these two systems. The lengths of PEO chains
quantitatively, ln[1 ꢁ (W
t
ꢁ W
0 0
)/W ] was plotted as a function of
reswelling time t (see the inset in Fig. 7). Based on the slopes of these
straight lines, the reswelling rate constants (k) were determined to
ꢁ3
ꢁ3
ꢁ3
ꢁ3
be 1.25 ꢀ 10 , 2.52 ꢀ 10 , 3.23 ꢀ 10 , 2.40 ꢀ 10 and 2.87 ꢀ
ꢁ3
10 for the control PNIPAAm, PNIPAAm–PEO5, PNIPAAm–PEO10,
PNIPAAm–PEO20 and PNIPAAm–PEO30 hydrogels, respectively.
The values of the reswelling rate constants for the PNIPAAm-g-PEO
hydrogels were two times greater than that of the control PNIPAAm

Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
183
1
00
0.0
-
-
-
-
-
0.5
1.0
1.5
2.0
2.5
8
6
4
2
0
0
0
0
0
o
4
6 C
o
4
0 C
0
50
100
150
200
250
300
Time (min)
o
3
6 C
o
3
4 C
PNIPPAm
o
3
2 C
PNIPAAm-PEO5
PNIPAAm-PEO10
PNIPAAm-PEO20
PNIPAAm-PEO30
o
25 C
0
1
2
3
0
80
160
240
320
400
480
560
10
10
10
R (nm)
10
Time (min)
h
Fig. 7. Plots of water uptake as functions of reswelling time for the control
PNIPAAm and the PNIPAAm-g-PEO hydrogels. The inset shows the plots of
Fig. 8. Plots of the distribution of hydrodynamic radii for the model PNIPAAm-b-
PEO diblock copolymer in 2% aqueous solution.
ln[1 ꢁ (W
t
0 e
ꢁ W )/W ] versus time t used to obtain the reswelling rate constants at
2
4 °C.
The plots of the hydrodynamic radius of the PNIPAAm-b-PEO
diblock copolymer in the aqueous solution (0.2 g/l) at various tem-
peratures are shown in Fig. 8. At 25 °C (i.e., below the LCST of PNI-
hydrogel. In marked contrast to the case of the control PNIPAAm
hydrogel, the linear fitting of the experimental data yielded correla-
tion coefficients of 0.930 or greater for the PNIPAAm-g-PEO hydro-
gels, indicating that the deswelling process can be described using
a first-order kinetics equation (see the inset of Fig. 7). This observa-
tion suggests that the reswelling mechanism could be changed with
the inclusion of the graft PEO chains into the PNIPAAm networks
PAAm), the scattering of the aqueous solution displayed
unimodal distribution of particles at a hydrodynamic radius of
= 141 nm. This distribution was assignable to the scattering of
a
R
h
the random coils of the PNIPAAm-b-PEO diblock copolymer. When
the temperature was increased to 32 °C or above, the hydrody-
namic radius of the scattering object was significantly diminished
(
see infra).
h
to R = 72 nm. The decreased hydrodynamic radius is attributed to
the occurrence of the coil-to-global transition of the PNIPAAm
block, i.e., the PNIPAAm block was collapsed into the hydrophobic
globules, whereas the PEO block remained hydrated. The dynamic
laser scattering results indicate that the PNIPAAm-b-PEO diblock
copolymer was self-assembled into nano-objects (e.g., micelles)
in water at temperatures greater than the LCST of PNIPAAm. Each
self-assembled nano-object was composed of a hydrophobic core
(viz. a collapsed and hydrophobic PNIPAAm globule) and hydro-
philic coronas (i.e., PEO chains).
The presence of a hydrated PEO corona is important for the sta-
bilization of the self-assembled nanoobjects without precipitation
from water. In the case of the control PNIPAAm hydrogel, the PNI-
PAAm chains in the crosslinked networks were collapsed, and the
dense and hydrophobic skin layer was formed at the surface of
the control dehydrated PNIPAAm hydrogels; thus the permeability
of water at the surface of the hydrogels was decreased. This skin
layer is responsible for the very slow deswelling and reswelling
processes of the bulk polymer network. For the PNIPAAm-g-PEO
hydrogels, the dense and hydrophobic skin layers at the surfaces
of hydrogels were disrupted by hydrophilic PEO microdomains,
as evidenced by the DSC and SAXS results. The hydrophilic PEO
microdomains could behave as tunnels for water molecules that
passed from the interior (and/or surface) to the surface (and/or
interior) of the hydrogels and thus facilitated the diffusion of water
molecules into the crosslinked networks. Therefore, the deswelling
and reswelling processes of the hydrogels were accelerated, as
shown in Figs. 6 and 7. In addition to the effect of the hydrated
component (viz. PEO), the fast deswelling and reswelling behavior
of the PNIPAAm-g-PEO hydrogels is attributed to the comb-like
architecture of the copolymer networks [32,34,54]. It is proposed
3
.6. Interpretation of hydrogel behavior
The absorbed water in PNIPAAm hydrogels can be released upon
heating the hydrogels to temperatures above the volume-phase
transition (VPT) temperature. The VPT behavior stems from the
coil-to-global transition of PNIPAAm chains in the vicinity of the
lower critical solution temperature (LCST) (i.e., 32 °C) in aqueous
solution. When the swollen hydrogels were heated to the VPT tem-
perature, the PNIPAAm chains gradually became insoluble in water
from the surface to the interior of the hydrogels; the surface of the
hydrogels was transformed into a dense hydrophobic layer.
Because of the restriction of collective diffusion of water at the
hydrophobic surfaces, the un-modified PNIPAAm hydrogels would
display a slow deswelling and reswelling behavior. When the PNI-
PAAm-g-PEO copolymer networks were subjected to the swelling
experiments, both the PNIPAAm and the PEO chains were capable
of forming stable hydration interactions with water at tempera-
tures below the VPT temperatures. When the hydrogels were
heated to temperatures above the VPT temperatures, the PNIPAAm
chains in the crosslinked networks underwent the coil-to-global
collapse transition and became insoluble with water, whereas the
PEO chains still remained hydrophilic and hydrated. This behavior
can be accounted for by the thermally induced self-assembly
behavior of the model PNIPAAm-b-PEO diblock copolymer in water.
In this work, the PNIPAAM-b-PEO diblock copolymer with a PEO
weight fraction of 12.4 wt.% was synthesized and was thus taken
as a linear model of the PNIPAAm-g-PEO10 hydrogel.
A dilute aqueous solution of the PNIPAAm-b-PEO diblock
copolymer was subjected to dynamic light scattering experiments.

1
84
Y. Zheng, S. Zheng / Reactive & Functional Polymers 72 (2012) 176–184
that the high mobility of the graft PEO chains in the networks was
favorable for the transport of water in the hydrogels. Similar find-
ings have been reported for comb-like PNIPAAm networks in
which numerous pendent PNIPAAm chains were grafted onto PNI-
PAAm networks [32,54].
[12] T.-A. Asoh, T. Kaneko, M. Matsusaki, M. Akashi, Macromol. Biosci. 6 (2006)
59–965.
9
[
[
13] P.W. Zhu, D.H. Napper, J. Colloid Interf. Sci. 164 (1994) 489–494.
14] P.W. Zhu, D.H. Napper, J. Colloid Interf. Sci. 177 (1996) 343–352.
[15] S. Luo, J. Xu, Z. Zhu, C. Wu, S. Liu, J. Phys. Chem. 110 (2006) 9132–9139.
[
[
[
16] J. Shan, J. Chen, M. Nuopponen, H. Tenhu, Langmuir 20 (2004) 4671–4676.
17] A. Gutowska, Y.H. Bae, J. Feijen, S.W. Kim, J. Control. Release 22 (1992) 95–104.
18] M.H. Stenzel, T.P. Davis, J. Polym. Sci., Part A: Polym. Chem. 40 (2002) 4498–
4
512.
4
. Conclusion
[
[
19] K. Turner, P.W. Zhu, D.H. Napper, Colloid Polym. Sci. 274 (1996) 622–627.
20] X.S. Wu, A.S. Hoffman, P. Yager, J. Polym. Sci., Part A: Polym. Chem. 30 (1992)
2
121–2129.
Poly(N-isopropylacrylamide)-graft-poly(ethylene oxide) (PNI-
[
[
21] R.-X. Zhuo, W. Li, J. Polym. Sci., Part A: Polym. Chem. 41 (2002) 152–159.
22] N. Kato, Y. Sakai, S. Shibata, Macromolecules 36 (2003) 961–963.
[23] X.-Z. Zhang, Y.-Y. Yang, T.-S. Chung, K.-X. Ma, Langmuir 17 (2001) 6094–6099.
[24] T. Serizawa, K. Wakita, M. Akashi, Macromolecules 35 (2002) 10–12.
25] Y. Chen, M. Sone, K. Fuchise, R. Sakai, R. i Kakuchi, Q. Duan, J. Sun, A. Narumi, T.
Satoh, T. Kakuchi, React. Funct. Polym. 69 (2009) 463–469.
26] M. Teodorescu, I. Negru, P.O. Stanescu, C. Draghici, A. Lungu, A. Sarbu, React.
Funct. Polym. 70 (2010) 790–797.
PAAm-g-PEO) copolymer networks were prepared via the revers-
ible addition–fragmentation chain transfer polymerization of
N-isopropylacrylamide with trithiocarbonate-terminated PEO as a
macromolecular chain transfer agent and with N,N-methylenebis-
acrylamide as the crosslinking agent. The PNIPAAm-g-PEO net-
works were characterized using FTIR spectroscopy, DSC and
SAXS. The PNIPAAm-g-PEO copolymer networks were found to
be microphase-separated. The hydrogel behavior of PNIPAAm-
g-PEO networks was investigated using swelling, deswelling and
reswelling tests. With the inclusion of PEO chains, the swelling
ratios of the hydrogels were significantly enhanced. In terms of
the swelling, deswelling and reswelling tests, the PNIPAAm-g-
PEO hydrogels displayed faster responses to external temperature
changes than did the control PNIPAAm hydrogel. Water-soluble
PEO microdomains could act as hydrophilic tunnels that allow
water molecules to pass through; the PEO microdomains thus
accelerate the diffusion of water molecules.
[
[
[
[
27] L. Li, X. Du, J. Deng, W. Yang, React. Funct. Polym. 71 (2011) 972–979.
28] T. Si, Y. Wang, W. Wei, P. Lu, G. Ma, Z. Su, React. Funct. Polym. 71 (2011) 728–
7
35.
[29] Y. Li, T. Tanaka, J. Chem. Phys. 92 (1990) 1365–1371.
[
[
30] H.K. Ju, S.Y. Kim, Y.M. Lee, Polymer 42 (2001) 6851–6857.
31] M. Annaka, T. Matsuura, M. Kasai, T. Nakahira, Y. Hara, T. Okano,
Biomacromolecules 4 (2003) 395–403.
[32] Guiseppi-Elie, N.F. Sheppard Jr., S. Brahim, D. Narinesingh, Biotechnol. Bioeng.
5 (2001) 475–484.
7
[
33] A. Nykaenen, M. Nuopponen, A. Laukkanen, S.-P. Hirvonen, M. Rytelae, O.
Turunen, H. Tenhu, R. Mezzenga, O. Ikkala, J. Ruokolainen, Macromolecules 40
(2007) 5827–5834.
34] Y. Kaneko, S. Nakamura, K. Sakai, T. Aoyagi, A. Kikuchi, Y. Sakurai, T. Okano,
Macromolecules 31 (1998) 6099–6105.
[
[
[
35] X.-Z. Zhang, R.-X. Zhuo, Langmuir 17 (2001) 12–16.
36] K. Haraguchi, R. Farnworth, A. Ohbayashi, T. Takehisa, Macromolecules 36
(
2003) 5732–5741.
Acknowledgments
[
[
37] K. Haraguchi, T. Takehisa, S. Fan, Macromolecules 35 (2002) 10162–10171.
38] H. Hayashi, K. Kono, T. Takagishi, Bioconjugate Chem. 9 (1998) 382–389.
Financial support from the Natural Science Foundation of China
Nos. 20474038, 50873059 and 51133003) and the National Basic
Research Program of China (No. 2009CB930400) are gratefully
acknowledged. The authors thank the Shanghai Synchrotron Radi-
ation Facility for partial support under project Nos. 0sr0260 and
[39] P. Schexnailder, G. Schmidt, Colloid Polym. Sci. 287 (2009) 1–11.
[
[
40] F. Yi, S. Zheng, Polymer 50 (2009) 670–678.
41] K.E. Zeng, Y. Fang, S. Zheng, J. Polym. Sci., Part B: Polym. Phys. 47 (2009) 504–
(
516.
[42] Y. Kaneko, S. Nakamura, K. Sakai, A. Kikuchi, T. Aoyagi, Y. Sakurai, T. Okano,
Polym. Gels Networks 6 (1998) 333–345.
[
43] Y. Kaneko, K. Sakai, A. Kikuchi, R. Yoshida, Y. Sakurai, T. Okano,
Macromolecules 28 (1995) 7717–7723.
1
0sr0126.
[
44] S.J. Kim, C.K. Lee, Y.M. Lee, K.B. Lee, Y.D. Park, S.I. Kim, J. Appl. Polym. Sci. 90
(
2003) 3922–3927.
References
[
[
45] K. Zeng, L. Wang, S. Zheng, J. Phys. Chem. B 113 (2009) 11831–11840.
46] X.Z. Zhang, R.X. Zhuo, Euro. Polym. J. 36 (2000) 2301–2303.
[
[
[
[
[
[
[
[
[
1] E.S. Gil, S.M. Hudson, Prog. Polym. Sci. 29 (2004) 1173–1222.
2] A. Kikuchi, T. Okano, Prog. Polym. Sci. 27 (2002) 1165–1193.
3] T. Hoare, R. Pelton, Macromolecules 40 (2007) 670–678.
4] X.-Z. Zhang, X.-D. Xu, S.-X. Cheng, R.-X. Zhuo, Soft Matter 4 (2008) 385–391.
5] Y. Ding, X. Ye, G. Zhang, Macromolecules 38 (2005) 904–908.
6] X. Yin, A.S. Hoffman, P.S. Stayton, Biomacromolecules 7 (2006) 1381–1385.
7] C. Wu, S. Zhou, Phys. Rev. Lett. 77 (1996) 3053–3055.
8] C. Wu, Macromolecules 30 (1997) 574–576.
[47] G. Odian, Principles of Polymerization, fourth ed., John Wiley and Sons, New
Jersey, 2004.
[48] T. Takigawa, H. Araki, K. Takahashi, T. Masuda, J. Chem. Phys. 113 (2000) 7640–
7645.
[49] P.M. Xulu, G. Filipcsei, M. Zrinyi, Macromolecules 33 (2000) 1716–1719.
[50] Nakamoto, T. Motonaga, M. Shibayama, Macromolecules 34 (2001) 911–917.
[51] K. Haraguchi, T. Takehisa, Adv. Mater. 14 (2002) 1120–1124.
[52] F. Liu, G. Tao, R. Zhuo, Polym. J. 25 (1993) 561–567.
[53] X. Xu, J. Huang, J. Polym. Sci., Part A: Polym. Chem. 44 (2005) 467–476.
[54] K. Van Durme, G. Van Assche, V. Aseyev, J. Raula, H. Tenhu, B. Van Mele,
Macromolecules 40 (2007) 3765–3772.
9] C. Wu, Polymer 39 (1998) 4609–4619.
[
[
10] M. Nakayama, T. Okano, React. Funct. Polym. 71 (2011) 235–244.
11] R. Yoshida, K. Uchida, Y. Kaneko, K. Sakai, A. Kikuchi, Y. Sakurai, T. Okano,
Nature (London) 374 (1995) 240–242.