Comblike thermoresponsive polymers with sharp transitions: Synthesis, characterization, and their use as sensitive colorimetric sensors

Article abstract of DOI:10.1021/ma201874c

The synthesis and thermoresponsive behavior of two structural novel comblike polymers are presented, which are constituted by polymethacrylates main chain with dendritic oligoethylene glycol (OEG) side groups spaced with a linear hydrophobic alkyl [PG1(A)

Full text of DOI:10.1021/ma201874c

ARTICLE
pubs.acs.org/Macromolecules
Comblike Thermoresponsive Polymers with Sharp Transitions:
Synthesis, Characterization, and Their Use as Sensitive Colorimetric
Sensors
Lianxiao Liu, Wen Li,* Kun Liu, Jiatao Yan, Guixia Hu, and Afang Zhang*
Department of Polymer Materials, College of Materials Science and Engineering, Shanghai University, Chengzhong Street 20,
Shanghai 201800, China
S Supporting Information
b
ABSTRACT: The synthesis and thermoresponsive behavior of two
structural novel comblike polymers are presented, which are constituted
by polymethacrylates main chain with dendritic oligoethylene glycol
(OEG) side groups spaced with a linear hydrophobic alkyl [PG1(A)] or
hydrophilic OEG unit [PG1(G)]. The design of this comblike architec-
ture is to retain the unique thermoresponsive behavior of OEG-based
dendritic polymers and, on the other side, to eliminate the tremendous
synthesis effort for the dendronized polymer analogues. Their thermo-
responsive behavior was investigated with UV/vis and temperature-
varied 1H NMR spectroscopy to determine their apparent LCSTs and
follow chain dehydration process, respectively. These polymers show sharp and fast transitions with small hystereses. The phase
transition temperatures are located in between 27 and 34 °C, which is in the vicinity of physiological temperature, and these
transition temperatures are independent of polymer concentration. The thermoresponsiveness of these polymers is also compared
with the corresponding macromonomers as well as the densely packed dendronized polymer analogues reported previously,
focusing on chemical structure and architecture effects. It was found that the more hydrophobic polymer PG1(A) could form denser
aggregates than that of the more hydrophilic polymer PG1(G). On the basis of the exceptional thermoresponsive behavior of these
comblike polymers, this architecture is utilized for fabricating polymer sensors. Random copolymerization of the macromonomers
with the monomer bearing solvatochromic dye moiety (Disperse Red 1) affords the thermoresponsive copolymers which act as
sensitive dual-sensors for both temperature and pH value.
’ INTRODUCTION
candidates in many applications, such as drug delivery,8 smart
surface,9 and water dispersants.10
Polyethylene glycols (PEGs) and their analogues oligoethy-
lene glycols (OEGs) have been well investigated for various
material applications due to their unique properties, including
biocompatibility, low toxicity, and good water solubility.1 They
were incorporated into different part of polymers to enhance
the functionality.2 Especially, polymers carrying short OEG
side chains endeavor phase transitions when heating their
aqueous solutions to their lower critical solution temperatures
(LCSTs), which makes them one of interesting thermorespon-
sive polymers.3 In contrast to the intriguing thermoresponsive
polymer poly(N-isopropylacrylamide) (PNiPAM), which
shows sharp phase transition but exhibits unavoidable large
hysteresis upon cooling due to the strong intermolecular
hydrogen bonding,4 OEG-based thermoresponsive polymers
are superior in several aspects: (1) they are nontoxic; (2)
thermally induced phase transitions normally exhibit small
hysteresis; (3) the LCST of these polymers can be tuned easily
by various routes, including modification of the terminal groups,5
the length of the OEG chains,6 or the copolymer compositions.7
Thus, OEG-based thermoresponsive polymers attract more and
more attention recently, and they are considered to be good
One of the most active research directions recently is to
construct OEG-based thermoresponsive polymers of different
architectures. Comblike homo- and copolymers with polymetha-
crylate as the backbone and short OEG as the side chains were
synthesized by Ishizone et al. first via anionic polymerization6a or
later by Lutz et al. via atom transfer radical polymerization
(ATRP) strategy.7b,11 Zhao and co-workers reported the synth-
esis of polyacrylates and polystyrenes with OEG pendent groups
through nitroxide-mediated radical polymerization (NMRP).6b,12
These work discovered that polymer composition and molecular
weight, terminal groups of the side chains, and polymer end-
groups all show influence on the polymers’ LCSTs. Because of the
simple structure of these polymers, the architecture effects13 on
their LCSTs were not brought into much attention. Recently,
hyperbranched copolymers from OEG-based polymethacrylates
were synthesized by Davis and co-workers.14 They pointed out
Received: August 17, 2011
Revised:
September 29, 2011
Published: October 10, 2011
r
2011 American Chemical Society
8614
dx.doi.org/10.1021/ma201874c Macromolecules 2011, 44, 8614–8621
|

Macromolecules
ARTICLE
Figure 1. Chemical structures of the ethoxyl-terminated comblike homopolymers PG1(A) and PG1(G) and copolymers PG1(A)-co-DR1 and
PG1(G)-co-DR1 from present work as well as the ethoxyl-terminated dendronized polymethacrylates PG2(A) and PG2(G) reported in previous
work.5b,16a Nomenclature: A and G represent interior with alkyl (A) and oligoethylene glycol (G) linkage, respectively; PG1 and PG2 represent the
polymers carrying the first (G1) and second generation (G2) of dendrons, respectively.
the polymer architecture shows big influence on their phase
transitions, as the LCSTs of the hyperbranched polymers are
normally 5À10 deg lower than that of the linear counterparts.
Alternatively, we reported recently a series of thermoresponsive
polymers with OEG linkages based on dendronized polymer
concept (from PG1 to PG3).5b,15 These polymers combine the
structural characteristics from dendronized polymers, including
the bulkiness, cylindrical molecular shape, and high rigidity with a
more or less stretched backbone in the interior, together with the
unique properties of OEG units. They were prepared via con-
ventional radical polymerization in bulk and could be achieved
with high molar masses. These polymers show unprecedented
thermoresponsive properties with sharp phase transitions in both
heating and cooling processes as well as small hysteresis. Their
phase transition temperatures can be tuned in the range of
32À65 °C, depending mostly on the peripheral units (either
methoxyl or ethoxyl). Furthermore, the dendritic architecture
takes great effects on their thermoresponsiveness and features
these polymers with heterogeneous dehydration of the peripheral
units at a lower temperature than that of the interior parts.16 This
behavior affords a possibility to fabricate a molecular nanocapsule.
But the tedious multiple step synthesis remains the major
disadvantage to these polymers, which limit their widely applica-
tions to different research areas.
polymers: below LCST, the polymers are water-soluble and polar,
while upon heated to elevated temperature, the polymers are
dehydrated and become hydrophobic and, in most cases, forming
aggregates by entropy-driving collapse. By incorporating solvato-
chromic dyes18 into the polymers covalently, temperature-induced
minor environmental polarity changes can initiate the specific
change of optical or emissive properties of the dyes. Thus,
sharpness and reversibility of the phase transitions of the thermo-
responsive polymers make them extremely suitable for loading the
solvatochromic dyes to have a reliable and accurate sensing. By
choosing different chromophores, polymer-based sensors can be
classified into fluorescent and visible sensors. The majority of such
polymer sensors are based on PNiPAM19 and, more recently, on
OEG-based polymers.20 Up to now, it still remain a challenge to
develop a better thermoresponsive polymer for loading sensor that
its LCST shows minor or negligible influence after loading dye, and
it can efficiently prevent fluorescent dye from self-quenching.
In present report, we present the synthesis of OEG-based
polymers PG1(A) and PG1(G) with comblike architecture
which contain first generation (G1) dendrons spaced with a
linear hydrophobic alkyl or hydrophilic OEG units in the side
chain (Figure 1). Their structures and compositions are similar to
the second generation (G2) dendronized polymer analogues
PG2(A) and PG2(G), except the lower density of the OEG
dendritic side units. These polymers show similar thermore-
sponsive behaviors as their dendronized polymer counterparts,
but their synthesis is greatly simplified. Their thermorespon-
siveness was thus investigated with UV/vis and temperature-
In the past two decades, it has seen renaissance of interest in
developing polymer sensors based on thermoresponsive poly-
mers.17 The principle is to make use of the microenvironmental
changes from the phase transitions of the thermoresponsive
8615
dx.doi.org/10.1021/ma201874c |Macromolecules 2011, 44, 8614–8621

Macromolecules
ARTICLE
varied proton NMR spectroscopy and compared with the
macromonomers and dendronized counterparts, and the struc-
ture and architecture effects were examined in details. Based on
the characteristic thermoresponsive behavior of these comb-
like polymers, copolymers PG1(A)-co-DR1 and PG1(G)-co-
DR1 containing 5 mol % of the solvatochromic dye (Disperse
Red 1, DR1) were synthesized to fabricate novel dual sensors
which can show sensitive response to both temperature and
pH value.
Polymerization of MG1(A) in Bulk. Monomer MG1(A) (0.50 g, 0.60
mmol) and AIBN (2.50 mg) were added into a Schlenk tube. The
mixture was thoroughly deoxygenated by several freezeÀpumpÀthaw
cycles and then stirred at 65 °C for 4 h. After cooling to rt, the polymer
was dissolved in DCM and purified by silica gel column chromatography
with DCM as an eluent. PG1(A) was yielded (0.35 g, 70%) as a highly
viscous oil. 1H NMR (CDCl3): δ = 0.80À1.10 (m, 3H, CH3),
1.19À1.23 (m, 9H, CH3), 1.27À1.34 (m, 10H, CH2), 1.62 (br, 4H,
CH2), 3.45À3.74 (m, 32H, CH2), 3.79À3.98 (m, 8H, CH2), 4.11À4.15
(m, 6H, CH2), 4.37 (br, 2H, CH2), 6.56 (br, 2H, CH).
Polymerization of MG1(G) in Bulk. Monomer MG1(G) (0.30 g, 0.36
mmol) and AIBN (1.50 mg) were added into a Schlenk tube. The
mixture was thoroughly deoxygenated by several freezeÀpumpÀthaw
cycles and then stirred at 65 °C for 3 h. After cooling to rt, the polymer
was dissolved in DCM and purified by silica gel column chromatography
with DCM as an eluent. PG1(G) was yielded (0.18 g, 60%) as a highly
viscous oil. 1H NMR (CDCl3): δ = 0.85À1.08 (m, 3H, CH3), 1.18À1.22
(m, 9H, CH3), 1.27À1.33 (m, 2H, CH2), 3.49À3.74 (m, 40H, CH2),
3.78À3.85 (m, 6H, CH2), 4.10À4.14 (m, 8H, CH2), 4.43 (br, 2H, CH2),
6.56 (br, 2H, CH).
Synthesis of Copolymer PG1(A)-co-DR1. Monomer MG1(A) (0.60 g,
0.72 mmol), MDR1 (14.40 mg, 0.04 mmol), and AIBN (3 mg) were
dissolved in dry DMF (0.10 mL) inside a Schlenk tube. The solution was
thoroughly deoxygenated by several freezeÀpumpÀthaw cycles and
then stirred at 65 °C for 8 h. After cooling to rt, the polymer was
dissolved in little DCM and purified by silica gel column chromatogra-
phy with DCM/hexane (1:3, v/v) as an eluent. PG1(A)-co-DR1 was
yielded (0.33 g, 46%) as a highly viscous red color oil. 1H NMR
(CDCl3): δ = 0.86À1.09 (m, CH3), 1.18À1.22 (m, CH3), 1.27À1.34
(m, CH2), 1.62 (br, CH2), 3.33À3.92 (m, CH2), 4.11À4.14 (m, CH2),
4.36 (br, CH2), 6.55 (br, CH), 6.87 (br, CH), 7.92 (br, CH), 8.30
(br, CH).
Synthesis of Copolymer PG1(G)-co-DR1. Monomer MG1(G) (0.60
g, 0.72 mmol), MDR1 (14.40 mg, 0.04 mmol), and AIBN (3 mg) were
dissolved in dry DMF (0.1 mL) inside a Schlenk tube. The solution was
thoroughly deoxygenated by several freezeÀpumpÀthaw cycles and
then stirred at 65 °C for 6 h. After cooling to rt, the polymer was
dissolved in little DCM and purified by silica gel column chromatogra-
phy with DCM/hexane (1:3, v/v) as an eluent. PG1(G)-co-DR1 was
yielded (0.49 g, 82%) as a highly viscous red color oil. 1H NMR
(CDCl3): δ = 0.86À1.09 (m, CH3), 1.18À1.22 (m, CH3), 1.27À1.34
(m, CH2), 3.49À3.85 (m, CH2), 4.10À4.12 (m, CH2), 4.43 (br, CH2),
6.55 (br, CH), 6.87 (br, CH), 7.92 (br, CH), 8.30 (br, CH).
’ EXPERIMENTAL SECTION
Materials. Compounds 1 and 2 were synthesized according to the
previous report.15b Dichloromethane (DCM) was distilled from CaH2
for drying. Azobis(isobutyronitrile) (AIBN) was recrystallized twice
from methanol. 0.1 M HCl was used for making the pH 1 solution. Other
reagents and solvents were purchased at reagent grade and used without
further purification. All reactions were run under a nitrogen atmosphere.
Macherey-Nagel precoated TLC plates (silica gel 60 G/UV254, 0.25
mm) were used for thin-layer chromatography (TLC) analysis. Silica gel
60 M (Macherey-Nagel, 0.04À0.063 mm, 200À300 mesh) was used as
the stationary phase for column chromatography.
1
Instrumentation and Measurements. H NMR spectra were
recorded on a Bruker AV 500 (500 MHz) spectrometer. Gel permeation
chromatography (GPC) measurements were carried out on a Waters
GPC e2695 instrument with 3 column set (Styragel HR3 + HR4 + HR5)
equipped with refractive index detector (Waters 2414), and DMF
(containing 1 g LÀ1 LiBr) as eluent at 45 °C. The calibration was
performed with poly(methyl methacrylate) standards in the range ofMp =
2540À936 000 (Polymer Standards Service-USA, Inc.). UV/vis turbidity
measurements were carried out on a PE UV/vis spectrophotometer
(Lambda 35) equipped with a thermostatically regulated bath. Aqueous
polymer solutions were placed in the spectrophotometer (path length
1 cm) and heated or cooled at a rate of 0.2 °C minÀ1. The absorptions of
the solution at λ = 500 nm were recorded every 5 s. The LCST is
determined the one at which the transmittance at λ = 500 nm had reached
50% of its initial value.
3,4,5-Tris{2-[2-(2-ethoxyethoxy)ethoxy]ethoxy}benzyloxyoctyl
Methacrylate [MG1(A)]. MAC (0.41 g, 3.92 mmol) was added dropwise
to a mixture of 1 (2.00 g, 2.61 mmol), TEA (1.32 g, 13.05 mmol), and
DMAP (0.20 g) in dry DCM (40 mL) at 0 °C over 5 min. The mixture
was stirred for 4 h at rt and then quenched with MeOH. After washing
successively with aqueous NaHCO3 solution and brine, the organic
phase was dried over MgSO4. Purification by column chromatography
with DCM/MeOH (30:1, v/v) afforded MG1(A) (2.20 g, 95%) as
colorless oil. 1H NMR (CDCl3): δ = 1.17À1.20 (m, 9H, CH3),
1.32À1.37 (m, 8H, CH2), 1.56À1.68 (m, 4H, CH2), 1.92À1.93
(m, 3H, CH3), 3.40À3.43 (m, 2H, CH2), 3.48À3.73 (m, 30H, CH2),
3.77 (t, 2H, CH2), 3.83 (t, 4H, CH2), 4.10À4.15 (m, 8H, CH2), 4.36
’ RESULTS AND DISCUSSION
Synthesis. The ethoxyl-terminated dendritic structure is
selected as the corresponding dendritic polymers show LCSTs
in the vicinity of physiological temperature. The macromonomer
route is applied here for the synthesis of these polymers in order
to achieve well-defined structures. Starting from two ethoxyl-
terminated dendron alcohols 1 and 2 with alkyl and OEG tails,
respectively,15b the reaction with MAC in the presence of TEA as
base and DMAP as catalyst afforded directly their corresponding
macromonomers MG1(A) or MG1(G) in high yields (92 or
95%) and on gram scales. Both macromonomers are colorless
viscous liquid at room temperature; thus, their free radical
polymerization was able to be carried out in bulk with AIBN as
the initiator. The polymerization media became solidified in less
than 30 min for both cases, indicating the fast polymerization
kinetics. The polymerizations were then stopped after 3À4 h to
avoid the possible gelation. After purification by column chro-
matography with DCM as eluent, PG1(A) and PG1(G) were
a
b
(s, 2H, CH2), 5.52À5.53 (m, 1H, CH2 ), 6.07À6.08 (m, 1H, CH2 ),
6.55 (s, 2H, CH).
2-{2-[2-(3,4,5-Tris{2-[2-(2-ethoxyethoxy)ethoxy]ethoxy}benzyloxy)-
ethoxy]ethoxy}ethyl Methacrylate [MG1(G)]. MAC(0.48 g, 4.55 mmol)
was added dropwise to a mixture of 2 (2.33 g, 3.03 mmol), TEA (1.53 g,
15.15 mmol), and DMAP (0.20 g) in dry DCM (40 mL) at 0 °C over
5 min. The mixture was stirred for 4 h at rt and then quenched with MeOH.
After washing successively withaqueous NaHCO3 solutionand brine, the
organic phase was dried over MgSO4. Purification by column chroma-
tography with DCM/MeOH (30:1, v/v) afforded MG1(G) (2.30 g,
92%) as colorless oil. 1H NMR (CDCl3): δ = 1.20À1.23 (m, 9H, CH3),
1.95À1.96 (m, 3H, CH3), 3.51À3.56 (m, 6H, CH2), 3.59À3.77 (m,
34H, CH2), 3.80 (t, 2H, CH2), 3.86 (t, 4H, CH2), 4.12À4.17 (m, 6H,
CH2), 4.30À4.32 (m, 2H, CH2), 4.46 (s, 2H, CH2), 5.57À5.58 (m, 1H,
a
b
CH2 ), 6.13À6.14 (m, 1H, CH2 ), 6.58 (s, 2H, CH).
8616
dx.doi.org/10.1021/ma201874c |Macromolecules 2011, 44, 8614–8621

Macromolecules
ARTICLE
Table 1. Conditions for and Results from the Polymerization of the Macromonomers MG1(A) and MG1(G) as Well as the
Copolymerization of MG1(A) and MG1(G) with MDR1
GPC results
a,b
entries
PG1(A)
monomers
MG1(A)
time (h)
yield (%)
Mn  10À5
DPn
PDI
LCST (°C)c
4
3
8
6
70
60
46
82
1.7
3.0
1.1
2.0
204
2.0
2.2
3.2
2.3
27.4
33.7
25.5
32.8
PG1(G)
MG1(G)
358
PG1(A)-co-DR1
PG1(G)-co-DR1
MG1(A)/MDR1d
MG1(G)/MDR1d
133/7
240/12
a DPn = apparent number-average degree of polymerization, which is calculated from the relative molar masses Mn (and the ratio of OEG unit and DR1
for the copolymers). b The copolymer composition ratios were analyzed with 1H NMR spectroscopy (Figures S9 and S10) by counting the proton
integrations (I) at 6.55 ppm (signal a) from OEG dendrons and at 7.92 ppm (signal c) from DR1 moieties, and calculated according to [OEG
unit]:[DR1] = [I6.55]/2:[I7.92]/4. c The apparent LCST of the polymers were determined as the temperature at 50% of the initial transmittance. d The
feed molar ratio of the macromonomers to MDR1 is 95:5.
Scheme 1. Synthesis of the Macromonomersa
a Reagents and conditions: (a) MAC, DMAP, TEA, DCM, À5 to 25 °C, 4 h [92%, 95%, and 65% for MG1(A), MG1(G), and MDR1, respectively].
DCM = dichloromethane, DMAP = N,N-dimethylaminopyridine, MAC = methacryloyl chloride, TEA = triethylamine.
achieved in yields of 60À70%. All the macromonomers and
cooling processes, resembling to the dendronized counterparts
PG2(A) and PG2(G), which should come from the simultaneous
dehydration of dense OEG branching units.22 (2) The apparent
LCSTs of these comblike polymers are dependent slightly on
their interior structures: the LCST of PG1(A) with a hydro-
phobic tail is 27.4 °C, about 6 deg lower than that of PG1(G)
which possesses a hydrophilic tail. The low density of grafted
dendrons here affords the side units high mobility which facil-
itates interior part evolved in the dehydration. (3) Although all
these polymers possess the same dendron in the periphery as
their dendronized counterparts, their apparent LCSTs are 2À
4 deg lower than these of the latter. This can be understood that
back-folding of peripheral ethoxyl units for the case of dendro-
nized polymers PG2(A) and PG2(G) enhanced by the high
dense peripheral dendritic units23 leads to the more hydrophilic
OEG units exposed onto the molecular surface, thus causing the
increase of LCST. (4) The apparent LCSTs of macromonomers
are quite different from those of their corresponding comblike
polymers. MG1(A) shows a LCST at 15.5 °C, which is 12 deg
1
polymers were characterized by H NMR spectroscopy (see
Supporting Information for the spectra). The molar masses of
the polymers were determined by GPC with DMF as eluent, and
the results are summarized in Table 1. Both PG1(A) and
PG1(G) with high molar masses were obtained.
Thermoresponsive Behavior. Both PG1(A) and PG1(G)
show thermoresponsive behavior: their aqueous solutions are
transparent below 25 °C but turn into turbid when heated to
elevated temperatures. Interestingly, the corresponding macro-
monomers MG1(G) and MG1(A) also exhibit thermorespon-
siveness. UV/vis spectroscopy was thus applied to investigate the
thermoresponsive behavior of all the monomers and polymers
and to determine their apparent LCSTs21 and examine the
architecture effects. The turbidity curves measured at 500 nm
were plotted in Figure 2a, which leads to the following con-
clusions: (1) comblike polymers PG1(A) and PG1(G) show
unique thermoresponsive behavior with sharp phase transitions
(Δ < 1.2°) and small hysteresis (Δ < 1°) during heating and
8617
dx.doi.org/10.1021/ma201874c |Macromolecules 2011, 44, 8614–8621

Macromolecules
ARTICLE
Figure 2. (a) Plots of transmittance vs temperature for 0.25 wt % aqueous solutions of MG1(A), MG1(G), PG1(A), and PG1(G). Heating and cooling
rate = 0.2 °C/min. (b) Dependence of LCST on the concentration of MG1(A), PG1(A), and PG1(G). For turbidity curves, see Figures S1 and S2a in the
Supporting Information.
Figure 3. Temperature-dependent 1H NMR spectra of PG1(A) (LCST = 27.2 °C) (a, 1.0 wt %) and PG1(G) (LCST = 33.4 °C) (b, 1.0 wt %) in D2O.
lower than its corresponding polymer PG1(A), while the LCST
of MG1(G) is above 80 °C, which is about 46 deg higher than
that of its corresponding polymer PG1(G). Furthermore,
MG1(A) exhibits sharp phase transitions; in contrast, the phase
transitions of MG1(G) are less sharp and show a larger hyster-
esis. Comparing the macromonomer to its corresponding poly-
mer, their overall hydrophilicities are nearly the same. By
polymerization, the OEG dendrons are aligned alone the poly-
mer main chain, which shades the linkage [either alkyl tail inside
PG1(A) or OEG tail inside PG1(G)] to a certain degree and thus
leads the polymer’s thermoresponsiveness dependent more on
the peripheral dendrons. Therefore, the LCSTs of PG1(A) and
PG1(G) are similar as they possess the same periphery dendron.
In sharp contrast, the tails in the macromonomers MG1(A) and
MG1(G) are exposed equally outside, which will join the
dehydration altogether; therefore, MG1(A) with a hydrophobic
tail starts dehydration at much lower temperature than its
polymer, while MG1(G) with a hydrophilic tail starts dehydration
at much higher temperature than its polymer. So, the architecture
shows significant influence on their thermoresponsiveness.
The influence of the concentration (from 0.1% to 2%) of
PG1(A) and PG1(G) on their LCSTs was also studied, and the
results are plotted in Figure 2b. Overall, the LCST of the polymer
decrease slightly with the increase of concentration, though the
decrease is less than 1 deg when polymer concentration increased
from 0.1 to 2%, proving the polymer concentration shows
negligible influence on their LCST. For comparison, the con-
centration influence on the LCST of the macromonomer MG1-
(A) was also investigated. Though the molecule is small, its LCST
show only minor concentration dependence (from 0.1 to 2%).
This is quite different from the OEG-based dendrimers, where
LCST shows obvious concentration dependence.24 The most
unusual phenomenon for MG1(A) is that at a diluted concentra-
tion (from 0.1 to 0.25%) the LCST increases first from 14.8 to
15.5 °C, but with concentration from 0.5% to 2% the LCST
decreases. For the phase transitions curves, see Figure S2a. All
these suggest the macromonomer and the polymer adopt a
different thermoresponsive mechanism:thedehydration ofneigh-
boring OEG units along main chain for the polymers facilitates
the aggregation, while the dehydrated macromonomers need to
diffuse to meet each other in order to form aggregates.
1H NMR spectroscopy was used to follow the dehydration
process of the polymers, and the spectra from 24 to 35 °C for
PG1(A) and from 25 to 40 °C for PG1(G) were recorded
(Figure 3). The proton signals of OEG moieties, which appears
at δ = 3.2À4.5, are resolved at lower temperatures but broaden
increasingly with the increase of solution temperature, simulta-
neously their intensities decrease, indicating the dehydration of
8618
dx.doi.org/10.1021/ma201874c |Macromolecules 2011, 44, 8614–8621

Macromolecules
ARTICLE
Coincidently, the molar ratios of the dye moiety within the
copolymers are nearly the same as these fed in the monomer
mixtures. These copolymers also show thermoresponsive beha-
vior as their homopolymer counterparts, but their LCSTs are
slightly lower (Δ = 1°À2°) due to the introduction of the
hydrophobic dye DR1 moieties (see Figure S2b for the turbidity
curves). It is interesting to point out that incorporation of the dye
moieties inside the comblike copolymers only shows minor
influence on their LCSTs, and this is in sharp contrast to the
conventional OEG-based polymers where LCST decreased more
than 10 deg after introduction of the same dye at 5 mol %.20a
Dissolving the copolymers in water at pH 1 or 7 at room
temperature yielded the solutions with red color (Figure 4).
Visually, all copolymer solution color at both pH conditions is
quite different from the purple color of aqueous solution from
free DR1 (at pH = 1), indicating the obvious interaction of DR1
moieties with the OEG polymer. The red color from PG1(A)-co-
DR1 solutions is less intensive than that from PG1(G)-co-DR1,
which suggests the more hydrophobic environment enhanced
the interaction. The pH value shows weak influence on the
solution color below the LCSTs since the color at pH 1 is nearly
the same as that at pH 7 for both polymers. Upon increasing
solution temperature above their LCSTs, different solvatochro-
mic shift happened for copolymer solutions at different pH
values. The turbid solutions all changed their color, which is a
direct evidence for the sensing ability of these copolymers. It is
necessary to mention that light scattering of the colloidal above
LCST may also contribute partially to the color difference of the
solutions below and above the LCST.
Figure 4. Photographs of the copolymer solutions below and above the
phase transition temperature at pH 1 and 7. Copolymer concentrations:
0.60 mg mLÀ1
.
OEG units. The temperatures to initiate the chain dehydration
for PG1(A) and PG1(G) are 26.5 and 33 °C, respectively, which
are slightly lower than their corresponding LCST. This is
understandable as the former is an indication for the dehydration
of OEG units, while the latter is an indication for aggregation of
the dehydrated or collapsed OEG units. Comparison of the
NMR spectra from both PG1(A) and PG1(G) leads to another
interesting point. The spectra from PG1(A) change dramatically
when solution temperature increased above the dehydration
temperature, and the signals from aromatic ring (δ = 6.5) and
alkyl tail (δ = 1.1À1.6) almost disappeared. But the spectra from
PG1(G) are always much resolved than these from PG1(A) at
the same temperature. Even at elevated temperature, for example
40 °C, which is much higher than the initial dehydration
temperature, the spectra from PG1(G) are still less broader,
and the signal from aromatic ring (δ = 6.5) can still be visualized.
Since the signals’ sharpness and intensities are indication of the
proton mobility under the circumstance, we thus can prelimina-
rily conclude that the aggregates formed from PG1(A) are denser
than those from PG1(G).
Synthesis of Dual Sensors for Temperature and pH Value.
The unique thermoresponsive behaviors of these comblike
polymers make them attractive candidates for fabricating various
responsive materials. It is known that Disperse Red 1 (DR1) is a
solvatochromic dye which can change its color with the solvent
polarity.25 We thus consider covalently incorporating this dye
into the comblike polymers may fabricate sensitive colorimetric
sensors. Furthermore, the sizable dimension of these dendritic
polymers may afford the polymer sensors special characteristics.
Herein, DR1 with polymerizable unit (MDR1) (Figure 1) was
prepared from compound 3 and was radically copolymerized
with MG1(A) and MG1(G) to afford random copolymers
PG1(A)-co-DR1 and PG1(G)-co-DR1, respectively. The feed
molar ratio of the macromonomer to MDR1 is 95:5. The
copolymer compositions were analyzed with 1H NMR spectros-
copy (Figures S9 and S10) by counting the proton integrations at
6.55 ppm (signal a) from OEG dendrons and at 7.92 ppm (signal c)
from DR1 moieties, and the results are summarized in Table 1.
The visual inspection is further confirmed by UV/vis spec-
troscopy and investigated in more detail. The temperature-
dependent UV/vis spectra of these copolymers at different pH
values were recorded (Figures S3 and S4), and typical spectra
below and above their LCSTs are shown in Figure 5a. The plots
of the wavelength with maximum absorbance (λmax) versus
temperature are shown in Figure 5b. Below LCST (20 °C), the
aqueous solutions of copolymers PG1(G)-co-DR1 at either pH 1
or pH 7 show λmax at about 500 nm, which is blue shift when
comparing to that from free DR1 in water at pH 1 (λmax
=
515 nm).20a For the more hydrophobic copolymer PG1(A)-co-
DR1 the blue shift is even enhanced, and the aqueous solutions at
either pH 1 or pH 7 show λmax at about 488 nm. Interestingly, the
solution pH show negligible influence on the blue shift below
LCST since the aqueous solutions for either PG1(G)-co-DR1 or
PG1(A)-co-DR1 give nearly the same λmax at both pH 1 and pH
7, similar to previous report.20a These results indicate the
interaction of the OEG polymer with DR1 moieties dominates
the state of dye in the solution at low temperature, and this
interaction enhances with the increase of polymer hydrophobi-
city. With increase of solution temperature to above their LCSTs,
the λmax for the copolymer solutions at both pH 1 and 7 all shows
obvious shifts. At solution pH 1, the λmax for both copolymer
PG1(G)-co-DR1 (at 40 °C) and PG1(A)-co-DR1 (at 32 °C) red
shifts to 526 nm. While at solution pH 7, the λmax for copolymer
PG1(G)-co-DR1 and PG1(A)-co-DR1 blue shifts to 483 and
478 nm, respectively. The more hydrophobic copolymer PG1-
(A)-co-DR1 always shifts more strongly than the more hydro-
philic one at both pH conditions. The results here are partially
different from the report from Hoogenboom et al.,20a where no
λmax shift happened for the OEG-based polymer sensors when
solution pH is 7. On the basis of above observation, we can
conclude that either polymer dehydration or acidic condition
8619
dx.doi.org/10.1021/ma201874c |Macromolecules 2011, 44, 8614–8621

Macromolecules
ARTICLE
Figure 5. (a) UV/vis spectra of PG1(A)-co-DR1 and PG1(G)-co-DR1 at solution pH 1 and 7 below and above their LCSTs. (b) Plots of λmax vs
temperature for copolymer sensor solutions at solution pH 1 and 7. For more temperature varied UV spectra, see Figures S3 and S4 in the Supporting
Information. Copolymer concentrations: 0.20 mg mLÀ1
.
cannot lead to λmax red shift, and this can happen only in polymer
dehydrated state at acidic solutions. Since the red shift of λmax is
an indication of the protonation of the dye moieties, it is clear
that DR1 moieties can only be protonated at β-nitrogen atom in
the dehydrated polymer media, and the more hydrophobic
polymer facilitates the protonation at a lower temperature. In
contrast, at pH 7 where no driving force for the protonation of
β-nitrogen exists due to the pKa of DR1,20a dehydration of
copolymers leads to the formation of more hydrophobic phase,
incorporating 5 mol % of hydrophobic dye unit shows a minor
influence on the phase transition temperature of the polymers.
These polymer sensors are responsive to both pH value and
temperature. At pH 7 where no driving force for the protonation
of β-nitrogen exists, more hydrophobic dehydrated copolymers
facilitate further blue shift of λmax, while at pH 1 the dye moieties
can be protonated at the β-nitrogen atom in the dehydrated
polymer media, and the more hydrophobic polymer enhances
the protonation at a lower temperature. The copolymer solutions
exhibit big red λmax shift upon minor increase of solution tem-
perature at acidic conditions, which makes them highly sensitive
sensors. These findings galantine these comblike polymers ideal
candidates as polymer-based sensors, which may find broad
interest in various material applications.
which induces further blue shift of λmax
.
Furthermore, the sharp transitions of these comblike copoly-
mers carrying OEG dendrons afford the polymer sensors higher
sensitivity than the copolymers carrying short OEG chains
reported previously.20a The λmax of the solutions at pH 1 all
moved to much higher wavelength in a fast fashion with the
increase of the solution temperature above their LCST. For
example, 2 deg temperature difference (from 27 to 29 °C) causes
the λmax to shift 18 nm (from 499 to 517 nm) for PG1(A)-co-
DR1. While for previously reported OEG-based polymer sen-
sors, the same red shift (Δλmax = 18 nm) can only be realized in
about 5 deg. This interesting property may afford this kind of
polymer sensors promising applications.
’ ASSOCIATED CONTENT
S
Supporting Information. Plots of concentration-depen-
b
dent transmittance vs temperature for PG1(A), PG1(G) and
MG1(A), temperature-varied UV spectra of PG1(A)-co-DR1
and PG1(G)-co-DR1 at pH 1 and 7, as well as proton NMR
spectra of macromonomers and (co)polymers. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ CONCLUSIONS
’ AUTHOR INFORMATION
Two structurally novel comblike polymers were efficiently
synthesized via free radical polymerization through macromo-
nomer strategy. These polymers carrying OEG dendron in the
side chains spaced with either hydrophobic alkyl chain or
hydrophilic OEG chain are all thermoresponsive and show sharp
phase transitions with small hysteresis. In contrast to conven-
tional linear thermoresponsive polymers, the concentration
shows negligible influence on their phase transition temperature.
Comparing with the corresponding macromonomers as well as
the densely packed dendronized polymer analogues, both chemi-
cal structure and architecture show significant effects on their
thermoresponsiveness. The more hydrophobic polymer PG1(A)
could form denser aggregates than that from the more hydro-
philic polymer PG1(G) due to the enhanced hydrophobic
interaction. On the basis of the characteristic thermoresponsive
behavior of these comblike polymers, the architecture is utilized
to fabricate polymer sensors by covalently incorporating solva-
tochromic dye (Disperse Red 1). It is interesting to find that
Corresponding Author
*Phone: +86-21-69982829. Fax: +86-21-69982827. E-mail: wli@
shu.edu.cn (W.L.), azhang@shu.edu.cn (A.Z.).
’ ACKNOWLEDGMENT
We thank Dr. Hongmei Deng from the Instrumental Analysis
of Research Center (Shanghai University) for her assistance in
NMR measurements. This work is financially supported by
National Natural Science Foundation of China (Nos. 21034004,
21104043, and 20974020) and the Science and Technology
Commission of Shanghai (No. 10520500300).
’ REFERENCES
(1) (a) Knop, K.; Hoogenboom, R.; Fischer, D.; Schubert, U. S. Angew.
Chem., Int. Ed. 2010, 49, 6288–6308. (b) Thompson, M. S.; Vadala, T. P.;
Vadala, M. L.; Lin, Y.; Riffle, J. S. Polymer 2008, 49, 345–373.
8620
dx.doi.org/10.1021/ma201874c |Macromolecules 2011, 44, 8614–8621

Macromolecules
ARTICLE
(2) (a) Gasteier, P.; Reska, A.; Petra, S.; Salber, J.; Offenh€ausser, A.;
M€oller, M.; Groll, J. Macromol. Biosci. 2007, 7, 1010–1023. (b) Lapienis,
G. Prog. Polym. Sci. 2009, 34, 852–892. (c) Osada, K.; Kataoka, K. Adv.
Polym. Sci. 2006, 202, 113–153. (d) Joralemon, M. J.; McRae, S.;
Emrick., T. Chem. Commun. 2010, 46, 1377–1393.
(23) Ballauff, M.; Likos, C. N. Angew. Chem., Int. Ed. 2004, 43,
2998–3020.
(24) Li, W.; Zhang, A.; Chen, Y.; Feldman, K.; Wu, H.; Schluter,
A. D. Chem. Commun. 2008, 5948–5950.
(25) Lacroix, P. G.; Malfant, I.; Iftime, G.; Razus, A. C.; Nakatani, K.;
Delaire, J. A. Chem.—Eur. J. 2000, 6, 2599–2608.
(3) Lutz, J. F. J. Polym. Sci., Part A: Polym. Chem. 2008, 46,
3459–3470.
(4) (a) Cheng, H.; Shen, L.; Wu., C. Macromolecules 2006,
39, 2325–2329. (b) Keerl, M.; Smirnovas, V.; Winter, R.; Richtering,
W. Angew. Chem., Int. Ed. 2008, 47, 338–341.
(5) (a) Ishizone, T.; Seki, A.; Hagiwara, M.; Han, S. Macromolecules
2008, 41, 2963–2967. (b) Li, W.; Zhang, A.; Schl€uter, A. D. Chem.
Commun. 2008, 5523–5525.
(6) (a) Han, S.; Hagiwara, M.; Ishizone, T. Macromolecules 2003,
36, 8312–8319. (b) Hua, F.; Jiang, X.; Li, D.; Zhao, B. J. Polym. Sci., Part
A: Polym. Chem. 2006, 44, 2454–2467.
(7) (a) Jones, J. A.; Novo, N.; Flagler, K.; Pagnucco, C. D.; Carew, S.;
Cheong, C.; Kong, X. Z.; Burke, N. A. D.; St€over, H. D. H. J. Polym. Sci.,
Part A: Polym. Chem. 2005, 43, 6095–6140. (b) Lutz, J. F.; Hoth, A.
Macromolecules 2006, 39, 893–896. (c) Qiao, Z.-Y.; Du, F.-S.; Zhang, R.;
Liang, D.-H.; Li, Z.-C. Macromolecules 2010, 43, 6485–6494.
(8) (a) Hu, Z.; Cai, T.; Chi, C. Soft Matter 2010, 6, 2115–2123. (b)
Lutz, J.-F. Adv. Mater. 2011, 23, 2237–2243.
(9) Gillich, T.; Benetti, E. M.; Rakhmatullina, E.; Konradi, R.; Li, W.;
Zhang, A.; Schl€uter, A. D.; Textor, M. J. Am. Chem. Soc. 2011,
133, 10940–10950.
(10) Chen, G.; Wright, P. M.; Geng, J.; Mantovani, G.; Haddleton,
D. M. Chem. Commun. 2008, 1097–1099.
€
(11) Lutz, J. F.; Akdemir, O.; Hoth, A. J. Am. Chem. Soc. 2006,
128, 13046–13047.
(12) Jiang, X.; Zhao, B. J. Polym. Sci., Part A: Polym. Chem. 2007,
45, 3707–3721.
(13) (a) Jia, Z.; Chen, H.; Zhu, X.; Yan, D. J. Am. Chem. Soc. 2006,
128, 8144–8145. (b) Gao, M.; Jia, X.; Kuang, G.; Li, Y.; Liang, D.; Wei, Y.
Macromolecules 2009, 42, 4273–4281. (c) Gao, M.; Jia, X.; Li, Y.; Liang,
D.; Wei, Y. Macromolecules 2010, 43, 4314–4323.
(14) Luzon, M.; Boyer, C.; Peinado, C.; Corrales, T.; Whittaker, M.;
Tal, L.; Davis, T. P. J. Polym. Sci., Part A: Polym. Chem. 2010,
48, 2783–2792.
(15) (a) Li, W.; Zhang, A.; Feldman, K.; Walde, P.; Schl€uter, A. D.
Macromolecules 2008, 41, 3659–3667. (b) Li, W.; Wu, D.; Schl€uter,
A. D.; Zhang, A. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 4530–4534.
(c) Li, W.; Zhang, A. Sci. China Chem. 2010, 53, 2509–2519.
(16) (a) Junk, M. J. N.; Li, W.; Schl€uter, A. D.; Wegner, G.; Spiess,
H. W.; Zhang, A.; Hinderberger, D. Angew. Chem., Int. Ed. 2010,
49, 5683–5687. (b) Junk, M. J. N.; Li, W.; Schl€uter, A. D.; Wegner,
G.; Spiess, H. W.; Zhang, A.; Hinderberger, D. J. Am. Chem. Soc. 2011,
133, 10832–10838.
(17) (a) Hu, J.; Liu, S. Macromolecules 2010, 43, 8315–8330. (b) Liu,
X.-Y.; Cheng, F.; Liu, Y.; Li, W.-G.; Chen, Y.; Pan, H.; Liu, H.-J. J. Mater.
Chem. 2010, 20, 278–284. (c) Pietsch, C.; Schubert, U. S.; Hoogenboom,
R. Chem. Commun. 2011, 47, 8750–8765.
(18) Reichardt, C. Chem. Rev. 1994, 94, 2319–2358.
(19) (a) Iwai, K.; Matsumura, Y.; Uchiyama, S.; de Silva, A. P. J. Mater.
Chem. 2005, 15, 2796–2800. (b) Koopmans, C.; Ritter, H. J. Am. Chem. Soc.
2007, 129, 3502–3503. (c) Gota, C.; Okabe, K.; Funatsu, T.; Harada, Y.;
Uchiyama, S. J. Am. Chem. Soc. 2009, 131, 2766–2767.
(20) (a) Pietsch, C.; Hoogenboom, R.; Schubert, U. S. Angew. Chem.,
Int. Ed. 2009, 48, 5653–5656. (b) Pietsch, C.; Vollrath, A.; Hoogenboom,
R.; Schubert, U. S. Sensors 2010, 10, 7979–7990.
(21) Precisely speaking, this transition temperature should be re-
ferred to as lower critical aggregation temperature (LCAT) because this
term describes more precisely the effect observed by turbidity measure-
ments (cloud point determinations). Though strictly speaking the term
LCST is not proper in the present connection, it is nevertheless much
more often used than LCAT; thus, we stick to it throughout the article.
(22) Bolisetty, S.; Schneider, C.; Polzer, F.; Ballauff, M.; Li, W.;
Zhang, A.; Schl€uter., A. D. Macromolecules 2009, 42, 7122–7128.
8621
dx.doi.org/10.1021/ma201874c |Macromolecules 2011, 44, 8614–8621