Switching preorganization and thermoresponsive behavior of a water-soluble polymer via light-tunable hydrogen bonding

Article abstract of DOI:10.1039/c3sm00020f

We report a general approach that can switch preorganization and thermoresponsive behaviour of a water-soluble polymer via light-tuned hydrogen bonding (H-bonding) "dimerization" or "polymerisation". To this end, well-defined copolymers of 2-hydroxy-3-(4-

Full text of DOI:10.1039/c3sm00020f

Soft Matter
View Article Online
View Journal | View Issue
PAPER
Switching preorganization and thermoresponsive
behavior of a water-soluble polymer via light-tunable
Cite this: Soft Matter, 2013, 9, 4036
hydrogen bonding
Fen Wu, Xianghua Tang, Lei Guo, Ke Yang and Yuanli Cai*
We report a general approach that can switch preorganization and thermoresponsive behaviour of a
water-soluble polymer via light-tuned hydrogen bonding (H-bonding) “dimerization” or
“
polymerisation”. To this end, well-defined copolymers of 2-hydroxy-3-(4-hydroxyiminomethyl)-
phenoxypropyl methacrylate with oligo(ethylene glycol) methacrylate (OEGMA) were synthesized via
RAFT copolymerisation of 3-(4-formylphenoxy)-2-hydroxypropyl methacrylate with OEGMA under visible
ꢀ
light irradiation at 25 C and fully converting the aldehyde groups into the targeted oxime groups.
Photoisomerization and stability of these oxime groups, and the effect of oxime configurations on
preorganization and thermoresponsive behaviour were studied using dynamic light scattering and
1
temperature-variable
H
NMR. The results demonstrated that photoisomerization of E-oximes
Received 3rd January 2013
Accepted 5th February 2013
ꢀ
equilibrated at 76% Z-type formation. These oxime groups were stable against heating at 80 C in
water for 8 h. Moreover, the self-restrained H-bonding “dimerization” of E-oximes led to the small-sized
micelles with negligible hysteresis, while the extensible H-bonding “polymerisation” of Z-type ones led
to the enlarged micelles with remarkable hysteresis.
DOI: 10.1039/c3sm00020f
www.rsc.org/softmatter
16,17
bonding rearrangements.
However, to the best of our
Introduction
awareness, switching preorganization or thermoresponsive
behavior of a water-soluble polymer via light-tuned hydrogen
bonding inside or between the polymer chains is unprece-
dented. Clearly, exploring new approaches that can switch
Thermoresponsive water-soluble polymers have become an
increasingly important class of materials that are quite prom-
ising in emerging biomedical and intelligent materials. These
1–3
smart polymers can adapt to surrounding aqueous environ- preorganization and thermoresponsive behavior of a water-
ments and undergo an abrupt phase transition in response to a soluble polymer via the light-tuned hydrogen bonding is
small change in temperature. Introducing photosensitivity into important, but also quite challenging to date.
water-soluble polymers may induce well-dened self-assembly
As is well known, adjusting the conguration may change
via preorganization and/or thermoresponsive micellization in the hydrogen bonding of organic compounds. For example,
4
–7
water that is crucial for bio-/nano-related applications. To
E-oximes are interlinked via self-restrained hydrogen bonding
H-bonding) “dimerization”, whereas Z-type ones are inter-
linked via chain-like H-bonding “polymerisation”, which
8,9
date, reversible photoisomerization of trans-/cis-azobenzene
(
10,11
18
or spiropyran-merocyanine,
and the irreversible cleavage of
have been employed for this purpose. brings about strikingly different melting points of benzaldox-
These reported thermoresponsive behaviors are based on the ime from 35 C (E-type) to 130 C (Z-type)
light-tuned hydrophilicity of such water-soluble polymers.
12,13
photosensitive groups
ꢀ
ꢀ
19,20
or 4-methoxy-
4–15
ꢀ
ꢀ
21,22
benzaldoxime from 63 C (E-type) to 132 C (Z-type).
Actually, light-induced rearrangements of hydrogen bonding Moreover, these oximes interconvert between E-/Z-congura-
20
commonly proceed inside the precisely-preorganized photo- tions upon UV irradiation or heating. Thus it is reasonable to
sensitive proteins. For example, during the light-powered
proton pumping of the proteorhodopsin inside the halobacteria
purple membrane, all-trans retinyl isomerizes into 13-cis on
envisage that introducing such tunable hydrogen bonding into
a water-soluble polymer inevitably gives rise to the tunable
association of the polymer chains in water thus leading to the
sunlight irradiation, which induces the transportation of imino changes in the thermoresponsive behavior, because of the
protons neighboring retinyl outwardly via chain-like hydrogen variable “dimerization” or “polymerisation” inside or between
polymer chains as illustrated in Fig. 1.
Herein we report a general approach that can readily switch
the preorganization and thermoresponsive behavior of water-
Jiangsu Key Laboratory of Advanced Functional Polymer Design and Application,
Department of Polymer Science and Engineering, College of Chemistry, Chemical
soluble polymers via light-tuned hydrogen bonding of oxime
groups (Fig. 1). To this end, well-dened random copolymers of
Engineering and Materials Science, Soochow University, Suzhou 215123, China.
E-mail: ylcai@suda.edu.cn; Fax: +86-512-6588-4419; Tel: +86-512-6588-4419
4
036 | Soft Matter, 2013, 9, 4036–4044
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
mol) and ethanol (40.0 mL) were charged in a dried ask. The
ꢀ
solution was reuxed in a nitrogen gas atmosphere at 85 C for
1
0 h. The mixture was concentrated by rotary evaporation, dis-
solved in chloroform (40 mL), and subsequently washed using a
.0 wt% NaOH (aq.), saturated NaCl (aq.) and nally water until
it was neutralized. The solution was dried over anhydrous
MgSO . Aer being ltered, the solvent was removed under
reduced pressure. The crude product was recrystallized from
4
4
ꢀ
ꢀ
ethyl ether at ꢁ18 C and dried in vacuum at 25 C to give a
1
white solid product. Weight: 20.10 g and yield: 60%. H NMR (d,
in CDCl ): 9.87 ppm (1H, CHO); 7.00, 7.84 ppm (4H, C H ); 6.14,
3
6 4
Fig. 1 Schematic illustration for the H-bonding “dimerization” (E-oximes) or
5
4
1
.60 ppm (2H, CH ]C(CH )); 4.31–4.37 ppm (2H, COOCH CH);
2
3
2
“
polymerisation” (Z-oximes) inside or between the polymer chains that can be
.13 ppm (2H, CH(OH)CH
2
O); 2.74 ppm (1H, CH
2 2
CH(OH)CH );
switched via the isomerization of oxime groups upon UV irradiation.
.93 ppm (3H, CH ]C(CH )).
2 3
2
-hydroxy-3-(4-hydroxyiminomethyl)phenoxypropyl methacry-
ꢀ
RAFT polymerisation under visible light radiation at 25 C
late with oligo(ethylene glycol) methacrylate [P(HHPPMA-
ran-OEGMA)] were synthesized via RAFT copolymerisation of
3
Typically, FPHPMA (1.00 g, 3.78 mmol), OEGMA (1.796 g, 3.78
mmol), CPFDB (12.7 mg, 0.058 mmol), TPO (6.0 mg, 0.017
mmol) and DMF (1.87 g) were charged in a 25 mL ask. The
-(4-formylphenoxy)-2-hydroxypropyl methacrylate (FPHPMA)
ꢀ
23–26
and OEGMA under visible light irradiation at 25 C,
and

ask was capped with rubber septa and then immersed in a
then the aldehyde groups were fully converted into the targeted
oxime groups via reacting with hydroxylamine. Photo-
isomerization and stability of their oxime groups, and the effect
of E-/Z-oxime congurations on the preorganization and ther-
moresponsive behavior were studied by dynamic light scat-
ꢀ
thermostatic water bath at 25 C. The solution was deoxygen-
ated by purging with nitrogen gas for 30 min, and irradiated
ꢁ
2
ꢀ
with mild visible light at I420nm ¼ 200 mW cm at 25 C for
9
0 min. The light intensity was measured using a UV-A radi-
ometer tted with a 420 nm detector. This polymerisation was
quenched on exposure to air and adding traces of hydroquinone
inhibitor. One portion of the solution was directly diluted in
1
tering and temperature-variable H NMR.
1
Experimental section
chloroform-d. Based on H NMR studies, 49.0% FPHPMA and
Materials
47.5% OEGMA monomers were polymerized. The polymer was
precipitated from a large excess of ethyl ether under stirring at
Glycidyl methacrylate (GMA) and oligo(ethylene glycol) mono-
methyl ether methacrylate (OEGMA, M
ꢀ
ꢁ
1
25 C. Aer being ltered, the solid polymer was washed with
n
¼ 475 g mol , M
/M
w n
¼
ꢀ
ethyl ether 3 times and dried in a vacuum oven at 25 C for 24 h.
1.03) were purchased from Aldrich. OEGMA was passed through
1
ꢁ1
Weight: 1.175 g and yield: 87.5%; H NMR: M
P(FPHPMA31-ran-OEGMA31); GPC: M
¼ 45.3 kg mol and M
¼ 1.15.
n
¼ 23.1 kg mol
,
a basic alumina column to remove inhibitor; GMA was dried over
ꢁ
1
n
w
/
˚
4
A molecular sieves overnight and then distilled under reduced
M
n
ꢀ
pressure. Both monomers were stored at ꢁ20 C prior to use. 2-
Cyanoprop-2-yl(4-uoro)dithiobenzoate (CPFDB) chain transfer
Synthesis of P(HHPPMA-ran-OEGMA) copolymers
27
agent was synthesized according to the reported procedures.
(
1
2,4,6-Trimethylbenzoyl)diphenylphosphine oxide (TPO, 97%) Typically, P(FPHPMA31-ran-OEGMA31) ( H NMR: M
n
¼ 23.1 kg
¼ 1.15; 3.000 g, 4.03
HCl (1.126 g, 16.12 mmol)
ꢁ
1
ꢁ1
photoinitiator was purchased from Runtec Co., 4-hydroxybe- mol ; GPC: M
n
¼ 45.3 kg mol , M
w n
/M
nzaldehyde was purchased from Langrui Co., and hydroxylamine mmol aldehyde groups) and HONH
2
hydrochloride (HONH
DMSO-d ) and heavy water (D
these agents were used as received. Water was freshly deionized stirring at 25 C, and further stirred overnight at 25 C. The
2
HCl), chloroform-d, dimethylsulfoxide-d
6
were dissolved in DMF (15 mL) in a 25 mL ask. NaOH (8.1 mol
ꢁ
1
(
6
2
O) were purchased from Aldrich;
L
in water, 2.0 mL) was added dropwise into this ask under
ꢀ
ꢀ
to R > 18.25 MU cm prior to DLS studies.
copolymer was puried by dialysis from water at pH 8.0 (tri-
ethylamine added, measured by a PHS-3C digital pH meter) for
2
days and freeze-dried overnight. The solids were dissolved in
Light source
DMSO-d to assess the conversion of aldehydes into oximes
using H NMR.
6
A 500 W mercury lamp emitting separately at l ¼ 254, 302, 313,
1
365, 405, 436, 545, and 577 nm was directly employed for the
photoisomerization of the oxime groups. Aer ltering UV light
using JB400 lters, this light was used to activate the RAFT
process.
Photoisomerization of oxime groups
1
ꢁ1
Typically, PHHPPMA59 ( H NMR: M
n
¼ 16.4 kg mol ; GPC:
ꢁ
1
M
n
¼ 30.0 kg mol , M
w
/M
n
¼ 1.13; 0.100 g, 0.36 mmol of oxime
groups) was dissolved in anhydrous THF (10.0 mL) in a 25 mL
Synthesis of FPHPMA
quartz ask. This ask was capped with rubber septa, and then
ꢀ
GMA (17.48 g, 0.123 mol), 4-hydroxybenzaldehyde (15.00 g, immersed in a thermostatic water bath at 25 C. The solution
0.123 mol), tetramethylammonium bromide (0.947 g, 0.062 was deoxygenated by purging nitrogen gas for 40 min and
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4036–4044 | 4037

View Article Online
Soft Matter
Paper
ꢁ2
irradiated with UV light at I365nm ¼ 700 mW cm , as measured
using a UV-A radiometer tted with a 365 nm detector. The
samples were collected at predetermined intervals. The degrees
1
of isomerization were assessed by H NMR. The procedures for
the photoisomerization of the copolymers were the same as
mentioned above and were also conducted at the oxime
ꢁ
1
concentrations of 36.0 mmol L
.
Gel Permeation Chromatography (GPC)
GPC measurements were performed on a PL-GPC 120 integrated
system tted with a refractive index detector. The columns
(
30 cm PL gel Mixed-D, 2 in series) were eluted with DMF that
ꢁ1
contained 0.01 mol L LiBr. A series of near-monodisperse
polystyrene (PS) standards (0.580–300.0 kg mol ) were used for
calibration. All calibration and analysis were performed at 80 C
and a ow rate of 1.0 mL min
ꢁ1
ꢀ
ꢁ1
.
Dynamic Light Scattering (DLS)
DLS studies were performed on a BI-200SM Brookhaven setup
tted with a 100 mW solid-state laser at lem ¼ 532 nm, a BI-
00SM goniometer and a BI-9000 digital correlator. The laser

2
was adjusted to 10.0 mW. Temperature was controlled using a
BI-TCD controller. Typically, P(HHPPMA -ran-OEGMA )
3
1
31
1
Fig. 2 (a) H NMR spectra (from upper to bottom) and (b) GPC traces (from right
(
10.0 mg) was dissolved in water (10.0 mL) in a 50 mL ask and
to left) of P(FPHPMA41-ran-OEGMA19), P(FPHPMA36-ran-OEGMA25
P(FPHPMA31-ran-OEGMA31) copolymers.
)
and
ꢀ
stirred overnight at 5 C. The dust was removed by passing
through a 0.45 mm lter at 5 C. The solution was moved to a
DLS cuvette, placed in a sample cell, and kept at 8 C for 20 min.
The solution was heated at 1.0 C min up to a predetermined
temperature and stabilized for 5 min. The scattered light was (t and w, in CPFDB chain-ends) overlapped with the C
ꢀ
ꢀ
ꢀ
ꢁ1
6
H
4
CHO
ꢀ
collected at 90 for 5 min. The hydrodynamic diameters (D ) signals (b and c, in FPHPMA units) at d ¼ 7.0–7.8 ppm, the
h
were calculated using cumulants analysis in CONTIN routine.
All data were averaged over three times.
C
6
H
4
CHO signal (a) at d ¼ 9.8 ppm was quite clear. Moreover,
the signals of protons e, d, f, and k in both units were detected
at d ¼ 3.8–4.5 ppm. Accordingly, the apparent individual
1
(
DPFPHPMA, DPOEGMA) and the overall (DP) degrees of polymer-
Temperature-variable H NMR
isation were assessed according to eqn (1)–(3).
1
H NMR studies were performed on a 400 MHz Bruker AV-400
2I
a
NMR spectrometer tted with a temperature controller. The
DPFPHPMA ¼ I
(1)
(2)
(3)
polymer sample (4.0 mg) was dissolved in D O (4.0 mL) and
bþw ꢁ 2I
a
2
ꢀ
stirred overnight at 5 C. The dust was removed using a 0.45 mm
DPOEGMA
DPFPHPMA
I
eþdþfþk ꢁ 5I
a

lter. The solution (0.5 mL) was moved to a NMR cuvette. Aer
¼
ꢀ
2I
a
being stabilized at 20 C for 2 h, the solution was heated at
ꢀ
ꢁ1
1.0 C min to a predetermined temperature and stabilized
1
for 5 min. The H NMR spectrum was recorded under scanning
2 times.
DP ¼ DPFPHPMA + DPOEGMA
3
As shown in Fig. 2b, the GPC traces were unimodal and
symmetrical, and slightly shied to the higher molecular weight
side on the increase of OEGMA fractions, most presumably due
Results and discussion
Synthesis of P(HHPPMA-ran-OEGMA) copolymers
Firstly, P(FPHPMA-ran-OEGMA) copolymers were synthesized to the higher molecular weight and more signicant steric effect
via RAFT copolymerisation of FPHPMA and OEGMA under of OEGMA than FPHPMA units. In order to gure out the effect
ꢀ
visible light radiation at 25 C. As shown in Fig. 2a, except the of oxime congurations on preorganization and thermores-
3 3 2 3
signal of CHCl in CDCl , the signals of CH ]CCH (mono- ponsive behavior, the copolymers whose FPHPMA fractions
mers) at d ¼ 5.56 ppm and other impurities could not be over 50% were selected as the parent copolymers. As shown in
detected, suggesting that the monomers and other impurities Table 1, the selected parent copolymers were well-dened with
were removed. Moreover, the integral ratios were consistent almost the same chain lengths (DP ¼ 61 ꢂ 1) at narrow
with the assigned proton ratios of copolymers, suggesting intact molecular weight distributions (M
w
/M
n
¼ 1.12–1.15) but
structures of the copolymers. Although the signals of C
6
H
4
F
different compositions.
4
038 | Soft Matter, 2013, 9, 4036–4044
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
Table 1 Structure parameters of aldehyde-functional P(FPHPMA-ran-OEGMA)
Photoisomerization of oxime groups
The photoisomerization proceeded via the UV irradiation of the
selected for the synthesis of the oxime-functional copolymers
ꢁ1
M
n,GPC
M
n,
1
H NMR
copolymers at the same oxime concentrations of 36.0 mmol L
ꢁ
1
ꢁ1
a
a
ꢀ
No. (kg mol ) M
w
/M
n
(kg mol ) DP Composition
in a nitrogen gas atmosphere at 25 C. As shown in Fig. 4a, the
signals a and b in E-C
while the signals a and b in Z-C
6
H
4
CH]NOH were gradually attenuated,
1
2
3
43.4
44.8
45.3
1.12
1.15
1.15
20.1
21.3
23.1
60 P(FPHPMA41-ran-OEGMA19
61 P(FPHPMA36-ran-OEGMA25
62 P(FPHPMA31-ran-OEGMA31
)
)
)
0
0
6
H
4
CH]NOH were gradually
enhanced upon irradiation with UV light, suggesting that
E-oximes gradually isomerized to Z-type ones. Moreover, the
signals of impurities possibly caused by the rearrangement or
cycloaddition of oximes could not be detected under such
conditions.
a
Overall degrees of polymerisation (DP) and compositions were
1
assessed by H NMR studies.
2
0
As shown in Fig. 4b, this photoisomerization equilibrated at
The targeted P(HHPPMA-ran-OEGMA) copolymers were 76% Z-type formation. More specically, the photo-
achieved via adding NaOH (aq.) to DMF solution of isomerization of E-type oxime groups was clearly more rapid in
P(FPHPMA-ran-OEGMA) and HONH HCl, and then stirring at the homopolymer chains than in the copolymer chains, which
2
ꢀ
2
9
8
5 C for 6 h. As shown in Fig. 3, the C
6
H
4
CHO signal at d ¼ were most presumably caused by the light-screening effect of
CH]NOH signals at d ¼ OEGMA units that were adjacent to HHPPMA units.
.05, 7.93, 7.50, 7.29, and 6.95 ppm were clearly detectable.
6 4
.8 ppm disappeared, and the C H
The proton signals of impurities that were possibly caused
by the side reactions were not detectable. Moreover,
I8.05+7.29 : I7.50+7.93 : I6.95 was assessed to be 1 : 2 : 2, within
analysis errors. These results suggested that their aldehyde
groups were fully converted into the targeted oxime groups, i.e.
the parent copolymers were converted into the targeted
copolymers.
Stability of oxime groups against heating in aqueous media
Clearly, excellent stability of these oxime groups against heating
in water was crucial because the thermoresponsive behavior of
such water-soluble polymers should be studied in the heated
water. Accordingly, the stability of oxime groups against heating
in aqueous media was studied by sequentially adding heavy
6
water to the copolymer solutions in DMSO-d and heating up to
More importantly, the integral ratios of I8.05 : I7.29 and
ꢀ
1
8
0 C, and then analyzing using H NMR.
As shown in Fig. 5a, for P(HHPPMA31-ran-OEGMA31
[E] : [Z] ¼ 90 : 10, in DMSO-d ), although the proton signal g
I
7.50 : I7.93 were constant at 9 : 1, suggesting that the oxime
conguration ratios remained at the same value of [E]/[Z] ¼
0 : 10. These E/Z-oxime ratios were in good agreement with
)
(
6
9
in CH CH(OH)CH disappeared upon adding D O, the other
2
2
2
those of the benzaldoxime compound synthesized via reaction
of benzaldehyde with HONH HCl in water–ethanol on adding
2
28
sodium acetate at room temperature.
1
Fig. 4 (a) The evolution of the H NMR spectrum of P(HHPPMA41-ran-OEGMA19
)
1
ꢀ
Fig.
3
H
NMR spectra of P(HHPPMA41-ran-OEGMA19), P(HHPPMA36-ran-
(from top to bottom).
upon UV irradiation at 25 C and (b) the kinetic curves for photoisomerization of
oxime groups.
OEGMA25), and P(HHPPMA31-ran-OEGMA31) in DMSO-d
6
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4036–4044 | 4039

View Article Online
Soft Matter
Paper
Fig. 6 The evolution of (a) light scattering intensity, (b) hydrodynamic diameter
1
2
Fig. 5
H NMR spectra of P(HHPPMA31-ran-OEGMA31) at [E] : [Z] ¼ 90 : 10 (a)
(D
h
, solid) and polydispersity (m
2
/G , hollow) of P(HHPPMA31-ran-OEGMA31
)
ꢀ
ꢁ1
and 26 : 74 (b). From bottom to top: in DMSO-d
6
at 20 C; in DMSO-d
6
and D
2
O
([E] : [Z] ¼ 90 : 10, 1.0 mg mL in H
2
O) upon heating. Inset: those of the cor-
ꢀ
ꢀ
(
1 : 1 v/v) at 20 C; in DMSO-d
6
and D
2
O (1 : 1 v/v) at 80 C for 4 h; and in DMSO-
responding aldehyde-functional parent copolymer.
ꢀ
d
6
and D
2
O (1 : 1 v/v) at 80 C for 8 h.
suggesting that the copolymer had started to phase-separate
from water. This critical temperature was the so-called cloud
point (CP). Interestingly, although the oxime groups were more
water-soluble than aldehyde groups, converting these aldehyde
groups into water-soluble oxime groups brought about the
signals of this copolymer were fully detectable. These results
suggested that the copolymer molecularly dissolved in DMSO-
d
6
–D
2
O media. Moreover, these proton signals did not change
ꢀ
upon heating up to 80 C for 4 or 8 h, as indicated by the
undetectable additional side-reaction signals. These results
demonstrated that both E- and Z-type oxime groups were stable
against heating in water.
ꢀ
remarkable CP-lowering down to DT ¼ ꢁ5.6 C. These results
demonstrated that the hydrogen bonding of oxime groups
effectively enhanced the association of these polymer chains,
thus promoting the phase transition in response to heating.
As shown in Fig. 5b, for the Z-isomerized copolymer at
ꢀ
Further heating the solution above 49 C resulted in a second
[
E] : [Z] ¼ 26 : 74, the proton signals of the copolymer also
leveling-off of the light scattering intensity. This copolymer self-
assembled into micelles with the intensity-average hydrody-
remained the same aer heating the aqueous solution up to
8
ꢀ
0 C for 4 or 8 h. These results suggested that both E- and Z-
2
namic diameter D
(
h
¼ 52 nm and low polydispersity m
2
/G ¼ 0.05
type oxime groups were stable under such heating conditions.
This excellent stability facilitated the studies on the thermor-
esponsive behavior of such copolymers in the heated water.
Fig. 6b). Moreover, these micelles were remarkably smaller
than the irregular aggregates of the corresponding aldehyde-
functional parent copolymer (D
h
¼ 1300 nm). Moreover, this
critical temperature was clearly lower than that of the parent
Preorganization and thermoresponsive behavior of oxime-
functional P(HHPPMA-ran-OEGMA)
ꢀ
copolymer, down to DT ¼ ꢁ6.1 C. Thus it was named as the
critical temperature for the formation of compact micelles (T ).
c
Dynamic light scattering (DLS) was employed to gure out the
As summarized in Table 2, at the same oxime conguration
preorganization and thermoresponsive behavior of such water- ratios at [E] : [Z] ¼ 90 : 10, increasing [HHPPMA] : [OEGMA]
soluble polymers in aqueous media. In order to elucidate the from 31 : 31 to 36 : 25 led to the remarkable CP-decrease from
ꢀ
ꢀ
ꢀ
ꢀ
hydrogen bonding effect of the oxime groups, the aldehyde- 35.5 C down to 18 C, and also T
c
from 49.2 C to 29.2 C.
¼ 52 nm down
functional parent copolymers were also studied for comparison. The size of micelles slightly decreased from D
Firstly, P(HHPPMA31-ran-OEGMA31) at [E] : [Z] ¼ 90 : 10 (1.0 mg to 48 nm.
h
ꢁ1
mL in water) was selected for DLS studies.
Further increasing [HHPPMA] : [OEGMA] to 41 : 19 resulted
As shown in Fig. 6a, the light scattering intensity remained in the copolymer already associated or preorganized into
ꢀ
ꢀ
2
below 1.0 kcps upon heating from 2 C to 30 C, suggesting that micelles at D ¼ 36 nm and a low polydispersity m /G ¼ 0.06
h
2
the copolymer molecularly dissolved in water. Further heating prior to heating the aqueous solution, even at a low temperature
up to 35.5 C led to a sharp increase of light scattering intensity, of 2 C, which led to a high light scattering intensity of
ꢀ
ꢀ
4
040 | Soft Matter, 2013, 9, 4036–4044
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
Table 2 Critical temperatures (including CP and T
c
), light scattering intensity of coacervation or dissociation of micelles were not observed upon
ꢁ1
1
.0 mg mL P(HHPPMA-ran-OEGMA) ([E] : [Z] ¼ 90 : 10) in water and D
h
of their
ꢀ
ꢀ
heating from 2 C to 40 C. These results demonstrated that the
micelles maintained the excellent stability around the physio-
compact micelles over T
c
ꢀ
logical temperature of 37 C over a wide range of E-/Z-congu-
Intensity
ꢀ
ꢀ
Copolymer
CP ( C)
T
c
( C) (kcps)
D
h
(nm) rations. More importantly, the compact micelles enlarged from
D
h
¼ 34 nm to D
h
¼ 43 nm and nally to D ¼ 61 nm (Fig. 7b) on
h
a
a
P(HHPPMA41-ran-OEGMA19
P(HHPPMA36-ran-OEGMA25
P(HHPPMA31-ran-OEGMA31
)
)
)
NA
NA
121
174
228
34
48
52
the increase of Z-oxime fractions, which brought about
the remarkable increase of light scattering intensity upon
heating to the same temperatures (Fig. 7a). These results
demonstrated that Z-type oxime groups could enhance the
intermolecular association of the polymer chains effectively
during the preorganization into micelles in water, even at a low
18.6
35.5
29.1
49.2
a
NA: not applicable.
6
1
6.6 kcps. The micelles just slightly shrank on heating from 2 to
5 C. More importantly, these micelles remained at the same
ꢀ
temperature of 2 C.
ꢀ
In contrast, as shown in Fig. 8a, for P(HHPPMA31-ran-
size at D
h
¼ 34 nm, i.e. the coacervation could not be detected
OEGMA31) over a wide range of [E] : [Z] ¼ 90 : 10 to 26 : 74
ꢀ
ꢀ
ꢁ1
on further heating from 15 C to 40 C (see Fig. 7). These results
suggested that these compact micelles were quite stable against
(1.0 mg mL in water), the light scattering intensity remained
ꢀ
ꢀ
below 1.0 kcps on heating the solutions from 2 C to 35.5 C,
suggesting that the copolymers were molecularly dissolved in
water, rather than preorganized into micelles as observed
above. Further heating led to the starting-up of phase transition
ꢀ
heating around the physiological temperature of 37 C.
Accordingly, it could be concluded that this water-soluble
copolymer exhibited a well-dened preorganization character in
water. The fascinating stability of these micelles is certainly
desirable for the potential applications in emerging biomedical
or nano-structured materials.
ꢀ
at a CP ¼ 35.5 C, as indicated by the abrupt increase of light
scattering intensity. Varying E-/Z-oxime ratios did not change
c
the critical temperatures including CP and T , suggesting the
negligible effect of E-/Z-oxime congurations on the hydration/
dehydration behaviour.
The oxime-conguration effect on preorganization and
thermoresponsive behavior
However, as shown in Fig. 8b, the size of compact micelles
clearly enlarged from D ¼ 52 nm up to 60 nm and nally to
h
As shown in Fig. 7a, for P(HHPPMA41-ran-OEGMA19) (1.0 mg
ꢁ1
74 nm on increasing Z-oxime fractions from 10% to 44%, and to
mL
in water), although [E] : [Z] increased from 90 : 10 to
74%, which brought about the remarkable increase of light
58 : 42 and nally to 26 : 74, the light scattering intensity
scattering intensity on heating up over the critical temperatures
maintained the same evolving tendency on heating. Either
Fig. 7 The heating-induced evolution of (a) the light scattering intensity of
Fig. 8 The heating-induced evolution of (a) the light scattering intensity of
ꢁ1
ꢁ1
1
.0 mg mL
hydrodynamic diameter (D
[E] :[ Z]): 90 : 10 (triangle), 58 : 42 (diamond), and 26 : 74 (square).
P(HHPPMA41-ran-OEGMA19) copolymers in
H
2
O and (b) the
1.0 mg mL
P(HHPPMA31-ran-OEGMA31) copolymers in
) of their thermoresponsive micelles. E/Z-oxime ratios
([E] : [Z]): 90 : 10 (triangle), 56 : 44 (diamond), and 26 : 74 (square).
2
H O and (b) the
h
) of these preorganized micelles. E/Z-oxime ratios
hydrodynamic diameter (D
h
(
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4036–4044 | 4041

View Article Online
Soft Matter
Fig. 8a). This enlarging tendency of thermoresponsive micelles cleavage of photosensitive groups,
Paper
12,13
(
in which photoreactions
on the increase of Z-oxime fractions was in reasonably good changed LCST behavior which directly brought about micelli-
agreement with the enlarging tendency of preorganized zation or dissociation of the polymer chains in water. In
micelles of P(HHPPMA -ran-OEGMA ). These results sug- contrast, varying Z-/E-oxime ratios, in this case, switched the
4
1
19
gested that the Z-oxime groups could effectively enhance the association of the polymer chains but without changing the
intermolecular association of the polymer chains during the LCST behavior.
thermoresponsive micellization in water.
1
The effect of oxime congurations was further investigated
using heavy water (D O) as a solvent. As shown in Fig. 9, varying
H NMR evidence for thermoresponsive behavior
2
1
Temperature-variable H NMR was employed to clarify the effect
of E-/Z-oxime congurations on the thermoresponsive behavior.
the oxime congurations ([E] : [Z]) also resulted in the
enlargement of preorganized micelles of P(HHPPMA -ran-
4
1
P(HHPPMA31-ran-OEGMA31) copolymers at [E] : [Z] ¼ 90 : 10
OEGMA ) and thermoresponsive micelles of P(HHPPMA -ran-
1
9
31
ꢁ1
and 24 : 76 (1.0 mg mL in D
2
O) were selected for this purpose.
OEGMA31) in D
2
2
O, which were similar to those observed in H O.
As shown in Fig. 10a, for the copolymer at a [E] : [Z] ¼ 90 : 10,
the signals CH C(CH
) at d ¼ 0.8–1.9 ppm (polymer backbones)
and COOCH
In contrast, on heating above T
of D O solutions were lower than those of H
the formation of smaller micelles in D O than in H
c
, light scattering intensities
O solutions due to
O. More
2
3
2
2
2
at d ¼ 4.1 ppm (unit spacers) gradually attenuated
2
2
ꢀ
ꢀ
ꢀ
on heating from 21 C to 38 C (DLS: CP ¼ 35.5 C), and then
specically, the sizes of P(HHPPMA41-ran-OEGMA19) compact
micelles were at a Dh ¼ 31 nm ([E] : [Z] ¼ 90 : 10), 36 nm
ꢀ
remained essentially the same on further heating up to 44 C.
These results suggested that the phase transition was induced
by the hydrophobic association of the less polar polymer
backbones and unit spacers. However, the OC H CH]NOH
6 4
(
[E] : [Z] ¼ 58 : 42) or 45 nm ([E] : [Z] ¼ 26 : 74) in D O, smaller
2
than the corresponding micelles in H O (D ¼ 34, 43 or 61 nm).
2
h
Similarly, the sizes of P(HHPPMA31-ran-OEGMA31) compact
micelles were at a D
¼ 41 nm ([E] : [Z] ¼ 90 : 10), 46 nm
[E] : [Z] ¼ 56 : 44) or 56 nm ([E] : [Z] ¼ 26 : 74) in D O, also
smaller than the corresponding micelles in H O (D
¼ 52, 60 or
4 nm). These results conrmed that varying Z-/E-oxime
signals at d ¼ 6.5–8.4 ppm (E-oxime groups) were clearly
h
ꢀ
detectable on heating from 21 to 44 C. This indicated that
(
2
E-oxime groups remained in the hydrated state during the
thermoresponsive phase transition.
2
h
7
In contrast, for the copolymer at a [E] : [Z] ¼ 24 : 76
congurations effectively tuned the intermolecular association
of the polymer chains, inside either preorganized or thermor-
esponsive micelles.
(
Fig. 10b), heating the solution also resulted in dehydration of
the polymer backbones and unit spacers, as indicated by
attenuation of their proton signals. Similarly, the proton signals
Clearly, the mechanism for this photosensitivity was totally
different from those via changing the hydrophilicity of polymer
chains through the reversible photoisomerization of trans-/cis-
of oxime groups remained essentially the same on heating from
ꢀ
2
1 to 44 C. However, the signals of these oxime groups were
clearly weaker than those of the former (Fig. 10a). These results
8,9
10,11
azobenzene or spiropyran-merocyanine,
or the irreversible
1
Fig. 9 The configuration-dependent evolution of the light scattering intensity of
Fig. 10
H NMR spectra evolution of P(HHPPMA31-ran-OEGMA31) at a [E] : [Z] ¼
ꢁ1
ꢁ1
1
.0 mg mL P(HHPPMA-ran-OEGMA) in D
2
O on heating.
90 : 10 (a) or 24 : 76 (b) (1.0 mg mL in D
2
O) on heating.
4
042 | Soft Matter, 2013, 9, 4036–4044
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
suggested that the Z-oxime groups were partially dehydrated, caused by the intra- and intermolecular hydrogen bonding of
34–36
i.e. Z-oxime groups tended to be interlinked via chain-like the amides.
These results suggested the chain-like H-
H-bonding “polymerisation”, which inevitably enhanced the bonding “polymerisation” of Z-oxime groups, which effectively
intermolecular association of the polymer chains, thus leading enhanced the intermolecular association of the polymer chains,
to the enlarged thermoresponsive micelles.
thus leading to the enlarged micelles and remarkable hysteresis
in response to cooling.
Hydrogen bonding modulated by E-/Z-oxime congurations
As shown in Fig. 11a, for P(HHPPMA -ran-OEGMA ) at a Conclusions
3
1
31
ꢁ
1
[
E] : [Z] ¼ 90 : 10 (1.0 mg mL in H
2
O), the cooling-induced light
This paper reported an approach that can switch preorganiza-
tion and thermoresponsive behavior of a water-soluble polymer
via light-tunable hydrogen bonding. To this end, well-dened
P(HHPPMA-ran-OEGMA) copolymers were synthesized. The
photoisomerization and stability of oxime groups, and the effect
of E-/Z-oxime congurations on the preorganization and ther-
moresponsive behavior were studied by DLS and temperature-
scattering intensity evolution was essentially overlapped with
that evolved on heating, suggesting that these micelles could be
dissociated and dissolved in water rapidly on cooling the heated
2
9
30
solution. As previously observed by the groups of Lutz, Li,
31
32
Wu and our group, the rapid dissociation of micelles also
occurred on heating–cooling the solutions of random copoly-
mers of OEGMA with 2-(2-methoxyethoxy)ethyl methacrylate,
oligo(ethylene glycol) acrylate with 2-(5,5-dimethyl-1,3-dioxan-
1
variable H NMR.
The results demonstrated that the photoisomerisation of
2
-yloxy)ethyl acrylate, and pyrrolidone-based poly(meth)acry-
E-oximes into Z-type ones equilibrated at 76% Z-type formation.
lates. It was well known that thermoresponsive phase transition
of such polymers was induced mainly by the hydrophobic asso-
ciation of less or apolar polymer backbones and unit spacers.
It suggested the self-restrained H-bonding “dimerization” of E-
oxime groups, which did not or did just slightly contribute to the
intermolecular association of the polymer chains, thus leading to
the small-sized micelles and the rapid dissociation of micelles or
negligible hysteresis in response to cooling.
ꢀ
These oxime groups were stable in water at 80 C for 8 h.
Moreover, the self-restrained H-bonding “dimerization” of
E-oximes led to the small-sized micelles with negligible hyster-
esis, whereas the extensible H-bonding “polymerization” of
Z-oximes brought about the enlarged micelles with remarkable
hysteresis. This provided a general approach toward light-
tunable preorganization and thermoresponsive behavior of
water-soluble polymers.
29–32
However, for the copolymer at a [E] : [Z] ¼ 24 : 76 (Fig. 11b),
the cooling-induced phase transition signicantly lagged
behind the heating-induced phase transition, suggesting the Acknowledgements
dissociation of these thermo-induced micelles was more energy-
This work was supported by National Natural Science Founda-
consuming. This hysteresis also occurred during heating–cool-
tion of China (20874081, 21074104 and 21274097), Priority
Academic Program Development of Jiangsu Higher Education
Institutions, and Scientic Research Fund of Hunan Provincial
Education Department (10A116). We thank Dr Lican Lu at
Xiangtan University for his assistance for the synthesis.
33
ing the solutions of poly(N-isopropyl acrylamide), which was
Notes and references
´
´
1
R. A. Alvarez-Puebla, R. Contreras-Caceres, I. Pastoriza-
Santos, J. P ´e rez-Juste and L. M. Liz-Marz ´a n, Angew. Chem.,
Int. Ed., 2009, 48, 138.
2
M. A. C. Stuart, W. T. S. Huck, J. Genzer, M. M u¨ ller, C. Ober,
M. Stamm, G. B. Sukhorukov, I. Szleifer, V. V. Tsukruk,
M. Urban, F. Winnik, S. Zauscher, I. Luzinov and S. Minko,
Nat. Mater., 2010, 9, 101.
3
4
5
H. Ajiro, Y. Takahashi and M. Akashi, Macromolecules, 2012,
4
5, 2668.
D. Roy, J. N. Cambre and B. S. Sumerlin, Prog. Polym. Sci.,
010, 35, 278.
2
J.-M. Schumers, C.-A. Fustin and J.-F. Gohy, Macromol. Rapid
Commun., 2010, 31, 1588.
6
7
Y. Zhao, Macromolecules, 2012, 45, 3647.
G. Pasparakis, T. Manouras, P. Argitis and M. Vamvakaki,
Macromol. Rapid Commun., 2012, 33, 183.
ꢁ1
8 S. Dai, P. Ravi and K. C. Tam, So Matter, 2009, 5, 2513.
9 F. D. Jochum and P. Theato, Chem. Commun., 2010, 46, 6717.
Fig. 11 The evolution of the light scattering intensity of 1.0 mg mL
P(HHPPMA31-ran-OEGMA31) in H O on a heating–cooling cycle.
2
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4036–4044 | 4043

View Article Online
Soft Matter
Paper
10 H. Lee, W. Wu, J. K. Oh, L. Mueller, G. Sherwood, L. Peteanu, 25 C. Cao, K. Yang, F. Wu, X. Wei, L. Lu and Y. Cai,
T. Kowalewski and K. Matyjaszewski, Angew. Chem., Int. Ed.,
007, 46, 2453.
1 J. A. Nam, A. Al-Nahain, S. Lee, I. In and S. Y. Park, Macromol.
Rapid Commun., 2012, 33, 1958.
Macromolecules, 2010, 43, 9511.
26 K. Yang, X. Wei, F. Wu, C. Cao, J. Deng and Y. Cai, So
Matter, 2011, 7, 5861.
27 M. Benaglia, E. Rizzardo, A. Alberti and M. Guerra,
Macromolecules, 2005, 38, 3129.
2
1
1
1
1
2 J. Jiang, X. Tong and Y. Zhao, J. Am. Chem. Soc., 2005, 127,
8
290.
28 J.-L. Boucher, M. Delaforge and D. Mansuy, Biochemistry,
1994, 33, 7811.
3 J. Babin, M. Pelletier, M. Lepage, J.-F. Allard, D. Morris and
Y. Zhao, Angew. Chem., Int. Ed., 2009, 48, 3329.
4 P. Ravi, S. L. Sin, L. H. Gan, Y. Y. Gan, K. C. Tam, X. L. Xia
and X. Hu, Polymer, 2005, 46, 137.
5 J. He, X. Tong and Y. Zhao, Macromolecules, 2009, 42, 4845.
6 N. Hampp, Chem. Rev., 2000, 100, 1755.
7 J. M. Walter, D. Greeneld, C. Bustamante and J. Liphardt,
Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 2408.
8 X. Lin, W. Xu, J. Wang and C. Liu, Chem. J. Chin. Univ., 2006,
¨
29 J.-F. Lutz, O. Akdemir and A. Hoth, J. Am. Chem. Soc., 2006,
128, 13046.
30 Z. Qiao, F. Du, R. Zhang, D. Liang and Z. Li, Macromolecules,
2010, 43, 6485.
31 H. Lai, G. Chen, P. Wu and Z. Li, So Matter, 2012, 8,
2662.
32 J. Deng, Y. Shi, W. Jiang, Y. Peng, L. Lu and Y. Cai,
Macromolecules, 2008, 41, 3007.
1
1
1
1
2
7, 897.
33 H. Cheng, L. Shen and C. Wu, Macromolecules, 2006, 39,
2325.
1
2
2
2
2
9 F. K. Gameron, J. Phys. Chem., 1898, 2, 409.
0 A. Padwa, Chem. Rev., 1977, 77, 37.
1 H. R. Carveth, J. Phys. Chem., 1899, 3, 437.
2 E. L. Skau and B. Saxton, J. Phys. Chem., 1933, 37, 197.
34 E. I. Tiktopulo, V. N. Uversky, V. B. Lushchik, S. I. Klenin,
V. E. Bychkova and O. B. Ptitsyn, Macromolecules, 1995, 28,
7519.
3 L. Lu, H. Zhang, N. Yang and Y. Cai, Macromolecules, 2006, 35 H. Mao, C. Li, Y. Zhang, S. Furyk, P. S. Cremer and
9, 3770. D. E. Bergbreiter, Macromolecules, 2004, 37, 1031.
4 Y. Shi, G. Liu, H. Gao, L. Lu and Y. Cai, Macromolecules, 2009, 36 B. Sun, Y. Lin, P. Wu and H. W. Siesler, Macromolecules,
2, 3917. 2008, 41, 1512.
3
2
4
4
044 | Soft Matter, 2013, 9, 4036–4044
This journal is ª The Royal Society of Chemistry 2013