Hydrophobically modified sulfobetaine copolymers with tunable aqueous UCST through postpolymerization modification of poly(pentafluorophenyl acrylate)

Article abstract of DOI:10.1021/ma402391a

Polysulfobetaines, polymers carrying highly polar zwitterionic side chains, present a promising research field by virtue of their antifouling properties, hemocompatibility, and stimulus-responsive behavior. However, limited synthetic approaches exist to produce sulfobetaine copolymers comprising hydrophobic components. Postpolymerization modification of an activated ester precursor, poly(pentafluorophenyl acrylate), employing a zwitterionic amine, 3-((3-aminopropyl)dimethylammonio)propane-1-sulfonate, ADPS, is presented as a novel, one-step synthetic concept toward sulfobetaine (co)polymers. Modifications were performed in homogeneous solution using propylene carbonate as solvent with mixtures of ADPS and pentylamine, benzylamine, and dodecylamine producing a series of well-defined statistical acrylamido sulfobetaine copolymers containing hydrophobic pentyl, benzyl, or dodecylacrylamide comonomers with well-controllable molar composition as evidenced by NMR and FT-IR spectroscopy and size exclusion chromatography. This synthetic strategy was exploited to investigate, for the first time, the influence of hydrophobic modification on the upper critical solution temperature (UCST) of sulfobetaine copolymers in aqueous solution. Surprisingly, incorporation of pentyl groups was found to increase solubility over a wide composition range, whereas benzyl groups decreased solubility - an effect attributed to different entropic and enthalpic contributions of both functional groups. While UCST transitions of polysulfobetaines are typically limited to higher molar mass samples, incorporation of 0-65 mol % of benzyl groups into copolymers with molar masses of 25.5-34.5 kg/mol enabled sharp, reversible transitions from 6 to 82 C in solutions containing up to 76 mM NaCl, as observed by optical transmittance and dynamic light scattering. Both synthesis and systematic UCST increase of sulfobetaine copolymers presented here are expected to expand the scope and applicability of these smart materials.

Full text of DOI:10.1021/ma402391a

Article
pubs.acs.org/Macromolecules
Hydrophobically Modified Sulfobetaine Copolymers with Tunable
Aqueous UCST through Postpolymerization Modification of
Poly(pentafluorophenyl acrylate)
Peter A. Woodfield,† Yicheng Zhu,† Yiwen Pei,†,‡ and Peter J. Roth*,†
†Centre for Advanced Macromolecular Design (CAMD), School of Chemical Engineering, University of New South Wales (UNSW),
Sydney, NSW 2052, Australia
‡Polymer Electronics Research Centre, School of Chemical Sciences, University of Auckland, Private Bag 92019, Auckland 1142, New
Zealand
S
* Supporting Information
ABSTRACT: Polysulfobetaines, polymers carrying highly
polar zwitterionic side chains, present a promising research
field by virtue of their antifouling properties, hemocompati-
bility, and stimulus-responsive behavior. However, limited
synthetic approaches exist to produce sulfobetaine copolymers
comprising hydrophobic components. Postpolymerization
modification of an activated ester precursor, poly-
(pentafluorophenyl acrylate), employing a zwitterionic
amine, 3-((3-aminopropyl)dimethylammonio)propane-1-sul-
fonate, ADPS, is presented as a novel, one-step synthetic
concept toward sulfobetaine (co)polymers. Modifications were
performed in homogeneous solution using propylene carbo-
nate as solvent with mixtures of ADPS and pentylamine,
benzylamine, and dodecylamine producing a series of well-defined statistical acrylamido sulfobetaine copolymers containing
hydrophobic pentyl, benzyl, or dodecylacrylamide comonomers with well-controllable molar composition as evidenced by NMR
and FT-IR spectroscopy and size exclusion chromatography. This synthetic strategy was exploited to investigate, for the first
time, the influence of hydrophobic modification on the upper critical solution temperature (UCST) of sulfobetaine copolymers
in aqueous solution. Surprisingly, incorporation of pentyl groups was found to increase solubility over a wide composition range,
whereas benzyl groups decreased solubilityan effect attributed to different entropic and enthalpic contributions of both
functional groups. While UCST transitions of polysulfobetaines are typically limited to higher molar mass samples, incorporation
of 0−65 mol % of benzyl groups into copolymers with molar masses of 25.5−34.5 kg/mol enabled sharp, reversible transitions
from 6 to 82 °C in solutions containing up to 76 mM NaCl, as observed by optical transmittance and dynamic light scattering.
Both synthesis and systematic UCST increase of sulfobetaine copolymers presented here are expected to expand the scope and
applicability of these smart materials.
improve the performance of optoelectronic devices.23,24
A
INTRODUCTION
■
further intriguing aspect of these materials is that certain
polysulfobetaines, predominantly the methacroyloxyethyl di-
methylammonio propanesulfonate25−27 and methacrylamido-
propyl dimethylammonio propanesulfonate26 derivatives, have
been shown to display an upper critical solution temperature
(UCST) in watera phenomenon known only for very few
types of water-soluble (co)polymers.28,29 Because of strong
intra- and interpolymer attractions (in polybetaines: electro-
static interactions), these hydrophilic polymers are insoluble in
water below a critical solution temperature but become soluble
when the temperature-weighed entropy of mixing (which favors
Polybetaineselectrically neutral polymers carrying positive
and negatives charges in every repeat unithave been known
for several decades1−6 and have been receiving increased
attention in recent years as a very promising class of materials.
An important subclass of polybetaines is constituted by
polysulfobetaines, featuring a sulfonate anion and typically a
quaternary ammonium cation, affording a largely pH-
independent zwitterionic character.7 A very tightly bound
hydration layer around each zwitterionic group has been
proposed to be the reason for the material’s exceptional
nonspecific protein absorbance and hemocompatibility,8 which
has been applied in antibiofouling surfaces,9−14 e.g. for
medicinal purposes,15−17 in separation science,18,19 as well as
for gene20 and drug21,22 delivery. Moreover, sulfobetaine-based
zwitterionic conjugated polyelectrolytes have been shown to
Received: November 18, 2013
Revised: January 12, 2014
Published: January 16, 2014
© 2014 American Chemical Society
750
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
a single phase) outbalances these enthalpic attractions. For
polysulfobetaines, this positive thermoresponsive behavior is
accompanied by an antipolyelectrolyte effect:26,27,30 addition of
electrolytes, e.g. sodium chloride, screens the charges of the
zwitterionic side chains from one another, reducing their
interactions, thereby increasing solubility (decreasing the
UCST). Exhibiting double responsiveness (temperature and
salt), the polysulfobetaine family offers great potential in the
smart materials arena, and several polysulfobetaine species have
been used to produce thermoresponsive gels,31 mechanically
tough, highly stretchable thermoresponsive nanocomposite
gels,32 schizophrenic diblock copolymers1,5,33 with temper-
ature-independent hemocompatibility,34 thermoresponsive hy-
brids containing biological materials,35 surfaces with thermo-
switchable wettability,36,37 and surfaces with switchable
wettability triggered by an electrical stimulus (exploiting the
antipolyelectrolyte effect).38
Hydrophobic modification of polysulfobetaines can therefore
be assumed to bring about the desired increase of UCST cloud
points. Because of the strongly polar nature of sulfobetaine
monomers, however, their copolymerization with hydrophobic
comonomers poses synthetic challenges. While many copoly-
mers of sulfobetaine monomers with hydrophilic comonomers
such as acrylamide,53−55 methacrylamide,56 N-isopropylacryl-
amide,31,57,58 N,N-dimethylacrylamide,18,32 and acrylic acid17,54
have been reported (and some of which were shown to
expectedlyhave increased solubility, i.e. lower UCST
transitions, than comparable homopolymers),31,57,59 the liter-
ature on hydrophobically modified sulfobetaine copolymers is
sparse. In a considerable synthetic effort, Koberle and
̈
Laschewsky60,61 prepared a series of acrylic and methacrylic
sulfobetaine monomers carrying alkyl substituents of varying
lengths on the quaternary nitrogen atom in varying geometrical
configurations. Several of the corresponding homopolymers,
including several propanesulfonate derivatives, were found to
be soluble only in (hot) water. Direct copolymerization of
sulfobetaine monomers with hydrophobic comonomers has
only been described for very few systems: free radical
polymerization with ethyl acrylate in ethanol (found to
precipitate most copolymers),62−64 or with vinyl acetate in an
aqueous emulsion polymerization,22,65 and with n-butyl
acrylate.66−70 The copolymerization behavior of latter system,
sulfobetaine monomer−n-butyl acrylate, has been investigated
by several research groups in ethanol (becomes heteroge-
neous),68 trifluoroethanol,69 and acetonitrile−water 96:4.70
Strehmel et al.70,71 found that using ionic liquids as solvents
significantly improved the copolymerization behavior of this
system, though isolation of pure zwitterionic homopolymers
from the ionic liquid was not possible.70 Different strategies
toward hydrophobically modified polysulfobetaines have
included copolymerization of a tertiary amine functional
monomer with n-butyl acrylate,72 ethyl acrylate,73 or acryloni-
trile19 followed by quaternization of the tertiary amines to give
sulfobetaine species, and copolymerization of styrene and N-
vinylpyrrolidone with the reactive monomer phthalimido
acrylate, which was converted, in two steps, into a tertiary
amine species and then a sulfobetaine by quaternization.74
While sulfobetaine copolymers comprising larger amounts of
N-vinylpyrrolidone,74 ethyl acrylate,64 n-butyl (meth)-
acrylate,66−68,72 and acrylonitrile19 were reported to be
insoluble in water, a temperature-dependent solubility or a
potential increase of a UCST transition was, however, not
investigated in these studies.
Though of great interest,28,29 the responsive behavior of
polysulfobetaine suffers from two major drawbacks, limiting
applicability: (i) As with most UCST systems, the phase
separation temperature depends strongly on the molar mass of
polymers, with transition temperatures decreasing with
decreasing molar masses.26,27 A recent study by Willcock et
al.39 demonstrated that for linear chains of poly-
(methacroyloxyethyl dimethylammonio propanesulfonate),
the most commonly described polysulfobetaine, relatively
high molar masses of Mw = 258 and 448 kg/mol were
necessary to reach cloud points (optically observed transition
temperatures) of 26 and 43 °C, respectively, with lower molar
mass samples displaying lower cloud points and a sample of Mw
= 29 kg/mol having no measurable cloud point. (ii) The
antipolyelectrolyte effect typically requires working under
strictly salt-free conditions in order to observe UCST
transitions. Because of these influences, UCSTs of polysulfo-
betaines of lower molar masses (e.g., <50 kg/mol) and/or in
solutions containing electrolytes may be impractically low or
may not be observable at all. In order to widen the applicability
of these “smart” materials, it would therefore be of considerable
interest to be able to tune the UCST, especially to increase the
phase separation temperature, so as to enable UCST transitions
for lower molar mass polysulfobetaines and in solutions
containing salts, e.g., physiological environment.
Tuning of aqueous UCST transitions was recently described
for polyacrylamide and poly(N-acryloyl glycinamide) for which
copolymerization with the (more) hydrophobic monomers
acrylonitrile and styrene or n-butyl acrylate, respectively, was
found to increase the phase separation temperatures.40
Hydrophobic modification of poly(methyl methacrylate)-
based copolymers,41 poly(ethylene glycol (meth)acrylate)-
based copolymers,42 and certain poly(2-alkyl-2-oxazoline)s43
has been successfully applied to increase their respective UCST
transition temperatures in ethanol−water mixtures. Likewise,
increasing the content of hydrophobic segments in polymers
displaying a lower critical solution temperature (LCST,
soluble−insoluble transition upon heating) is known to
decrease their solubility (observable through decreasing
LCSTs). This method is well-established in the literature, and
both postpolymerization modification44−48 and (azeotropic)
copolymerization49−52 have lent themselves perfectly to the
production of libraries of well-defined copolymers comprising
varying amounts of hydrophobic comonomer units displaying
tunable LCST transitions.
Herein, we present a simple one-step synthesis of
sulfobetaine copolymers incorporating different hydrophobic
comonomers through postpolymerization modification75,76 of
an activated ester precursor.77 We optimized reaction
conditions to convert poly(pentafluorophenyl acrylate), in
homogeneous solution, with a zwitterionic amine and with
mixtures of the zwitterionic amine and hydrophobic amines
producing well-defined sulfobetaine acrylamido copolymers
incorporating a controllable amount of hydrophobic como-
nomers. In a first description of this synthetic strategy, we
investigated the influence of hydrophobic comonomers on the
UCST phase behavior and found that while alkyl functional
comonomerscounterintuitivelyincreased solubility over a
wide composition range, aromatic comonomers brought about
the desired decreased solubility, enabling sharp, reproducible
UCST transitions of lower molar mass (∼30 kg/mol)
copolymers between 6 and 82 °C and in aqueous solutions
751
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
Poly(pentafluorophenyl acrylate), pPFPA. Monomer PFPA
(4.5 g, 18.9 mmol, 130 equiv), RAFT agent BPTC (35.2 mg, 0.145
mmol, 1 equiv), initiator AIBN (2.4 mg, 0.015 mmol, 0.1 equiv), and
acetonitrile (6 mL) were combined in a flask equipped with a stir bar.
The mixture was sealed with a rubber septum and purged with
nitrogen for 20 min before being placed into a preheated oil bath at 70
°C for 10 h. Polymerization was stopped by quenching the reaction
with liquid nitrogen. A sample (100 μL) was withdrawn, diluted with
CDCl3 (550 μL), and analyzed by 19F NMR spectroscopy which
indicated a monomer conversion of 90% by comparison of the signal
at −157.1 ppm (bs, polymer para-F) with the signal at −158.1 ppm (t,
monomer para-F). The polymer was isolated as a slightly yellow
powder (3.78 g, 84%) by two precipitations in methanol followed by
containing up to 76 mM sodium chloride. We envisage that
these findings will increase the scope of polysulfobetaines in the
smart materials arena, and setting the stage for contact angle
tuning of polysulfobetaine surfaces38 and for the incorporation
of zwitterionic segments into more complex polymeric
architectures78 will advance many of the aforementioned
applications.
EXPERIMENTAL SECTION
■
Materials. All reagents were purchased from Sigma-Aldrich and
were used as received unless stated otherwise. Propylene carbonate
(99.7%, anhydrous) and dimethylformamide (DMF, 99.8%, anhy-
drous) were stored in a glovebox. Azobis(isobutyronitrile) (AIBN)
was recrystallized from methanol and stored at −24 °C. Amberlyst
A26 anion exchange resin was washed with mQ water three times prior
to use. The synthesis of benzyl propyl trithiocarbonate (BPTC) was
described elsewhere.79
drying in vacuum. Mn,theor (from conversion) = 28.1 kg/mol, DPtheor
=
117, Mn,NMR (by end-group analysis) = 25.7 kg/mol, DPNMR = 107,
Mn,SEC (THF) = 13.8 kg/mol, ĐM = Mw,SEC/Mn,SEC = 1.40. 19F NMR
(282 MHz, CDCl3), δ/ppm = −153.2 (bm, 2 F, ortho), −157.1 (bs, 1
F, para), −162.4 (bs, 2 F, meta). FT-IR: ν/cm−1 = 1780 (carbonyl
CO stretch), 1520 (aryl CC bend).
Methods. NMR spectroscopic measurements in D2O and CDCl3
were performed on Bruker Avance 300 MHz and Bruker Avance III
500 MHz instruments in 5 mm NMR tubes. The internal solvent
signals were used as reference (δ(CDCl3) = 7.26 ppm, δ(D2O) = 4.79
ppm). Measurements on copolymers containing 35, 53, and 65 mol %
of benzylacrylamide were measured on D2O solutions containing 0.1,
0.2, and 0.5 M NaCl, respectively. A copolymer containing 85 mol %
of pentylacrylamide was measured in CDCl3 solution. NMR
spectroscopic measurements in trifluoroethanol-d3 were performed
on a Bruker Avance III 600 MHz instrument in 3 mm NMR tubes.
The internal solvent signal was used as reference (δ(CF3CD2OH) =
5.22 ppm).
Size exclusion chromatography (SEC) in tetrahydrofuran (THF)
was performed on a Shimadzu system with four 300 × 7.8 mm2 linear
phenogel columns (105, 104, 103, and 500 Å) operating at a flow rate of
1 mL/min. The system was calibrated with a series of narrow molar
mass distribution polystyrene (PS) standards with molar masses
ranging from 0.58 to 1820 kg/mol. Aqueous SEC was performed on a
Shimadzu system with two Agilent Aquagel columns with 0.1 M NaCl
solution containing 0.2 mass % sodium azide as eluent at a flow rate of
0.8 mL/min. This system was calibrated with a series of narrow molar
mass distribution poly(ethylene glycol) (PEG) standards. Chromato-
grams were analyzed by Cirrus SEC software version 3.0.
Fourier transform infrared spectroscopy (FT-IR) was performed on
a Bruker IFS 66/S instrument under attenuated total reflectance
(ATR), and data were analyzed on OPUS software version 4.0.
Inductively coupled plasma atomic emission spectroscopy (ICP-
OES) was performed on a PerkinElmer OPTIMA 7300 ICPOES
instrument after sample digestion in concentrated nitric acid.
Turbidity measurements were performed on a Varian Cary 300
Scan spectrophotometer equipped with a Cary temperature controller
and a Peltier heating element in quartz cuvettes of 10 mm path length
at a wavelength of 520 nm with heating/cooling rates of 1 °C/min.
Unless otherwise noted, polymer concentrations were 10 g/L. For
clear solutions the baseline was corrected to zero absorbance, A.
Transmittance, T = 10−A, was plotted against temperature, and cloud
points were determined at 50%.
tert-Butyl (3-(Dimethylamino)propyl)carbamate. was pre-
pared in analogy to a literature procedure.80 Briefly, N,N-
dimethylaminopropylamine (9.125 mL, 72.5 mmol) was dissolved in
methanol (38 mL), and the mixture was cooled to 0 °C. Di-tert-butyl
dicarbonate (15.0 g, 68.8 mmol) was added, and the mixture was
stirred overnight while warming to room temperature. After
evaporating the solvent, water (60 mL) was added, and the product
was extracted with ethyl acetate (4 × 70 mL) yielding 8.7 g (78%) of a
1
viscous oil. H NMR (300 MHz, CDCl3), δ/ppm = 3.02 (t, 2 H,
−CH2NH−), 2.27 (t, 2 H, (CH3)2NCH2−), 2.17 (s, 6 H,
(CH3)2N−), 1.61 (m, 2 H, −CH2CH2CH2−), 1.37 (s, 9 H,
−C(CH3)3). 13C NMR (75 MHz, CDCl3), δ/ppm = 156.6 (CO),
78.8 (−C(CH3)3), 56.5 ((CH3)2NCH2−), 44.4 ((CH3)2N−), 38.2
(−CH2NH−), 27.8 (−C(CH3)3), 27.0 (−CH2CH2CH2−).
3-((3-((tert-Butoxycarbonyl)amino)propyl)dimethyl-
ammonio)propane-1-sulfonate. tert-Butyl (3-(dimethylamino)-
propyl)carbamate (8.6 g, 42.5 mmol), propane sultone (7.3 g, 60
mmol), and anhydrous DMF (50 mL) were combined in a round-
bottom flask and stirred at room temperature for 5 days. The product
was precipitated into diethyl ether (300 mL) and isolated by
centrifugation and drying in vacuum. The product was used in the
1
next step without further purification. H NMR (300 MHz, D2O), δ/
ppm = 3.40, 3.28 (2 m, 2 × 2 H, −CH2N+(CH3)2CH2−), 3.11 (t, 2 H,
−NHCH2CH2CH2N+−), 3.03 (s, 6 H, −N+(CH3)2−), 2.91 (t, 2 H,
− C H 2 S O 3 − ) , 2 . 1 4 , 1 . 8 9 ( 2 m ,
2
×
2
H ,
−CH2CH2N+(CH3)2CH2CH2−), 1.36 (s, 9 H, (CH3)3O2CNH−).
3-((3-Aminopropyl)dimethylammonio)propane-1-sulfonate
Hydrochloride, ADPS·HCl. The product of the previous step, 3-((3-
((tert-butoxycarbonyl)amino)propyl)dimethylammonio)propane-1-
sulfonate, was dissolved in mQ water (32 mL) in a beaker, and
concentrated hydrochloric acid (15 mL) was added. The mixture was
stirred overnight, and water and excess acid were evaporated. The
1
product was recrystallized from water−ethanol. H NMR (300 MHz,
D2O), δ/ppm = 3.55, 3.49 (2 m, 2 × 2 H, −CH2N+(CH3)2CH2−),
3.18 (s, 6 H, −N+(CH3)2−), 3.13 (t, 2 H, H3N+CH2−), 3.03 (t, 2 H,
−
−CH2SO3 ), 2.56 (m, 4 H, −CH2CH2N+(CH3)2CH2CH2−). 13C
NMR (75 MHz, D2 O), δ/ppm
= 62.4, 60.6 (2 t,
−CH2N+(CH3)2CH2−), 50.8 (t, −N+(CH3)2−), 47.1 (−CH2SO3 ),
−
Dynamic light scattering (DLS) measurements were performed on a
Malvern Zetasizer Nano ZS at a scattering angle of 173° and were
analyzed by Malvern Zetasizer Software version 6.20.
−
36.3 (H3N+CH2−), 20.5 (H3N+CH2CH2−), 18.1 (−CH2CH2SO3 ).
3-((3-Aminopropyl)dimethylammonio)propane-1-sulfonate,
ADPS. The corresponding hydrochloride, ADPS·HCl, was dissolved in
mQ water and treated with Amberlyst A26 anion exchange resin until
a pH of ∼11 was reached. The resin was filtered off, and the aqueous
solution was freeze-dried. Yield based on tert-butyl (3-
(dimethylamino)propyl)carbamate (product of step 1) = 61%. 1H
NMR (300 MHz, D2O), δ/ppm = 3.52, 3.42 (2 m, 2 × 2 H,
−CH2N+(CH3)2CH2−), 3.16 (s, 6 H, −N+(CH3)2−), 3.03 (t, 2 H,
−CH2SO3−), 2.75 (t, 2 H, H2NCH2−), 2.26 (m, 2 H,
Pentafluorophenyl Acrylate, PFPA. Pentafluorophenol (18.4 g,
100 mmol), triethylamine (14.6 mL, 105 mmol), and dichloromethane
(500 mL) were mixed and cooled to 0 °C. Acryloyl chloride (8.94 mL,
110 mmol) was added dropwise, and the mixture was stirred and let
warm to room temperature overnight. 18.9 g (80%) of product was
obtained after washing the organic phase with water (5 × 400 mL). 1H
NMR (300 MHz, CDCl3), δ/ppm = 6.67 (dd, 1 H), 6.31 (dd, 1 H),
6.12 (dd, 1 H). 13C NMR (75 MHz, CDCl3), δ/ppm = 161.8 (CO),
1
−
142.9, 141.1, 139.8, 137.6, 135.6 (5 m (large JC−F coupling), 5 F),
−CH2CH2SO3 ), 1.96 (m, 2 H, NH2CH2CH2−). 13C NMR (75
135.6, 125.4 (CC). 19F NMR (282 MHz, CDCl3), δ/ppm = −152.7
MHz, D2O), δ/ppm = 62.3, (m, 2 C,−CH2N+(CH3)2CH2−), 50.6 (t,
−
(m, 2 F, ortho), −158.1 (t, 1 F, para), −162.5 (m, 2 F, meta).
2 C,−N+(CH3)2−), 47.3 (1 C, −CH2SO3 ), 37.6 (1 C, H2NCH2−),
752
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
−
25.1 (1 C, H2NCH2CH2−), 18.1 (1 C, −CH2CH2SO3 ). FT-IR: ν/
cm−1 = 3350 (amine N−H stretch), 2990−2870 (C−H stretch), 1490
(CH2 bend), 1210 (C−N stretch), overlapping with 1170 (SO
stretch), 1035 (SO stretch).
comparison of the aromatic peaks (5 H) to the sum of the zwitterionic
group signals from 3.75−2.65 (12 H).
RESULTS AND DISCUSSION
■
Poly[(3-((3-acrylamidopropyl)dimethylammonio)propane-
1-sulfonate)], pADPS. The target homopolymer was prepared as
follows. pPFPA (44.6 mg, 0.188 mmol of PFP units, 1 equiv) was
dissolved in propylene carbonate (1.5 mL) at 60 °C, and hydroxyethyl
acrylate (5 μL) was added. In parallel, zwitterionic amine ADPS (63.1
mg, 0.281 mmol, 1.5 equiv) was dissolved in propylene carbonate (1.5
mL) at 60 °C. After dissolving, the amine solution was quickly added
into the polymer solution, and the mixture was stirred at 40 °C
overnight. The solution was transferred into a dialysis bag (MWCO
3500 Da) and dialyzed against mQ water for 3 days. The product (48.2
mg, 92%) was isolated by freeze-drying as a fluffy white solid and was
Synthesis of Zwitterionic Amine. In order to incorporate
sulfobetaine functionality into an activated ester precursor, the
zwitterionic amine 3-((3-aminopropyl)dimethylammonio)-
propane-1-sulfonate, ADPS, was synthesized from N,N-
dimethylaminopropylamine in three synthetic stepsBOC
protection of the primary amine, quaternization with propane
sultone, and BOC deprotectionfollowed by treatment with
hydroxide loaded anion exchange resin to yield the free base
1
(Scheme 1). H and 13C NMR spectra of ADPS hydrochloride
1
kept in a tightly sealed container to avoid moisture uptake. H NMR
Scheme 1. Synthesis of Amine-Functional Sulfobetaine
ADPS
−
(500 MHz, D2O), δ/ppm = 3.53 (bs, 2 H, −CH2CH2CH2SO3 ), 3.40
(bs, 2 H, −NHCH2CH2CH2−), 3.27 (bs, 2 H, −NHCH2−), 3.16 (bs,
6 H, −N+(CH3)2−), 3.01 (bt, 2 H, −CH2SO3−), 2.24 (bs,
−CH2CH2SO3−), 2.04 (bs, −NHCH2CH2−) overlapping 2.11
(backbone −CH2CHR−), 1.74, 1.64, 1.50 (backbone −CH2CHR−).
13C NMR (75 MHz, D2O), δ/ppm = 176.6 (−CONH−), 62.1,
(−CH2 N+ (CH3 )2 CH2 −), 61.6 (−CONHCH2 −), 50.9
−
(−N+(CH3)2−), 47.4 (−CH2SO3 ), 42.4 (backbone −CH2CHR−),
36.4 (backbone −CH2CHR−), 22.1 (−CONHCH2CH2−), 18.2 (1 C,
−
−CH2CH2SO3 ). FT-IR: ν/cm−1 = 1650 (amide CO stretch), 1550
(amide N−H bend), 1170 (side chain C−N stretch and SO
stretch), 1035 (SO stretch).
Statistical Copolymers p(ADPS-co-pentylacrylamide), p-
(ADPS-co-benzylacrylamide), and p(ADPS-co-dodecylacryla-
mide) (p(ADPSx-co-Penty), p(ADPSx-co-Bzy), and p(ADPSx-co-
Dodecy)). A general procedure is given. pPFPA (44.6 mg, 0.188 mmol
of PFP units, 1 equiv) was dissolved in propylene carbonate (1.5 mL)
at 60 °C, and hydroxyethyl acrylate (5 μL) was added. In parallel,
zwitterionic amine ADPS (x × 0.281 mmol) and a hydrophobic amine
(pentylamine, benzylamine, or dodecylamine) (y × 0.281 mmol, x + y
= 1; 1.5 equiv of amines to PFP ester) were dissolved in propylene
carbonate (1.5 mL) at 60 °C. After dissolving, the amine solution was
quickly added into the polymer solution, and the mixture was stirred at
40 °C overnight. Purification: Pentylacrylamide copolymers p(ADPSx-
co-Penty) with y ≤ 0.65 and benzylacrylamide copolymers p(ADPSx-
co-Bzy) with y ≤ 0.25 were purified by dialysis (MWCO 3500 Da)
against mQ water for 3 days, followed by freeze-drying. Copolymer
p(ADPS0.15-co-Pent0.85) containing 85 mol % of pentylacrylamide
comonomer units was purified by two precipitations from propylene
carbonate into water, followed by centrifugation, washing with water,
and drying. Benzylacrylamide copolymers p(ADPSx-co-Bzy) with y >
0.25 were purified by dialysis (MWCO 3500 Da) in warm (up to 70
°C) water in a beaker heated by a hot plate and equipped with a stirrer
bar and temperature controller and closed with an inverted plastic
funnel connected to a reflux condenser. Samples were isolated by
freeze-drying. Elemental analysis by inductively coupled plasma atomic
emission spectroscopy (ICP-OES) run on select samples confirmed a
very low concentration of sodium (below 0.01 mass %) and potassium
(below detection limit) after purification by (warm) dialysis.
Dodecylacrylamide copolymer p(ADPS0.60-co-Dodec0.40) was purified
by dialysis in ethanol and then dialysis in water, during which the
polymer precipitated inside the dialysis bag. Yields of all copolymers
were typically between 85 and 90%. 1H NMR (300 MHz, D2O, pentyl
side chains), δ/ppm = 1.54 (−NHCH2CH2CH2CH2CH3), 1.34
(−NHCH2CH2CH2CH2CH3), 0.93 (−NHCH2CH2CH2 CH2CH3).
1H NMR (300 MHz, D2O, benzyl side groups, 0−0.5 M NaCl), δ/
and ADPS are shown in Figures S1−S4 of the Supporting
Information. ADPS was found to be soluble in water, acetic
acid, trifluoroethanol, warm dimethylformamide (DMF), warm
dimethyl sulfoxide (DMSO), warm propylene carbonate (PC),
and warm N-methylpyrrolidone (NMP) and was insoluble in
methanol, ethanol, 2-propanol, 1,8-diazabicyclo[5.4.0]undec-7-
ene (DBU), pyridine, triethylamine, and dichloromethane.
Synthesis of Sulfobetaine Homopolymer. With the
zwitterionic amine ADPS in hand, we investigated its reaction
with an activated polyacrylate precursor with the aim of
producing the sulfobetaine homopolymer poly[(3-((3-
acrylamidopropyl)dimethylammonio)propane-1-sulfonate],
pADPS (Scheme 2, top). Based on the poor solubility of the
zwitterionic amine in solvents of medium polarity and the
highly polar nature of the target polysulfobetaine product, the
use of a water-soluble activated ester precursor, such as the
(meth)acrylate derivatives of N-hydroxysulfosuccinimide so-
dium salt, sodium 4-hydroxy-3-nitrobenzenesulfonate, or
sodium 2,3,5,6-tetrafluoro-4-hydroxybenzenesulfonate,81 in
aqueous solution would have seemed reasonable. Because of
the poor availability and higher cost of these materials, however,
we sought to employ a common, well-established activated
ester, poly(pentafluorophenyl acrylate), pPFPA, in spite of its
apparent hydrophobicity. pPFPA with a degree of polymer-
ization of 117, a molar mass of Mn,theor = 28.1 kg/mol, and a
polydispersity index of 1.40 was prepared by RAFT polymer-
ization82 using a trithiocarbonate chain transfer agent. This
reactive precursor was insoluble in water, alcohols (including
longer chain alcohols), and DMSO, dissolved in warm DMF
and warm propylene carbonate, while it was well-soluble in
solvents of lower polarity, e.g. THF and chloroform. We
initially explored the reaction of pPFPA with ADPS in DMF−
ppm = 7.65, 7.53 (m, −CH2C6H5), 4.45 (−CH2C6H5). 1H NMR (300
MHz, trifluoroethanol-d3, dodecyl side chains), δ/ppm = 1.57
(−NHCH2CH2−), 1.35 (−CH2−), 0.94 (−CH3). Molar compositions
of pentylacrylamide and dodecylacrylamide copolymers were
determined by comparison of the alkyl −CH3 signal (0.9 ppm, 3 H)
to the broad triplet of the −CH2SO3− (3.0 ppm, 2 H) segment. Molar
compositions of the benzylacrylamide copolymers were determined by
753
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
Scheme 2. Synthesis of Hydrophobically Modified Sulfobetaine (Co)polymers in Propylene Carbonate
water, adding an aqueous solution of ADPS into a DMF
solution of pPFPA. Precipitation occurred within seconds, and
a polymeric material was isolated by dialysis in water. FT-IR
analysis (not shown) revealed hydrolysis of pentafluorophenyl
(PFP) esters as a main reaction pathway yielding a poorly
defined sulfobetaine−acrylic acid copolymer that was not
further analyzed. Further experiments involving water or
trifluoroethanol likewise resulted in solvolysis of the activated
ester precursor, suggesting protic solvent systems to be less
ideal and implying a need for optimization. Further experiments
revealed propylene carbonate to be an expedient solvent for
this problem. A highly polar aprotic solvent, propylene
carbonate, has previously been employed in sulfobetaine
synthesis,3,83 and we found it to enable a homogeneous
reaction between pPFPA and ADPS. Reactions were carried out
in anhydrous propylene carbonate (to avoid hydrolysis) by
stirring overnight at 40 °C with a 1.5-fold excess of amines to
activated ester groups. A small amount of hydroxyethyl acrylate
was added to scavenge any thiols released by aminolysis of the
trithiocarbonate RAFT end groups in a thiol−Michael addition
reaction.84 Products were isolated by excessive dialysis in
ultrapure water and subsequent lyophilization. The success of
the reaction was confirmed by 1H, 13C, and 19F NMR
spectroscopy, FT-IR spectroscopy, and size exclusion chroma-
tography.
A 19F NMR spectroscopic measurement of a reaction sample
withdrawn before purification showed only the sharp signals of
released pentafluorophenol and complete disappearance of the
broad PFP ester signals of the pPFPA precursor, indicating
complete reaction of the PFP esters and suggesting a
quantitative integration of sulfobetaine side groups into the
backbone (Figure 1).
FT-IR measurements were performed of the pPFPA
precursor, the ADPS reagent and the sulfobetaine (co)polymer
products. Representative curves are shown in Figure 2. The
spectrum of the pPFPA precursor showed the characteristic
754
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
Figure 1. 19F NMR spectra of pPFPA (A), after reaction with ADPS
before purification showing the signals of free pentafluorophenol (B),
and of a sulfobetaine polymer after purification indicating complete
removal of pentafluorophenol by dialysis (C).
Figure 3. 1H NMR spectra of homopolymer pADPS (500 MHz, D2O)
and of pentylacrylamide copolymers p(ADPSx-co-Penty) (300 MHz,
D2O) with peak assignments.
Figure 2. FT-IR spectra of the pPFPA precursor (top), the ADPS
reagent (middle), and the pADPS homopolymer product (bottom).
The characteristic bands of the activated ester CO stretch (gray),
the amide CO stretch (red), and the C−N stretch (left green) and
SO stretch (two bands, one overlapping with C−N, green) are
marked.
Figure 3), suggesting successful preparation of pADPS.
Noteworthy, the signals originating from groups furthest from
the backbone showed best resolution, with the methylene
group adjacent to the sulfonate (marked i in Figure 3)
appearing as a broad triplet, while groups closer to the
backbone, most notably the methylene segment adjacent to the
amide functionality (denoted c in Figure 3), gave rise to
broader signals. This was interpreted to reflect a somewhat
poorer hydration around the hydrophobic backbone. 13C NMR
spectroscopy (chemical shifts given in the Experimental
Section) likewise conformed to the targeted structure.
Size exclusion chromatograms of pPFPA (measured in THF,
polystyrene standard) and pADPS (measured in 0.1 M aqueous
NaCl, poly(ethylene glycol) standard) are shown in Figure 4
(bottom). Both curves were of very similar shape, being
monomodal with light tailing toward lower molar masses and
had similar polydispersity indices (both 1.40, see Table 1), as to
be expected from a quantitative postpolymerization modifica-
tion reaction. Coincidentally, both polymers had comparable
apparent molar masses, in spite of the different calibrations
used. Additionally, the curve of pADPS did not show any
shoulder toward higher molar masses, suggesting that unwanted
thiol−thiol coupling reactions following an aminolysis of RAFT
end groups had not occurred.84
absorption bands of the activated carbonyl group and the
perfluorinated aromatic at 1780 and 1515 cm−1, respectively.
Strong bands in the spectrum of ADPS were assigned to C−N
stretching vibrations of the quaternary ammonium group (1210
and 1170 cm−1) and the SO stretching vibrations of the
sulfonate groups (1170 cm−1, overlapping with C−N band, and
1035 cm−1). Most characteristically, a spectrum of the product
pADPS showed amide bands at 1650 cm−1 (CO stretch) and
1550 cm−1 (N−H bend), with no remaining signals of the
activated ester carbonyl group discernible, suggesting a
complete conversion of esters into amides. Additionally, signals
assigned to the ammonium and sulfonate groups (see above)
were clearly visible, indicating the presence of zwitterionic
groups in the polymeric product. An absorbance ∼3400 cm−1
was assigned to traces of water due to the hygroscopic nature of
the zwitterionic product.
Further evidence of complete conversion of the pPFPA
precursor and conversion into the target polysulfobetaine
pADPS was obtained by NMR spectroscopic measurements. A
1
section of a H NMR spectrum of pADPS in D2O solution is
shown in Figure 3 (bottom). The observed signals and their
integration conformed to the expected structure (as assigned in
Synthesis of Sulfobetaine Copolymers. After successful
synthesis of the sulfobetaine homopolymer pADPS by way of
755
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
dialysis in water. The resulting copolymers of the series
containing pentylacrylamide and benzylacrylamide comono-
mers are denoted p(ADPSx-co-Penty) and p(ADPSx-co-Bzy),
respectively, where y defines the molar ratio of hydrophobic
1
comonomers, as determined by H NMR spectroscopy. For
most copolymers 1H NMR analysis was performed on solutions
in D2O and confirmed on solutions in trifluoroethanol-d3
(TFE-d3), with excellent agreement between both measure-
1
ments (see Table 1). H NMR spectra of the p(ADPSx-co-
Penty) series measured in D2O are plotted in Figure 3 showing
the characteristic signals of the zwitterionic and the pentyl side
1
chains with ratios changing with molar composition. H NMR
spectra of both series measured on TFE-d3 solutions are shown
in Figures S5 and S6 of the Supporting Information. Simple
2
linear regression (not shown) with high correlation (Rpent
=
2
0.986; RBz = 0.993) between the molar percentage of
hydrophobic amines in feed and the molar percentage of
hydrophobic amide comonomer units in the products (from 1H
NMR spectroscopy in D2O) was found for both series,
signifying the high amount of control over copolymer
composition that this method offers. Postpolymerization
modification with mixtures of reagents is generally considered
to yield statistical copolymers,83 though we note that recent
research on effects of neighboring groups to cause nonstatistical
incorporation of functionality is limited. As thermoresponsive
copolymers prepared herein displayed sharp transitions without
evidence of micellization (vide infra), we excluded the presence
of block-like segments within the products and assumed a
largely statistical distribution of hydrophobic and zwitterionic
side chains along the copolymer backbones. SEC was measured
of copolymers containing 4, 8, and 20 mol % of
pentylacrylamide and of those containing 5, 9, and 25 mol %
of benzylacrylamide in 0.1 M aqueous NaCl solution (see
Figure 4). Very similar shapes, similar apparent molar masses,
and similar polydispersity indices (see Table 1) were found for
all copolymers, indicating that the characteristic molar mass
distribution of the pPFPA precursor was retained in the
modified copolymers. We noted, however, a more pronounced
Figure 4. SEC curves of pPFPA precursor (eluent THF, polystyrene
standard, bottom curves), and sulfobetaine homo- and copolymers
(eluent 0.1 M aqueous NaCl, poly(ethylene glycol) standard).
postpolymerization modification of pPFPA in propylene
carbonate, we next took advantage of this synthetic approach
to prepare a library of statistical hydrophobically modified
sulfobetaine copolymers. Expecting different behavior of alkyl
versus aromatic functionality within a zwitterionic copolymer,
we first chose pentylamine and benzylamine as hydrophobic
modifiers. Reactions were carried out by adding mixtures of
benzylamine (5−60 mol % of total amines in feed) or
pentylamine (5−85 mol % of total amines in feed) and
ADPS dissolved in propylene carbonate to solutions of pPFPA
in propylene carbonate, followed by stirring overnight and
Table 1. Homo- and Copolymers Derived from pPFPA (DP = 117, ĐM = Mw/Mn = 1.40) with ADPS and Hydrophobic
Coamines
coamine
a
b
c
d
e
code
type
feed (mol %)
found (mol %)
Mn,theor (kg/mol)
ĐM
cloud point (°C)
pADPS
none
0
5
0 (D2O)
34.5
33.9
33.1
31.3
28.2
24.0
20.9
33.7
33.2
31.1
29.7
27.2
25.5
32.6
1.40
1.41
1.42
1.44
8.5
2.5
p(ADPS0.96-co-Pent0.04
p(ADPS0.92-co-Pent0.08
p(ADPS0.80-co-Pent0.20
p(ADPS0.61-co-Pent0.39
p(ADPS0.35-co-Pent0.65
p(ADPS0.15-co-Pent0.85
)
)
)
)
)
)
pentyl
pentyl
pentyl
pentyl
pentyl
pentyl
benzyl
benzyl
benzyl
benzyl
benzyl
benzyl
dodecyl
3.7 (D2O); 3.8 (TFE-d3)
g
10
20
30
60
85
5
8.3 (D2O); 8.3 (TFE-d3)
s
19.7 (D2O); 19.3 (TFE-d3)
38.7 (D2O); 37.7 (TFE-d3)
65.3 (D2O); 70.7 (TFE-d3)
84.6 (CDCl3)
s
s
s
f
n.d.
n.d.
n.d.
1.38
1.40
1.47
n.d.
n.d.
n.d.
n.d.
h
ins
p(ADPS0.95-co-Bz0.05
p(ADPS0.91-co-Bz0.09
p(ADPS0.75-co-Bz0.25
p(ADPS0.65-co-Bz0.35
p(ADPS0.47-co-Bz0.53
p(ADPS0.35-co-Bz0.65
)
)
)
)
)
)
5.2 (D2O); 4.6 (TFE-d3)
7.2
10
20
30
45
60
33
9.2 (D2O); 8.2 (TFE-d3)
6.4
24.7 (D2O); 23.2 (TFE-d3)
34.8 (D2O, 0.1 M NaCl); 36.8 (TFE-d3)
52.8 (D2O, 0.2 M NaCl); 53.5 (TFE-d3)
65.3 (D2O, 0.5 M NaCl); 72.5 (TFE-d3)
34.3 (TFE-d3)
14.5
32.2
63.6
82.4
ins
p(ADPS0.60-co-Dodec0.34
)
a
b
Amount of coamine in reaction. Amount of coacrylamide in copolymer determined by 1H NMR spectroscopy, solvent given in parentheses; TFE
c
= trifluoroethanol. Molar mass of copolymer calculated from DP of precursor and copolymer composition from NMR assuming full conversion.
d
e
f
Polydispersity index determined by aqueous size exclusion chromatography. Cloud point measured in mQ water at a concentration of 10 g/L. Not
g
h
determined. Soluble between 0 and 90 °C at a concentration of 10 g/L. Insoluble between 0 and 90 °C at a concentration of 10 g/L.
756
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
Figure 5. Influence of hydrophobe type and content on aqueous solution properties of sulfobetaine copolymers. (A) Turbidity curves for
homopolymer pADPS and benzylacrylamide copolymer series. Solid lines are cooling curves, and dashed lines are heating curves. (B) Phase diagram
of cloud point vs hydrophobe content.
tailing toward lower molar masses for the measured copolymers
containing the highest amount of hydrophobic groups, i.e.,
p(ADPS0.80-co-Pent0.20) and p(ADPS0.75-co-Bz0.25), resulting in
slightly higher polydispersity indices (ĐM = 1.44 and 1.47,
respectively), possibly due to interactions with the column
material, accompanied by an overall shift toward lower apparent
molar masses, reflecting lower absolute molar masses of
copolymers with higher hydrophobe content. Absolute molar
masses of the (co)polymers were estimated from the degree of
polymerization of the pPFPA precursor and from molar
bag, indicating insolubility in both solvents. The dried
polymeric material was found to be insoluble in chloroform,
to dissolve partially in DMSO−water 5:1, and to dissolve well
1
in trifluoroacetic acid−H2O 10:1 and trifluoroethanol-d3. A H
NMR spectrum measured of the latter solution conformed to a
sulfobetaine copolymer containing 34 mol % of dodecylacryl-
amide. Though no signals of lower molar mass impurities were
discernible, the purity of this product was, however,
questionable, due to precipitation during dialysis. While a
copolymer containing both zwitterionic and dodecyl side chains
may have limited applicability due to poor solvent compati-
bility, our observations nevertheless indicated that the synthetic
method presented here allows also for the preparation of
copolymers with strongly disparate side chains including larger
hydrophobic groups.
1
compositions determined by H NMR spectroscopy. Values,
given in Table 1, ranged from 20.9 kg/mol for p(ADPS0.15-co-
Pent0.85) to 34.5 kg/mol for the pADPS homopolymer.
With sulfobetaine copolymers containing varying (small)
amounts of pentyl- and benzylacrylamide successfully prepared
through postpolymerization modification of pPFPA in
propylene carbonate, we additionally investigated the scope of
this synthetic method to produce two further copolymer
species: (i) Hydrophobic copolymers containing minor
amounts of zwitterionic side chains. Combining improved
antifouling properties imparted by the zwitterionic segments
with the mechanical properties of a hydrophobic matrix, such
materials have recently gained interest in the production of
fibers67 and ultrafiltration membranes.19,66 We thus prepared
p(ADPS0.15-co-Pent0.85) by the same method as described
Aqueous Phase Behavior of Hydrophobically Modi-
fied Sulfobetaine (Co)polymers. The homopolymer pADPS
dissolved quickly in ultrapure water at room temperature and
showed a cloud point of 8.5 °C at a concentration of 10 g/L
(see Table 1 and black curve in Figure 5A). This low cloud
point was in agreement with cloud points measured for similar
(methacrylato)sulfobetaine copolymers,39 taking into account
that the molar mass of pADPS of 34.5 kg/mol was considerably
lower than that of literature examples.26,27,31,85 With post-
polymerization modification producing (co)polymers with
identical degrees of polymerization (and thus comparable
molar masses), the p(ADPSx-co-Pent/Bzy) series presented an
ideal platform to investigate the influence of statistically
distributed hydrophobic moieties on the phase behavior with
the goal of increasing the UCST transition to a more practical
range while keeping the main chain degree of polymerization
constant. The aqueous solution behavior of sulfobetaine
copolymers containing between 4−65 mol % of pentylacryl-
amide and 5−65 mol % of benzylacrylamide comonomers was
investigated by turbidity (at a concentration of 10 g/L) and
dynamic light scattering. Consider first the pentylacrylamide
series. p(ADPS0.94-co-Pent0.04) had a cloud point of 2.5 °C, 6 °C
lower than the homopolymer (see Table 1 and Figure 5B).
p(ADPS0.92-co-Pent0.08) was soluble in water upon cooling to 0
°C. However, below 2 °C the transmittance of this sample
decreased very slightly to 98.5%, suggesting the presence of a
theoretical cloud point below 0 °C (see Figure S7 in
Supporting Information). These observations indicated that,
surprisingly, incorporation of pentyl side chains into poly-
sulfobetaines increased their solubility, in contrast to the well-
1
above, using precipitation into water for purification. H NMR
spectroscopy in CDCl3 confirmed successful synthesis and the
targeted molar composition (see Table 1). The copolymer was
soluble in methanol, ethanol, propylene carbonate, THF, and
chloroform and was insoluble in water (swelling slightly after
prolonged heating to 95 °C and also when cooled to 0 °C),
acetone, and hexane. These observations suggested successful
synthesis of a hydrophobic copolymer comprising zwitterionic
functionality using our protocol. (ii) Copolymers containing
larger hydrophobic groups. In order to investigate whether the
synthesis would also allow for the production of copolymers
containing larger hydrophobic groups, the synthesis of a
copolymer containing 33 mol % of dodecylacrylamide was
investigated (see Scheme 2, bottom). After adding a mixture of
ADPS and dodecylamine dissolved in propylene carbonate to a
solution of pPFPA in propylene carbonate, clouding (but no
precipitation) occurred, suggesting limited solubility of the
copolymer product in propylene carbonate. The reaction
mixture was then subjected to dialysis in ethanol and then
water. In both cases, precipitation occurred inside the dialysis
757
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
Figure 6. Concentration influence on the cloud point: (A) turbidity curves of p(ADPS0.65-co-Bz0.35) in water at varying concentration and (B)
temperature−concentration phase diagrams of three (co)polymers showing a largely flat region above concentrations of ∼10 g/L.
established effect of hydrophobic modification to decrease the
solubility of polymers with an LCST in water.44−52 Copolymers
containing up to 65 mol % of pentylacrylamide comonomers
were found to be soluble in water over the entire temperature
range between 0 and 90 °C, showing an increased solubility
over a wide composition range. As mentioned above, the sister
copolymer comprising 85 mol % of pentylacrylamide was found
to be insoluble over the entire investigated temperature range,
with thermal phase behavior possibly occurring at molar
compositions between these two values.
micellar structures with the hydrophobic chains segregated in
the cores. Previous research on LCST-type polymers
demonstrated a reduced influence of hydrophobic end groups
on lowering the solubility if the end groups self-assembled and
were thus not exposed to the solvent.47 In the current case,
however, there was no evidence of such self-assembly. Notably,
1
the H NMR signals of the pentyl side groups were quite well
resolved in D2O (Figure 3), showing solvation of the
hydrophobic groups. An excellent agreement between molar
compositions determined through NMR analysis in D2O and
TFE-d3 (and the feed ratios of pentyl vs zwitterionic groups,
Table 1) suggested that the pentyl groups were not segregated
in D2O solution. Additionally, DLS of pentylacrylamide
copolymers in aqueous solution gave no evidence of self-
assembled structures, suggesting unimerically dissolved copoly-
mers (see Figure S9 in the Supporting Information). With no
previous systematic studies on UCST behavior of hydrophobi-
cally modified sulfobetaine copolymers available, we propose
the following explanation. (i) Thermodynamic considerations.
As mentioned in the Introduction, UCST behavior relies on a
balance between enthalpic polymer−polymer interactions
favoring phase separation and mixing entropy favoring
dissolution. Entropic and enthalpic contributions from the
incorporated pentyl and benzyl side groups can thus be
expected have an influence on the phase separation temper-
ature. In a recent study on entropic trends during hydration of
common functional groups,86 the authors described that rigid
molecules (such as benzyl groups) cause a higher loss of water
orientational entropy during hydrophobic hydration compared
to flexible groups, while alkyl chains have higher rotational
entropy and can, in fact, experience an increase in internal
entropy through hydration by stabilization of higher energy
gauche transformations in solution. This would suggest a higher
gain in entropy during the dissolution of a pentyl-functional
copolymer compared to a similar benzyl-functional copolymer,
thus making the pentyl-modified species more soluble than the
corresponding benzyl analogue. It should also be considered
that the aromatic benzyl groups have the potential to form
attractive interactions with each other (π−π),87 as well as with
the ammonium groups (cation−π),88 and, though less likely,
the sulfonate groups (anion−π).89 Such enthalpic interac-
tionsgranted, their experimental observation may be
difficultwould likewise suggest a decreased solubility of the
benzyl-functional polymers compared to their pentyl counter-
A very different behavior was observed for the benzylacryl-
amide copolymer seriesturbidity measurements are shown in
Figure 5A, cloud points are listed in Table 1, and a plot of cloud
points vs benzylacrylamide content is presented in Figure 5B.
Incorporation of small amounts of benzylacrylamide, i.e., 5 and
9 mol %, were found to slightly decrease the transition
temperature, resulting in cloud points of 7.2 and 6.4 °C,
respectively. Higher content of benzyl groups, however, caused
the UCST transition temperature to increase, as intended, with
cloud points increasing in an almost linear fashion from 14.5 °C
found for p(ADPS0.75-co-Bz0.25) to 82.4 °C observed for
p(ADPS0.35-co-Bz0.65) (Figure 5B). Of note, a similar trend
slightly increased solubility at low hydrophobe contentwas
found by Seuring et al.40 for the variation of UCST cloud points
of poly(N-acryloylglycinamide) with increasing butyl acrylate
content. Transitions observed for benzylacrylamide−sulfobe-
taine copolymers occurred sharply and were fully reversible and
reproducible, with a hysteresis of ∼7 °C between cooling and
heating curves observed at cooling/heating rates of 1 °C/min
(see Figure S8 in Supporting Information). For the sample of
highest benzyl content, 65 mol %, complete redissolution was
slow during heating, with small aggregates remaining in an
otherwise clear solution until the sample was annealed to 90 °C
(∼8 °C above the cloud point) for 15 min, after which a
reproducible cooling curve was obtained.
The difference between benzyl and pentyl side chains within
pADPS was striking. While benzylacrylamide units expectedly
decreased solubilitythat is, above a content of ∼9 mol %,
incorporation of pentyl groups increased the solubility of
sulfobetaine copolymers over a composition range from 0 to at
least 65 mol %, though both acrylamide units are hydrophobic,
their respective homopolymers polypentylacrylamide and
polybenzylacrylamide being insoluble in water. It is conceivable
that the pentylacrylamide copolymers might self-assemble into
758
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
parts. (ii) Kinetic considerations. Insolubility of sulfobetaine
(co)polymers at low temperature and in the absence of
electrolytes is based on an interlocking of zwitterionic side
groups of different polymer chains, creating a semicrystalline
structure. The presence of flexible pentyl chains can be assumed
to sterically interrupt this interlocking to a higher extent than
the presence of rigid benzyl groups, thereby reducing the
crystallinity and increasing the solubility of pentyl-functional
copolymers. We therefore hypothesize that incorporation of
any rigid hydrophobic functionality into sulfobetaine copoly-
mers will decrease their solubility, allowing for a tuning of their
UCST cloud points. If, on the other hand, temperature-
independent water solubility of a hydrophobically modified
sulfobetaine copolymer is desired, flexible functionality that
does not contribute to polymer−polymer attractions would
appear to be the appropriate choice.
Having successfully increased the UCST of polysulfobetaines
enabling tunable, sharp transitions of copolymers with molar
masses ranging from 25.5 to 34.5 kg/mol (see Table 1), we
next investigated the phase behavior of the p(ADPSx-co-Bzy)
series in more detail. Temperature−composition phase
diagrams were measured for three (co)polymers with benzyl
contents of 0, 35, and 53 mol % (see Figure 6). With very low
transition temperatures, the homopolymer pADPS only had
measurable cloud points above a concentration of 4 g/L, while
the benzyl-modified copolymers with higher transition temper-
atures showed the expected cloud point decrease with
decreasing concentration, measured down to 0.5 g/L,
demonstrating again the benefit of hydrophobic modification.
An “inverted LCST” around 10 g/L as previously described for
similar systems27 was not observed for the present (co)-
polymers. The phase diagrams had largely flat regions above a
concentration of ∼10 g/L, offering a reliable phase behavior
above this concentration. The concentration coordinate of the
UCSTs (maxima of the binodal curves) decreased with
increasing benzyl content lying at or above 24 g/L for the
homopolymer, at ∼16 g/L for p(ADPS0.65-co-Bz0.35) and at ∼10
g/L for p(ADPS0.47-co-Bz0.53).
temperature-dependent derived count rate of a sample of
p(ADPS0.47-co-Bz0.53) in water at a concentration of 1.5 g/L is
shown together with the optical transmittance of the same
sample. By turbidity, a cloud point (50% transmittance) of 44.9
°C was found with an onset of transmittance decrease just
around 55 °C. By DLS, an increase of count rate, i.e. increased
scattering, was found to occur in the same temperature range,
agreeing well with the turbidity measurement. DLS analysis also
confirmed that the polymer chains appeared to be unimerically
dissolved just above the phase transition having volume-average
diameters of 9.40 and 9.25 nm at temperatures of 60 and 55 °C,
respectively (see inset of Figure 7). This is in contrast to amide-
based polymers with aqueous UCSTs, such as poly(N-acryloyl
glycinamide), which are known to contain aggregates above the
UCST, in spite of appearing as clear solutions.28
As mentioned in the Introduction, zwitterionic (co)polymers
are known to display an antipolyelectrolyte effectaddition of
electrolytes increasing solubilitywhich has been exploited in
applications38 but which also poses limitations to applicability
of polysulfobetaines, typically necessitating working in ultrapure
water. We therefore probed the tolerance of the phase
transition of a sample with a UCST increased through
benzylacrylamide incorporation, p(ADPS0.47-co-Bz0.53), toward
the presence of a model electrolyte, sodium chloride, at
constant polymer concentration of 10 g/L. In pure water, this
copolymer had a cloud point of 63.6 °C. At a NaCl
concentration of 1.5 mM the cloud point increased slightly to
65.2 °C. This effect has been shown to occur at salt levels below
a threshold concentration, when the amount of added salt is
not sufficient to completely screen the attractions between the
positive and negative charges in the polymers, with reported
NaCl concentrations at maximum solubility being between 0.7
and 2 mMconsistent with our observation.26 It is, at this
stage, important to note that the amount of electrolyte
contamination within the (co)polymer samples after exhaustive
dialysis in ultrapure water was lower than the lowest added
concentration of NaCl, with the sodium concentration of a
(co)polymer solution at a concentration of 10 g/L estimated to
be 0.06 mM from elemental analysis. At higher added NaCl
concentrations, the cloud point of p(ADPS0.47-co-Bz0.53) was
found to decrease, showing the expected antipolyelectrolyte
effect. Turbidity curves are shown in Figure 8A, and the trend
of cloud point vs NaCl concentration is plotted in Figure 8B.
While the cloud point of p(ADPS0.47-co-Bz0.53) initially showed
a rather steep decline, decreasing by 5.8 to 59.4 °C at a salt
concentration of 3.0 mM, the influence became somewhat less
pronounced at higher salt concentrations, with cloud points of
14.7 and 7.3 °C measured at salt concentrations of 60.7 and
75.8 mM (corresponding to 3.55 and 4.43 g/L), respectively. A
cloud point of 39.2 °C, close to body temperature, was found
for a salt concentration of 22.8 mM. Phase separations of
p(ADPS0.47-co-Bz0.53) in salt water were fully reversible and
reproducible with a hysteresis of ∼7 °C, similar to that found in
pure water (see Figure S8 in Supporting Information). These
measurements illustrated the potential of tuning the phase
separation temperature of hydrophobically modified sulfobe-
taine copolymers through the addition of salt, showing that the
incorporation of benzyl groups can enable sharp, reproducible
phase transitions in aqueous salt solutions. Yet, the salt
tolerance of p(ADPS0.47-co-Bz0.53) is unquestionably not high
enough for an exploitation of the UCST transition under
physiological conditions, typically involving (among many
other components) salt concentrations of ∼154 mM (9 g/L)
Dynamic light scattering (DLS) was used as a further method
to characterize the phase transition. In Figure 7, the
Figure 7. Scattering intensity (by dynamic light scattering, black
curve) and transmittance (by turbidity, blue curve) for a sample of a
benzylacrylamide−sulfobetaine copolymer showing concurrent de-
crease of transmittance and increase of scattering count rate and
volume-average diameter of scatterers (inset) suggesting unimers and
absence of aggregates just above the UCST transition.
759
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
Figure 8. Influence of salt concentration on the phase separation temperature of a benzylacrylamide−sulfobetaine copolymer: (A) turbidity curves
for increasing NaCl concentration; (B) variation of cloud point with increasing NaCl concentration, showing an initial increase below the threshold
salt concentration26 and a decrease above enabling UCST transitions up to a NaCl concentration of 75.6 mM (4.43 g/L).
NaCl. Nonetheless, this first detailed study into tuning the
UCST transition of sulfobetaine copolymers offers the
necessary understanding and tools to further develop
zwitterionic copolymers and to extend the scope of these
smart, blood-compatible materials into other realms of science.
zwitterionic segments, thus significantly widening the applic-
ability of these smart materials.
ASSOCIATED CONTENT
* Supporting Information
■
S
1H and 13C NMR spectra of the zwitterionic amine and its
hydrochloride, 1H NMR spectra of pentylacrylamide and
benzylacrylamide copolymers in TFE-d3, turbidity curves of
pentylacrylamide copolymers, reproducibility measurements of
copolymers in pure water and in NaCl solution, DLS
measurements of pentylacrylamide copolymers This material
is available free of charge via the Internet at http://pubs.acs.org.
CONCLUSION
■
We expanded the scope of activated esters in the polymer
chemistry arena by optimizing reaction conditions for
converting poly(pentafluorophenyl acrylate) with a zwitterionic
amine, thus providing simple synthetic access to a library of
well-defined hydrophobically modified zwitterionic (co)-
polymers. This strategy was applicable to copolymers over
the full range of molar compositions including hydrophobic
copolymers comprising minor amounts of zwitterionic
functionality, as well as copolymers containing large hydro-
phobic moieties, providing excellent control over the molar
composition of copolymers. In spite of hydrophobic
modification, a copolymer series containing varying amounts
of pentylacrylamide showed temperature-independent water
solubility over a wide compositional range, which was
attributed to an entropic contribution of the flexible pentyl
chains. Conversely, a copolymer series comprising varying
amounts of benzylacrylamide expectedly exhibited decreased
water solubility with increasing benzyl content, manifested in
increasing UCST transition temperatures. Benzylacrylamide
incorporation thus allowed, for the first time, a precise tuning of
the UCST cloud point of sulfobetaine copolymers and made
this rare phase behavior much more accessible with polymer
samples of intermediate molar masses (∼30 kg/mol) exhibiting
UCST cloud points in the range of 6−82 °C. Addition of
electrolytes had the opposite effect, causing decreasing cloud
points with increasing NaCl concentration. Both effects
antipolyelectrolyte effect and increased UCST by virtue of
benzyl incorporationwere found to compensate each other
well, allowing for sharp, fully reversible and reproducible UCST
transitions of benzyl-modified sulfobetaine copolymers in salt
solutions. The synthetic approach described here, moving
UCST transitions of polysulfobetaines into a much more
accessible temperature/molar mass range, is envisaged to allow
for exploitation of the full plethora of advantages that
postpolymerization modification offers including incorporation
of smart/biological functionality into polybetaines and
construction of complex polymeric architectures containing
AUTHOR INFORMATION
Corresponding Author
*E-mail: P.Roth@unsw.edu.au (P.J.R.).
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors acknowledge the NMR facility of the Mark
Wainwright Analytical Centre and Rhiannon Batchelor for
assistance with NMR measurements, Yanyan Jiang for
assistance with aqueous SEC measurements, and Janina Noy
for assistance with synthesis. P.J.R. acknowledges the Australian
Research Council (ARC) for funding through a Discovery Early
Career Researcher Award (DE120101547).
REFERENCES
■
(1) Arotcarena, M.; Heise, B.; Ishaya, S.; Laschewsky, A. J. Am. Chem.
Soc. 2002, 124, 3787.
(2) Kudaibergenov, S.; Jaeger, W.; Laschewsky, A. In Supramolecular
Polymers Polymeric Betains Oligomers; Springer: Berlin, 2006; Vol. 201,
p 157.
(3) Monroy Soto, V. M.; Galin, J. C. Polymer 1984, 25, 121.
(4) Monroy Soto, V. M.; Galin, J. C. Polymer 1984, 25, 254.
(5) Weaver, J. V. M; Armes, S. P.; Butun, V. Chem. Commun. 2002,
2122.
(6) Xuan, F.; Liu, J. Polym. Int. 2009, 58, 1350.
(7) Lowe, A. B.; McCormick, C. L. Chem. Rev. 2002, 102, 4177.
(8) Wu, J.; Lin, W.; Wang, Z.; Chen, S.; Chang, Y. Langmuir 2012,
28, 7436.
(9) Chang, Y.; Liao, S.-C.; Higuchi, A.; Ruaan, R.-C.; Chu, C.-W.;
Chen, W.-Y. Langmuir 2008, 24, 5453.
760
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
(10) Chang, Y.; Shih, Y.-J.; Lai, C.-J.; Kung, H.-H.; Jiang, S. Adv.
Funct. Mater. 2013, 23, 1100.
(11) Yuan, J.; Huang, X.; Li, P.; Li, L.; Shen, J. Polym. Chem. 2013, 4,
(49) Steinhauer, W.; Hoogenboom, R.; Keul, H.; Moeller, M.
Macromolecules 2010, 43, 7041.
(50) Taylor, L. D.; Cerankowski, L. D. J. Polym. Sci., Polym. Chem. Ed.
1975, 13, 2551.
5074.
(51) Hoogenboom, R.; Zorn, A.-M.; Keul, H.; Barner-Kowollik, C.;
Moeller, M. Polym. Chem. 2012, 3, 335.
(52) Steinhauer, W.; Hoogenboom, R.; Keul, H.; Moeller, M.
Macromolecules 2013, 46, 1447.
(53) Johnson, K. M.; Poe, G. D.; Lochhead, R. Y.; McCormick, C. L.
J. Macromol. Sci., Part A 2004, 41, 587.
(54) Kathmann, E. E. L.; Davis, D. D.; McCormick, C. L.
Macromolecules 1994, 27, 3156.
(55) Liaw, D. J.; Huang, C. C.; Kang, E. T. Colloid Polym. Sci. 1997,
275, 922.
(56) Liaw, D.-J.; Huang, C.-C.; Sang, H.-C.; Wu, P.-L. Polymer 2000,
(12) Muro, E.; Pons, T.; Lequeux, N.; Fragola, A.; Sanson, N.;
Lenkei, Z.; Dubertret, B. J. Am. Chem. Soc. 2010, 132, 4556.
(13) Zhan, N.; Palui, G.; Safi, M.; Ji, X.; Mattoussi, H. J. Am. Chem.
Soc. 2013, 135, 13786.
(14) Gunkel, G.; Huck, W. T. S. J. Am. Chem. Soc. 2013, 135, 7047.
(15) Lalani, R.; Liu, L. Biomacromolecules 2012, 13, 1853.
(16) Lin, P.; Lin, C.-W.; Mansour, R.; Gu, F. Biosens. Bioelectron.
2013, 47, 451.
(17) Newcomer, R. G.; Moussallem, M. D.; Keller, T. C. S.;
Schlenoff, J. B.; Sang, Q.-X. A. Biotechnol. Res. Int. 2011, 2011, 854068.
(18) Cao, F.; Tan, L.; Xiang, L.; Liu, S.; Wang, Y. J. Biomater. Sci.,
Polym. Ed. 2013, 24, 2058.
(19) Sun, Q.; Su, Y.; Ma, X.; Wang, Y.; Jiang, Z. J. Membr. Sci. 2006,
285, 299.
(20) Dai, F.; Liu, W. Biomaterials 2011, 32, 628.
(21) Blanco, M. D.; Rego, J.; Huglin, M. B. Polymer 1994, 35, 3487.
(22) Kostova, B.; Kamenska, E.; Rachev, D.; Simeonova, S.;
Georgiev, G.; Balashev, K. J. Polym. Res. 2013, 20, 1.
41, 6123.
(57) Chang, Y.; Chen, W.-Y.; Yandi, W.; Shih, Y.-J.; Chu, W.-L.; Liu,
Y.-L.; Chu, C.-W.; Ruaan, R.-C.; Higuchi, A. Biomacromolecules 2009,
10, 2092.
(58) Nedelcheva, A. N.; Novakov, C. P.; Miloshev, S. M.; Berlinova,
I. V. Polymer 2005, 46, 2059.
(59) Tian, H.-Y.; Yan, J.-J.; Wang, D.; Gu, C.; You, Y.-Z.; Chen, X.-S.
Macromol. Rapid Commun. 2011, 32, 660.
(23) Fang, J.; Wallikewitz, B. H.; Gao, F.; Tu, G.; Muller, C.; Pace,
̈
(60) Koberle, P.; Laschewsky, A.; Lomax, T. D. Makromol. Chem.,
̈
G.; Friend, R. H.; Huck, W. T. S. J. Am. Chem. Soc. 2010, 133, 683.
(24) Kumar, A.; Pace, G.; Bakulin, A. A.; Fang, J.; Ho, P. K. H.; Huck,
W. T. S.; Friend, R. H.; Greenham, N. C. Energy Environ. Sci. 2013, 6,
1589.
Rapid Commun. 1991, 12, 427.
(61) Koberle, P.; Laschewsky, A.; van den Boogaard, D. Polymer
̈
1992, 33, 4029.
(62) Bazuin, C. G.; Zheng, Y. L.; Muller, R.; Galin, J. C. Polymer
1989, 30, 654.
(25) Huglin, M. B.; Radwan, M. A. Polym. Int. 1991, 26, 97.
(26) Mary, P.; Bendejacq, D. D.; Labeau, M.-P.; Dupuis, P. J. Phys.
Chem. B 2007, 111, 7767.
(27) Schulz, D. N.; Peiffer, D. G.; Agarwal, P. K.; Larabee, J.; Kaladas,
J. J.; Soni, L.; Handwerker, B.; Garner, R. T. Polymer 1986, 27, 1734.
(28) Seuring, J.; Agarwal, S. Macromol. Rapid Commun. 2012, 33,
1898.
(29) Seuring, J.; Agarwal, S. ACS Macro Lett. 2013, 2, 597.
(30) Higgs, P. G.; Joanny, J. F. J. Chem. Phys. 1991, 94, 1543.
(31) Ning, J.; Kubota, K.; Li, G.; Haraguchi, K. React. Funct. Polym.
2013, 73, 969.
(32) Ning, J.; Li, G.; Haraguchi, K. Macromolecules 2013, 46, 5317.
(33) Virtanen, J.; Arotcarena, M.; Heise, B.; Ishaya, S.; Laschewsky,
A.; Tenhu, H. Langmuir 2002, 18, 5360.
(63) Mathis, A.; Zheng, Y.-L.; Galin, J.-C. Makromol. Chem., Rapid
Commun. 1986, 7, 333.
(64) Zheng, Y.-L.; Galin, M.; Galin, J.-C. Polymer 1988, 29, 724.
(65) Kostova, B.; Kamenska, E.; Dimitrov, N.; Rachev, D.; Georgiev,
G. J. Controlled Release 2010, 148, e44.
(66) Brown, R. H.; Duncan, A. J.; Choi, J.-H.; Park, J. K.; Wu, T.;
Leo, D. J.; Winey, K. I.; Moore, R. B.; Long, T. E. Macromolecules
2009, 43, 790.
(67) Brown, R. H.; Hunley, M. T.; Allen, J. M. H.; Long, T. E.
Polymer 2009, 50, 4781.
(68) Ehrmann, M.; Galin, J.-C. Polymer 1992, 33, 859.
(69) Gauthier, M.; Carrozzella, T.; Penlidis, A. J. Polym. Sci., Polym.
Chem. 2002, 40, 511.
(70) Strehmel, V.; Laschewsky, A.; Wetzel, H. e-Polym. 2006, 6, 131.
(71) Strehmel, V.; Wetzel, H.; Laschewsky, A.; Moldenhauer, E.;
Klein, T. Polym. Adv. Technol. 2008, 19, 1383.
(72) Vamvakaki, M.; Billingham, N. C.; Armes, S. P. Polymer 1998,
39, 2331.
(73) Mathis, A.; Zheng, Y. L.; Galin, J. C. Polymer 1991, 32, 3080.
(74) Berlinova, I. V.; Dimitrov, I. V.; Kalinova, R. G.; Vladimirov, N.
G. Polymer 2000, 41, 831.
(75) Gauthier, M. A.; Gibson, M. I.; Klok, H.-A. Angew. Chem., Int.
Ed. 2009, 48, 48.
(34) Shih, Y.-J.; Chang, Y.; Deratani, A.; Quemener, D.
Biomacromolecules 2012, 13, 2849.
(35) Tian, M.; Wang, J.; Zhang, E.; Li, J.; Duan, C.; Yao, F. Langmuir
2013, 29, 8076.
(36) Azzaroni, O.; Brown, A. A.; Huck, W. T. S. Angew. Chem., Int.
Ed. 2006, 45, 1770.
(37) Zhou, F.; Huck, W. T. S. Phys. Chem. Chem. Phys. 2006, 8, 3815.
(38) Pei, Y.; Travas-Sejdic, J.; Williams, D. E. Langmuir 2012, 28,
8072.
(39) Willcock, H.; Lu, A.; Hansell, C. F.; Chapman, E.; Collins, I. R.;
O’Reilly, R. K. Polym. Chem. 2014, in press.
(40) Seuring, J.; Agarwal, S. Macromolecules 2012, 45, 3910.
(41) Zhang, Q.; Schattling, P.; Theato, P.; Hoogenboom, R. Polym.
Chem. 2012, 3, 1418.
(42) Roth, P. J.; Collin, M.; Boyer, C. Soft Matter 2013, 9, 1825.
(43) Lambermont-Thijs, H. M. L.; Kuringen, H. P. C. v.; Put, J. P. W.
v. d.; Schubert, U. S.; Hoogenboom, R. Polymers 2010, 2, 188.
(44) Chua, G. B. H.; Roth, P. J.; Duong, H. T. T.; Davis, T. P.; Lowe,
A. B. Macromolecules 2012, 45, 1362.
(76) Gunay, K. A.; Theato, P.; Klok, H.-A. J. Polym. Sci., Polym. Chem.
̈
2013, 51, 1.
(77) Theato, P. J. Polym. Sci., Polym. Chem. 2008, 46, 6677.
(78) Boyer, C.; Stenzel, M. H.; Davis, T. P. J. Polym. Sci., Polym.
Chem. 2011, 49, 551.
(79) Quek, J. Y.; Roth, P. J.; Evans, R. A.; Davis, T. P.; Lowe, A. B. J.
Polym. Sci., Polym. Chem. 2013, 51, 394.
(80) Kim, G.; Yoo, C. E.; Kim, M.; Kang, H. J.; Park, D.; Lee, M.;
Huh, N. Bioconjugate Chem. 2012, 23, 2114.
(81) Benoiton, N. L. In Houben-Weyl Methods in Organic Chemistry;
Goodman, M., Toniolo, C., Moroder, L., Felix, A., Eds.; Thieme
Medical Publishers Inc.: Stuttgart, 2004; Vol. E22a, p 443.
(82) Moad, G.; Rizzardo, E.; Thang, S. H. Aust. J. Chem. 2012, 65,
985.
(45) Jochum, F. D.; zur Borg, L.; Roth, P. J.; Theato, P.
Macromolecules 2009, 42, 7854.
(46) Quek, J. Y.; Zhu, Y.; Roth, P. J.; Davis, T. P.; Lowe, A. B.
Macromolecules 2013, 46, 7290.
(47) Roth, P. J.; Jochum, F. D.; Forst, F. R.; Zentel, R.; Theato, P.
Macromolecules 2010, 43, 4638.
(48) Zhu, Y.; Quek, J. Y.; Lowe, A. B.; Roth, P. J. Macromolecules
2013, 46, 6475−6484.
(83) Boucher, E. A. Prog. Polym. Sci. 1978, 6, 63.
(84) Roth, P. J.; Boyer, C.; Lowe, A. B.; Davis, T. P. Macromol. Rapid
Commun. 2011, 32, 1123.
761
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762

Macromolecules
Article
(85) Shih, Y.-J.; Chang, Y. Langmuir 2010, 26, 17286.
(86) Irudayam, S. J.; Plumb, R. D.; Henchman, R. H. Faraday Discuss.
2010, 145, 467.
(87) Martinez, C. R.; Iverson, B. L. Chem. Sci. 2012, 3, 2191.
(88) Mahadevi, A. S.; Sastry, G. N. Chem. Rev. 2013, 113, 2100−
2138.
(89) Quinonero, D.; Garau, C.; Rotger, C.; Frontera, A.; Ballester, P.;
̃
̀
Costa, A.; Deya, P. M. Angew. Chem. 2002, 114, 3539.
762
dx.doi.org/10.1021/ma402391a | Macromolecules 2014, 47, 750−762