Pyrrolidone-functional smart polymers via nitroxide-mediated polymerization

Article abstract of DOI:10.1002/pola.27208

Nitroxide-mediated polymerization (NMP) of N-(2-methacryloyloxyethyl) pyrrolidone (MAEPYR) with 2-([tert-butyl[1-(diethoxyphosphoryl)-2,2- dimethylpropyl]amino]oxy)-2-methylpropanoic acid (BlocBuilder) initiator and N-tert-butyl-N-[1-diethylphosphono-(2,2-dimethylpropyl)] (SG1) nitroxide permitted controlled synthesis of poly(N-(2-methacryloyloxyethyl)-pyrrolidone- stat-9-(4-vinylbenzyl)-9H-carbazole) (poly(MAEPYR-stat-VBK)) statistical copolymers. With at least 5 mol % VBK, the dispersity D strok sign of the copolymers was below 1.4 at conversions less than 50%. At conversions higher than 50%, and at lower VBK feed content, there was a significant amount of termination reactions, which broadened the molecular weight distribution of the final polymers (= 1.4-2.3). The MAEPYR-rich statistical copolymers were subsequently tested for thermoresponsive behavior in aqueous media. The cloud point temperatures (CPTs) in aqueous solution were tuned by changing the VBK composition, solution concentration, and heating rate, and the transitions were thermally reversible with partial loss of reversibility at higher heating rates. The CPT decreased from 59.0 to 49.7 C with addition of only 1 mol % of VBK in the copolymer, and at more than 6 mol % VBK, the copolymer was water insoluble.

Full text of DOI:10.1002/pola.27208

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Pyrrolidone-Functional Smart Polymers via Nitroxide-Mediated
Polymerization
Xeniya Savelyeva, Milan Maric´
Department of Chemical Engineering, McGill Institute of Advanced Materials (MIAM), Centre for Self-Assembled
ꢀ
ꢀ
Chemical Structures (CSACS), McGill University, 3610 University Street, Montreal, Quebec, Canada H3A 2B2
Correspondence to: M. Maric´ (E-mail: milan.maric@mcgill.ca)
Received 4 March 2014; accepted 2 April 2014; published online 00 Month 2014
DOI: 10.1002/pola.27208
ABSTRACT: Nitroxide-mediated polymerization (NMP) of N-(2-
methacryloyloxyethyl) pyrrolidone (MAEPYR) with 2-([tert-
butyl[1-(diethoxyphosphoryl)-2,2-dimethylpropyl]amino]oxy)-2-
methylpropanoic acid (BlocBuilder) initiator and N-tert-butyl-N-
[1-diethylphosphono-(2,2-dimethylpropyl)] (SG1) nitroxide per-
mitted controlled synthesis of poly(N-(2-methacryloyloxyethyl)-
pyrrolidone-stat-9-(4-vinylbenzyl)-9H-carbazole) (poly(MAEPYR-
stat-VBK)) statistical copolymers. With at least 5 mol % VBK,
tures (CPTs) in aqueous solution were tuned by changing the
VBK composition, solution concentration, and heating rate,
and the transitions were thermally reversible with partial loss
of reversibility at higher heating rates. The CPT decreased from
59.0 to 49.7 ꢀC with addition of only 1 mol % of VBK in the
copolymer, and at more than 6 mol % VBK, the copolymer was
C
V
water insoluble.
2014 Wiley Periodicals, Inc. J. Polym. Sci.,
Part A: Polym. Chem. 2014, 00, 000–000
-
the dispersity D of the copolymers was below 1.4 at conver-
sions less than 50%. At conversions higher than 50%, and at
lower VBK feed content, there was a significant amount of ter-
mination reactions, which broadened the molecular weight dis-
KEYWORDS: BlocBuilder; cloud point temperature (CPT); copoly-
merization; living radical polymerization (LRP); N-(2-methacry-
-
tribution of the final polymers (D 5 1.4–2.3). The MAEPYR-rich
loyloxyethyl)-pyrrolidone
(MAEPYR);
nitroxide-mediated
statistical copolymers were subsequently tested for thermores-
ponsive behavior in aqueous media. The cloud point tempera-
polymerization (NMP); stimuli-sensitive polymers; water-solu-
ble polymers
The hydrophobic and hydrophilic moieties present in the
chain determine the LCST of the phase transition.7 The phase
transition can be adjusted subtly by the changes in a mono-
mer’s structure. For instance, Cai and coworkers reported a
decrease in cloud point temperature (CPT) of as much as 37
ꢀC from the acrylic-based pyrrolidone-functional homopoly-
mers (CPT ꢁ 66.5 ꢀC, poly[N-(3-acryloyloxypropyl) pyrroli-
done]) to the methacrylic-based pyrrolidone-functional
INTRODUCTION A functional material that can respond to
different external stimuli is a key toward developing a new
generation of intelligent or “smart” materials. The response
of such materials can be regarded as a change in shape,
solubility, and surface characteristics,1 usually with an
attendant reversibility. The stimuli for these transitions can
be a change in temperature,2 pH,3 ionic strength,3 light,4 or
presence of certain metabolic chemicals.5 Materials that can
undergo a phase transition in response to the changes in
the metabolic variables of biological fluids (temperature
and/or pH) are of special interest6 in a wide range of
applications, such as drug delivery, biosensors, tissue engi-
neering, coatings, textiles, and optical systems. As the most
widely used stimulus,7 temperature can be easily controlled
and applied both in vivo8,9 and in vitro.10,11 Many
temperature-sensitive polymers exhibit a lower critical solu-
tion temperature (LCST) in water, which is the critical point
where a polymer becomes water-insoluble upon heating.12
Poly(N-isopropylacrylamide) (PNIPAM) has become the
most studied temperature-sensitive polymer, particularly
because its LCST is near body temperature.13–17
ꢀ
homopolymers (CPT ꢁ 29.5 C, poly[N-(3-methylacryloyloxy-
propyl) pyrrolidone]).18 Poly(2-N-morpholinoethyl acrylate)
(PMEA)
and
poly(2-N-morpholinoethyl
methacrylate)
(PMEMA) were analyzed recently by our group for their
phase transition behavior. PMEA was completely water solu-
ble in the range of temperatures studied (20–95 ꢀC),
whereas PMEMA had a CPT at about 37 ꢀC.19 The higher
CPT of the latter polymer can be explained by the constraint
of hydrogen bonding due to rotation of the backbone methyl
groups.17 Poly(N-vinyl-pyrrolidone) (PVP) is another example
where the subtle changes in the molecular structure affect
the phase transition. PVP is soluble in aqueous solution over
a wide range of temperature; however, its analogs, where the
C
V
2014 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
molecular structure was modified by addition of a hydropho-
bic spacer, were shown to have a LCST.18,20 Polymers with
pyrrolidone functionalities are applied in biomedical
research owing to their water solubility,21,22 biocompatibil-
ity,20–22 low toxicity,21 and coordination capacity.18,20,22 PVP
can also be used as a blood plasma substitute,21–24 food
thickener,22 binding agent for drug delivery,20,23 and inhibi-
tor of hydrate formation.25,26
tion times.23 MAEPYR has been polymerized in a controlled
manner previously using RAFT18,40 and conventional radical
polymerization,41 but not by NMP. NMP does not require
removal of metallic impurities, as in the case of ATRP, nor
does it use sulfur-based chain transfer agents (CTAs) as in
the case of RAFT. It is thus valuable to see if NMP’s favorable
characteristics can be used for a controlled synthesis of
MAEPYR polymer without impeding phase transition proper-
ties in aqueous solution. However, to apply NMP via com-
mercially available BlocBuilder (2-([tert-butyl[1-(diethoxyph
osphoryl)-2,2-dimethylpropyl]amino]oxy)-2-methylpropanoic
acid), the methacrylate must be copolymerized with a como-
nomer to decrease the average equilibrium constant to keep
the polymerization controlled. Styrene and acrylonitrile are
examples of the most commonly used controlling comono-
mers.42,43 Our group has found that 9-(4-vinylbenzyl)-9H-
carbazole (VBK) (Scheme 1) can be used as a very effective
controlling comonomer.19,44–48 The VBK units not only
served to control the copolymerization with reduced amount
of comonomer compared with styrene but also imparted flu-
orescent, electron-donating functionality into the copolymers
and permitted tuning of the LCST owing to its relative
hydrophobicity.
Polymer microstructure influences directly the polymer
properties. In the case of temperature-responsive polymers,
the response to temperature changes is sharp for monodis-
perse samples, and it is less defined for polymers with broad
molecular weight distributions.18,27–29 Various polymeriza-
tion techniques, such as living ionic polymerization, have
been developed to allow the synthesis of well-defined poly-
mers with predictable molecular weight and precise micro-
structure. Ionic polymerization, however, requires very
stringent reaction conditions because the presence of any
impurity (e.g., air and moisture) will terminate the reaction.
This is where controlled radical polymerization (CRP)
becomes useful. It has moderate sensitivity to the impurities,
like conventional radical polymerization, and it has pseudo
“living” character as in ionic polymerization because termi-
nation reactions can be sufficiently suppressed for a signifi-
cant portion of the polymerization. CRP techniques do not
require rigorous purification conditions to obtain well-
defined polymers; therefore, small amounts of impurities are
tolerated.30 Among various techniques currently used for
CRP, reversible addition-fragmentation chain transfer poly-
merization (RAFT),31–34 nitroxide-mediated polymerization
(NMP),22,35–37 and atom transfer radical polymerization
(ATRP)38,39 are the most popular ones.39
In this study, NMP with BlocBuilder (Scheme 1) was exam-
ined to yield MAEPYR-rich copolymers when using VBK as a
controlling comonomer. The activity of the copolymers’ chain
end was assessed by reinitiating with a fresh batch of N,N-
dimethylacrylamide (DMAA) (Scheme 1). The resulting statis-
tical and block copolymers were solubilized in aqueous solu-
tions and their CPTs were measured. The effect of copolymer
composition, solution concentration, pH, and the addition of
poly(DMAA) block was studied and reported.
N-(2-methacryloyloxyethyl) pyrrolidone (MAEPYR) monomer
(Scheme 1) shares similar properties with N-vinylpyrroli-
done (VP) owing to the common pyrrolidone functionality.
Hadjichristidis and coworkers have attempted to polymerize
VP via NMP using 2,2,6,6-tetramethylpiperidinyl-1-oxy
(TEMPO) nitroxide.22,23 The final polymers were obtained at
low yields with high dispersity (1.7–2.2) mostly owing to the
prevalent termination reactions present at prolonged reac-
EXPERIMENTAL
Materials
Calcium hydride (90–95%, reagent grade, Sigma Aldrich),
basic alumina (Brockmann, type 1, 150 mesh, Sigma Aldrich),
lithium bromide (ReagentPlus, ꢂ99%, Sigma Aldrich), 1-(2-
hydroxyethyl)22-pyrrolidone (98%, Sigma Aldrich), dimethyl
sulfoxide (DMSO) (deuterated-d6, 99%, Sigma Aldrich),
SCHEME 1 Synthetic route used to obtain poly(N-(2-methacryloyloxyethyl) pyrrolidone-stat-9-(4-vinylbenzyl)-9H-carbazole) (poly(-
MAEPYR-stat-VBK)) statistical copolymers by NMP using BlocBuilder initiator followed by the chain extension with N,N-dimethyla-
crylamide (DMAA) to form poly(N-(2-methacryloyloxyethyl) pyrrolidone-stat-9-(4-vinylbenzyl)-9H-carbazole)-b-poly(N,N-
dimethylacrylamide) (poly(MAEPYR-stat-VBK)-b-poly(DMAA)) block copolymer.
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
methacryloyl chloride (97%, contains 200 ppm monomethyl
ether hydroquinone as stabilizer, Sigma Aldrich), N,N-dime-
thylformamide (DMF, >95% certified ACS, and 99.5% HPLC
grade, Acros Organics), ethyl ether anhydrous (>95%, BHT
stabilized/certified ACS, Fisher Scientific), magnesium sulfate
(anhydrous, certified powder, Fisher Scientific), sodium chlo-
ride (ꢂ99.5, certified ACS, Fisher Scientific), sodium carbonate
(anhydrous, ꢂ99.5, certified ACS, Fisher Scientific), and trie-
thylamine (99%, Fisher Scientific) were used as received. The
deuterated chloroform (CDCl3, >99%) was obtained from
Cambridge Isotopes Laboratory and also used as received.
Chloroform (>99%, reagent ACS grade, Fisher Scientific) and
N,N-dimethylacrylamide (DMAA, 99%, contains 500 ppm
monomethyl ether hydroquinone as stabilizer, Sigma Aldrich)
were purified by passage through a column of 5 wt % calcium
hydride relative to basic alumina and then stored in a sealed
flask under a head of nitrogen in a fridge until needed. MAE-
PYR was synthesized according to the literature40 with slight
variations as noted below and stored in a sealed flask under a
head of nitrogen in a refrigerator until needed. VBK was syn-
thesized according to the literature49 and stored in a refrigera-
tor away from the light source. BlocBuilder (99%) was
obtained from Arkema and used without further purification.
[tert-butyl[1-(diethoxyphosphoryl)-2,2-dimethylpropyl]amino]
oxidanyl, also known as SG1 (>85%), was kindly donated by
Noah Macy of Arkema and used as received. 2-Cyano-2-propyl
dodecyl trithiocarbonate (CPDTC) was obtained from Sigma
Aldrich and used without further purification. 2,20-Azobisisobu-
tyronitrile (AIBN, Vazo 67 from Du Pont) was recrystallized
from methanol.
3.46 ppm (2H, NCOCH2CH2CH2 in pyrrolidone ring), 2.35
ppm (2H, NCOCH2CH2CH2 in pyrrolidone ring), 2.01 ppm
(2H, NCOCH2CH2CH2 in pyrrolidone ring), and 1.91 ppm
(3H, CH2@CHCH3).
Polymer Synthesis
General
All the copolymerizations were performed in a 10-mL three-
necked round-bottom glass reactor equipped with a con-
denser, a temperature well with a thermocouple, and a stir
bar. The reactor was placed in a heating mantle on a stir
plate and a thermocouple was connected to a temperature
controller. The condenser was cooled using an Isotemp
3016D (Fisher Scientific) chiller unit to prevent any mono-
mer or solvent evaporation. The nitrogen purge was applied
for 20 min before the reaction, and the needle was inserted
into the rubber septum, used to cap the condenser, to vent
the purge. The reactor was then heated (at 5–7 ꢀC min21),
and the point when the temperature reached the set point
was arbitrarily chosen as t 5 0 min. The purge with the
reduced flow rate was kept throughout the reaction, and the
samples were periodically withdrawn by syringe to track the
polymerization progress. Both samples and the crude poly-
mer were precipitated in exceꢀss of cold diethyl ether, recov-
ered, and vacuum dried at 40 C overnight.
Synthesis of Poly(N-(2-methacryloyloxyethyl)-
pyrrolidone-stat-9-(4-vinylbenzyl)-9H-carbazole)
(Poly(MAEPYR-stat-VBK)) Statistical Copolymers
via NMP
Table 1 lists all the formulations studied. All the copolymer-
izations were conducted with BlocBuilder and SG1 initiators
in DMF (50 wt %) at 90 C. The molar ratio of SG1 relative
to BlocBuilder (r 5 [SG1]0/[BlocBuilder]0) was 0.1, the VBK
initial feed concentration (fVBK,0) was varied between 2 and
10 mol %, and the target molecular weight at complete con-
version was calculated to be 10, 25, or 50 kg mol21. The for-
Monomer Synthesis
The esterification reaction was adopted from Cai and
coworkers40 with slight variations. A 125-mL three-necked
jacketed reactor, equipped with a stir bar, and a dropper col-
umn were used. Measured amounts of 1-(2-hydroxyethyl)-2-
pyrrolidone (40.06 g, 0.31 mol), triethylamine (31.26 g, 0.31
mol), and anhydrous chloroform (35 mL) were added to the
reactor and stirred well at 5 ꢀC. The reaction mixture was
purged with nitrogen for 20 min. Methacryloyl chloride
(32.40 g, 0.31 mol) was diluted with anhydrous chloroform
(40 mL), and then the mixture was added to the reactor
dropwise over a 1-h period. The reaction was allowed to
proceed for at least 10 h at 10 ꢀC. The nitrogen purge
remained for the entirety of the reaction. For the purifica-
tion, white ammonium salt was removed by vacuum filtra-
tion. The solution was first concentrated by rotary
evaporation and then extracted thrice using first 5 wt %
Na2CO3 (200 mL), then saturated NaCl solution (100 mL)
and deionized water (200 mL). The solution was dried under
MgSO4 (3 g) and concentrated by rotary evaporation to
remove any solvent residuals. The monomer’s structure was
verified by nuclear magnetic resonance (1H NMR) spectros-
copy in deuterated chloroform and DMSO. The final yield
was 14.7 g (24%).
ꢀ
mulation of MV-1, with initial VBK molar composition (fVBK,0
)
2 mol %, is given as an example (Table 1). BlocBuilder
(0.048 g, 0.125 mmol), SG1 (0.004 g, 0.012 mmol), MAEPYR
(3.01 g, 15.28 mmol), VBK (0.085 g, 0.32 mmol), and DMF
(3.10 g, 42.37 mmol) were added to the reactor. The poly-
merization was performed at 90 ꢀC for 5.3 h. Samples of
0.1–0.15 mL were withdrawn during the reaction to track
the polymerization progress, and at the end of the reaction
the final yield was 0.99 g (88% conversion; final molar com-
position of VBK, FVBK 5 0.01) with Mn 5 10.2 kg mol21 and
Mw =Mn 5 1.58. The procedure for compositional analysis via
NMR is shown in the Characterization section. The gel per-
meation chromatographyꢀ (GPC) was calibrated using PMMA
standards in DMF at 50 C. The yield was determined gravi-
metrically and it was lower than expected. This might be
due to the choice of the non-solvent used for the precipita-
tion (diethyl ether in this case), because the low-molecular-
weight oligomers that contributed to the overall conversion,
as determined by NMR analysis, might have been washed
out.
1H NMR (d, in CDCl3): 6.08 and 5.56 ppm (2H, CH2@CHCH3),
4.25 ppm (2H, COOCH2CH2), 3.56 ppm (2H, COOCH2CH2),
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Experimental Formulations for Poly(N-(2-methacryloyloxyethyl)-pyrrolidone-stat29-(4-vinylbenzyl)29H-carbazole) (Poly(-
MAEPYR-stat-VBK)) Statistical Copolymerizations Performed in 50 wt % DMF Solution at 90 ꢀC
b
[BlocBuilder]0
[SG1]0
(mol L21
[VBK]0
(mol L21
[MAEPYR]0
[DMF]0
(mol L21
Mn,target
IDa
Symbol
(mol L21
)
)
)
(mol L21
)
)
(kg mol21
)
fVBK,0
c
MV-1
MV-2
MV-3
MV-4
MV-5
MV-6
MV-7
D
0.021
0.051
0.020
0.011
0.021
0.020
0.020
0.002
0.005
0.002
0.001
0.002
0.002
0.002
0.05
0.10
0.10
0.11
0.13
0.20
0.25
2.57
2.45
2.47
2.55
2.41
2.36
2.21
7.12
7.38
7.34
7.17
7.48
7.60
7.94
24.8
10.0
25.2
49.3
24.9
25.8
24.9
0.02
0.04
0.04
0.04
0.05
0.08
0.10
᭺
᭛
w
1
*
3
a
b
Experiments for poly(MAEPYR-stat-VBK) copolymerizations are
denoted MV-Z with M representing N-(2-methacryloyloxyethyl)-pyrroli-
done, V representing 9-(4-vinylbenzyl)-9H-carbazole, and Z representing
the experiment number.
The target molecular weight, Mn,target, was calculated according to
n,target 5 MI 1 Mm*[m]0/[I]0, where MI and Mm are the molecular weight
M
of the initiator and monomer, respectively, [m]0 and [I]0 are the initial
monomer and initiator concentrations, respectively.
c
fVBK,0 is the initial molar fraction of VBK in the feed.
Synthesis of Poly(N-(2-methacryloyloxyethyl)-pyrrolidone-
stat-9-(4-vinylbenzyl)-9H-carbazole)-block-Poly(N,N-
dimethylacrylamide) (Poly(MAEPYR-stat-VBK)-b-
The initial ratio of CTA relative to AIBN [CTA]0:[AIBN]0 was
5. AIBN (0.003 g, 0.019 mmol), CPDTC (0.036 g, 0.104
mmol), MAEPYR (2.53 g, 12.83 mmol), and DMF (2.53 g,
34.61 mmol) were added to the reactor. The polymerization
was stopped after 2 h. The final yield was 0.82 g (88% con-
version with Mn 5 10.9 kg mol21 and Mw =Mn 5 1.60).
Poly(DMAA)) Block Copolymers via NMP
Table 2 lists the formulations studied. The general procedure
for chain extension experiments was the same as for the
copolymerizations. The macroinitiator was synthesized
before the chain extension experiment, and a fresh batch of
DMAA monomer was used. A formulation for chain extension
Characterization
Polymer Characterization
Gel Permeation Chromatography. The number-average molec-
ꢀ
of MV-3 at 110 C is given as an example below. Macroinitia-
tor MV-3 (0.55 g, 0.04 mmol), DMAA (2.0 g, 20.4 mmol), and
DMF (1.9 g, 25.9 mmol) were added to the reactor. For the
specific example, polymerization was stopped after 2 h. Final
yield after fractionation was 0.43 g (molar composition of
-
ular weight, Mn , and the dispersity, D, or Mw =Mn, were deter-
mined by GPC (Water Breeze). GPC was equipped with
ultraviolet (UV 2487) and differential refractive index (RI
2414) detectors. The GPC was equipped with two ResiPore
(3 mm, MULTI pore type, 250 mm 3 4.6 mm) columns with
a ResiPore guard column (3 mm, 50 mm 3 4.6 mm) from
Polymer Laboratories. HPLC-grade DMF was used as a
mobile phase with a flow rate of 0.3 mL min21. The GPC
was calibraꢀted using poly(methyl methacrylate) standards in
DMF at 50 C.
DMAA,
F
DMAA 5 0.73) with Mn 5 19.9 kg mol21 and
Mw =Mn 5 2.26, as measured by GPC relative to PMMA stand-
ꢀ
ards in DMF solvent at 50 C. Copolymer composition deter-
mination by NMR is detailed in the Characterization section.
Synthesis of Poly(N-(2-methacryloyloxyethyl) Pyrrolidone
(Poly(MAEPYR)) Homopolymer via RAFT Polymerization
The polymerization was conducted using AIBN as a source
1
Nuclear Magnetic Resonance Spectroscopy (NMR). All H NMR
spectroscopy were performed in CDCl3 using a 400 MHz
ꢀ
of radicals and CPDTC as a CTA in DMF (50 wt %) at 75 C.
TABLE 2 Formulations for Chain Extension Experiments Using N,N-Dimethylacrylamide (DMAA) Monomer Performed in 50 wt %
DMF Solution at 110 ꢀC and Molecular Weight Characterization for Poly(MAEPYR-stat-VBK)-b-Poly(DMAA) Block Copolymers
IDa
Macroinitiator IDa
[Macroinitiator]0 (mol L21
)
[DMAA]0 (mol L21
)
[DMF]0 (mol L21
)
Mn,target,2 (kg mol21
)
b
MV-8
MV-9
MV-10
MV-2
MV-3
MV-5
0.015
0.016
0.009
1.7
5.0
4.8
10.7
6.3
11.3
30.7
52.3
6.5
a
Experiments for poly(MAEPYR-stat-VBK) copolymerizations and chain
extensions are denoted MV-Z with M representing N-(2-methacryloylox-
yethyl)-pyrrolidone, V representing 9-(4-vinylbenzyl)29H-carbazole, and
Z representing the experiment number.
molecular weight of the monomer, [m]0 and [macro]0 are the initial
monomer and macroinitiator concentrations, respectively. The target
molecular weight, Mn,target,block, of the final block can be calculated
according to Mn,target,block 5 Mn,macro 1 Mn,target,2, where Mn,macro is the
molecular weight of the macroinitiator used.
b
The target molecular weight, Mn,target,2, of the second block was calcu-
lated according to
Mn,target,2 5 Mm*[m]0/[macro]0, where Mm is the
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 1 Typical 1H NMR (CDCl3) spectra of the crude mixture during a poly(MAEPYR-stat-VBK) statistical synthesis. The region
shown is only where the peaks used for conversion determination occurred. Note that the peak due to the DMF solvent (ꢃ3 ppm)
is not shown in the region of interest. Also, the aromatic protons due to the carbazole ring are at higher chemical shifts. [Color fig-
ure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Varian Gemini. Conversion of the MAEPYR was determined
by comparing the area associated to the vinyl peaks (Fig. 1)
(H1,2 d 5 6.08 and 5.56 ppm) to the area associated to the
ethyl spacer hydrogens (H3,4 d 5 4.0 for the polymer and
d 5 4.25 ppm for the monomer). Conversion of the VBK was
determined by comparing the areas associated to the vinyl
peaks (H5–7 d 5 6.6, 5.7, and 5.2 ppm) to the area associated
to the methylene hydrogens (H8,9 d 5 5.5 ppm for the mono-
mer and d 5 5.4 ppm for the polymer). Conversion, X, of the
poly(MAEPYR-stat-VBK) copolymerizations was calculated
was calculated using the peaks corresponding to SG1 group
(d 5 24–27 ppm) and DEP (d 5 8–9 ppm). This ratio was
compared to the molar ratio of copolymer to DEP to obtain
the molar percentage of SG1-capped chains in the sample.
CPT Measurement
Two techniques were used to report CPT of a polymer in
aqueous solution: UV–Vis spectroscopy and dynamic light
scattering (DLS).
UV–Vis Spectroscopy. The CPTs were measured by UV–Vis
spectroscopy using a Cary 5000 UV2Vis2NIR spectrometer
from X 5 XMAEPYR
xMAEPYR,0 1 XVBK xVBK,0, where xMAEPYR,0
and xVBK,0 are the initial weight fractions of MAEPYR and
VBK, respectively.
(Agilent Technologies) equipped with
a
temperature-
controlled Peltier thermostatted (6 3 6) multicell holder.
The light absorbance was measured at a wavelength of 500
nm. The heating rate and temperature range observed varied
from one analysis to another, as these variables were stud-
ied. The CPT was determined as the temperature at which
the normalized absorbance reached 0.5 on a heating cycle.
The final composition of the statistical and/or block copoly-
mers was determined by the ratio of the ethyl spacer pro-
tons corresponding to MAEPYR units (H3,4, 2H, d 5 4.0), the
methylene protons corresponding to VBK units (H8,9, 2H,
d 5 5.4 ppm), and the methyl protons corresponding to
DMAA units (CAN(CH3)2, 6H, d 5 3.0–3.2 ppm,). For instance,
I
DMAA=6
, where IDMAA, IMAEPYR, and IVBK
MAEPYR=21IVBK =21IDMAA=6
Dynamic Light Scattering. The Malvern Zetasizer Nano ZS
equipped with a 633-nm red laser was used. The samples
were filtered ꢀusing a 0.2-mm filter and then heated in incre-
ments of 0.2 C, allowed to equilibrate for 1 min followed by
10–14 measurements, which were averaged together to give
one value at the corresponding temperature. All DLS meas-
urements were performed at a scattering angle (h) of 173ꢀ.
For more accurate measurement of the hydrodynamic radius,
the refractive index (RI) of a sample has to be estimated.
The RI that was used was that of PMMA.
F
DMAA5 I
are the integrated areas corresponding to DMAA, MAEPYR,
VBK polymer peaks, respectively, and FDMAA is the DMAA
composition in the final block copolymer.
31P NMR spectroscopy of isolated poly(MAEPYR-stat-VBK)
copolymers was recorded in CDCl3 solvent using a 200 MHz
Varian Gemini. The copolymer was carefully weighed and a
known amount of diethyl phosphate (DEP) was added as an
internal standard. The ratio of SG1-capped chains to DEP
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 2 (a) The semilogarithmic plots of scaled conversion (ln[1 2 X]21) (X 5 conversion) versus time; (b) the dispersity, D, ver-
sus X plots; (c) the number-average molecular weight (Mn ) versus X plots for poly(MAEPYR-stat-VBK) copolymerizations at vary-
ing VBK feed content, where the symbols are as follows: D (fVBK,0 5 0.02); ᭛ (fVBK,0 5 0.04); 1 (fVBK,0 5 0.05); * (fVBK,0 5 0.08); x
(fVBK,0 5 0.10).
-
RESULTS AND DISCUSSION
Based on the results presented in Figure 2, it showed that
the initial concentration of VBK clearly influenced the control
Statistical Copolymer Synthesis
of the polymerization. Table 1 shows all the formulations
The homopolymerization of methacrylates like MAEPYR
studied. Characteristic kinetic plots for poly(MAEPYR-stat-
using NMP is problematic owing to the high equilibrium con-
VBK) copolymerizations with various VBK compositions are
stant (K) between the dormant and active species, resulting
shown in Figure 2. The polymerization rate, as indicated by
in irreversible termination due to b-hydrogen transfer from
the slope, decreased with increasing VBK initial feed compo-
the propagating radical to the nitroxide.50 Recently, NMP of
sition. Number-average molecular weight (Mn ) increased lin-
methyl methacrylate (MMA) without any comonomer was
early with conversion up to 60–80% with final copolymers
successful up until 60% conversion by using a new alkoxy-
characterized by monomodal molecular weight distributions
amine based on 2,2-diphenyl-3-phenylimino-2,3-dihydroin-
-
even though dispersities were somewhat high (D < 1.7). The
dol-1-yloxyl (DPAIO) nitroxide51 with final dispersity close to
1.3–1.4. Grubbs and coworkers used N-phenylalkoxyamines
to homopolymerize MMA up to moderate conversions (up to
50%) while maintaining narrow molecular weight distribu-
tions throughout the entire reaction (dispersity of 1.12–
1.30).52 In this study, a commercially available initiator, Bloc-
Builder, and a controlling comonomer, VBK, are used to yield
MAEPYR-rich copolymers. The characteristic plots of the
number-average molecular weight Mn and dispersity Mw =Mn
versus conversion, X, show a plateau in Mn and a significant
increase in Mw =Mn (Fig. 2).
molecular weight data did not follow the theoretical line pre-
cisely [Fig. 2(c)] for two reasons likely. First, the difference
in hydrodynamic volume of the samples and the PMMA
standards used for the GPC calibration could be the source
for the difference. Second, the deviations from the theoretical
line at higher conversions might be due to the presence of
irreversible termination reactions. The properties of the final
statistical copolymers are listed in Table 3. At t 5 0 in Figure
2, there are conversions of 24–38%, which suggests that fast
polymerization occurred during the initial stage. BlocBuilder
begins to decompose at much lower temperatures53 com-
pared with the reaction temperature, and it would initiate
the reaction before the set point temperature, which was
taken as t 5 0. The heating rate was set so that the set point
temperature is reached within 10–15 min so as to minimize
the polymerization rate before the prescribed set point was
attained. It is interesting to note that in almost all the cases,
the controlling comonomer, VBK, was consumed faster than
the methacrylate monomer (MAEPYR), which explains the
higher VBK content in the final polymer when compared
with the initial feed composition. Only for the case of MV-1
is the composition much richer in methacrylate than
expected.
The first attempt to copolymerize MAEPYR with as little as 2
mol % VBK (MV-1, Table 1) with a target molecular weight
at complete conversion of 25 kg mol21 resulted in a rela-
tively fast polymerization, reaching 88% conversion in 5 h. It
followed first-order kinetics in the initial stages [Fig. 2(a)]
with a noticeable termination at later stages of the polymer-
ization. NMR analysis revealed that at 30 min of the reaction,
about 80% of VBK monomer (VBKm) was polymerized (Fig.
3). With almost no controlling comonomer remaining in the
mixture, the propagation rates increased drastically and
increased the probability of irreversible termination reac-
tions.42 The number-average molecular weight of MV-1 ver-
-
sus conversion is linear in Figure 2(c). The D increased at
Figure 4 shows the effect of different target molecular
weights on the kinetics, while keeping the initial feed
the early stage of polymerization and then decreased steadily
to 1.58 at the final conversion of 88%.
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 3 (a) Labeled protons (H1–4) of 9-(4-vinylbenzyl)-9H-carbazole monomer (VBKm) and the corresponding 1H NMR spectra
are shown in (b) for MV-1 in CDCl3 at various polymerization times. VBKm are the peaks corresponding to monomer only, and
VBKm1p are the peaks corresponding to both monomer and polymer. [Color figure can be viewed in the online issue, which is
available at wileyonlinelibrary.com.]
composition of the controlling comonomer constant. The the-
oretical molecular weight depends on the alkoxyamine con-
centration42,54 and can be calculated by eq (1).
radical effect (PRE)42 helps to control the reaction by
decreasing the polymerization rate. As the result, the final
polymers are characterized by relatively narrow molecular
-
weight distribution [D ꢃ 1.4, Fig. 4(b)]. In contrast, at low
alkoxyamine solution concentration, the PRE is less pro-
found. The polymerization rates are high, and the probability
of termination reactions increases, which is exemplified by a
½mꢄ
½Iꢄ00
Mn5MI1
ꢅXꢅMm
(1)
-
where Mn is the target molecular weight of the final polymer
at complete conversion (X 5 1); MI and Mm are the molecular
weight of the initiator and monomer, respectively; [m]0 and
[I]0 are the initial monomer and initiator concentrations,
respectively.
broad molecular weight distribution [D ꢃ 2.5, Fig. 4(b)].
Number-average molecular weights for all three experiments
in Figure 4 increased linearly with conversion up to ꢃ60%.
A typical set of GPC chromatograms for the poly(MAEPYR-
stat-VBK) statistical copolymerizations is shown in Figure
5(a), whereas Figure 5(b) shows GPC traces for chain exten-
sion experiments with DMAA. The growth of the chains with
For low target molecular weight, there is sufficiently high
alkoxyamine concentration in the solution, and the persistent
TABLE 3 Molecular Weight Characterization for Poly(MAEPYR-stat-VBK) Statistical Copolymers
IDa
Symbol
Mn,target (kg mol21 b
)
fVBK,0
FVBK
Xd
Mn (kg mol21 e
)
Mw/Mn
c
c
e
MV-1
MV-2
MV-3
MV-4
MV-5
MV-6
MV-7
D
24.8
10.0
25.2
49.3
24.9
25.8
24.9
0.02
0.04
0.04
0.04
0.05
0.08
0.10
0.01
0.05
0.06
0.06
0.06
0.16
0.16
0.88
0.76
0.53
0.59
0.83
0.37
0.80
10.2
5.9
1.58
1.43
1.71
2.32
1.56
1.59
1.56
᭺
᭛
w
1
*
8.4
24.1
11.8
8.5
3
10.3
a
c
Experiments for poly(MAEPYR-stat-VBK) copolymerizations are
denoted MV-Z with M representing N-(2-methacryloyloxyethyl)-pyrroli-
done, V representing 9-(4-vinylbenzyl)-9H-carbazole, and Z representing
the experiment number.
fVBK,0 is the initial molar fraction of VBK in the feed; FVBK is the molar
fraction of VBK in the final copolymer as determined by 1H NMR
spectroscopy.
d
Monomer conversion determined by 1H NMR spectroscopy.
b
e
The target molecular weight was calculated according to eq (1).
Number-average molecular weight (Mn ) and dispersity were deter-
mined by GPC.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
7

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 4 (a) The semilogarithmic plots of scaled conversion (ln[1 2 X]21) (X 5 conversion) versus time; (b) the dispersity, D, ver-
sus X plots, and (c) the number-average molecular weight (Mn ) versus X plots for poly(MAEPYR-stat-VBK) copolymerizations at
varying target molecular weight, where ᭺ (Mn,target 5 10.0 kg mol21); ᭛ (Mn,target 5 25.2 kg mol21); w (Mn,target 5 49.3 kg mol21).
-
time in both cases is indicated by the shift of the peaks in
the chromatogram to lower elution times (higher molecular
weight). The tailing on the low-molecular-weight side is
likely due to irreversible termination.
reinitiation. Fractionation was attempted to separate the
dead chains from the block copolymer, but with little suc-
cess; however, the DMAA content in the final polymer
decreased slightly (for instance, FDMAA 5 0.73 for MV-9 after
the fractionation and 0.90 before the fractionation). 31P NMR
spectroscopy was used to determine the amount of SG1
present at the chain ends (as phosphorus is present in
SG153). For instance, it was found that MV-2 was 13% living.
Such a low fraction of SG1-terminated chains explained the
poor reinitiation and broad dispersities of the final block
copolymers. However, 31P NMR spectroscopy might be mis-
leading because the concentration of phosphorus in the poly-
mer was very low and could lead to substantial error.44,55
The GPC chromatograms of the chain extensions shown in
Figure 5 did not indicate an obvious peak corresponding to
unreacted macroinitiator; this may be due to the molecular
weights/hydrodynamic volumes of the species not being
very different from one another.
Block Copolymer Synthesis
To test the ability of the final polymers to reinitiate a fresh
batch of monomer, chain extension experiments were per-
formed with DMAA monomer in DMF solvent (Scheme 1).
The formulations for all the chain extension experiments are
listed in Table 2. The monomodal shift in the GPC chromato-
grams [Fig. 5(b)] for MV-10 suggests that the macroinitiator
(MV-5) was significantly SG1-terminated. Similar behavior
was observed for the other two chain extension experiments
(MV-8 and MV-9). Table 4 lists the properties of all final
block copolymers. The dispersity values of the final chain-
extended samples are quite high, which is suggestive of poor
FIGURE 5 Gel permeation chromatograms (GPC) of (a) a characteristic copolymerization of MV-5 and (b) DMAA chain extension
(MV-10) done in a 50 wt % solution in DMF at 110 ꢀC from MV-5 macroinitiator.
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 4 Molecular Weight Characterization for Poly(MAEPYR-stat-VBK)-b-poly(DMAA) Block Copolymers
IDa
Macroinitiator IDa
Macroinitiator Mn (kg mol21
)
Macroinitiator Mw/Mn
Mn (kg mol21
)
Mw/Mn
FDMAA
b
b
b
b
c
MV-8
MV-9
MV-10
MV-2
MV-3
MV-5
5.9
1.43
1.71
1.56
12.4
19.9
53.5
2.37
2.26
2.96
0.92
0.73
0.89
8.4
11.8
a
b
Experiments for poly(MAEPYR-stat-VBK) copolymerizations and chain
extensions are denoted MV-Z with M representing N-(2-methacryloylox-
yethyl)-pyrrolidone, V representing 9-(4-vinylbenzyl)-9H-carbazole, and Z
representing the experiment number.
Number-average molecular weight (Mn ) and dispersity index were
determined by GPC.
c
FDMAA is the molar fraction of DMAA in the final block copolymer as
determined by 1H NMR spectroscopy.
Solution Properties of Statistical and Block Copolymers
The MAEPYR-rich statistical copolymers and block copoly-
mers were tested for thermoresponsive behavior in aqueous
media. Properties of stimuli-responsive polymers are con-
nected to the microstructure of the polymer; therefore, well-
defined polymers with narrow molecular weight distribu-
tions are essential for clear determination of CPT.18,27–29
Davis and coworkers synthesized MAEPYR homopolymers
via conventional polymerization.41 Although there is no
information given about the microstructure of the resulting
Effect of Solution Concentration on the Transition of
Statistical Copolymers
Particle size was measured by DLS (see the Experimental
section) for the statistical copolymers during the phase tran-
sition. Poly(MAEPYR-stat-VBK) statistical copolymers with
low VBK content (e.g., MV-1) were readily soluble at room
temperature in deionized water. The light scattering intensity
was low, as measured by DLS, from room temperature until
about 39 ꢀC [Fig. 6(a)]. During heating, the intensity
increased significantly when the temperature reached the
critical temperature, which is the CPT.20 From Figure 6(a) it
is seen that the intensity increased dramatically at 42 ꢀC.
The intensity is related to the particle size: larger particles
scatter more light, which would result in higher intensity.58
Hence, it is possible to determine the CPT by monitoring the
change in the population of particle sizes. Figure 6(c,b)
shows the temperature dependence of average hydrody-
namic radius, <Rh>, of poly(MAEPYR-stat-VBK) chains (MV-
1, Mn 5 10.2 kg mol21, FVBK 5 0.01) in one heating cycle as
measured by DLS, and normalized absorbance of the heating
ramp of the same sample as measured by UV–Vis spectros-
copy, respectively. The CPT at 50% normalized absorbance
was 49.7 and 40.9 ꢀC for 0.5 and 1 wt % solutions, respec-
tively, and the <Rh> increased at a temperature above 47
and 42 ꢀC, respectively. The <Rh> measurements showed
considerable scatter above the CPT, which is likely due to
sedimentation.40 The CPT by DLS was determined when the
intensity-weighted particle distribution was shifted from the
lower to the higher <Rh> as indicated in Figure 6(a). The
transparent solution became opaque when the temperature
was above the CPT owing to aggregation20,59 of polymer
chains. The temperature-induced phase separation was
reversible, and the polymer solution became clear again
when the temperature was below the CPT.
polymers, the CPTs in water (0.7 wt %) were measured at a
21
ꢀ
heating rate of 1 C min and the CPTs ranged between 29
and 34 ꢀC.41 Cai and coworkers reported CPTs of 2 wt %
solutions ranging from 52.8 to 71.5 ꢀC for MAEPYR homo-
polymers synthesized by RAFT polymerization with weight-
average molecular weights of 105.4–20.6 kg mol21, respec-
tively.40 Here, the effect of composition, solution concentra-
tion, and heating rate on CPT was studied.
Effect of Composition on the Transition of Statistical
Copolymers
Poly(MAEPYR-stat-VBK) statistical copolymers exhibited
tuneable CPT by varying VBK content. Increasing the content
of the more hydrophobic monomer, VBK, in the final copoly-
mer structure will decrease the CPT owing to the decreased
degree of hydrogen bonding with water.7,56,57 Poly(MAEPYR)
homopolymer (see Experimental section for details about the
synthesis and characterization) has
a
CPT of 59.0 ꢀC
(Mn 5 10.9 kg mol21), and the CPT decreased to 49.7 C (fꢀor
MV-1 with FVBK 5 1 mol %, Mn 5 10.2 kg mol21) and 39.6 C
(for MV-5 with FVBK 5 6 mol %, Mn 5 11.8 kg mol21) for pol-
y(MAEPYR-stat-VBK) statistical copolymers at 1 wt % solu-
tion concentration. Poly(MAEPYR) homopolymer (Mn 5 10.9
ꢀ
kg mol21) has a CPT of 53.5 C at 2 wt %, which is slightly
ꢀ
different from the values reported by Cai and coworkers
(71.5 ꢀC at 20.6 kg mol21) at the same concentration. The
difference may arise from different polymer samples used
(e.g., molecular weight is relative to either PMMA or PS
standards, for this and Cai’s work, respectively) owing to
higher dispersity values associated with polymers used in
this work (compared to Ð of 1.11–1.13 for polymers used in
Cai’s work) and/or owing to the conditions used for the CPT
determination (e.g., heating rates, and turbidity vs. light scat-
tering methods).
At lower solution concentration (Fig. 7) the CPT increases
because the formation of polymer aggregates is a much
slower process at low polymer concentrations.59 Depending
on the solution cꢀoncentration, MV-1 exhibited a difference of
ꢀ
as much as 16.6 C in CPTs (from 50.9 C for 0.3 wt % solu-
ꢀ
tion to 34.3 C for 4 wt % solution). These observations are
consistent with literature.19,59–62 For instance, the CPT for
poly(DMAEMA-stat-styrene) copolymers decreased by 7–10
ꢀC when the solution concentration increased from 0.1 to 0.3
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
9

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
wt %.60 Similarly, PNIPAM polymer samples showed a signif-
icant concentration dependence below 5 wt %, and the
curve reaches a plateau at higher concentrations (15–20 wt
%).59 The polymer samples, other than MV-1 and MV-5,
were water insoluble, and were not considered for the
analysis.
Hysteresis and Effect of Scanning Rate on the Transition
of Statistical Copolymers
One of the most important properties of “smart” polymers is
the reversibility of the phase transition. It was found that
the temperature-induced phase separation was reversible,
but the hysteresis effect was noticeable as the temperature
of the phase transition in the cooling process was different
from the one in the heating process. The hysteresis observed
was as low as 2 ꢀC and as high as 10 ꢀC when the rate of
heating and cooling was 1 ꢀC min21. Cai and coworkers
reported no profound hysteresis effect for one heating–cool-
ing cycle for one of the many poly(MAEPYR) homopolymers
they examined (Mw 5 105.4 kg mol21).40 In this study, the
Mn s of the copolymers studied were 10.2–11.8 kg mol21
(relative to PMMA standards) and the hysteresis effect
becomes larger as the molecular weight decreases: smaller
chains tend to form larger aggregates owing to interchain
association, which are more difficult to dissociate during the
cooling cycle owing to the increased degree of hydrogen
bonding.17,62 It is interesting to note that the poly(MAEPYR)
homopolymer (synthesized by RAFT for comparison, sꢀee
Experimental section) exhibited very little hysteresis of 2 C
when observed under theꢀsame conditions (1 wt % solution,
heating/cooling rate of 1 C min21) as MV-1 copolymer. The
observed difference for our samples suggests the presence of
weak van der Waals interactions18 and/or hydrophobic inter-
actions of the polymer’s backbone with the ethyl spacers
and the carbazole groups from the VBK comonomer.
The CPT strongly depends on the heating rate.57,59,63 As the
heating rate was increased, the CPT for MV-1 samples at 1
wt % concentration was found to increase as well. For exam-
ple, the CPT increased from 39.8 to 45.2 ꢀC for the heating
rates ranging from 0.2 to 4 ꢀC min21, respectively. Similar
behavior was observed previously in other systems.19,59,63
For instance, increasing the heating rate from 0.02 to 5 ꢀC
min21 resulted in about 11 ꢀC increase in CPT for PNIPAM,
as determined by turbidity methods.59 Poly(N,N-diethylacry-
ꢀ
lamide) (DEAAM) showed only a 6 C increase in CPT when
21 63
ꢀ
the heating rate was altered from 0.06 to 5 C min
.
Poly-
mer–polymer interactions responsible for the phase separa-
tion are time dependent and at higher heating/cooling rates
there is less time to react to the environmental change.57,63
As a result, the transitions become more diffusive19,63 with a
FIGURE 6 (a) The distribution of different particle size popula-
tions during DLS measurement of statistical poly(MAEPYR-stat-
VBK) copolymer MV-1 (Mn 5 10.2 kg mol21
,
-
D 5 1.58,
FVBK 5 0.01) at 1 wt % solution during a heating ramp; (b) the
normalized absorbance as measured by UV–Vis, and (c) aver-
age hydrodynamic radius (<Rh>) as measured by DLS of MV-1
at 0.5 wt % solution (triangles) and 1 wt % solution (circles)
during one heat cycle.
FIGURE 6.
10
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
by chain extending the copolymer with a hydrophilic mono-
mer. Many water-soluble monomers are polymerizable by
NMP, including N,N-dimethylacrylamide (DMAA),64 N-vinyl-
pyrrolidone (VP),23 2-N-morpholinoethyl acrylate (MEA),19
and 4-acryloylmorpholine (4AM).65 In this study, DMAA was
used in the synthesis of a hydrophilic block. With the tem-
perature changes, the block copolymer might gain or lose its
amphiphilicity, resulting in formation or dissolution of
micelles.7 There are two types of micelles that can be
formed. One type consists of the temperature-responsive
corona, and the other one has the temperature-responsive
inner core. In the case of poly(MAEPYR-stat-VBK)-b-poly
(DMAA) block copolymers, the hydrophilic DMAA block
would form the outer shell, and the inner core would consist
of the temperature-responsive copolymer resulting in stable
micelle formation below the LCST.45 Figure 8 shows the tem-
perature dependence of average hydrodynamic radius, <Rh>,
of poly(MAEPYR-stat-VBK)-b-poly(DMAA) chains (MV-8 and
MV-9) in one heating cycle as measured by DLS, and normal-
ized absorbance on the heating ramp of the same sample as
measured by UV–Vis spectroscopy. The increase of average
size near the CPT suggests the presence of intermicellar
aggregation,66 which is minimal owing to the hydrophilic
poly(DMAA) block. As a result, the final average hydrody-
namic radius for the block copolymer is smaller than that of
the statistical copolymers.45 Also, above the LCST, the hydro-
phobic effect in the core might disappear owing to the inter-
actions of the hydrophilic block, if it is too long, resulting in
micelle destruction.7 Addition of the hydrophilic poly(DMAA)
block slowed the rate of phase transition for the block
copolymers compared with the statistical copolymers. From
Figure 8 it is seen that the phase transition observed by UV–
Vis spectroscopy occurs over several tens of degrees (e.g.,
FIGURE 7 CPT dependence of MV-1 (D, Mn 5 10.2 kg mol21
,
,
D 5 1.58,
F
VBK 5 0.01) and MV-5 (1, Mn 5 11.8 kg mol21
-
-
D 5 1.56, FVBK 5 0.06) samples on composition and MV-1 sam-
ples on solution concentration as measured by UV–Vis spec-
troscopy with a heating ramp of 1 ꢀC min21
.
partial loss of reversibility.19 For very high heating rates (3–
4 C min21), the transition of MV-1 samples during the cool-
ꢀ
ing process was very broad, and the solution became clear
only at room temperature.
Transition of Block Copolymers
For the poly(MAEPYR-stat-VBK) statistical copolymers, CPT
was tuned by the inclusion of the more hydrophobic mono-
mer, VBK, which resulted in lower CPTs as VBK content
increased. It is possible to tune CPT in the opposite direction
ꢀ
45–85 C for MV-8 chains). Similar behavior was observed in
the literature.45,66 Dispersity of the final polymer influenced
FIGURE 8 Temperature dependence of the normalized absorbance (black line) as measured by UV–Vis during the heat cycle and
average hydrodynamic radius (<Rh>) (gray circles) as measured by DLS during the heat cycle of (a) MV-8 chains (Mn 5 12.4 kg
mol21, D 5 2.37, FDMAA 5 0.92) and (b) MV-9 (Mn 5 19.9 kg mol21, D 5 2.26, FDMAA 5 0.73), both in 1 wt % solution.
-
-
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
11

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 5 The Difference in the CPTs for Poly(MAEPYR-stat-
VBK)-b-Poly(DMAA) Block Copolymers as Determined by UV–
Vis Spectroscopy and DLS, and the Corresponding CPT Values
Determined on a Heating Ramp
CPT by
Macroinitiator UV–Vis
CPT by DLS
(ꢀC)
/
solution
IDa
IDa
(ꢀC)
(wt %)
MV-8 MV-2
MV-9 MV-3
MV-10 MV-5
MV-10 MV-5
60.3
69.4
82.9
66.2
50
39
48
n/a
1
1
1
2
a
Experiments for poly(MAEPYR-stat-VBK) copolymerizations and chain
extensions are denoted MV-Z with M representing N-(2-methacryloylox-
yethyl)-pyrrolidone (MAEPYR), V representing 9-(4-vinylbenzyl)29H-car-
bazole (VBK), and Z representing the experiment number. Composition
and molecular weight data of the final blocks can be found in Table 4.
FIGURE 9 Time dependence of the average hydrodynamic
radius (<Rh>) of 1 wt % statistical copolymer (MV-1, Mn 5 10.2
the phase transition as well. Smaller chains tend to form
larger aggregates owing to additional hydrogen bonding
present in the interchain associations, whereas larger chains
are likely to undergo only intrachain contractions.17 The two
types of interactions do not occur simultaneously. With the
kg mol21, D 5 1.58) and block copolymer (MV-10, Mn 5 53.5 kg
-
mol21, D 5 2.96) solutions at a constant temperature of 50 ꢀC.
-
profound for MV-10, indicating that the micelles formed
were relatively stable.
-
broad molecular weight distributions (e.g., D 5 2.26–2.96 for
MV-8, MV-9, and MV-10 polymers) the transition will become
more diffuse as a result of the difference in time required to
form aggregates for chains of different length.
CONCLUSIONS
The BlocBuilder/SG1 initiators were able to copolymerize
MAEPYR-rich compositions in a relatively controlled manner
with a minimum of 5 mol % VBK in the initial feed to give lin-
ear number-average molecular weight Mn versus conversion
(until about 60% conversion) with fairly narrow molecular
weight distributions of the final polymers (dispersity of 1.4–
1.7). The ability of the final copolymers to reinitiate a fresh
monomer batch was tested using DMAA monomer. In all cases
studied, the growth of the polymer chains was monitored via
GPC, where a shift of the peak to a lower elution time indicated
the increase in molecular weight. The final statistical and block
copolymers were tested for LCST-type behavior in aqueous
solutions. The effect of VBK composition shifted the CPT from
As predicted, the hydrophilic monomer shifted the CPT
upward, if compared to the CPT of the precursor MV-1. The
CPTs of the macroinitiators (MV-2 and MV-3 in this case)
were not determined owing to limited solubility of the poly-
mer in aqueous solution. The final CPTs for block copoly-
mers as determined by UV–Vis spectroscopy and DLS are
summarized in Table 5. There is a discrepancy between the
determined CPT values, which is solely owing to the analyti-
cal technique used. UV–Vis spectroscopy relies on turbidity
measurements, so it is sensitive only to the macroscopic
phase separation, which is a slow process at low solution
concentration.59 DLS, on the other hand, can detect the col-
lapse of a single polymer chain,59 before the macroscopic
phase separation, which occurs usually at a higher tempera-
ture. From Figure 8 it is seen that the transition, as deter-
mined by DLS, is sharper than the one determined by UV–
Vis; therefore, CPT values for block copolymers were
reported from the DLS measurements.
21
ꢀ
49.7 C (for MV-1 with Mn 5 10.2 kg mol and FVBK 5 0.01)
to 39.6 ꢀC (for MV-5 with Mn 5 11.8 kg mol21 and
F
VBK 5 0.06) at 1 wt % solution concentration. The CPT also
decreased with increasing solution concentration, and MV-1
exhibited a difference of as much asꢀ16.6 ꢀC in CPTs (from
ꢀ
50.9 C for 0.3 wt % solution to 34.3 C for 4 wt % solution).
The CPT was found to be a function of the heating/cooling
rate, and the transitions became more diffusive in nature at
higher rates. For the block copolymers, the particle size meas-
urements indicated that the block copolymers had a broader
transition compared with statistical copolymers and that the
block copolymer micelles were relatively stable upon pro-
longed heating above the CPT.
Stability of Statistical and Block Copolymers
The stability of the phase transition was tested by observing
the changes in particle size over a prolonged period of time.
The 1 wt % solutions of statistical and block copolymers,
MV-1 and MV-10, respectively, were held above the CPT at
ꢀ
50 C for 15 h (Fig. 9). The changes in <Rh> for MV-1 over
the 15-h period indicated that the system was not stable. At
the end of the experiment, the polymer settled at the bottom
of the vial, which suggested that macrophase separation had
occurred.18 In contrast, the changes in <Rh> are less
ACKNOWLEDGMENTS
The authors are very grateful for the funding support from the
NSERC Discovery Grant and the Eugenie Ulmer Lamothe (EUL)
12
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
29 P. Liu, H. Xie, H. Tang, G. Zhong, H. Zhang, J. Polym. Sci.
Part A: Polym. Chem. 2012, 50, 3664–3673.
Scholarship. The authors thank Bruce Lennox for the use of his
UV–Vis spectrometer. Finally, they thank Arkema, Inc. for help
in obtaining BlocBuilder (Mickael Havel) and SG1 (Noah Macy).
30 Handbook of Radical Polymerization. K. Matyjaszewski, T. P.
Davis, Eds.; Wiley-Interscience: Hoboken, NJ, 2002; pp 920.
31 G. Moad, E. Bicciocchi, M. Chen, J. Chiefari, C. Guerrero-
Sanchez, M. Haeussler, S. Houshyar, D. Keddie, E. Rizzardo, S.
H. Thang, J. Tsanaktsidis, In Progress in Controlled Radical
Polymerization: Mechanisms and Techniques; K. Matyjaszew-
ski, B. S. Sumerlin, N. Tsarevsky, Eds.; Oxford University Press:
Oxford UK, 2012; Vol. 1100, pp 243–258.
REFERENCES AND NOTES
1 B. Jeong, A. Gutowska, Trends Biotechnol. 2002, 20, 360.
2 T. Okano, Adv. Polym. Sci. 1993, 110, 179–197.
3 B. R. Twaites, C. D. Alarcon, D. Cunliffe, M. Lavigne, S.
Pennadam, J. R. Smith, D. C. Gorecki, C. Alexander, J. Con-
trolled Release 2004, 97, 551–566.
32 G. Moad, E. Rizzardo, S. H. Thang, Acc. Chem. Res. 2008,
41, 1133–1142.
4 D. Roy, J. N. Cambre, B. S. Sumerlin, Prog. Polym. Sci. 2010,
35, 278–301.
33 G. Moad, J. Chiefari, Y. K. Chong, J. Krstina, R. T. A.
Mayadunne, A. Postma, E. Rizzardo, S. H. Thang, Polym. Int.
2000, 49, 993–1001.
5 N. Lomadze, H. M. Schneider, Tetrahedron Lett. 2005, 46,
751–754.
34 G. Moad, R. T. A. Mayadunne, E. Rizzardo, M. Skidmore, S.
H. Thang, ACS Symp. Ser. 2003, 854, 520–535.
6 C. D. H. Alarcon, S. Pennadam, C. Alexander, Chem. Soc.
Rev. 2005, 34, 276–285.
35 M. Cunningham, M. Lin, J. A. Smith, J. Ma, K. McAuley, B.
Keoshkerian, M. Georges, Prog. Colloid Polym. Sci. 2004, 124,
88–93.
7 E. S. Gil, S. M. Hudson, Prog. Polym. Sci. 2004, 29, 1173–
1222.
36 R. B. Grubbs, Polym. Rev. 2011, 51, 104–137.
8 A. Chilkoti, M. R. Dreher, D. E. Meyer, D. Raucher, Adv. Drug
Delivery Rev. 2002, 54, 613–630.
37 C. J. Hawker, A. W. Bosman, E. Harth, Chem. Rev. 2001,
101, 3661–3688.
9 J. Weidner, Drug Discovery Today 2001, 6, 1239–1241.
10 M. Yamato, C. Konno, A. Kushida, M. Hirose, M. Utsumi, A.
Kikuchi, T. Okano, Biomaterials 2000, 21, 981–986.
38 K. Matyjaszewski, N. V. Tsarevsky, Nat. Chem. 2009, 1, 276–
288.
11 K. Uchida, K. Sakai, E. Ito, O. H. Kwon, A. Kikuchi, M.
Yamato, T. Okano, Biomaterials 2000, 21, 923–929.
39 K. Matyjaszewski, W. A. Braunecker, Prog. Polym. Sci. 2007,
32, 93–146.
12 R. J. Young, P. A. Lovell, Introduction to Polymers, 3rd ed.;
CRC Press: Boca Raton, FL, 2011; pp 312–314.
40 J. Deng, Y. Shi, W. Jiang, Y. Peng, L. Lu, Y. Cai, Macromole-
cules 2008, 41, 3007–3014.
13 Y. Maeda, T. Higuchi, I. Ikeda, Langmuir 2000, 16, 7503–
7509.
41 G. M. Iskander, L. E. Baker, D. E. Wiley, T. P. Davis, Polym.
Chem. 1998, 39, 4165–4169.
14 H. Chang, L. Shen, C. Wu, Macromolecules 2006, 39, 2325–
2329.
42 B. Charleux, J. Nicolas, O. Guerret, Macromolecules 2005,
38, 5485–5492.
15 Y. Ding, X. Ye, G. Zhang, Macromolecules 2005, 38, 904–
908.
43 J. Nicolas, S. Brusseau, B. Charleux, J. Polym. Sci. Part A:
Polym. Chem. 2010, 48, 34–47.
16 M. Heskins, J. E. Guillet, J. Macromol. Sci. Part A: Chem.
1968, 2, 1441–1455.
44 B. Lessard, E. J. Ling, M. S. T. Morin, M. Maric, J. Polym.
Sci. Part A: Polym. Chem. 2011, 49, 1033–1045.
17 Y. C. Tang, Y. W. Ding, G. Z. Zhang, J. Phys. Chem. B 2008,
112, 8447–8451.
45 B. Lessard, M. Maric, J. Polym. Sci. Part A: Polym. Chem.
2011, 49, 5270–5283.
18 J. Sun, Y. Peng, Y. Chen, Y. Liu, J. Deng, L. Lu, Y. Cai, Mac-
romolecules 2010, 43, 4041–4049.
46 C. Zhang, M. Maric, J. Polym. Sci. Part A: Polym. Chem.
2013, 51, 4702–4715.
19 B. H. Lessard, X. Savelyeva, M. Maric, Polymer 2012, 53,
5649–5656.
47 B. H. Lessard, E. J. Ling, M. Maric, Macromolecules 2012,
45, 1879–1891.
20 P. Liu, L. Xiang, Q. Tan, H. Tang, H. Zhang, Polym. Chem.
2013, 4, 1068–1076.
48 B. H. Lessard, M. Maric, Can. J. Chem. Eng. 2012, 91, 618–
629.
21 X. Liu, Y. Xu, Z. Wu, H. Chen, Macromol. Biosci. 2012, 13,
147–154.
49 W. Zhang, Y. Yan, N. Zhou, Z. Cheng, J. Zhu, C. Xia, X. Zhu,
Eur. Polym. J. 2008, 44, 3300–3305.
22 J. Nicolas, Y. Guillaneuf, C. Lefay, D. Bertin, D. Gigmes, B.
Charleux, Prog. Polym. Sci. 2013, 38, 63–235.
50 C. Dire, J. Belleney, J. Nicolas, D. Bertin, S. Magnet, B.
Charleux, J. Polym. Sci. Part A: Polym. Chem. 2008, 46, 6333–
6345.
23 P. Bilalis, M. Pitsikalis, N. Hadjichristidis, J. Polym. Sci. Part
A: Polym. Chem. 2006, 44, 659–665.
51 Y. Guillaneuf, D. Gigmes, S. R. A. Marque, P. Astolfi, L.
Greci, P. Tordo, D. Bertin, Macromolecules 2007, 40, 3108–
3114.
24 H. A. Ravin, A. M. Seligman, J. Fine, N. Engl. J. Med. 1952,
247, 921–929.
25 U. Karaaslan, M. Parlaktuna, Prepr. Pap. Am. Chem. Soc.
Div. Fuel Chem. 2002, 47, 355–358.
52 A. C. Greene, R. B. Grubbs, Macromolecules 2010, 43,
10320–10325.
26 N. Daraboina, P. Linga, Chem. Eng. Sci. 2013, 93, 387–394.
53 J. Nicolas, C. Dire, L. Mueller, J. Belleney, B. Charleux, S. R.
A. Marque, D. Bertin, S. Magnet, L. Couvreur, Macromolecules
2006, 39, 8274–8282.
27 S. Aoshima, H. Oda, E. Kobayashi, J. Polym. Sci. Part A:
Polym. Chem. 1992, 30, 2407–2413.
28 H. Mori, H. Iwaya, A. Nagai, T. Endo, Chem. Commun. 2005,
4872–4874.
54 J. Nicolas, B. Charleux, O. Guerret, S. Magnet, Macromole-
cules 2004, 37, 4453–4463.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
13

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
55 C. Lefay, J. Belleney, B. Charleux, O. Guerret, S. Magnet,
Macromol. Rapid Commun. 2004, 25, 1215–1220.
61 J. Y. Huang, K. Matyjaszewski, Macromolecules 2005, 38,
3577–3583.
56 H. M. L. Lambermont Thijs, R. Hoogenboom, C.-A. Fustin, C.
Bomal-D’Haese, J.-F. Ghy, U. S. Schubert, J. Polym. Sci. Part
A: Polym. Chem. 2009, 47, 515–522.
62 F. J. Hua, X. G. Jiang, D. J. Li, B. Zhao, J. Polym. Sci. Part
A: Polym. Chem. 2006, 44, 2454–2467.
63 I. Idziak, D. Avoce, D. Lessard, D. Gravel, X. X. Zhu, Macro-
molecules 1999, 32, 1260–1263.
57 H. Y. Liu, X. X. Zhu, Polymer 1999, 40, 6985–6990.
58 R. Xu, In Particle Characterization: Light Scattering Methods;
B. Scarlett, Ed.; Kluwer Academic: New York, 2002, pp 56–105.
64 D. Li, W. J. Brittain, Macromolecules 1998, 31, 3852–3855.
65 X. Savelyeva, B. H. Lessard, M. Maric, Macromol. React.
Eng. 2012, 6, 200–212.
59 C. Boutris, E. G. Chatzi, C. Kiparissides, Polymer 1997, 38,
2567–2570.
66 J. E. Chung, M. Yokoyama, T. Aoyagi, Y. Sakurai, T. Okano,
J. Controlled Release 1998, 53, 119–130.
60 C. Zhang, M. Maric, Polymers 2011, 3, 1398–1422.
14
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000