Synthesis and crosslinking of hyperbranched poly(n-nonyl acrylate) to form organogels

Article abstract of DOI:10.1002/pola.27700

(2-Bromo-n-nonan-1-oxycarbonyl)ethyl acrylate was synthesized as an inimer for self-condensing vinyl polymerization (SCVP) to produce hyperbranched poly(n-nonyl acrylate), either as a homopolymer or as a copolymer with n-nonyl acrylate. The inimer was hom

Full text of DOI:10.1002/pola.27700

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Synthesis and Crosslinking of Hyperbranched Poly(n-nonyl acrylate) to
Form Organogels
Yangtian Lu,1 Dibyendu Debnath,1 R. A. Weiss,2 Coleen Pugh1
1Department of Polymer Science, University of Akron, Akron, Ohio 44325-3909
2Department of Polymer Engineering, University of Akron, Akron, Ohio 44325-3909
Correspondence to: R. A. Weiss (E-mail: rweiss@uakron.edu) or C. Pugh (E-mail: cpugh@uakron.edu)
Received 30 October 2014; accepted 25 April 2015; published online 3 June 2015
DOI: 10.1002/pola.27700
ABSTRACT: (2-Bromo-n-nonan-1-oxycarbonyl)ethyl acrylate was
synthesized as an inimer for self-condensing vinyl polymeriza-
tion (SCVP) to produce hyperbranched poly(n-nonyl acrylate),
either as a homopolymer or as a copolymer with n-nonyl acry-
late. The inimer was homopolymerized and copolymerized by
atom transfer radical polymerization (ATRP) and activator gen-
erated by electron transfer ATRP to produce soluble polymers
mer of (2-bromo-n-nonan-1-oxycarbonyl)ethyl acrylate synthe-
sized in bulk. Crosslinked organogels that swell in
tetrahydrofuran were formed when the rate of crosslinking
decreased using acetonitrile solutions. Dynamic shear and
stress relaxation experiments demonstrated that the dry net-
work behaves as a covalently crosslinked soft gel, with a glass
transition at 250 8C according to differential scanning calorime-
-
C
V
with broad polydispersities (up to D 5 9.91), which is character-
try. 2015 Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym.
istic of hyperbranched polymers produced by SCVP. The result-
ing hyperbranched (co)polymers were crosslinked by atom
transfer radical coupling in both one-pot and two-step proce-
dures. The radical–radical crosslinking reaction is extremely
efficient, resulting in hard plastic particles from the homopoly-
Chem. 2015, 53, 2399–2410
KEYWORDS: gels; hydrophobic; hyperbranched polymer; inimer;
organogel; polyacrylates
INTRODUCTION Atom transfer radical polymerization (ATRP)1,2
is a versatile method for the polymerization of a wide range of
monomers, including (meth)acrylates,3 to produce novel well-
defined polymers. ATRP can also be applied to self-condensing
vinyl polymerizations (SCVPs),4 which provide a convenient,
one-batch synthesis of hyperbranched polymers.5–7 Recently, we
designed a new type of acrylate inimer8,9 that produces hyper-
branched polyacrylates by ATRP.10 These acrylate inimers were
designed to produce nearly exact architectural analogs of the
corresponding linear polyacrylate, with an ester group attached
to every other carbon atom along the polymer backbone and a
nonfunctionalized alkyl ester attached as a free pendant group.
Because of the chemical similarity of these branched polyacry-
zewski proposed an activator generated by electron transfer
(AGET) ATRP system that uses a standard ATRP initiator and
a higher oxidation state transition metal complex (X–Mtn11
/
L);12 the higher oxidation state transition metal complex is
converted to the activator (Mtn) by reaction with a reducing
agent such as ascorbic acid. This method has the characteris-
tics of normal ATRP, but with the added benefit that it uses
a more oxidatively stable catalyst complex in the reaction
mixture. The current article describes the use of an AGET
ATRP system, in addition to traditional ATRP, for the SCVP of
an acrylate inimer with a bulky n-nonyl ester substituent
(Scheme 1).
1
lates and their linear analogs, their H and 13C NMR spectra are
€
The simulations of the kinetics of SCVP proposed by Muller
nearly identical, with no unique resonances for the branch
points. Nevertheless, the hyperbranched architecture is sup-
ported by the characteristically broad molecular weight distribu-
tions throughout the polymerization;10 by the significant
underestimation in their molecular weights measured by gel per-
meation chromatography relative to a linear standard;10 and by
two-dimensional mass spectrometry characterization;11 however,
none of these techniques are quantitative.
and coworkers demonstrated that SCVP produces polymers
with a high degree of branching and numerous initiating
sites at high conversion.13,14 Therefore, SCVPs of the acrylate
inimers produce hyperbranched polyacrylates with many
halogen atoms throughout each molecule. We previously lim-
ited the acrylate inimer conversions to 80–90% to prevent
crosslinking by inevitable radical–radical coupling.10 The
high concentration of halogen atoms throughout the hyper-
branched structure should provide an ideal route for cross-
linking these polymers by radical–radical coupling reactions
The ATRP of these acrylate inimers is slow, especially with
bulky ester substituents. Recently, Jakubowski and Matyjas-
C
V
2015 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410
2399

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 1
Synthesis and crosslinking of hyperbranched poly(n-nonyl acrylate); only intramolecular crosslinking is shown.
[Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
to prepare crosslinked systems such as gels. Polymeric gels
are solvent-swollen systems in which the polymer is physi-
cally or chemically crosslinked to form a three-dimensional
network. Gels have attracted considerable attention because
of their many applications, including the encapsulation and
delivery of drugs.15,16 Covalent crosslinking affects many
important physical properties such as the modulus, thermal
response, solvent/water ingress, and fracture behavior.17
These crosslinked networks do not dissolve in any solvent,
but rather swell on solvent adsorption.
therefore very slow,19 this study will demonstrate that the
rate of radical–radical coupling in hyperbranched poly(n-
nonyl acrylate)s is highly efficient and useful for crosslinking.
Although the radical–radical coupling reaction can be influ-
enced by altering the solvent polarity, our objective is to syn-
thesize a hyperbranched polyacrylate-based organogel with
high absorption for organic solvents.
EXPERIMENTAL
Materials
Acrylic acid (99%; Aldrich), acryloyl chloride (99%; Aldrich),
anisole (99.7%; Aldrich), L-ascorbic acid (Fisher), cupric bro-
mide (99%; Aldrich), HBr (ꢀ48 wt %; Acros), NaNO2 (97%;
Fisher), NaHCO3 (99.7%; Mallinakrodt), NaBr (98%; Acros),
methanol (99.8%; Aldrich), n-nonanol (98%; Sigma),
N,N,N0,N00,N00-pentamethyldiethylenetriamine (PMDETA, 99%;
Aldrich), D,L-serine (98%; Aldrich), and p-toluenesulfonic acid
(Fisher) were used as received. Acrylic anhydride (19–55%
yield), 2-bromo-3-hydroxypropionic acid (40% yield), and
(2-bromo-2-methoxycarbonyl)ethyl acrylate (55–61% yield)
were synthesized as described previously.9 Azobisisobutyro-
nitrile (AIBN, 99.7%; Sigma Aldrich) was recrystallized from
methanol below 40 8C. Cuprous bromide (98%; Aldrich) was
purified by stirring it in glacial acetic acid overnight, washing
it several times with methanol, and then drying.30 Diethyl
ether was dried by distillation from purple sodium benzo-
phenone ketyl under N2. Benzene was distilled from CaH2.
Reagent-grade tetrahydrofuran (THF) was dried by distilla-
tion from purple sodium benzophenone ketyl under N2. Trie-
thylamine (99%; Aldrich) was distilled from KOH under
nitrogen.
The hyperbranched polyacrylates reported in this study with
hydrophobic n-nonyl substituents should be useful for pre-
paring organogels. Organogels18,19 are a special class of gels
that swell in the presence of an organic solvent. Although
the chemical composition of the gel determines which sol-
vents will cause swelling, the molecular architecture of the
gel determines its absorption capacity. For example, Wu
et al.20 reported that complexes of sulfonated poly[styrene-
block-(ethylene-ran-butylene)-block-styrene] and a polypro-
pylenimine dendrimer can absorb 22 times its weight in die-
sel and in toluene. Organogels are useful for a variety of
applications, including dermal drug delivery, such as with a
lecithin-based organogel;21 colorimetric detection of ions,
such as fluoride ions using a poly(aryl ether) dendritic
organogel connected to an anthracene chromophore through
acylhydrazone linkages;22 and fluorometric detection of small
molecules, such as aliphatic amines using a naphthalimide-
containing organogel.23 Thermoresponsive organogels can
also be formed by complexing a polyelectrolyte with an
oppositely charged surfactant.24
As the hyperbranched polyacrylates contain abundant halo-
gen atoms throughout their structure, they may produce
organogels by atom transfer radical coupling25 (ATRC) using
an efficient reducing agent (Scheme 2). Linear polystyrene
and poly(methyl acrylate) (PMA) with halogenated end
groups can be coupled by radical–radical coupling in high
efficiency (99%);26,27 even poly(methyl methacrylate) radi-
cals can be coupled if styrene is first added to the terminal
halide unit.28 Although the equilibrium constant for radical
formation is 100 times lower29 for PMA-Br than polystyrene-
Br, and the rate of linear PMA radical–radical coupling is
SCHEME 2
Mechanism for atom transfer radical coupling
(ATRC).
2400
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Techniques
of 0.205 N, after which the viscoelastic properties were
measured at 25 8C, covering an angular frequency (x) range
of 1023 to 102 rad/s. Stress relaxation experiments were
measured on similar samples as those used for the dynamic
shear experiments. A constant strain of 10% was used, and
the time-dependent torque, from which the time-dependent
shear stress was calculated, was measured from 0.02 s to as
long as 13.3 h at constant temperatures of 25, 50, 75, and
100 8C.
All reactions were performed under a N2 atmosphere using a
Schlenk line unless noted otherwise. In some cases, polymer-
izations were performed in what will be referred to as a
“vial reactor,” created by covering a 20-mL glass-threaded
vial with an upside–down rubber septum, piercing the sep-
tum with a short needle, and then covering the pierced sep-
tum with a 14/20 gas adapter, which can then be attached
to the Schlenk line. Thin-layer chromatography was per-
formed using silica gel plates (200 lm particle size w/
UV254; Sorbent Technologies) with a polyester backing.
Synthesis of n-Nonyl 2-Bromo-3-hydroxypropionate by
Esterification of 2-Bromo-3-hydroxypropionic Acid with
n-Nonanol
1H (300 MHz) and 13C (75 MHz) NMR spectra (d, ppm) were
recorded on a Varian Mercury 300 instrument. All spectra
were recorded in CDCl3, and the resonances were measured
relative to residual solvent resonances and referenced to tet-
ramethylsilane (0.00 ppm). Number–average (Mn) and
weight–average (Mw) molecular weights relative to linear
polystyrene [gel permeation chromatography (GPCPSt)] and
A solution of 2-bromo-3-hydroxypropionic acid (18 g, 0.11
mol), n-nonanol (19 g, 0.13 mol), and p-toluenesulfonic acid
(81 mg, 0.47 mmol) in dry benzene (50 mL) was stirred at
75–80 8C for 45 h while collecting the benzene–water azeo-
trope in a Dean-Stark trap. After cooling to room tempera-
ture, CH2Cl2 (200 mL) was added, and the organic phase
was washed three times with 4 wt % aqueous NaHCO3
(50 mL each) and once with saturated aqueous NaCl
(50 mL), and then dried over Na2SO4. After filtration and
removing the solvent by rotary evaporation, 32 g (69%) of
crude product was obtained as a slightly yellow liquid, and
the product was purified by column chromatography
(Rf 5 0.15) using silica gel as the stationary phase and
CH2Cl2/Et2O (95:5 v/v) as the eluant to obtain 22 g (48%
yield) of n-nonyl 2-bromo-3-hydroxypropionate as a colorless
oil. The 1H NMR spectrum in Figure 1 demonstrates that n-
nonyl 2-bromo-3-hydroxypropionate is contaminated with a
small amount of an ester (small resonances a0, b0, and c0)
-
polydispersities (D 5 Mw/Mn) were determined by GPC from
calibration curves of log Mn versus elution volume at 35 8C
using THF as solvent (1.0 mL/min), a guard column and set
of 50 Å, 100 Å, 104 Å, and linear (50–104 Å) Styragel 5 lm
columns, a Waters 486 tunable UV/vis detector set at
254 nm, a Waters 410 differential refractometer, and Millen-
ium Empower 2 software. The samples (ꢀ0.1 g/L) were dis-
solved overnight and filtered through a 0.45-lm PTFE filter.
A TA Q200 differential scanning calorimeter was used to
determine the glass transition temperature of a 3.99 mg
sample of dry gel, which was read as the middle of the
change in heat capacity. All heating and cooling rates were
10 8C/min. The transition temperature was calibrated using
an indium standard. Thermogravimetric analysis (TGA) of a
3.01 mg dry gel sample was performed with a TA Q50
instrument from room temperature to 700 8C at a heating
rate of 10 8C/min under a continuous flow (50 mL/min) of
air. Wide-angle X-ray diffraction was performed on a Rigaku
Rapid II diffractometer with an image-plate area detector
and a tube-anode X-ray generator (Cu Ka radiation) operated
at 40 kV and 30 mA (k 5 1.5418 Å). Powder diffraction pat-
terns were collected in reflection geometry on a fixed v-axis
(458) goniometer continuously spinning around the u-axis.
The intensity–2h plots were obtained via azimuthal integra-
tion of the two-dimensional diffraction pattern with a 2h
resolution of 0.10. Peak positions were calibrated using sili-
con powder as the standard.
formed by
a
second esterification with 2-bromo-3-
hydroxypropionic acid. This oligomer could not be separated
from n-nonyl 2-bromo-3-hydroxypropionate by column chro-
matography using a range of eluants: hexanes (Rf 5 0), 9:1
(v:v) hexanes/THF (Rf 5 0), CH2Cl2 (Rf 5 0.08), CHCl3
(Rf 5 0.10), 9:1 (v:v) CHCl3/Et2O (Rf 5 0.18), 9:1 (v:v)
CH2Cl2/Et2O (Rf 5 0.35), 1:1 (v:v) hexanes/ethyl acetate
(Rf 5 0.40), 1:1 (v:v) hexanes/THF (Rf 5 0.43), 1:1 (v:v)
CH2Cl2/Et2O (Rf 5 0.50), 9:1 (v:v) CH2Cl2/THF (Rf 5 0.5–0.8),
THF (Rf 5 0.60), and Et2O (Rf 5 0.68). n-Nonyl 2-bromo-3-
hydroxypropionate was therefore used in the next step with-
out further purification, and the oligomer was removed
when the inimer was purified.
Synthesis of (2-Bromo-2-n-nonan-1-oxycarbonyl)ethyl
Acrylate
Two solutions of acrylic anhydride (9.3 g, 74 mmol) in THF
(50 mL) and of triethylamine (7.4 g, 73 mmol) in THF
(50 mL) were simultaneously added dropwise over 4 min to
an ice-cooled solution of nonyl 2-bromo-3-hydroxypropionate
(21 g, 72 mmol) in THF (50 mL). After stirring the reaction
mixture at room temperature for 15 h, it was poured into ice
water (150 mL), and the resulting mixture was stirred for
1 h. THF was removed by rotary evaporation. As the product
formed an oil rather than precipitating as a solid from the
remaining aqueous mixture, the organic mixture was
extracted four times (50 mL each) with CH2Cl2. The combined
Dynamic shear and stress relaxation experiments were con-
ducted with an ARES G2 rheometer (TA Instruments) using
an 8-mm-diameter parallel-plate geometry. Frequency
sweeps with 1% strain were used to measure the linear
dynamic shear properties of the dry polymer. Strain sweeps
at a frequency of 0.25 rad/s indicated that the linear region
extended to ꢀ20% strain. Disk-shaped samples of approxi-
mately 8-mm-diameter were die-cut and placed between the
plates with a gap of 2.14 mm. The sample was equilibrated
for 30 min until the normal force achieved a constant value
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410
2401

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 1
1H NMR (300 MHz) spectrum of n-nonyl 2-bromo-3-hydroxypropionate and its impurity formed by condensation oligo-
merization. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
organic layers were washed four times with 4 wt % aqueous
NaHCO3 (50 mL each) and once with saturated aqueous NaCl
(50 mL), and then dried over Na2SO4. After filtration and sol-
vent removal by rotary evaporation, 16 g (63%) of crude
product was obtained as a yellow liquid, and the crude prod-
uct was purified by column chromatography (Rf 5 0.38) using
silica gel as the stationary phase and CHCl3/hexanes (95:5 v/
v) as the eluant to yield 13 g (52%) of (2-bromo-2-n-nonan-
1-oxycarbonyl)ethyl acrylate as a colorless liquid.
was removed by rotary evaporation. The white residue was
purified by column chromatography (Rf 5 0.57) using silica
gel as the stationary phase and CH2Cl2 as the eluant to yield
5.7 g (67%) of n-nonyl acrylate as a colorless liquid.
1H NMR: 0.86 (t, CH3, 3J 5 6.7 Hz), 1.28 (m, [CH2]6), 1.67
(quint, CH2CH2O2C, J 5 6.8 Hz), 4.13 (t, OCH2CH2, 3J 5 6.7
Hz), 5.78 (dd, CHHb@ trans to CO2, 2Jab 5 1.6 Hz, 3Jbc 5 10.4
Hz), 6.10 (dd, 5 CHc, 3Jac 5 17.3 Hz, 3Jbc 5 10.4 Hz), 6.37
2
3
(CHaH@ cis to CO2, Jab 5 1.6 Hz, Jac 5 17.3 Hz).
1H NMR: 0.87 (t, CH3, 3J 5 6.6 Hz,), 1.27 (m, [CH2]6), 1.66
(quint, CH2CH2O2C, J 5 6.8 Hz), 4.19 (m, CH2O2C), 4.43 (dd,
ATRP of (2-Bromo-2-methoxycarbonyl)ethyl Acrylate for
Screening Purposes
3
3
2
CHBr, J 5 5.9 Hz, J 5 8.1 Hz), 4.53 (dd, CHHO2C, J 5 11.4 Hz,
3J 5 5.9 Hz), 4.58 (dd, CHHO2C, J 5 11.4 Hz, J 5 8.1 Hz), 5.88
(dd, CHHb@ trans to CO2, 2Jab 5 1.3 Hz, 3Jbc 5 10.4 Hz), 6.11
(dd, 5 CHc, 3Jac 5 17.3 Hz, 3Jbc 5 10.4 Hz), 6.44 (CHaH@ cis to
CO2, 2Jab 5 1.3 Hz, 3Jac 5 17.3 Hz). 13C NMR: 14.4 (CH3), 22.6
(CH2CH3), 25.7 (CH2[CH2]2O2C), 28.3 (CH2CH2O2C), 29.1
(CH2[CH2]4H), 29.2 (CH2), 29.4 (CH2[CH2]3H), 31.8
(CH2CH2CH3), 40.7 (CHBr), 64.2 (CH2CHBr), 66.5 (CO2CH2CH2),
127.4 (CH@), 131.9 (CH2@), 165.0 (acrylate C@O), 167.6
(CO2CH2).
2
3
In a typical procedure, (2-bromo-2-methoxycarbonyl)ethyl
acrylate (0.33 g, 0.97 mmol) was added under a stream of
nitrogen to a well-stirred mixture of CuBr (7.5 mg, 52 lmol)
and PMDETA (11 mg, 62 lmol) in anisole (0.6 mL) in a dry
Schlenk tube, and the tube was sealed with a glass stopper.
After stirring at room temperature for 10 min, the inimer
mixture was degassed by five freeze–pump/30-min thaw
cycles, and the polymerization tube was backfilled with N2.
The polymerization mixture was stirred at 130 8C for 9 h
and then quenched by immersing the Schlenk tube into liq-
uid N2. The contents of the polymerization tube were
thawed, exposed to the atmosphere, diluted with THF
(7 mL), and precipitated into water/methanol (3/7 v/v,
30 mL). The precipitate was collected and reprecipitated
twice from THF (10 mL) into water/methanol (3/7 v/v,
30 mL) to yield 46 mg (14%) hyperbranched PMA as a
Synthesis of n-Nonyl Acrylate
A solution of acryloyl chloride (3.8 g, 41 mmol) in dry
CH2Cl2 (50 mL) was added dropwise over 30 min to a solu-
tion of n-nonanol (5.6 g, 39 mmol) and triethylamine (4.0 g,
40 mmol) in dry CH2Cl2 (100 mL). The reaction mixture was
stirred at room temperature for 48 h, and then, the solvent
2402
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 2
X-ray diffraction pattern of the brown-colored hyperbranched poly(n-nonyl acrylate) produced by atom transfer radical
polymerization of (2-bromo-2-n-nonan-1-oxycarbonyl)ethyl acrylate system in anisole at 90 8C for 12 h using CuBr and PMDETA as
the catalyst. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
slightly yellow powder; Mn 5 1.00 3 104 g/mol, D 5 2.76.
The product was not purified further to remove copper
catalyst(s).
aqueous solution of ascorbic acid (16 lL, 30 lmol) was
added using a gas-tight syringe, and the polymerization tube
was backfilled with N2. The polymerization mixture was
stirred at 75 8C for 5 h and then quenched by immersing the
Schlenk tube into liquid N2. The contents of the polymeriza-
tion tube were thawed, exposed to the atmosphere, diluted
with THF, and passed through a plug of basic activated alu-
mina to remove catalyst(s). The THF solution was concen-
trated to 5 mL and precipitated into methanol (25 mL). The
precipitate was collected and reprecipitated three times from
THF (5 mL) into methanol (25 mL) to yield 26 mg (48%) of
-
ATRP of (2-Bromo-2-n-nonan-1-oxycarbonyl)ethyl
Acrylate
A mixture of CuBr (6.2 mg, 42 lmol) and PMDETA (8.9 lL,
43 lmol) in anisole (0.5 mL) was sealed in a dry Schlenk
tube with a rubber septum and degassed by five freeze–
pump–thaw (5–15–10 min) cycles. Degassed (2-bromo-2-n-
nonan-1-oxycarbonyl)ethyl acrylate (0.53 g, 1.5 mmol) was
added to the catalyst mixture using a gas-tight syringe, and
the polymerization tube was backfilled with N2. The poly-
merization mixture was stirred at 90 8C for 12 h and then
quenched by immersing the Schlenk tube into liquid N2. The
contents of the polymerization tube were thawed, exposed to
the atmosphere, diluted with THF (5 mL), passed through a
plug of basic activated alumina to remove the catalyst(s),
and precipitated into methanol (25 mL). The precipitate was
collected and reprecipitated twice from THF (5 mL) into
methanol (25 mL) to yield 56 mg (48%) hyperbranched
poly(n-nonyl acrylate) as a brown viscous rubber. As the
brown color still remained after purification to remove cop-
per catalyst(s), the polymer was not injected into the GPC
for molecular weight analysis; X-ray diffraction (Fig. 2) con-
firmed that the sample contained trace amounts of CuBr,
CuO (black),31 and Cu2O (red),32 which evidently causes the
brown color.
hyperbranched poly(n-nonyl acrylate) as a white viscous
3
-
gum; Mn 5 9.89 3 10 g/mol, D 5 8.91.
AGET ATRP of (2-Bromo-2-n-nonan-1-oxycarbonyl)ethyl
Acrylate
In a typical procedure, (2-bromo-2-n-nonan-1-oxycarbonyl)-
ethyl acrylate (0.53 g, 1.5 mmol) was added under a stream
of nitrogen to a well-stirred mixture of CuBr2 (6.6 mg, 30
lmol) and PMDETA (6 lL, 31 lmol) in acetonitrile
(0.25 mL) in a dry Schlenk tube. The Schlenk tube was
sealed with a glass stopper, and its contents were degassed
by five freeze–pump–thaw (5–15–10 min) cycles. A saturated
FIGURE 3
X-ray diffraction pattern of the hyperbranched
poly(n-nonyl acrylate)s produced by activator generated by
electron transfer (AGET) atom transfer radical polymerization
of (2-bromo-2-n-nonan-1-oxycarbonyl)ethyl acrylate system at
60–75 8C using ascorbic acid as the reducing agent (Table 2,
Entries 7 and 8). [Color figure can be viewed in the online
issue, which is available at wileyonlinelibrary.com.]
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410
2403

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 3
Synthesis of (2-bromo-2-n-nonan-1-oxycarbonyl)ethyl acrylate.
In contrast to the sample prepared by ATRP (Fig. 2), copper
contaminants are not detectable by X-ray diffraction in the
hyperbranched polymers isolated from these AGET ATRP
(Fig. 3).
with a rubber septum and degassed by four freeze–pump–
thaw cycles. second Schlenk tube containing hyper-
branched poly(n-nonyl acrylate) (0.53 g, Mn 5 1.56 3 104 g/
A
-
mol, D 5 6.26) in anisole (20 mL) was sealed with a rubber
septum and degassed by four freeze–pump–thaw (5–15–10
min) cycles. After the addition of PMDETA (6.8 mL, 32 mmol)
to the polymer solution via a gas-tight syringe, an aliquot of
the polymer solution (5 mL, ꢀ13 mmol/L) was transferred
via a cannula to the catalyst solution. The Schlenk tube was
then backfilled with N2, and the contents were stirred at 75
8C for 92 h. The resulting green insoluble powder (43 mg,
32% yield) was collected in a glass frit.
AGET ATRP Copolymerization of n-Nonyl Acrylate and (2-
Bromo-2-n-nonan-1-oxycarbonyl)ethyl Acrylate
In a typical example, a solution of (2-bromo-2-n-nonan-1-oxy-
carbonyl)ethyl acrylate (0.27 g, 0.77 mmol) and n-nonyl acry-
late (1.6 g, 8.0 mmol) was added to a well-stirred mixture of
CuBr2 (33 mg, 0.16 mmol) and PMDETA (33 lL, 0.16 mmol)
in CH3CN (1.25 mL) in a dry vial. The vial reactor was sealed
with a rubber septum, and its contents were degassed by four
freeze–pump–thaw (5–15–10 min) cycles. Solid ascorbic acid
(27 mg, 0.16 mmol) was added to the frozen contents of the
vial reactor. After one additional freeze–pump–thaw (5–15–
10) cycle, the polymerization mixture was stirred at 75 8C for
8 days. The contents of the vial reactor were diluted with
THF, decanted from the crosslinked gel in the bottom of the
vial, and the solvent was removed in vacuo to yield 1.6 g
RESULTS AND DISCUSSION
Synthesis of the (2-Bromo-2-n-nonan-1-oxycarbonyl)ethyl
Acrylate Inimer
As outlined in Scheme 3, (2-bromo-2-n-nonan-1-oxycarbonyl)-
ethyl acrylate was synthesized by esterification of 2-bromo-
3-hydroxypropionic acid with n-nonyl alcohol, followed by
esterification of the resulting n-nonyl 2-bromo-3-
hydroxypropionate with acrylic anhydride, similar to our
previous synthesis of (2-bromo-2-methoxycarbonyl)ethyl
acrylate.9 Figure 4 presents the 1H NMR spectrum of the n-
nonyl inimer. In addition to the resonances at 5.88 ppm (dd,
CHH@ trans to CO2), 6.11 ppm (dd, @CH), and 6.44 ppm
(CHH@ cis to CO2) due to the acrylate double bond, the
(86%) yield of hyperbranched copolymer as a brownish vis-
4
-
cous oil (Mn 5 5.11 3 10 g/mol, D 5 6.31).
Crosslinking of Hyperbranched Poly(n-nonyl acrylate) In Situ
(2-Bromo-2-n-nonan-1-oxycarbonyl)ethyl acrylate (0.51 g,
1.5 mmol) was added under a stream of nitrogen to a well-
stirred mixture of CuBr2 (6.4 mg, 29 lmol) and PMDETA
(6.6 lL, 31 lmol) in CH3CN/H2O (9/1 v/v, 0.5 mL) in a dry
Schlenk tube. The Schlenk tube was sealed with a rubber
septum, and its contents were degassed by five freeze–
pump–thaw (5–15–10 min) cycles. A saturated aqueous solu-
tion of ascorbic acid (24 lL, 45 lmol) was added using a
gas-tight syringe. The polymerization mixture was stirred at
75 8C for 1.8 h. As crosslinking occurred, the polymerization
was quenched by immersing the Schlenk tube into liquid N2.
The contents of the polymerization tube were thawed,
exposed to the atmosphere, and 0.52 g of crosslinked hyper-
branched poly(n-nonyl acrylate) was isolated as a yellowish
gummy material by filtration and drying in vacuo for 2 days.
Copper catalysts and soluble materials were extracted three
times from the gummy material by immersing it in a solu-
tion of triethylamine (three drops) in THF (10 mL) for 4
days each to obtain 0.28 g (>55% yield) of a yellowish
crosslinked polymer; the yield is underestimated because the
amount of bromine atoms lost is not known.
1
most characteristic H resonances of the inimer are those at
4.43 ppm due to the CHBr methine, and at 4.53 ppm and
4.58 ppm due to the diastereotopic methylene protons alpha
to the brominated carbon. Figure 5 presents the correspond-
ing 13C NMR spectrum. In addition to the resonances at
127.4 ppm (CH@), 131.9 ppm (CH2@), and 165.0 ppm (acry-
late C@O), the most characteristic 13C resonances of the
inimer are those at 40.7 ppm (CHBr) and 66.5 ppm
(CH2O2C).
Polymerizations
We previously demonstrated that the methyl acrylate inimer
could be polymerized by both ATRP and reverse ATRP,8
although the conditions were not optimized. We therefore
briefly screened the ATRP of the methyl bromoinimer as a
function of time and concentration to determine when the
growing polymer starts to crosslink and gel under standard
ATRP conditions. As demonstrated by Entries 1 and 2 in
Table 1, the hyperbranched PMA remained soluble at short
polymerization times (ꢁ9 h), but underwent radical–radical
crosslinking when the polymerization time increased to 14 h.
In solution using anisole as the solvent, crosslinking evi-
dently involved significant intermolecular crosslinking, as
Crosslinking of Preformed Hyperbranched Poly(n-nonyl
acrylate) (Two-Step Procedure)
A mixture of CuBr (6.5 mg, 31 mmol) and AIBN (19 mg, 0.12
mmol) in anisole (1 ml) was sealed in a dry Schlenk tube
2404
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 4
1H NMR (300 MHz) spectrum of (2-bromo-2-n-nonan-1-oxycarbonyl)ethyl acrylate. [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
well as intramolecular crosslinking, as the polymerization
resulted in the formation of an insoluble gel. Surprisingly, in
the absence of solvent, the polymerization resulted in small,
hard plastic particles at long reaction times, evidently due to
extensive intramolecular crosslinking. The lack of macrogel
formation under more concentrated (bulk) conditions may
be explained by the high rate of radical–radical coupling,
which rapidly resulted in insoluble particles that were not
able to further react with other partially crosslinked mole-
cules in the system.
The bulky n-nonyl substituent should decrease the rates of
reactions of the new n-nonyl acrylate inimer relative to those
of the methyl inimer. As summarized in Table 2, we therefore
briefly screened ATRP and AGET ATRP conditions to deter-
mine under what conditions the growing hyperbranched
poly(n-nonyl acrylate) starts to crosslink and gel. The brown-
ish color of the sample in Entry 1 indicates that copper salts
remain in the polymer produced in anisole at 90 8C within
12 h under standard ATRP conditions; X-ray diffraction con-
firmed that this sample contained trace amounts of CuBr,
FIGURE 5
13C NMR (75 MHz) spectrum of (2-bromo-2-n-nonan-1-oxycarbonyl)ethyl acrylate.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410
2405

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Cursory Results from the Atom Transfer Radical Poly-
merization of (2-Bromo-2-methoxycarbonyl)ethyl Acrylate at
130 8Ca
trolled crosslinking by radical–radical coupling, and therefore
little soluble polymer. When water was eliminated from the
solvent system (Entry 7), the polymerization rate decreased,
and a relatively high yield of soluble polymer was obtained
at a short reaction time. (2-Bromo-2-n-nonan-1-oxycarbonyl)-
ethyl acrylate also copolymerized well with n-nonyl acrylate
using the same conditions. For example, a copolymerization
of 1:10 inimer/monomer produced an 86% yield of hyper-
branched copolymer that was still soluble after 8 days at 75
8C in acetonitrile using 1:1:1 [CuBr2]/[PMDETA]/[AA] and 55
equivalents of acrylate groups.
Entry
Solvent
Reaction Time
Observation
1
2
3
Anisole
Anisole
–
9 h
14 h
Hyperbranched polymerb
Swollen gel
4 days
Hard particles
a
[Inimer]:[CuBr]:[PMDETA] 5 17–19:1:1.
b
n 5 1.00 3 104 g/mol, D 5 2.76.
-
M
Figure 6 presents the 1H NMR spectrum of the hyper-
branched poly(n-nonyl acrylate) reported as Entry 1 in Table
3. As is typical of these hyperbranched polyacrylates,9,10 the
terminal vinyl resonances appear at 5.75–6.5 ppm; the
CO2CH2CHBr and CHBr resonances of the acrylate ester por-
tion of the polymer overlap at 4.5 ppm; and the CH2O2C res-
onance of the n-nonyl ester pendant groups appear at 3.9
ppm. The backbone protons resonate in the region from 1 to
3.5 ppm, under the prominent resonances at 0.8 ppm (CH3),
1.25 (A[CH2]6A) ppm, and 1.7 ppm (CH2CH2O2C) due to the
n-nonyl ester pendant groups. All of the resonances are
broad, as expected for a polymer, and no unique resonances
are detected for the branch points; therefore, the degree of
branching cannot be calculated for these hyperbranched
polyacrylates by NMR spectroscopy.10
CuO, and Cu2O (Fig. 2). The very low yields of insoluble
powders in Entries 2–4 demonstrate that DMF is not a good
solvent for this polymerization under both standard ATRP
and AGET ATRP conditions at 60–75 8C, as highly crosslinked
powders formed before the polymerization reached high con-
version, evidently due to the generation of a high concentra-
tion of radicals in this very polar solvent. Entries 5 and 6
demonstrate that the rate of reduction of Cu(II) with Cu(0)
was too slow for an effective AGET ATRP, as no polymer was
obtained. However, Entries 7 and 8 demonstrate that ascor-
bic acid is a promising reducing agent for AGET ATRP of the
n-nonyl inimer in acetonitrile and a mixture of acetonitrile
and water, as crosslinking does not occur until high conver-
sions. In these two cases, no copper was detected in the iso-
lated samples by X-ray diffraction (Fig. 3).
One-Pot In Situ versus Two-Step Crosslinking
The results of the AGET ATRP of 50 equivalents of (2-bromo-
2-n-nonan-1-oxycarbonyl)ethyl acrylate relative to CuBr2 in
solutions of 9:1 (v/v) acetonitrile and water at 75 8C are
summarized in Table 3. The broad polydispersities are char-
acteristic of hyperbranched polymers produced by SCVPs
due to propagation by both step and chain mechanisms.13,14
Comparison of Entries 1–3 demonstrates that the highest
yield of soluble hyperbranched polymer was produced using
a ratio of 1:1 [CuBr2]:[ascorbic acid]. Higher concentrations
of ascorbic acid (Entry 4) and inimer (Entries 5 and 6)
resulted in faster polymerization rates that led to uncon-
As summarized in Table 4, we compared two approaches for
crosslinking the hyperbranched poly(n-nonyl acrylate)s by
ATRC of the activated carbon–bromine bonds that are pres-
ent throughout the hyperbranched structure. In the first
route, the hyperbranched polymer was crosslinked in one
pot, in situ during its synthesis due to the inevitable radical–
radical coupling at high inimer conversion. The second route
used a two-step process in which the carbon–bromine bonds
of a preformed hyperbranched polymer were activated under
ATRC conditions, and the resulting radicals were coupled.
TABLE 2 Cursory Results from the ATRP and AGET ATRP of (2-Bromo-2-n-nonan-1-oxycarbonyl)ethyl Acrylate under Various
Conditionsa
[In]:[CuBr]:
[CuBr2]b
Reducing
Agent
[Reducing Agent]/
[CuBr2]
Temperature
% Yield
Entry
Solvent
Anisole
(8C)
Time (h)
(Observations)
1
2
3
4
5
6
7
8
35/1/0
47/1/0
52/0/1.6
50/0/1
43/0/0.9
50/0/1
40/0/1
50/0/1
–
–
–
90
60
70
75
75
25
60
75
12
40
13
18
24
74
15
27
45 (brown soluble polymer)
(Insoluble hard particles)
2 (hard particles)
2 (hard particles)
0
–
DMF
AA
AA
Cu(0)
Cu(0)
AA
AA
1.6
1
DMF
DMF
1.3
1.5
2.3
2
9:1 CH3CN/H2O
DMSO
0
5:1 CH3CN/H2O
CH3CN
13 (pseudo-crosslinked)c
Crosslinked gel
a
c
Inimer concentration 5 [In] 5 3 mmol/mL; AA, ascorbic acid.
Pseudo-crosslinked refers to a poorly soluble, swollen gel that dis-
b
Relative to
(PMEDTA).
1
equiv N,N,N0,N00,N00-pentamethyldiethylenetriamine
solved almost completely after extensive time in a good solvent; in this
4
-
case, Mn 5 7.04 3 10 , D 5 4.62.
2406
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 3 Results from AGET ATRP of (2-Bromo-2-n-nonan-1-oxycarbonyl)ethyl Acrylate under Various Conditions in 9:1 (v/v)
CH3CN/H2O at 75 8Ca
[CuBr2]/
[Inimer]
% Yieldb
Mn 3 1024
-
D
Entry
[PMDETA]/[AA]
(mmol/mL)
Time (h)
1
1:1:1
1:1:0.5
1:1:0.8
1:1:1.3
1:1:1
3
3
3
3
8
6
6
20
120
23
18
0.5
1
25
2
1.56
0.445
5.09
2.56
12.1
10.4
9.89
6.26
1.46
2.86
2.70
3.41
3.11
8.91
2
3
16
c
4
c
c
5
6
7d
1:1:1
1:1:1
5
48
a
c
[Inimer]/[CuBr2] 5 50; AA, ascorbic acid.
After precipitation and three reprecipitations to remove any unreacted
Small amount of soluble polymer extracted from crosslinked material.
In acetonitrile without water.
b
d
inimer.
Surprisingly, the two-step reaction produced insoluble, highly
crosslinked powders rather than a lighter crosslinked swol-
len gel from both a homopolymer of the inimer (Entry 1b)
and a copolymer (Entry 2b) in which the number of cross-
linkable sites were diluted using 10 equivalents of n-nonyl
acrylate. As exemplified in Figure 7(b), the resulting powders
were not soluble in THF or water (which were used to
extract copper catalysts and/or ligand); the greenish color in
THF and bluish color in water is due to the extraction of
copper ions. The lack of macrogel formation again indicates
that the rate of radical–radical coupling was too high, and
the crosslinked particles rapidly precipitated out of the solu-
tion before they were able to react intermolecularly with
other partially crosslinked molecules in the system.
onstrated by the example in Figure 7(a), the resulting cross-
linked poly(n-nonyl acrylate)s act as organogels and swell in
THF. The swelling ratio of the gel, as determined by the
weight of the swollen gel divided by the weight of the dried
gel, increased as the crosslink density decreased as a result
of copolymerization with the acrylate monomer. No cross-
linking occurred when the density of crosslinkable sites was
further decreased by copolymerization of the inimer with 50
equivalents of acrylate monomer.
Characterization of Crosslinked Poly(n-nonyl acrylate)
The thermal behavior of a representative dry gel (Table 2,
Entry 8) resulting from a hyperbranched homopolymer of
poly(n-nonyl acrylate) was characterized by differential scan-
ning calorimetry (Fig. 8) and thermal gravimetric analysis
(Fig. 9). As shown in Figure 8, the dry gel exhibits a low-
temperature glass transition at 250 8C, evidently due to the
long n-nonyl ester substituents, which decrease chain pack-
ing. The Tg of the dry gel is similar to that of linear poly(n-
nonyl acrylate) (Tg 5 257 8C).33 However, the steady change
in the baseline at lower temperature indicates that the glass
Crosslinking is evidently slower in the one-pot method using
AGET ATRP conditions. In this case, a gel formed from the
entire reaction mixture using both a homopolymerization
(Entry 1a), which therefore results in a high concentration of
crosslinkable sites, and a copolymerization of inimer diluted
with 10 equivalents of n-nonyl acrylate (Entry 2a). As dem-
FIGURE 6
merization of (2-bromo-2-n-nonan-1-oxycarbonyl)ethyl acrylate in 9:1 (v/v) CH3CN/H2O at 75 8C (Table 3, Entry 1).
1H NMR (300 MHz) spectrum of hyperbranched poly(n-nonyl acrylate) produced by AGET atom transfer radical poly-
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410
2407

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 4 Results of One-Pot versus Two-Step Crosslinking of Hyperbranched Poly(n-nonyl acrylate) with Varying Concentrations
of Crosslinkable Sites
Two-Step
One-Step Crosslinking
Entries
[Inimer]/[M]a
Crosslinking
Product
Swelling Ratiob
1a and 1b
2a and 2b
3a
1/0
Gel
4
Plastic particles
Plastic particles
–
1/10
1/50
Gel
34
No crosslinking
a
b
M 5 monomer (n-nonyl acrylate).
Swelling ratio 5 weight of swollen gel/weight of dry crosslinked
polymer.
transition may cover a broader range of temperatures, possi-
bly due to both the nonuniform structure of hyperbranched
polymers and the inherent irregularity of crosslinking. As
shown by the TGA scan in Figure 9, the dry gel is quite ther-
mally stable in air up to 150 8C; therefore, its viscoelastic
behavior can be studied at temperatures up to at least 100
8C without degradation occurring. The dry gel starts to
slowly degrade at temperatures above 150 8C, with 3 wt %
loss at 275 8C. Degradation is rapid at temperatures above
325 8C.
without defects. The G00 frequency dependence in Figure 10
is expected to have a slope of 1 for a material exhibiting vis-
cous flow; the higher slope of the G00 frequency dependence
demonstrates that the dry gel is a rubbery-like solid and not
a liquid.
The data in Figure 10 cannot distinguish between a network
containing covalent crosslinks and a physical network simply
of chain entanglements or weak physical crosslinks. The
stress relaxation data in Figure 11 resolve this uncertainty. If
the network structure for the sample characterized in Figure
10 was only due to physical crosslinks or chain entangle-
ments, the stress in a relaxation experiment should go to
zero as time goes to infinity. The stress relaxation results
shown in Figure 11 were run for nearly 14 h, and the stress
is clearly approaching a finite value, Ginf, as expected for a
covalently crosslinked polymer.34 Moreover, the theory of
The viscoelastic behavior of the same dry gel at 1% strain is
shown by the dynamic shear data in Figure 10. The elastic
modulus (G0) exhibited a relatively weak frequency depend-
ence, whereas the loss modulus (G00) showed a stronger fre-
quency dependence in the frequency range of 0.001–100
rad/s. Moreover, G0 > G00 for all frequencies studied at 25 8C.
This indicates that the material is a viscoelastic solid at 25
8C, and the data in Figure 10 are consistent with a time-
dependent network. The magnitude of G0 (103 to 104 Pa) is
consistent with that of a soft gel. The frequency dependence
of G0 in Figure 10 is indicative of a weak network that would
arise as a result of physical, reversible crosslinks; defects in
a covalently crosslinked network,34 such as dangling chains
formed during the crosslinking step or random crosslinking
of reactive ends of propagating hyperbranched polymer; or a
combination of chemical and physical crosslinks with or
rubber elasticity predicts that the modulus, in this case, Ginf
,
should scale linearly with temperature.35
Ginf was determined for the data in Figure 11 by fitting a
stretched exponential36 to the data,
d
GðtÞ5a1b½expð2t=cÞ ꢂ;
(1)
where a is the equilibrium relaxation modulus (Ginf), which
is zero for a liquid and finite for a solid, c is a characteristic
FIGURE 7
(a) Crosslinked gel of hyperbranched poly(n-nonyl acrylate) formed in situ during the copolymerization of 1:10 inim-
er:monomer before (left) and after swelling (right; swelling ratio 5 34) in THF. (b) Insoluble crosslinked hard particles (red circles)
of hyperbranched poly(n-nonyl acrylate) (Table 4, Entry 1b) formed by crosslinking preformed homopolymer after 6 h swelling
time in THF (5 mL) and in water (5 mL). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.
com.]
2408
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 10
Variation of storage modulus (G0) and loss modu-
lus (G00) of dry gel (Table 2, Entry 8) as a function of angular
frequency (x) at 25 8C. [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
FIGURE 8
Differential scanning calorimetry trace (10 8C/min;
third heating scan, which is equivalent to the first and second
heating scans) of dry gel (Table 2, Entry 8).
CONCLUSIONS
relaxation time, d is a stretching parameter, which is a func-
tion of the distribution of relaxation times (0 < d < 1), and
b 5 G(inf) - G(0). This equation is often used to represent
relaxation data where there is a wide distribution of relaxa-
tion times. The stretched exponential provides a simple fit to
data with many relaxation times, although it sacrifices the
ability to determine specific relaxation times; that is, instead
a single relaxation time is determined. The solid lines in Fig-
ure 11 are the model fits, and the parameters for the four
sets of data are summarized in Table 5.
The acrylate inimer, (2-bromo-2-n-nonan-1-oxycarbonyl)ethyl
acrylate, can be polymerized and copolymerized with n-nonyl
acrylate to high molecular weight in a relatively short time
by AGET ATRP (co)polymerization in acetonitrile at 75 8C
using 1:1:1 CuBr2]/[PMDETA]/[ascorbic acid]. The resulting
hyperbranched poly(n-nonyl acrylate)s are readily cross-
linked via the many bromine atoms throughout the structure.
However, crosslinking of preformed hyperbranched polymers
generate hard plastic powders under ATRC conditions, evi-
dently due to rapid radical–radical coupling, which does not
allow partially crosslinked material to remain in solution
long enough to undergo sufficient intermolecular
The values of a 5 Ginf increased with temperature, and the
characteristic relaxation time, c, was independent of temper-
ature until the temperature reached 100 8C. The theory of
rubber elasticity predicts that the modulus, in this case, Ginf
,
should scale linearly with temperature.37 The inset in Figure
11 shows that the temperature dependence of Ginf is linear,
which is consistent with rubber elasticity theory. This result
confirms that these samples are covalently crosslinked.
FIGURE 11
Stress relaxation behavior of dry gel (Table 2,
Entry 8) at four different temperatures. The inset shows the
temperature dependence of Ginf. The data curves correspond
from top to bottom at long time (104 s): 100 8C (top) and 75 8C,
50 8C, and 25 8C (bottom). [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
FIGURE 9
Thermal gravimetric analysis scan (10 8C/min) of
dry gel (Table 2, Entry 8) in air.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410
2409

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
10 C. Pugh, A. Singh, R. Samuel, K. M. Bernal Ramos, Macro-
molecules 2010, 43, 5222–5232.
TABLE 5 Fitting Parameters for Using eq 1 fGðtÞ5a1b ½exp
d
ð2t=cÞ ꢂg Used to Fit the Data in Figure 11
11 K. Chaicharoen, M. J. Polce, A. Singh, C. Pugh, C.
Wesdemiotis, Anal. Bioanal. Chem. 2008, 392, 595–607.
25 8C
50 8C
75 8
100 8C
12 W. Jakubowski, K. Matyjaszewski, Macromolecules 2005, 38,
4139–4146.
a (Pa)
b (Pa)
d
1,917
943
2,179
767
2,302
420
2,552
275
€
13 A. H. E. Muller, D. Yan, M. Wulkow, Macromolecules 1997,
30, 7015–7023.
0.950
1.163
0.950
0.9538
0.950
0.8896
0.872
0.8541
€
14 D. Yan, A. H. E. Muller, K. Matyjaszewski, Macromolecules
c (s)
1997, 30, 7024–7033.
15 T. M. Don, M. L. Huang, A. C. Chiu, K. H. Kuo, W. Y. Chi,
Mater. Chem. Phys. 2008, 107, 266–273.
crosslinking. To form a gel rather than small plastic particles,
the rate of crosslinking must be decreased, which is more
readily achieved in one pot in solution at high inimer conver-
sions during the synthesis of hyperbranched polymers and
copolymers. The resulting crosslinked polymers are hydro-
phobic due to the n-nonyl ester substituents and act as orga-
nogels that swell in THF. The swelling ratio increases as the
number of crosslinks in the gel decreases, which was
16 P. R. Ninawe, S. J. Parulekar, Ind. Eng. Chem. Res. 2012, 51,
1741–1755.
17 C. M. Sahagun, S. E. Morgan, ACS Appl. Mater. Interfaces
2012, 4, 564–572.
18 P. Terech, R. G. Weiss, Chem. Rev. 1997, 97, 3133–3159.
19 D. Dasgupta, J. M. Guenet, Macromol. Chem. Phys. 2012,
214, 1885–1892.
20 Y. Wu, T. Zhang, Z. Xu, Q. Guo, J. Mater. Chem. A 2015, 3,
1906–1909.
achieved by copolymerization with
a standard acrylate
monomer (n-nonyl acrylate). Without solvent, the dry net-
work behaves as a viscoelastic solid with a glass transition
at 250 8C. The linear temperature dependence of Ginf con-
firms that the dry networks are covalently crosslinked.
21 R. Kumar, O. P. Katare, AAPS Pharm. Sci. Tech. 2005, 6,
E298–E310.
22 P. Rajamalli, E. Prasad, Org. Lett. 2011, 13, 3714–3717.
23 X. Cao, T. Zhang, A. Gao, K. Li, Q. Cheng, L. Song, M.
Zhang, Org. Biomol. Chem. 2014, 12, 6399–6405.
ACKNOWLEDGMENTS
24 Y. Liu, A. Lloyd, G. Guzman, K. A. Cavicchi, Macromolecules
2011, 44, 8622–8630.
This work was supported by the National Science Foundation
through CHE-1112326 and CMMI-1300212. The authors thank
Jing Wang and Professor Stephen Z. D. Cheng for the X-ray scat-
tering analyses.
25 S. Yurteri, I. Cianga, Y. Yagci, Macromol. Chem. Phys. 2003,
204, 1771–1783.
26 T. Sarbu, K. Y. Lin, J. Ell, D. J. Siegwart, J. Spanswick, K.
Matyjaszewski, Macromolecules 2004, 37, 3120–3127.
27 T. Sarbu, K. Y. Lin, J. Spanswick, R. R. Gil, D. J. Siegwart, K.
Matyjaszewski, Macromolecules 2004, 37, 9694–9700.
REFERENCES AND NOTES
28 C. F. Huang, Y. Ohta, A. Yokoyama, T. Yokozawa, Macromo-
lecules 2011, 44, 4140–4148.
1 J. S. Wang, K. Matyjaszewski, Macromolecules 1995, 28,
7572–7573.
29 M. N. C. Balili, T. Pintauer, Inorg. Chem. 2010, 49, 5642–
5649.
2 T. E. Patten, J. Xia, T. Abernathy, K. Matyjaszewski, Science
1996, 272, 866–868.
30 R. N. Keller, H. D. Wycoff, Inorg. Synth. 1946, 2, 1–4.
3 M. J. Ziegler, K. Matyjaszewski, Macromolecules 2001, 34,
31 J. F. Xu, W. Ji, Z. X. Shen, S. H. Tang, X. R. Ye, D. Z. Jia, X.
Q. Xin, J. Solid State Chem. 1999, 147, 516–519.
415–424.
4 J. M. J. Frechet, M. Henmi, I. Gitsov, S. Aoshima, M. R.
Leduc, R. B. Grubbs, Science 1995, 269, 1080–1083.
32 M. Salavati-Niasari, F. Davar, Mater. Lett. 2009, 63, 441–443.
33 A. R. Katritzky, S. Sild, V. Lobanov, M. Karelson, J. Chem.
Inf. Comput. Sci. 1998, 38, 300–304.
5 K. Matyjaszewski, S. G. Gaynor, A. Kulfan, M. Podwika, Mac-
romolecules 1997, 30, 5192–5194.
34 M. Rubinstein, R. Colby, Polymer Physics, 1st ed.; Oxford
University Press; New York, 2003; Chapter 7, p 293.
6 K. Matyjaszewski, S. G. Gaynor, Macromolecules 1997, 30,
7042–7049; (b) K. Matyjaszewski, J. Pyun, S. G. Gaynor, Mac-
romol. Rapid Commun. 1998, 19, 665–670.
35 A. V. Tobolsky, D. W. Carlson, N. Indictor, J. Polym. Sci.
1961, 54, 175–192.
7 C. Gao, D. Yan, Prog. Polym. Sci. 2004, 29, 183–275.
36 G. Williams, D. C. Watts, Trans. Faraday Soc., 1970, 66, 80–
85.
8 C. Pugh, A. Singh, WO2008045299-A1, 2008; (b) C. Pugh, A.
Singh, U.S. Patent 8,524,942, 2013.
37 M. Rubinstein, R. Colby, Polymer Physics, 1st ed.; Oxford
University Press; New York, 2003; Chapter 9, p 362.
9 C. Pugh, B. Raveendra, A. Singh, R. Samuel, G. Garcia, Syn-
lett. 2010, 13, 1947–1950.
2410
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 53, 2399–2410