Synthesis and properties of hyperbranched poly(ether sulfone)s prepared by self-polycondensation of novel AB2 monomer

Article abstract of DOI:10.1002/pola.26172

Hyperbranched poly(ether sulfone)s were prepared by the self-polycondensation of the novel AB₂ monomer, 4-(3,5- hydroxyphenoxy)-4'-fluorodiphenylsulfone. The high-molecular-weight polymers were isolated in good yields. The degree of branching (

Full text of DOI:10.1002/pola.26172

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Synthesis and Properties of Hyperbranched Poly(ether sulfone)s Prepared
by Self-Polycondensation of Novel AB2 Monomer
Mitsutoshi Jikei, Daisuke Uchida, Yuuki Haruta, Yuuki Takahashi, Kazuya Matsumoto
Department of Applied Chemistry, Akita University, 1-1, Tegatagakuen-machi, Akita-shi, Akita 010-8502, Japan
Correspondence to: M. Jikei (E-mail: mjikei@gipc.akita-u.ac.jp)
Received 13 March 2012; accepted 3 May 2012; published online
DOI: 10.1002/pola.26172
ABSTRACT: Hyperbranched poly(ether sulfone)s were prepared
by the self-polycondensation of the novel AB2 monomer, 4-
(3,5-hydroxyphenoxy)-40-fluorodiphenylsulfone. The high-mo-
lecular-weight polymers were isolated in good yields. The
degree of branching (DB) of the resulting polymers was investi-
gated by the preparation of dendritic and linear model com-
pounds. The DB determined by gated decoupling 13C NMR
measurements was in the range 0.17–0.41 and was dependent
on the base used for the self-polycondensation. It was found
that cesium fluoride was an effective base to form the polymer
having the DB of 0.41. The resulting hyperbranched poly(ether
sulfone)s showed good solubility in organic solvents. The solu-
bility and the glass transition temperature of the polymers
were influenced by the terminal functional groups. The unique
thermal crosslinking phenomenon was observed during
the DSC measurements of the hydroxyl-terminated hyper-
C
V
branched poly(ether sulfone) under air condition. 2012 Wiley
Periodicals, Inc. J Polym Sci Part A: Polym Chem 000: 000–000,
2012
KEYWORDS: crosslinking; degree of branching; hyperbranched;
poly(ether sulfones); polycondensation; thermal crosslinking
INTRODUCTION Hyperbranched polymers are unique branched
macromolecules, which have consecutive branching points in
their repeating units and many terminal functional groups.1–4
As analog to dendrimers, hyperbranched polymers have a
three-dimensional architecture in contrast to linear poly-
mers. Many functional groups present in the hyperbranched
polymers could allow fine-tuning of the material properties,
such as solubility and thermal properties. The viscosity of
hyperbranched polymers is generally low even if the molecu-
lar weight is high enough due to the lack of chain entangle-
ments. One of the potential advantages of the hyperbranched
polymers lies in its production process. Hyperbranched poly-
mers can be prepared by the one-step polymerization of the
designed monomers. Therefore, mass production of hyper-
branched polymers can be easily achieved in contrast to den-
drimers. It should be mentioned that hyperbranched poly-
mers contain an irregular structure as the linear units,
which is different from the perfect-branched dendrimers.
ing points to the rigid aromatic polymers results in an
improved solubility, low viscosity, and low crystallinity. As
the chain entanglements are reduced, the mechanical proper-
ties of the hyperbranched polymers become lower than the
corresponding linear polymers. Therefore, the application of
the hyperbranched engineering plastics as bulk materials
seems to be limited. On the other hand, it has been observed
for the copolymers of hyperbranched aromatic polyamides29
and polyimides30 that the small number of the branching
components dramatically changes the properties. The fact
suggests that the hyperbranched polymers have the potential
as additives, such as a viscosity modifier, crystallization
inhibitor, solubility controller, and compatibilizer.
There are only a few papers describing the study of hyper-
branched poly(ether sulfone)s. Hay and coworker reported
the hyperbranched poly(ether sulfone)s prepared from AB2
and A2B monomers.26 Fossum and coworker reported the
hyperbranched poly(ether sulfone) from the self-polyconden-
sation of 3,5-difluuoro-40-hydroxydiphenyl sulfone as an AB2
monomer.27 In this study, we report the synthesis and prop-
erties of hyperbranched poly(ether sulfone)s through the
nucleophilic aromatic substitution of the novel AB2 mono-
mer. The simple AB2 monomer was prepared from bis(p-flu-
orophenyl) sulfone and 3,5-dimethoxyphenol as the starting
materials. The hyperbranched poly(ether sulfone)s were pre-
pared by the self-polycondensation of the AB2 monomer. It
There are several articles describing the preparation of
hyperbranched engineering plastics. Miller and Neenan first
reported the preparation of hyperbranched poly(aryl ether)s
by a nucleophilic aromatic substitution.5 Hyperbranched
poly(ether ketone)s,6–8 polyamides,9–15 polyimides,16–21
poly(phenylene oxide)s,22,23 poly(phenylene sulfide)s,24,25
and poly(ether sulfone)s26–28 have already been reported in
the literature. It is reported that the introduction of branch-
C
V
2012 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
was found that the reaction condition affected the degree
of branching (DB) of the resulting polymer. A unique ther-
mal crosslinking was also observed during the DSC
measurements.
water, and dried under vacuum at 100 ꢀC for 5 h. The
isolated yield was 89%.
1H NMR (DMSO, ppm): 9.6 (2H, br, OH), 8.0 (2H, m, aro-
matic), 7.9 (2H, d, aromatic), 7.5 (2H, t, aromatic), 7.1 (2H, d,
aromatic), 6.1 (1H, t, aromatic), 5.9 (2H, d, aromatic). 13C
NMR (DMSO, ppm): 167.2, 163.9, 162.0, 160.1, 156.9, 138.5,
135.3, 131.1, 130.9, 118.9, 117.9, 100.2, 98.9. Anal. Cald for
EXPERIMENTAL
Materials
Bis(p-fluorophenyl) sulfone and phloroglucinol were pur-
chased from Wako Pure Chemical Industry and used without
further purification. 3,5-Dimethoxyphenol was purchased
from Tokyo Chemical Industry and used without further pu-
rification. Cesium fluoride was purchased from Sigma-
Aldrich, and dried under vacuum at 150 ꢀC for 2 h before
use. N,N-Dimethylacetamide (DMAc) and N-methyl-2-pyrroli-
dinone (NMP; Sigma-Aldrich) were purified by distillation
before use. Boron tribromide was purchased from Nacalai
Tesque, and used without further purification. 5-Benzyloxyr-
esorcinol was prepared by a procedure reported in litera-
ture.31 All other reagents and solvents were purchased from
Kanto Chemical, and used without further purification.
C
20H17F7O5S: C,ꢀ59.99; H, 3.64. Found: C, 59.44; H, 3.74. mp
¼ 161.5–162.5 C.
Self-Polycondensation of the AB2 Monomer
Using Potassium Carbonate
In a three-necked flask equipped with a nitrogen inlet, a
Dean-Stark trap, and a condenser, the AB2 monomer (0.48 g,
1.33 mmol), potassium carbonate (0.46 g, 3.3 mmol), DMAc
(5 mL), and toluene (5 mL) were charged under nitrogen.
ꢀ
The mixture was stirred at 160 C for 90 min under flowing
nitrogen. The azeotrope of toluene and water was removed
from the Dean-Stark trap. The flask was subsequently heated
at 160 ꢀC for an additional 10 h under nitrogen. After cool-
ing, the mixture was poured into an aqueous HCl solution
(0.1 mol/L). The precipitateꢀwas recovered by filtration and
dried under vacuum at 100 C for 5 h. The polymer was iso-
lated as a brown powder in 91% yield.
Preparation of 4-(3,5-Dimethoxyphenoxy)-40-
0
fluorodiphenylsulfone (AB2 Monomer)
In a three-necked flask equipped with a Dean-Stark trap, a
condenser, and
a nitrogen inlet, difluorodiphenylsulfone
1H NMR (DMSO, ppm): 10.6, 10.3, 10.1, 8.0–7.7, 7.3–7.0, 7.0–
6.8, 6.6, 6.4–6.2, 6.1, 5.9, 5.8, 5.7. 13C NMR (DMSO, ppm):
162.5, 162.0, 161.3, 160.9, 157.6, 157.4, 156.5, 156.0, 137.0,
136.5, 136.0, 135.9, 131.7, 130.5, 119.0, 117.5, 116.8, 110.9,
109.6, 106.5, 104.4, 103.4, 101.0, 100.2, 98.9, 97.7.
(6.55 g, 25.6 mmol), 3,5-dimethoxyphenol (3.97 g, 25.6
mmol), potassium carbonate (4.58 g, 33.1 mmol), toluene
(20 mL), and NMP (100 mL) were charged under nitrogen.
The flask was heated at 160 ꢀC for 90 min under nitrogen
flow. The azeotrope of toluene and water was removed from
the Dean-Stark trap. The flask was subsequently heated at
Self-Polycondensation of the AB2 Monomer
Using Cesium Fluoride
ꢀ
160 C for an additional 4 h under nitrogen. After cooling to
room temperature, the NMP was removed by reduced distil-
lation (ca. 1 mmHg). A small amount of chloroform was
added to the flask and the insoluble inorganic salts were
removed by filtration. The filtrate was evaporated to remove
the chloroform and the crude product was dried under vac-
uum at room temperature. The dried crude product was
purified by flash column chromatography in chloroform/hex-
ane (2/1, v/v). After removing the solvents, the white pow-
der was isolated in 49% yield.
In a three-necked flask equipped with a nitrogen inlet and a
condenser, the AB2 monomer (0.48 g, 1.33 mmol), cesium
fluoride (0.47 g, 3.1 mmol), and DMAc (5 mL) were charged
under nitrogen. The mixture was stirred at 160 ꢀC for 10 h
under nitrogen. After cooling, the mixture was poured into
an aqueous HCl solution (0.1 mol/L). The precipitate was
recovered by filtration and dried under vacuum at 100 ꢀC
for 5 h. The polymer was isolated as a brown powder in
94% yield.
1H NMR (DMSO, ppm): 8.1 (2H, m, aromatic), 7.9 (2H, d, aro-
matic), 7.5 (2H, t, aromatic), 7.1 (2H, d, aromatic), 6.4 (1H, t,
aromatic), 6.3 (2H, d, aromatic), 3.7 (6H, s, methyl). 13C NMR
(DMSO, ppm): 168.6, 163.8, 162.2, 156.1, 138.5, 135.0,
131.0, 130.8, 118.5, 117.8, 100.0, 97.9, 57.2.
Self-Polycondensation of the AB2 Monomer Under
Diluted Conditions
The polymerization was carried out using the same proce-
dure as described above. The AB2 monomer (0.188 g, 0.5
mmol) was dissolved in DMAc (5 mL) to achieve an approxi-
mately five times diluted solution.
Preparation of 4-(3,5-Hydroxyphenoxy)-40-
fluorodiphenylsulfone (AB2 Monomer)
Preparation of the Dendritic Model Compound
In a three-necked flask eq0uipped with a nitrogen inlet and a
dropping funnel, the AB2 monomer (10.00 g, 25.8 mmol)
was dissolved in dichloromethane (140 mL) under nitrogen.
After cooling the flask in an ice-bath, boron tribromide (14.7
mL, 0.156 mol) was dropwise added to the flask through the
In a three-necked flask equipped with a nitrogen inlet,
phloroglucinol (63.1 mg, 0.50 mmol), the AB2 monomer
0
(0.583 g, 1.5 mmol), cesium fluoride (0.273 g, 1.8 mmol),
and DMAc (5 mL) were charged under nitrogen. The mixture
was stirred at 160 ꢀC for 10 h. After cooling to room tem-
perature, the mixture was poured into an aqueous HCl solu-
tion (0.1 mol/L). The precipitate was recovered by filtration
and dried under vacuum at 100 ꢀC. The crude product
was purified by flash column chromatography in
ꢀ
dropping funnel. After stirring at 0 C for 20 h, the solution
was slowly poured into ice water (700 mL). The mixture
was stirred and warmed to remove the dichloromethane.
The precipitate was recovered by filtration, washed with
2
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 1 Preparation of the AB2 monomer.
chloroform/hexane (13/1, v/v). After removing the solvents,
a yellow powdery product was isolated in 97% yield.
Measurements
The 1H and 13C NMR spectra were recorded using a JEOL
JNM-ECX 500 NMR spectrometer. The inherent viscosity was
measured in NMP at 30 ꢀC at a concentration of 0.5 g/dL.
The gel permeation chromatography (GPC) measurements
with NMP containing lithium bromide (0.01 mol/L) as the
eluent were carried out using a Wyatt DAWN HELEOSII 8þ,
a Wyatt Optilab rEX differential refractometer, and three
polystyrene-divinylbenzene columns (three Shodex LF-804
columns). The specific refractive index increment (dn/dc) at
658 nm was measured using a Wyatt Optilab rEX differential
refractometer. The dn/dc of the hyperbranched poly(ether
sulfone) in NMP containing lithium bromide (0.01 mol/L)
was determined to be 0.1499. The DSC measurements were
carried out by a Rigaku Thermo plus DSC 8230. The heating
1H NMR (DMSO, ppm): 7.9 (m, aromatic), 7.2 (d, aromatic),
7.1 (d, aromatic), 6.9 (s, aromatic), 6.4 (m, aromatic), 6.4 (m,
aromatic), 6.3 (m, aromatic), 3.7 (s, methyl). 13C NMR
(DMSO, ppm): 162.1, 162.0, 161.0, 157.9, 156.6, 136.4,
135.8, 130.4, 119.2, 118.4, 109.6, 103.5, 99.6, 98.0, 56.1.
Preparation of the Linear Model Compound
In a three-necked flask equipped with a nitrogen inlet, 5-
0
benzyloxyresorcinol (0.123 g, 0.50 mmol), the AB2 monomer
(0.388 g, 1.0 mmol), cesium fluoride (0.182 g, 1.2 mmol),
and DMAc (5 mL) were charged under nitrogen. The mixture
was stirred at 160 ꢀC for 10 h. After cooling to room tem-
perature, the mixture was poured into an aqueous HCl solu-
tion (0.1 mol/L). The precipitate was recovered by filtration
and dried under vacuum at 100 ꢀC. The crude product was
purified by flash column chromatography in chloroform/hex-
ane (13/1, v/v). After removing the solvents, a yellow pow-
dery product was isolated in 17% yield.
ꢀ
rate was set at 10 C/min. The TG/DTA measurements were
carried out by a Rigaku Thermo plus TG 8230 at a heating
ꢀ
rate of 10 C/min.
RESULTS AND DISCUSSION
The AB2 monomer, 4-(3,5-hydroxyphenoxy)-40-fluorodiphe-
nylsulfone, was prepared from difluorodiphenylsulfone and
3,5-dimethoxyphenol as the starting materials. As shown in
Scheme 1, difluorodiphenylsulfone was reacted with 3,5-
dimethoxyphenol via nucleophilic aromatic substitution to
form 4-(3,5-methoxyphenoxy)-40-fluorodiphenylsulfone. The
crude product contained monosubstituted and disubstituted
compounds and unreacted difluorodiphenylsulfone. The
monosubstituted compound, 4-(3,5-methoxyphenoxy)-40-fluo-
rodiphenylsulfone, was isolated by flash column chromato-
graphy. As the methoxy groups can be converted to hydroxyl
groups, 4-(3,5-methoxyphenoxy)-40-fluorodiphenylsulfone is
1H NMR (DMSO, ppm): 10.1 (s, OH), 7.9 (d, aromatic), 7.2
(m, aromatic), 6.40 (s, aromatic), 6.38 (d, aromatic), 6.31 (d,
aromatic), 6.30 (d, aromatic), 3.7 (s, methyl). 13C NMR
(DMSO, ppm): 162.1, 162.0, 161.5, 160.9, 157.8, 157.4,
136.2, 135.8, 130.4, 119.2, 118.4, 104.4, 103.4, 99.6, 98.0,
56.1.
Preparation of the Nitrobenzene-Terminated Polymer
In a two-necked flask equipped with a nitrogen inlet, the
resulting polymer (56 mg), potassium carbonate (50 mg, 0.4
mmol), and DMAc (2 mL) were charged under nitrogen. p-
Fluoronitorobenzene (0.02 mL, 0.2 mmol) was added drop-
wise to the flask, and the mixture was stirred at 120 ꢀC for
24 h. After cooling, any unreacted p-fluoronitrobenzene was
removed by precipitation in methanol. The formed precipi-
tate was recovered by filtration and dried under vacuum at
100 ꢀC for 10 h. A brown powder (46 mg, 61%) was
isolated.
0
0
recognized as an AB2 monomer. The reaction of the AB2
monomer and boron tribromide efficiently proceeded to
form 4-(3,5-hydroxyphenoxy)-40-fluorodiphenylsulfone (AB2
monomer). The structure of the AB2 monomer was con-
firmed by NMR and elemen0 tal analysis. Figure 1 shows the
1H NMR spectra of the AB2 and AB2 monomers. The 0peak at
3.7 ppm, attributed to the methoxy group of the AB2 mono-
mer, disappeared from the spectrum of the AB2 monomer.
Moreover, a broad peak at 9.6 ppm in the AB2 monomer
spectrum was attributed to the hydroxyl groups.0 The peaks
at 6.4 and 6.3 ppm in the spectrum of the AB2 monomer
were clearly shifted to a higher magnetic field in the
1H NMR (DMSO, ppm): 8.2 (br, aromatic), 7.9 (br, aromatic),
7.3 (br, aromatic), 6.8 (br, aromatic). 13C NMR (DMSO, ppm):
162.0, 161.5, 160.0, 158.0, 156.9, 143.0, 136.1, 130.8, 126.2,
119.0, 109.1.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 1 1H NMR spectra of the AB2 and AB2 monomers (DMSO-d6).
0
spectrum of the AB2 monomer. All the other peaks were rea-
sonably assigned to the proposed structure and the integra-
tion ratio of the peaks supported the assignment. Figure 2
The self-polycondensation of the AB2 monomer was carried
out in the presence of a base at 160 C in DMAc (Scheme 2).
ꢀ
When potassium carbonate was used as a base, water gener-
ated by the formation of potassium phenolate was removed
by azeotropic distillation with toluene. The results of the
self-polycondensation are summarized in Table 1. The hyper-
branched poly(ether sulfone)s were isolated in good yield in
all experiments. The absolute molecular weights determined
by GPC with a laser light scattering detector indicated the
formation of high-molecular-weight polymers. The GPC
shows the 13C NMR spectra of the AB2 and AB2 monomers.
0
The0 peak at 57 ppm, attributed to methoxy carbon of the
AB2 monomer, disappeared in the spectrum of the AB2
monomer. All the other peaks were assigned to the proposed
structures. It should be noted that the peaks attributed to
the aromatic ring connected to the fluoro group were split
due to the 19F nuclear resonance effect.
FIGURE 2 13C NMR spectra of the AB2 and AB2 monomers (DMSO-d6).
0
4
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 2 Self-polycondensation of the AB2 monomer.
ꢀ
curves of the samples prepared at 160 C are shown in Fig-
ure 3. It is clear that the polymer prepared in the presence
of potassium carbonate showed a large shoulder peak in the
high-molecular-weight region. The shoulder peak was not
observed for the polymers prepared in the presence of
cesium fluoride. The shoulder peak might be caused by the
undesired side reactions. The fluoride anions generated
during the self-polycondensation might cause the side reac-
tions.32 As a result, the molecular weight distribution (Mw/
Mn) of the polymer prepared in the presence of potassium
carbonate was much higher than the polymer prepared with
cesium fluoride. The high-molecular-weight polymer was
also isolated by the polymerization at 120 ꢀC for 40 h. The
inherent viscosity of the polymer prepared with potassium
carbonate was clearly higher than the polymer prepared
with cesium fluoride.
polymer and the model compounds. The dendritic model
showed one peak at 109.7 ppm attributed to the tri-substi-
tuted phloroglucinol unit (peak DC). The linear model
showed two peaks at 104.4 and 103.4 ppm attributed to the
disubstituted phloroglucinol unit (peaks L2C and L1C). The
resulting polymer contained peaks attributed to the dendritic
and linear units. The peaks that originated from the terminal
unit were assigned by the peaks of the AB2 monomer. The
DBs of the resulting polymers were calculated on the basis
of the integration ratio of these peaks and listed in Table 1.
It is clear that the DB of the polymer prepared with cesium
fluoride is higher than the polymer prepared with potassium
carbonate. The difference in DBs could give influence on the
solution viscosity. The polymer prepared with cesium fluo-
ride had a lower inherent viscosity and larger DB. It is well
known that hyperbranched polymers show low solution vis-
cosity in comparison with linear polymers even the molecu-
lar weight is high enough.
DB is one of the key factors to define the molecular struc-
ture of hyperbranched polymers. To determine the DB, the
dendritic and the linear model compounds were prepared as
shown in Scheme 3. The dendritic model contains the tri-
substituted phloroglucinol unit and no hydroxyl groups. The
linear model contains the disubstituted phloroglucinol unit
Generally, the DB of hyperbranched polymers prepared by
the self-polycondensation of the AB2 monomers statistically
approaches 0.5. Most DBs reported in the literature are in
the range of 0.4–0.6. Therefore, the polymer prepared with
potassium carbonate has a noticeably low DB. The low DB
(0.15) was also reported in the case of the hyperbranched
poly(ether ketone) prepared from 3,5-dihydroxy-40-fluoro-
benzophenone.6 When the conventional nucleophilic aro-
matic substitution was carried out, the hydroxyl groups are
converted to the corresponding phenolate anions which act
as strong nucleophiles. On the other hand, potassium pheno-
lates formed in situ have a limited solubility in organic sol-
vents. In this study, the AB2 monomer contains the dihydrox-
yphenyl unit as a B2 function. When both of the hydroxyl
groups are converted to the corresponding phenolates, the
solubility of the diphenolates might be very low. Therefore,
the concentration of the diphenolates in the polymerization
mixture could be much lower than the corresponding mono-
phenolate. Consequently, the formation of the linear units
1
and one hydroxyl group. Figure 4 shows the H NMR spectra
of the resulting polymer, the dendritic model and the linear
model. In the spectrum of the dendritic model, the peak at
6.9 ppm is attributed to the aromatic proton of the tri-sub-
stituted phloroglucinol unit (peak DH). The peaks at 6.4 and
6.3 ppm in the spectrum of the linear model were attributed
to the aromatic protons of the disubstituted phloroglucinol
(peaks L1H and L2H). The peaks labeled a and b were attrib-
uted to the aromatic protons of the dimethoxy-substituted
phloroglucinol unit. According to the 1H NMR spectra, the
resulting polymer apparently contained the dendritic and lin-
ear units. Unfortunately, these peaks were broad and some
undefined peaks were also observed. Therefore, it is difficult
to determine the DB from these 1H NMR spectra. Figure 5
shows the gated decoupling 13C NMR spectra of the resulting
TABLE 1 Self-Polycondensation of the AB2 Monomer to Form Hyperbranched Poly(ether
sulfone)s
Temperature
(ꢀC)
Yield
(%)
ginh
(dL/g)b
DBc
a
No
Base
Mw
Mw/Mn
1
2
3
a
K2CO3
CsF
160
160
120
91
94
89
2.7 ꢁ 105
5.6 ꢁ 104
2.7 ꢁ 104
5.7
1.8
2.0
0.32
0.19
0.17
0.17
0.39
0.41
CsF
Absolute molecular weight determined by GPC-MALLS in NMP.
Measured in NMP at a concentration of 0.5 g/dL at 30 ꢀC.
Degree of branching determined by gated decoupling 13C NMR measurements.
b
c
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
nolate anions could not be formed during the reaction. The
phenolate-ion-like intermediates show a better solubility in
organic solvents, which eventually results in the formation of
the higher DB polymer. The solubility difference between the
dissociated phenolate and the phenolate-ion-like intermedi-
ate was also observed for the preparation of the dendritic
model compound. Only the cesium fluoride-mediated cou-
0
pling of phloroglucinol with the AB2 monomer produced a
high yield of the dendritic model.
It should be mentioned that there are several unknown
peaks in the 1H and 13C NMR spectra of the resulting poly-
mers. Ether exchange reactions and intramolecular cycliza-
tions are two possible reasons for the structural defects. The
13
1H and C NMR spectra of the polymer prepared at 120 C
ꢀ
were almost consistent with the ones of the polymer pre-
pared at 160 ꢀC. This fact suggests that the ether exchange
reactions are not the main reason for the unknown peaks.
The polymerization under diluted conditions was carried out
in order to investigate the contribution of the intramolecular
1
cyclization. H NMR spectra of the polymers prepared at nor-
mal or diluted conditions are shown in Figure 6. Several
unknown peaks in the range of 7.2–6.2 ppm, which are
marked by the arrows in Figure 6(b), became stronger than
those of the polymer prepared under normal concentrations.
Therefore, these peaks could be attributed to the intramolec-
ular cyclized products. Fossum and coworker have already
pointed out that the cyclization reaction was accompanied
by the propagation reaction for the self-polycondensation of
3,5-difluoro-40-hydroxydiphenyl sulfone to form the hyper-
branched poly(ether sulfone).27 We assume that the
unknown peaks mainly originated from the structural defects
formed by the cyclization reactions.
FIGURE 3 GPC curves of the resulting polymers prepared at
160 ꢀC in the presence of potassium carbonate (a) and cesium
fluoride (b).
might be preferred by the reaction of the monophenolate
with the AB2 monomer, especially in the early stage of the
polymerization. On the other hand, it is reported that the
nucleophilic aromatic substitution mediated by cesium fluo-
ride proceeds through the formation of phenolate-ion-like
intermediates.33 In other words, completely dissociated phe-
SCHEME 3 Preparation of the dendritic and the linear model compounds.
6
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 4 1H NMR spectra of the resulting polymer (a), the dendritic model (b), and the linear model (c) (DMSO-d6).
The hyperbranched poly(ether sulfone) is soluble in organic
solvents, such as DMF, DMAc, NMP, DMSO, tetrahydrofuran,
and acetone and insoluble in chloroform. The 5% weight-
loss temperature of the hyperbranched poly(ether sulfone)
was 183 ꢀC which is noticeably lower than the original
hydroxyl-terminated polymer. The hydroxyl groups may play
an important role in the strong intra-and intermolecular
interactions.
ꢀ
determined by TG/DTA measurements was 385 C. The glass
When the DSC measurements of the hydroxyl-terminated
polymer prepared with cesium fluoride were carried out in
air, the Tg gradually increased with the measurement times.
Figure 7 shows the DSC curves of the hydroxyl-terminated
polymer. When the DSC measurements were carried out
transition temperature (Tg) deteꢀrmined by the DSC measure-
ments under nitrogen was 217 C. It is well known that the
solubility and glass transition temperature are influenced by
the terminal function of the hyperbranched polymer. The
hydroxyl groups in the hyperbranched poly(ether sulfone)
were chemically modified to the nitrophenyl ether by the
reaction with p-fluoronitrobenzene. The nitrobenzene-termi-
nated polymer became soluble in chloroform but insoluble
in acetone. The Tg of the nitrobenzene-terminated polymer
ꢀ
under nitrogen, the reproducible Tg at 217 C was observed.
On the other hand, the Tg increased with the measurement
times when the measurements were carried out in air. Figure
8 shows the plots of the Tg s of the hyperbranched
FIGURE 5 Gated decoupling 13C NMR spectra of the resulting polymer (a), the dendritic model (b), and the linear model (c).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000
7

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 6 1H NMR spectra of the resulting polymers prepared under the normal (a) and the diluted (b) conditions (DMSO-d6).
poly(ether sulfone) determined by the DSC measurements in
air. When the heating-cooling scan in air from 50 to 300 ꢀC
was carried out five times, the Tg increased to 236 ꢀC. The
sample having the Tg at 236 ꢀC became insoluble in all of
the organic solvents. It is clear that thermal crosslinking
reactions occur during the heating–cooling scan in air. Figure
8 also shows the effect of the DB on the thermal crosslinking
reaction. The polymer prepared with potassium carbonate
also showed an increase in the Tg but the trend was weak.
The location or geometry of the hydroxyl groups may influ-
ence the crosslinking reaction. It should be pointed out that
the nitrobenzene-terminated polymer did not show the Tg
increase after the heating–cooling scans from 50 to 260 ꢀC
for five times in air. Figure 9 shows the TGA curves of the
polymer before and after heating at 300 ꢀC for 30 min. The
hydroxyl-terminated polymer prepared with cesium fluoride
showed the gradual weight loss from 150 to 350 ꢀC. The 5
and 10% weight loss temperature were 339 and 404 ꢀC,
respectively. The curve (b) in Figure 9 represents the weight
loss of the polymer after the heating at 300 ꢀC for 30 min.
The gradual weight loss observed before the heating was not
ꢀ
observed. The 5 and 10% weight loss were 406 and 424 C,
respectively. It is clear that the thermal stability of the
hydroxyl-terminated hyperbranched poly(ether sulfone) is
improved by the heating at 300 ꢀC. IR spectra of the
hydroxyl-terminated hyperbranched poly(ether sulfone)s
before and after the heating–cooling scans in air are shown
in Figure 10. Some of the peaks in the IR spectrum of the
sample after the heating scans became broad. It is difficult
to assign the each change in the spectra to the certain func-
tional group. The broad peaks from 1100 to 1200 /cm might
suggest the formation of aromatic ether linkages.
CONCLUSIONS
Hyperbranched poly(ether sulfone)s were prepared by the
self-polycondensation of 4-(3,5-hydroxyphenoxy)-40-fluorodi-
phenylsulfone as an AB2 monomer. The high-molecular-
weight polymers were isolated in good yields. DB of the
FIGURE 8 The Tg increase of the hydroxyl-terminated hyper-
branched poly(ether sulfone) by the heating-cooling scans in
air from 30 to 300 ꢀC.
FIGURE 7 DSC curves of the hyperbranched poly(ether sul-
fone) (OH terminal) at a heating rate of 10 ꢀC/min.
8
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
the solubility was influenced by the terminal functional
groups. The unique thermal crosslinking phenomenon was
observed during the DSC measurements of the hydroxyl-ter-
minated hyperbranched poly(ether sulfone) in air. The cross-
linking behavior was also influenced by the DB. The thermal
crosslinking may be useful for application in solvent-less
coatings.
ACKNOWLEDGMENTS
This work is partially supported by a Grant-in-Aid for Scientific
Research (C) from the Ministry of Education, Culture, Sports, Sci-
ence, and Technology (No. 22550106) and Nissan Motor Co., Ltd.
REFERENCES AND NOTES
1 Yan, D.; Gao, C.; Frey, H. In Hyperbranched Polymers; Yan,
D.; Gao, C.; Frey, H., Eds.; Wiley: New Jersey, 2011; Chapter 1,
pp 1–26.
2 Jikei, M.; Kakimoto, M. Prog. Polym. Sci. 2001, 26, 1233–1285.
3 Gao, C.; Yan, D. Prog. Polym. Sci. 2004, 29, 183–275.
4 Voit, B. I.; Lederer, A. Chem. Rev. 2009, 109, 5924–5973.
FIGURE 9 TGA curves of hydroxyl-terminated hyperbranched
poly(ether sulfone) prepared with cesium fluoride before (a)
and after (b) heating at 300 ꢀC for 30 min.
5 Miller, T. M.; Neenan, T. X. J. Am. Chem. Soc. 1993, 113,
356–357.
6 Hawker, C. J.; Chu, F. Macromolecules 1996, 29, 4370–4380.
7 Shu, C.-F.; Leu, C.-M. Macromolecules 1999, 32, 100–105.
8 Shu, C.-F.; Leu, C.-M.; Huang, F.-Y. Polymer 1999, 40,
6591–6596.
resulting polymers was determined by gated decoupling 13C
NMR measurements. The DB of the polymer prepared with
potassium carbonate was lower than that prepared with ce-
sium fluoride. The low DB might be caused by the poor solu-
bility of the phenolate anions formed in situ during the poly-
merization. The resulting hyperbranched poly(ether
sulfone)s showed good solubilities in organic solvents, and
9 Kim, Y. H. J. Am. Chem. Soc. 1992, 114, 4947–4948.
10 Yang, G.; Jikei, M.; Kakimoto, M. Macromolecules 1999, 32,
2215–2220.
11 Ishida, Y.; Sun, A. C. F.; Jikei, M.; Kakimoto, M. Macromole-
cules 2000, 33, 2832–2838.
12 Jikei, M.; Chon, S.-H.; Kakimoto, M.; Kawauchi, S.; Imase,
T.; Watanabe, J. Macromolecules 1999, 32, 2061–2064.
13 Russo, S.; Boulares, A. Macromol. Symp. 1998, 128, 13–20.
14 Haba, O.; Tajima, H.; Ueda, M.; Nagahata, R. Chem. Lett.
1998, 333–334.
15 Yang, G.; Jikei, M.; Kakimoto, M. Macromolecules 1998, 31,
5964–5966.
16 Thompson, D. S.; Markoski, L. J.; Moore, J. S. Macromole-
cules 1999, 32, 4764–4768.
17 Thompson, D. S.; Markoski, L. J.; Moore, J. S.; Sendijarevic,
I.; Lee, A.; McHugh, A. J. Macromolecules 2000, 33, 6412–6415.
18 Orlicki, J. A.; Thompson, D. S.; Markoski, L. J.; Sill, K. N.;
Moore, J. S. J. Polym. Sci. Part A: Polym. Chem. 2002, 40,
936–946.
19 Yamanaka, K.; Jikei, M.; Kakimoto, M. Macromolecules
2000, 33, 1111–1114.
20 Yamanaka, K.; Jikei, M.; Kakimoto, M. Macromolecules
2000, 33, 6937–6944.
21 Yamanaka, K.; Jikei, M.; Kakimoto, M. Macromolecules
2001, 34, 3910–3915.
22 In, I.; Lee, H.; Kim, S.-Y. Macromol. Chem. Phys. 2003, 204,
1660–1664.
23 Zhang, J.; Wang, H.; Li, X. Polymer 2006, 47, 1511–1518.
24 Jikei, M.; Hu, Z.; Kakimoto, M.; Imai, Y. Macromolecules
1996, 29, 1062–1064.
FIGURE 10 IR spectra of the hydroxyl-terminated hyper-
branched poly(ether sulfone)s before (a) and after (b) the five
heating–cooling scans from 30 to 300 ꢀC.
25 Mellace, A.; Hanson, J. E.; Griepenburg, J. Chem. Mater.
2005, 17, 1812–1817.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000
9

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
26 Mart´ınez, C. A.; Hay, A. S. J. Polym. Sci. Part A: Polym.
Chem. 1997, 35, 2015–2033.
30 Markoski, L. J.; Thompson, J. L.; Moore, J. S. Macromole-
cules 2000, 33, 5315–5317.
27 Himmelberg, P.; Fossum, E. J. Polym. Sci. Part A: Polym.
Chem. 2005, 43, 3178–3187.
31 Nagvekar, D. S.; Gibson, H. W. Org. Prep. Proced. Int. 1997,
29, 240–242.
28 Takeuchi, M.; Jikei, M.; Kakimoto, M. Chem. Lett. 2003, 32,
242–243.
32 Carlier, V.; Devaux, J.; Legras, R.; Bunn, A.; McGrail, P. T.
Polymer 1994, 35, 415–422.
29 Jikei, M.; Fujii, K.; Yang, G.; Kakimoto, M. Macromolecules
33, 6228–6234.
33 Imai, Y.; Ishikawa, H.; Park, K.-H.; Kakimoto, M. J. Polym.
Sci. Part A: Polym. Chem. 1997, 35, 2005–2061.
10
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 000, 000–000