Preparation, characterization, and properties of poly(aryl ether sulfone) systems with double-decker silsesquioxane in the main chains by reactive blending

Article abstract of DOI:10.1002/pola.27058

A series of copoly(aryl ether sulfone)s containing double-decker-shaped silsesquioxane (DDSQ) in the main chain was prepared. Toward this end, a novel diphenol polyhedral oligomeric silsesquioxane macromer was synthesized by hydrosilylation between 3,13-d

Full text of DOI:10.1002/pola.27058

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Preparation, Characterization, and Properties of Poly(aryl ether sulfone)
Systems with Double-Decker Silsesquioxane in the Main Chains by
Reactive Blending
Weihai Zhang, Jiuduo Xu, Xuesong Li, Guanghe Song, Jianxin Mu
College of Chemistry, Engineering Research Center of High Performance Plastics, Ministry of Education, Jilin University,
Changchun 130012, China
Correspondence to: J. Mu (E-mail: jianxin_mu@jlu.edu.cn)
Received 24 September 2013; accepted 29 November 2013; published online 18 December 2013
DOI: 10.1002/pola.27058
ABSTRACT: A series of copoly(aryl ether sulfone)s containing
double-decker-shaped silsesquioxane (DDSQ) in the main chain
was prepared. Toward this end, a novel diphenol polyhedral
oligomeric silsesquioxane macromer was synthesized by
hydrosilylation between 3,13-dihydro octaphenyl double-decker
silsesquioxane (denoted dihydro DDSQ) and eugenol. The pol-
y(aryl ether sulfone)s were synthesized from diphenol DDSQ,
bisphenol A (BPA), and 4-fluorophenyl sulfone using a one-
step high-temperature solution method. By adjusting the ratio
of diphenol DDSQ to BPA, copolymers with variable DDSQ
content in the main chains were obtained. With increased
DDSQ content in the main chain, the glass transition tempera-
ture decreased based on differential scanning calorimetry, and
anti-degradation was enhanced based on thermogravimetric
analysis. Moreover, the dielectric constant j of pure polymer
(3.19 at
1 MHz) initially increased to 4.04 (DDSQ molar
ratio 5 10%), and then decreased to 2.68 at 1 MHz (DDSQ molar
ratio 5 100%). Crystallization behavior, solubility, and surface
C
V
hydrophobicity were also investigated.
2013 Wiley Periodi-
cals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2014, 52, 780–
788
KEYWORDS: dielectric properties; double-decker-shaped silses-
quioxane (DDSQ); high performance polymers; macromono-
mers; nanocomposites; polysiloxanes; surface hydrophobicity;
WAXS
materials. Given that the number and type of functional
groups on POSS cages can be readily varied, POSS molecules
are excellent building blocks for preparing organic–inorganic
hybrid nanocomposites with lots of structures, such as lin-
ear,7–9 star-shaped,10–12 as well as network types.13–17 More-
over, using POSS derivatives as a monomer has proven to be
an efficient method of designing nanocomposite material and
side chains18 or end groups19 in hybrid polymers. Surface
and interfacial properties lie at the heart of many potential
POSS applications, but information on the surface properties
of POSSs is limited.20–22 Furthermore, POSSs have a low
dielectric constant because of their porosity.23–25 Incorpora-
tion of a POSS into a polymer can reduce the dielectric con-
stant of a nanocomposite.26
INTRODUCTION Organic–inorganic hybrid materials are
attracting considerable interest because they offer the oppor-
tunity to develop high-performance materials that combine
many desirable properties of conventional organic and inor-
ganic components, such as thermal stability, solubility, and
processability.1 Silsesquioxanes exist in a variety of struc-
tures, from random polymers to more ordered arrange-
ments,2,3 In particular, polyhedral oligomeric silsesquioxanes
(POSSs) are a class of nanoscale molecule having a well-
defined cube-like inorganic core (Si8O12) with its corners
connected to eight organic functional groups.4 The Si8O12
core can have a wide variety of functional groups attached
to each vertex of the core, leading to the use of POSSs in var-
ious applications.5 POSSs are cube-octameric molecules of
nanoscale dimensions that may be functionalized with reac-
tive groups suitable for the synthesis of new organic–inor-
ganic hybrids, thereby providing the opportunity to design
and build materials with extremely well-defined dimensions
possessing and nanophase behavior.6 The rigid inorganic
core of POSSs provides the strength and oxidative stability of
a ceramic, whereas synthesis control over the organic coro-
nae (R) provides processability and compatibility with other
Recently, POSS-containing polymers have become the focus
of many studies because these hybrid materials possess the
combined properties of organic and inorganic components
with the inclusion of well-defined nanosized building
blocks.27–34 Chemical approaches have been demonstrated to
be efficient in introducing POSSs into polymers by the for-
mation of chemical linkages between POSS cages and
C
V
2013 Wiley Periodicals, Inc.
780
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 780–788

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
polymer matrices.35 In a majority of the previous studies,
POSS macromers bearing single polymerizable (or reactive)
groups were either grafted onto polymer chains by radical
copolymerization (reactive grafting) or capped to the ends of
polymer chains.36–39 Given that the double-decker-shaped
silsesquioxane (DDSQ) possesses precisely two reactive
hydrosilane groups, linear organic–inorganic hybrid polymers
should be obtainable without cross-linking.40–45
nitrogen or air purge of 100 mL min21. Infrared (IR) spectra
were recorded with a Nicolet Impact 410 FTIR spectropho-
tometer by KBr pellet with a measurement from 400 cm21
to 4000 cm21. Gel permeation chromatography (GPC) analy-
sis was carried out using Waters 410 instrument with dime-
thylformamide (DMF) as an eluent and polystyrene as the
standard. Hydrogen-1 nuclear magnetic resonance (1H NMR)
€
was performed on a Bruker Advance 300 spectrometer (300
29
€
MHz), and Si NMR was performed on a Bruker 510 NMR
Poly(arylene ether sulfone)s (PAESs) are important engineer-
ing thermoplastics with excellent resistance to chlorinated
disinfectants, hydrolysis, and oxidation, as well as excellent
mechanical properties, such as thermal stability and tough-
ness.46 To the best of our knowledge, no previous reports
exist on the modification of PAES using DDSQ in the main
chain.
spectrometer (500 MHz). The solutions were measured using
tetramethylsilane as an internal reference. The MALDI-TOF-
MS experiment was performed using a Shimadzu/AMIMA-
CFR. The matrix 2,5-dihydroxybenzoic acid 98% was dis-
solved in THF (10 mg mL21) and mixed with the sample
solution (0.5 mg mL2121.0 mg mL21 in THF) at a v/v ratio
of 1:1. The samples were dried in air for at least 30 min and
the measurements were taken in linear mode with a UV
laser (337 nm). The spectra were calibrated using bradykinin
dissolved in THF (10 mg mL21) and mixed with the sample
solution 0.5 mg mL2121.0ꢀmg mL21. The j of the thin films
coated with silver by vacuum evaporation were obtained
using a Hewlett-Packard 4285A apparatus at room tempera-
ture and at frequencies from 103 Hz to 106 Hz. The crystalli-
zation behaviors of the hybrid polymer and control PAES
were examined using a Rigaku D/max-2500 X-ray diffractom-
eter with CuKa radiation (k 5 0.154 nm) as the X-ray source.
Contact angle measurements were performed at room tem-
In this article, we synthesize of a novel DDSQ aromatic dis-
phenol macromer by hydrosilylation of eugenol with
dihydro-DDSQ and its subsequent reaction with difluoro-
phenyl sulfones and bisphenol A (BPA) to produce a series
of linear containing DDSQ in the main chain hybrid PAESs.
However, previous reports about low dielectric constant (j)
materials focused on macroscopic influence on j of homo-
polymers with DDSQ in the main chain. In contrast, in this
article, both the free volume of chain entanglement and free
volume of POSS self-cage are discussed in terms of their
microcosmic effects on the j of hybrid copolymers. The ther-
mal and surface properties as well as crystallization behavior
were also investigated.
perature using
a POWEREACH/JC2002C2 contact angle
meter with pure water as probe liquid. The dispersion of the
POSS clusters in the polymer matrix was observed from SSX-
550 Shimadzu scanning electron microscope (SEM).
EXPERIMENTAL
Monomer Synthesis
Materials
Synthesis of 3,13-Di(2-methoxy-4-propylphenol)octaphenyl
DDSQ
Phenyltrimethoxysilane was purchased from Energy-
Chemical and used without further purification. Methyldi-
chlorosilane (AR) and BPA (AR) were purchased from Alad-
din Industrial Corporation. Anhydrous magnesium and
anhydrous calcium chloride were purchased from Guangdong
Xilong Chemical. Triethylamine (AR) was purchased from
Tianjin Fuyu Fine Chemical. Tetrahydrofuran (THF) (AR) was
purchased from Tianjin Tiantai Fine Chemicals. Toluene
(AR), chloroform (AR), ethyl acetate (AR), sodium hydroxide
(AR), potassium carbonate (AR), N,N-dimethylacetamide
(DMAc; AR), dimethyl sulfoxideand (DMSO; AR), and 2-
propanol (AR) were purchased from Beijing Chemical Works.
N-methyl-2-pyrrolidinone (NMP; AR) and potassium carbon-
ate (AR) were purchased from Tianjin BODI chemicals. NMP
was further purified by distillation under reduced pressure
over calcium hydride. Before use, the THF was refluxed
above sodium and then distilled.
3,13-Di(2-methoxy-4-propylphenol)octaphenyl DDSQ (diphe-
nol DDSQ) was synthesized by hydrosilylation reaction
between 3,13-dihydrooctaphenylhexacyclodecasiloxane (dihy-
dro-DDSQ) synthesized according to the method reported by
Seino et al.1 and eugenol, catalyzed by Karstedt’s catalyst, as
shown in Scheme 1. Typically, a 50 mL two-necked round-
bottomed flask equipped with a magnetic stirrer, a nitrogen
inlet, and an anhydrous calcium chloride drying tube were
connected to a Schlenk line to degas by a repeated exhaust-
ing–refilling process using pure nitrogen. DDSQ (2.3 g, 2
mmol), eugenol (0.69 g, 4.2 mmol), and toluene (20 mL)
were then added. With 30 min of vigorous stirring, the reac-
tion was conducted with the catalysis of Karstedt’s catalyst
ꢀ
(5 drops of 0.2 mol %) in toluene at 95 C for 24 h to yield
diphenol DDSQ. The mixture was then held for 12 h at 105
ꢀC with stirring to insure that the hydrosilylation reached
completion. The solvent and excess eugenol were removed
under reduced pressure to yield the nearly colorless solids
with a yield of 92%. 1H NMR (300 MHz, DMSO-d6, d): 0.30
(s, 3H, Si-CH3), 0.77 (t, 2H, SiCH2CH2CH2A), 1.67 (t, 2H,
SiCH2CH2CH2A), 2.42 (t, 2H, SiCH2CH2CH2A), 3.52 (s, 3H,
CH3AO), 6.32–6.35 (d, 1H, Ar H), 6.50–6.55 (m, 2H, AR H),
7.17–7.751 (m, 20H, AR H of Si-Ph), 8.62 (s,1H,AOH). 13C
Methods
Differential scanning calocalorimetry (DSC) analysis was per-
formed using a Mettler-Toledo Instrument DSC 821e-modu-
lated thermal analyzer at a heating rate of 10 ꢀC min21
under a nitrogen purge of 200 mL min21. Thermogravimet-
ric analysis (TGA) was performed on a PerkinElmer Pryis 1
TGA analyzer at a heating rate of 10 ꢀC min21 under a
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 780–788
781

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
and then a substantial amount of the resulting water and tol-
uene were removed. The reaction mixture was heated to 180
ꢀ
ꢀC–190 C. After 6 h, another 10 mL of DMSO was added to
the viscous reaction mixture. The polymerization was com-
plete after another 2 h. The viscous solution was then
poured into distilled water. The flexible threadlike polymer
was pulverized into powder using a blender. The polymer
was then refluxed in deionized water and ethanol several
times to remove the salts and solvents and dried at 120 ꢀC
for 24 h to obtain a constant weight. The yield was 92%. All
the organic–inorganic PSF with DDSQ in the main chains
were subjected to GPC to measure their molecular weights.
The GPC profiles were shown in Table 2. In all cases, high-
molecular-weight products were obtained. In fact, the GPC
result of 5%-DDSQPSF displayed longer elution time. This
observation suggests that there could be some components
that possessed lower molecular weights in the resulting
polymers, which were not easy to grow into higher-
molecular-weight species. Similar results of polymerization
have been reported in other organic–inorganic copolymers
with POSS in the main chains.1,42,45,47
SCHEME 1 Synthesis of 3,13-di(2-methoxy-4-propylphenol)oc-
taphenyl DDSQ.
NMR (300 MHz, DMSO-d6, d): 15.72, 24.58, 37.79, 55.24,
112.17, 115.15, 120.21, 128.14, 129.91, 131.00, 132.50,
133.24, 144.36, and 147.21. 29Si NMR (500MHz; DMSO-d6;
d): 217.33, 277.32, 278.56, 278.80, 278.88, and 279.48.
MALDI-TOF-MASS spectroscopy [m/z (%)]: 1481.6 (M1,
43%); 1505.0 ([M 1 Na] 1, 100), calculated: 1505.2 Da.
Film Preparation
A total of 1 g copolymer was dissolved into 10 mL DMAc,
keeping under magnetic stirring for 24 h, and was then fil-
tered. The filtrate was poured on a 8 cm 3 8 cm glass plate.
The programmed temperature was increased from 60 ꢀC to
21
ꢀ
ꢀ
120 C at a heating rate of 10 C h and dried for 48 h to
remove the solution completely. A series of novel PAES
hybrid films were finally obtained. The blending films were
prepared in a same procedure, with a mass content of 3,13-
di(2-methoxy-4-propylphenol)octaphenyl DDSQ in polymer
equal to that in hybrid films.
Polymer Synthesis
Synthesis of Hybrid Silicon PAESs
The novel poly(ether sulfones of bisphenol A) (PSF) copoly-
mers with POSS in the main chains were synthesized by aro-
matic substituted reaction between diphenol [BPA and/or
diphenol DDSQ] and 4-fluorophenyl sulfone as depicted in
Scheme 2. By adjusting the molar ratio of diphenol DDSQ to
BPA, a series of DDSQ/PSF copolymers was obtained. Typi-
cally, for the preparation of 50%-DDSQPSF copolymer, to a
100 mL three-necked flask equipped with a mechanical stir-
rer, a nitrogen inlet, and a Dean-Stark trap with a condenser
were added BPA (1.140 g, 5 mmol), 4-Fluorophenyl sulfone
(2.542 g, 10 mmol), diphenol DDSQ (0.7410 g, 5 mmol),
anhydrous K2CO3 (1.794 g, 13 mmol), DMSO (18 mL), and
toluene (7 mL). The system was allowed to reflux for 3 h,
RESULTS AND DISCUSSION
Structure Characterization
In this article, direct polymerization between difluoromethyl sul-
fone and diphenol via nucleophilic substitution reaction cata-
lyzed by Lewis bases (e.g., potassium carbonate) was employed
to afford PAES. The diphenol-functionalized POSS macromer
(viz., diphenol DDSQ) was synthesized via a three-step approach,
as shown in Scheme 1. First, octaphenyldicycloocatasiloxane
SCHEME 2 Synthesis of hybrid poly(aryl ether sulfone)s.
782
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 780–788

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 1 1H NMR spectrum of 3,13-di(2-methoxy-4-propylphe-
FIGURE 2 MALDI-TOF-MS spectrum of 3,13-di(2-methoxy-4-
nol)octaphenyl DDSQ.
propylphenol)octaphenyl DDSQ.
and using deuterated dimethylsulfoxide as a solvent were dis-
played in Figure 4. The peaks from 6.5 ppm to 8.0 ppm were
assigned to the protons of aromatic ring, the single peaks at 3.45
ppm and 0.64 ppm were assigned to alkyl in the main chain, and
the single peaks at 0.15 ppm and 1.63 ppm were assigned to
methyl.
tetrasodium silanolate [Na4O14Si8(C6H5)8] was synthesized by
following the method reported by Seino et al.1 Second, the silyla-
tion reaction between Na4O14Si8(C6H5)8 and methyldichlorosi-
lane (HSiCH3Cl2) was conducted to obtain 3,13-
dihydrooctaphenyl DDSQ (i.e., dihydro-DDSQ). Thereafter, the
hydrosilylation reaction between dihydro-DDSQ and eugenol
was performed to afford 3,13-di(2-methoxy-4-propylphenol)oc-
taphenyl double-decker silsesquioxane (viz., diphenol DDSQ).
The 29Si NMR spectrum of di(2-methoxy-4-propylphenol)octa-
phenyl double-decked silsesquioxane (diphenol DDSQ) is shown
in Figure 3. The resonance at 217.33 ppm was ascribed to the
silicon nucleus connected to methyl group whereas the signals
of resonance in the range 277.3 ppm to 279.4 ppm were
ascribed to other silicon nuclei of the POSS cage. The resonance
at 217.33 ppm, 277.32 ppm, and 278.8 ppm were ascribed to
the silicon nucleus of the POSS cage in trans configuration and
the resonance at 217.33 ppm, 277.32 ppm, 278.56 ppm, and
279.48 ppm were assigned to the silicon nucleus of the POSS
cages in cis configuration.48 The 1H NMR spectra of di(2-
methoxy-4-propylphenol)octaphenyl DDSQ are shown in Figure
1. Signals of resonance appeared at 0.77 ppm, 1.67 ppm, 2.41
ppm, 3.52 ppm, 6.32–6.35 ppm, and 6.50–6.55 ppm, which were
assigned to the protons of 2-methoxy-4-propylphenol groups.
The ratio of the integration intensity of signals at 0.77 ppm, 1.67
ppm, and 2.41 ppm was 1:1:1, which well agreed with the value
calculated according to the structural formula, suggesting that
the hydrosilylation was complete. The 3,13-di(2-methoxy-4-pro-
pylphenol)octaphenyl DDSQ was subjected to MALDI-TOF-MS
spectroscopy to measure its molecular weight and the mass
spectrum is presented in Figure 2. The POSS macromer pos-
sessed the molecular weight of M 5 1482.0 (viz., 1505.0–23),
which perfectly matched the value calculated according to the
structural formula (Fig. 3). The results of 29Si NMR, 1H NMR, and
mass spectroscopy indicated that diphenol DDSQ was success-
fully obtained.The structure of the synthesized 50%-DDSQPSF
IR (KBr) spectrum of pure PSF polymer is shown in Figure
5(a): 1013 cm21 (Ar), 1102 cm21, and 1153 cm21 (O@S@O
symA), 1240 cm21 (CAOAC), 1296 cm21, and 1320 cm21
(O@S@O asymA). The results of IR and 1H NMR confirmed
that the copolymer was synthesized successfully.
The structure of the prepared DDSQ/PSF hybrid films were
confirmed by IR (KBr) spectra, as shown in Figure 5(b–f):
1255 cm21 and 830 cm21could be assigned to SiAC band,
1
1
FIGURE 3 29Si NMR spectrum of 3,13-di(2-methoxy-4-propyl-
polymer was confirmed by H NMR and FTIR spectra. The H
NMR spectra obtained with a 500 MHz Bruker 510 spectrometer
phenol)octaphenyl DDSQ.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 780–788
783

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 4 1H NMR spectrum of 50%-DDSQPSF copolymer in
d6-DMSO.
and 1071 cm21 and 480 cm21 could be attributed to SiAO
band of the DDSQ. From the comparison with IR spectrum of
pure PSF polymer Figure 5(a), the typical absorption band at
1255 cm21, 830 cm21 of SiAC band, and 1071 cm21, 480
cm21 of SiAO band in the skeleton of DDSQ disappeared. In
summary, Figure 5 confirms that DDSQ nanoparticles were
successfully introduced into the main chains of the polymer
by covalent bonds.
FIGURE 6 WAXD profiles of (a) pure 3,13-dihydrooctaphenyl
DDSQ, (b) 0%-DDSQPSF, (c) 5%-DDSQPSF, (d) 10%-DDSQPSF,
(e) 20%-DDSQPSF, (f) 50%-DDSQPSF, and (g) 100%-DDSQPSF
hybrids films. [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
2h 5 7.3ꢀ observed in the diffraction curve of hybrid films
indicated that the DDSQ molecules still retained their nano-
scale dimensions in the polymer main chain. The peak inten-
sity of 2h 5 7.3ꢀ increased as the amount of DDSQ in the
polymer increased. A similar phenomenon is observed for
hybrid films in other systems.49–52 In addition, the disap-
pearing crystalline structure also indicated the successful
introduction of the POSSs into the main chain of the
copolymer.
Crystallization Behavior of Pure Dihydro-DDSQ and
DDSQ/PSF Hybrids Films
In Figure 6, the solid state pure dihydro-DDSQ exhibited
multiple sharp diffraction peaks at 2h 5 6.1ꢀ, 7.0ꢀ, 8.0ꢀ, 8.7ꢀ,
10.1ꢀ, 11.4ꢀ, 12.0ꢀ, 16.6ꢀ, 18.3ꢀ, 19.1ꢀ, 19.8ꢀ, 20.5ꢀ, and
24.9ꢀ, respectively, suggesting that the pure dihydro-DDSQ
formed a crystalline structure. The pure 0%-DDSQPSF only
exhibited a diffuse peak at 2h 5 18.3ꢀ, and the rest of
DDSQPSF hybrid films exhibited two diffuse peaks at
2h 5 18.3ꢀ and 2h 5 7.3ꢀ, respectively. The diffuse peak at
Dielectric Constant of Pure PSF, DDSQ/PSF Hybrid Films,
and Blending Films
Figure 7(a) presents the j of pure PSF and DDSQ/PSF hybrid
films at a frequency ranging from 1000 Hz to 1 MHz, and
Figure 7(b) summarizes the variation of their j along with
the contents of POSSs at 1 MHz. Figure 7(a) shows that
when the molar ratio of diphenol DDSQ to 4-fluorophenyl
sulfone was 5:5 or higher, j of organic–inorganic copolymers
were lower than those of pure PSF. Nonetheless, j values of
the copolymers were higher than that of pure PSF. The con-
tent of POSSs in the main chain molar percent was 10%, and
j was the largest at nearly 4.04. This article proposed that
the following opposite tendencies affect the
j of the
organic–inorganic copolymers. On the one hand, the intro-
duction of vacant cage-like structure of the POSS macromer
into the main chains of copolymers simultaneously resulted
in the increase in the free volume of the main chain in the
copolymers, which reduced j. On the other hand, the intro-
duction of the bulky and vacant cage-like structure of the
POSS macromer resulted in the polymer chains unable to
undo chain entanglements because the POSSs existed in
place of the phenoxy structural units in the main chains of
the copolymers, causing the free volume of chain to
decrease. This decrease resulted in the decreased of the
amount of copolymers. Moreover, as the content of POSSs in
FIGURE 5 IR spectra of (a) 0%-DDSQPSF, (b) 5%-DDSQPSF, (c)
10%-DDSQPSF, (d) 20%-DDSQPSF, (e) 50%-DDSQPSF, (f) 100%-
DDSQPSF. [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
784
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 780–788

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 7 (a) Frequency dependence of dielectric constants of DDSQ/PSF hybrid films; (b) plot of dielectric constants at 1 MHz as
a function of the content of POSS in the organic–inorganic copolymers. [Color figure can be viewed in the online issue, which is
available at wileyonlinelibrary.com.]
TABLE 1 Dielectric Constant of blending films
Blending
Blending
Blending
Blending
Blending
Samples
0%-DDSQPSF
3.19
polymer 1
polymer 2
polymer 3
polymer 4
polymer 5
Dielectric constant
3.14
3.08
3.37
–
–
Dielectric constants were measured at 1 MHz.
FIGURE 8 SEM photographs of (a) blending polymer 1 film; (b) blending polymer 2 film; (c) blending polymer 3 film; and (d) the
image of blending polymer 4. [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 2014, 52, 780–788
785

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Summary of the Properties of the POSS–PSF
Samples
Tg (ꢀC)a
Td5 (ꢀC)b
Contact angle (ꢀ)
Dielectric constant (j)c
Residue (%)d
Mn
Mw/Mn
0%-DDSQPSF
5%-DDSQPSF
10%-DDSQPSF
20%-DDSQPSF
50%-DDSQPSF
100%-DDSQPSF
194
184
174
167
156
130
444.5
430.2
416.4
398.6
376.4
377.3
78.2
81.3
87.4
91.5
92.5
97.5
3.19
3.32
4.04
3.56
2.78
2.68
0.0
5.2
92,795
48,107
44,708
31,586
11,270
9861
1.20
4.18
1.25
1.84
1.49
1.32
8.7
15.6
28.8
36.6
a
c
DSC at a heating rate of 10 ꢀC min21 in nitrogen.
The dielectric constant was measured at 1 MHz.
b
d
Temperature at which 5% weight loss was recorded by thermogravim-
The yields of degradation residues were taken as the values at 800 ꢀC
etry at a heating rate of 10 ꢀC min21 in air.
in air.
the main chain increased, the trend of chain entanglements
worsened, and the trend of free volume of organic–inorganic
POSS cage changed. When the POSS content in the main
chain was low, the increased free volume of organic–inor-
ganic POSS cage could not counteract the decreased free vol-
ume resulting from chain stretching. However, with the
molar ratio content of POSSs in the main chain increasing by
10% or higher, the free volume of POSS cage was increased
by POSS cage introduction, in which one part was used to
counteract the decreased free volume of chain stretching and
the other part was used to enhance the free volume of copol-
ymer; thus, the decrease in j. This article proposed that if
the POSS content in the main chain is low, free volume
formed by chain entanglements becomes dominant, whereas
if the POSS content in the main chain is high, free volume of
POSSs itself dominates.
mer films had lower j while blending polymer could not
form a film.
Surface Properties and Solubility of PSF Copolymers
Organosilicon compound of the POSS moiety was introduced
into the PSF copolymers, which was of low free energy. The
surface hydrophobicity of the PSF copolymers is expected to
improve compared with pure PSF. This effect can be readily
examined by measuring contact angles. In this article, the
static surface contact angles were measured with water as
probe liquids, and the results are summarized in Table 2. The
water contact angle of pure PSF was ca. 78.2ꢀ. For 100%-
DDSQPSF, the water contact angle was as high as 97.5ꢀ, indi-
cating that the hydrophobicity of the materials was signifi-
cantly enhanced. Compared with pure PSF, the water contact
angles of the PSF copolymers were dramatically enhanced.
The water contact angles increased as the POSS content in
the copolymers increased (Fig. 9). The water contact angles
became almost impartial to the molar ratio of DDSQ, suggest-
ing that the surfaces of the organic–inorganic copolymers
were saturated by the organosilicon moieties (viz. POSSs).
The dielectric constant of blending of POSS and poly(aryl
ether sulfone) with different contents at 1 MHz was sum-
marized in Table 1. The mass percentage of POSS in blending
polymer 1, 2, 3, 4, 5 was equal to that in 5%-DDSQPSF,
10%-DDSQPSF, 20%-DDSQPSF, 50%-DDSQPSF, 100%-
DDSQPSF, respectively. Meanwhile, due to hydrogen bonding
interactions of 3,13-di(2-methoxy-4-propylphenol)octaphenyl
DDSQ in blending polymers, blending films occurred too
obvious phase separation to form a film when the mass con-
tent of POSS was increasing to equal to that in 50%-
DDSQPSF or more. Phase separation of blending polymer 4
was shown in Figure 8(d). The dielectric constant (at 1
MHz) of blending polymer 1 and blending polymer 2 was
decreasing with the content of POSS increasing. It was
mainly due to the POSS hollow cage introduction. However,
the j of blending polymer 3 was higher than blending poly-
mer 1 and blending polymer 2. That was because of numer-
ous phenolic hydroxyl groups of 3,13-di(2-methoxy-4-
propylphenol)octaphenyl DDSQ in blending polymer 3 set-
tling to the surface of film, shown in Figure 8(c), while the
POSS of blending polymer 1 and blending polymer 2 was
uniformly dispersed in blending films, shown in Figure
8(a,b). Blending polymer films had lower j than hybrid poly-
mer films with the same mass content of POSS, due to chain
entanglements caused by the POSS in hybrid polymer. How-
ever, when the mass content of POSS was high, hybrid poly-
FIGURE 9 Plot of surface water contact angles as a function of
the content of POSS in the organic–inorganic copolymers.
[Color figure can be viewed in the online issue, which is avail-
able at wileyonlinelibrary.com.]
786
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 780–788

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 3 Solubility of POSS-PSF
Samples
DMAc
DMSO
DMF
THF
Chloroform
Acetone
Methanol
0%-DDSQPSF
5%-DDSQPSF
10%-DDSQPSF
20%-DDSQPSF
50%-DDSQPSF
100%-DDSQPSF
1
12
12
12
12
12
12
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
1
12
12
12
12
12
12
12
12
2
12
12
12
2
12
12
12
12
2
2
Solubility under a mass of liquid: 0.1 g mL21; 1 soluble at room temperature; 2 insoluble even on heating; 12 partially soluble or swelling on
heating.
The solubility of PSF copolymers is shown in Table 3. With
the introduction of nanoscale POSSs into the main chain of
polymers, the solubility of copolymers worsened.
POSS structure and the straight chain alkyl led to a lower Tg
value. As a consequence, the Tg values of the copolymers
decreased. The thermal stability of the organic–inorganic
copolymers was evaluated using TGA and the TGA curves
are shown in Figure 10. All copolymers displayed two steps
in the degradation progress. The first step originated from
the degradation of active alkyl chain and the second step ori-
ginated from the degradation of frameworks, such as the
aromatic ring and POSS cage. The temperatures at the maxi-
mum rate of degradation of all the copolymers were signifi-
cantly higher than those of pure PSF, suggesting that the
presence of POSS moiety in the main chains retarded the
random-chain scission. The organic–inorganic copolymers
displayed higher residuals of degradation than pure PSF, and
the yields of the degradation residues increased as the POSS
content increased. The increased yields of degradation resi-
dues were ascribed to the ceramic formation from POSS moi-
ety during thermal decomposition.
Thermal Properties of PSF Copolymers
The organic–inorganic PSF copolymers were subjected to
DSC to measure their Tg. The values of Tg are summarized in
Table 2. The data indicated that the Tg values of the materi-
als changed with the content of DDSQ in the main chain of
copolymers. The Tg of 0%-DDSQPSF was about 192 ꢀC, and
Tg of 100%-DDSQPSF was about 130 ꢀC. This article pro-
posed that the following opposite tendencies affected the Tg
of the organic–inorganic copolymers. On the one hand, the
nanoreinforcement of POSS cages in the main chains resulted
in the enhanced Tg. On the other hand, the introduction of
POSSs into the main chains of copolymers simultaneously
resulted in the increase in free volume in the copolymers.
The increased free volume was responsible for the introduc-
tion of the bulky and vacant cage-like SiAO structure of the
POSS macromer, which resulted in the polymer chains being
unable to densely pack in the glassy state because the bulky
and cage-like POSSs exist in place of the phenoxy structural
units in the main chains of the copolymers. Introducing the
CONCLUSIONS
In this article, a novel 3,13-di(2-methoxy-4-propylphenol)oc-
taphenyl double-decker silsesquioxane, which is a difunc-
tional POSS macromer, was successfully synthesized. The
diphenol POSS macromere was used to prepare organic–
inorganic PSF copolymers with POSSs in the main chains via
direct polymerization based on the nucleophilic substitution
reaction. By adjusting the molar ratio of BPA to di(2-
methoxy-4-propylphenol)octaphenyl DDSQ, a series of PSF
copolymers with variable POSS contents was successfully
obtained. These PSF copolymers were characterized by NMR
spectroscopy and GPC. DSC showed that the Tg of the
organic–inorganic PSF copolymers was relatively dependent
on the composition of the copolymers. TGA indicated that
the stability of the copolymers was significantly improved in
terms of the temperatures at the maximum rate of degrada-
tion and the yields of degradation residues. Owing to the
presence of POSSs in the main chains, the surface hydropho-
bicity of the organic–inorganic copolymers was significantly
improved compared with those of the pure PSF, which only
related to the POSS amount in the main chain. Hybrid films
could get a lower j than blending films. Meanwhile, in
hybrid films, j was affected by the combined effects of free
volume of chain entanglement and the POSS cavity of the
FIGURE 10 TGA curves of PSF and the POSS-containing PSF
copolymers in air. [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 2014, 52, 780–788
787

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
24 K. Su, D. R. Bujalski, K. Eguchi, G. V. Gordon, D. L. Ou, P.
Chevalier, S. Hu, R. P. Boisvert, Chem. Mater. 2005, 17, 2520–
2529.
main chain. When the POSS content in main chain was low,
the free volume produced by chain entanglements domi-
nated, and j was decided by the extent of chain entangle-
ments. In contrast, when the POSS content in the main chain
was high, the free volume of POSS cavity dominated, and j
was determined by the amount of POSS cavity. Moreover,
with the nanoscale POSS introduction into main chain of
polymer, the solubility worsened.
25 W. C. Liu, Y. Y. Yu, W. C. Chen, J. Appl. Polym. Sci. 2004,
91, 2653–2660.
26 R. Q. Su, T. E. Muller, J. Prochazka, J. A. Lercher, Adv.
Mater. 2002, 14, 1367–1369.
27 M. G. Voronkov, V. I. Lavrent’yev, Top. Curr. Chem. 1982,
102, 199–236.
28 J. J. Schwab, J. D. Lichtenhan, Appl. Organomet. Chem.
1998, 12, 707–713.
ACKNOWLEDGMENTS
The authors gratefully acknowledge the financial support from
the International Science and Technology Cooperation program
of China (2012DFR50300).
29 G. Li, L. Wang, H. Ni, C. U. Pittman, Jr., J. Inorg. Organo-
met. Polym. 2002, 11, 123–154.
30 Y. Abe, T. Gunji, Prog. Polym. Sci. 2004, 29, 149–182.
31 S. H. Phillips, T. S. Haddad, S. J. Tomczak, Curr. Opin. Solid
State Mater. Sci. 2004, 8, 21–29.
REFERENCES AND NOTES
32 P. D. Lickiss, F. Rataboul, Adv. Organomet. Chem. 2008, 57,
1 M. Seino, T. Hayakawa, Y. Ishida, M. Kakimoto, Macromole-
cules 2006, 39, 3473–3475.
1–116.
33 R. M. Laine, M. F. Roll, Macromolecules 2011, 44, 1073–
2 M. G. Voronkov, V. I. Lavrent’yev, Top. Curr. Chem. 1982,
102, 199–236.
1109.
34 S. W. Kuo, F. C. Chang, Progr. Polym. Sci. 2011, 36, 1649–
3 R. H. Baney, M. Itoh, A. Sakakibara, T. Suzuki, Chem. Rev.
1995, 95, 1409–1430.
1696.
35 Y. J. Leea, J. M. Huangb, S. W. Kuoa, F. C. Chang, Polymer
2005, 46, 10056–10065.
4 P. T. Mather, H. G. Jeon, A. R. Uribe, T. S. Haddad, J. D.
Lichtenhan, Macromolecules 1999, 32, 1194–1203.
36 J. D. Lichtenhan, Y. A. Otonari, M. J. Carr, Macromolecules
1995, 28, 8435–8437.
5 A. Provatas, J. G. Matisons, Trends Polym. Sci. 1997, 5, 327–
332.
37 K. Zeng, Y. Fang, S. Zheng, J. Polym. Sci., Part B: Polym.
Phys. 2009, 47, 504–516.
6 A. J. Waddon, L. Zheng, R. J. Farris, E. B. Coughlin, Nano
Lett. 2002, 10, 1149–1155.
38 B. S. Kim, P. T. Mather, Macromolecules 2002, 35, 8378–
8384.
7 H. Xu, B. Yang, J. Wang, S. Guang, C. Li, Macromolecules
2005, 38, 10455–10460.
39 W. Zhang, A. H. E. Mueller, Macromolecules 2010, 43, 3148–
3152.
8 H. Xu, B. Yang, J. Wang, S. Guang, C. Li, J. Polym. Sci.: Part
A 2007, 45, 5308–5317.
40 S. Wu, T. Hayakawa, R. Kikuchi, S. J. Grunzinger, M.
Kakimoto, Macromolecules 2007, 40, 1353–1362.
9 L. Zheng, R. J. Farris, E. B. Coughlin, J. Polym. Sci., Part A:
Polym. Chem. 2001, 39, 2920–2928.
41 S. Wu, T. Hayakawa, R. Kikuchi, S. J. Grunzinger, M.
Kakimoto, Macromolecules 2007, 40, 5698–5705.
10 S. Muthukrishnan, F. Plamper, H. Mori, A. H. E. Muller, Mac-
romolecules 2005, 38, 10631–10642.
42 S. Wu, T. Hayakawa, M. Kakimoto, H. Oikawa, Macromole-
cules 2008, 41, 3481–3487.
11 J. Huang, X. Li, T. Lin, C. He, K. Y. Mya, Y. Xiao, J. Li, J.
Polym. Sci., Part B: Polym. Phys. 2004, 42, 1173–1180.
43 M. A. Hoque, Y. Kakihana, S. Shinke, Y. Kawakami, Macro-
molecules 2009, 42, 3309–3315.
12 S. Xiong, Y. Xiao, J. Ma, L. Zhang, X. Lu, Macromol. Rapid
Commun. 2007, 28, 281–285.
44 A. C. Kucuk, J. Matsui, T. Miyashita, Langmuir 2011, 27,
6381–6388.
13 G. Deng, Y. Chen, Macromolecules 2004, 37, 18–26.
14 H. Liu, S. Zheng, Macromol. Rapid Commun. 2005, 26, 196–
200.
45 L. Wang, C. Zhang, S. Zheng, J. Mater. Chem. 2011, 21,
19344–19352.
15 Y. Liu, F. Meng, S. Zheng, Macromol. Rapid Commun. 2005,
26, 920–925.
46 R. N. Johnson, A. G. Farnham, R. A. Clenndinning, W. H.
Hale, C. N. Merriam, J. Poly. Sci. 1967, 5, 2375–2398.
16 Y. Zhao, Y. Chen, C. Chen, F. Xi, Polymer 2005, 46, 5808–5819.
47 K. Wei, L. Wang, S. Zheng, Polym. Chem. 2013, 4, 1491–
17 S. Devaraju, M. R. Vengatesan, M. Alagar, High Perform.
Polym. 2011, 23, 99–111.
1501.
48 V. Ervithayasuporn, X. Wang, Y. Kawakami, Chem. Com-
mun. 2009, 5130–5132.
18 C. M. Leu, Y. T. Chang, K. H. Wei, Chem. Mater., 2003, 15,
3721–3727.
49 C. M. Leu, G. M. Reddy, K. H. Wei, C. F. Shu, Chem. Mater.
2003, 15, 2261–2265.
19 B. S. Kim, P. T. Mather, Polymer 2006, 47, 6202–6207.
20 V. A. Ogarev, Colloid J. 2001, 63, 445–452.
50 C. M. Leu, Y. T. Chang, K. H. Wei, Macromolecules 2003, 36,
9122–9127.
21 H. Kobayashi, Macromol. Chem. 1993, 194, 2569–2577.
22 R. Knischka, F. Dietsche, R. Hanselmann, H. Frey, R.
Mulhaupt, P. J. Lutz, Langmuir 1999, 15, 4752–4756.
51 C. H. Chou, S. L. Hsu, S. W. Yeh, H. S. Wang, K. H. Wei,
Macromolecules 2005, 38, 9117–9123.
23 T. C. Chang, Y. S. Mor, P. T. Liu, Thin Solid Films 2001, 398–
399, 523–526.
52 C. H. Chou, S. L. Hsu, K. Dinakaran, M. Y. Chiu, K. H. Wei,
Macromolecules 2005, 38, 745–751.
788
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 780–788