Polysiloxanes containing polyhedral oligomeric silsesquioxane groups in the side chains; synthesis and properties

Article abstract of DOI:10.1016/j.polymer.2010.01.066

Linear polysiloxanes having different amounts of polyhedral oligomeric silsequioxane (POSS) side groups were prepared from the hydrosilylation reaction of poly(ethylhydrosiloxane) with different amount of POSS and 1-octene using platinum(0)-divinyl tetram

Full text of DOI:10.1016/j.polymer.2010.01.066

Polymer 51 (2010) 2296–2304
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Polysiloxanes containing polyhedral oligomeric silsesquioxane groups
in the side chains; synthesis and properties
*
Hyun-Soo Ryu, Dong-Gyun Kim, Jong-Chan Lee
Department of Chemical and Biological Engineering, Seoul National University, 599 Gwanakno, Gwanak-gu, Seoul 151-744, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Linear polysiloxanes having different amounts of polyhedral oligomeric silsequioxane (POSS) side groups
were prepared from the hydrosilylation reaction of poly(ethylhydrosiloxane) with different amount of
Received 28 August 2009
Received in revised form
26 January 2010
Accepted 31 January 2010
Available online 6 February 2010
POSS and 1-octene using platinum(0)-divinyl tetramethyldisiloxane as
a catayst. 1-Octene and
platinum(0)-divinyl tetramethyldisiloxane were found to be very important for the preparation of the
linear polymer without any cross-linked structures. The linear polysiloxanes with POSS side groups are
soluble in various organic solvents. When the content of POSS-containing monomeric unit is larger than
10 mol%, free standing films can be prepared from a routine solution casting method, although this
polysiloxane is not cross-linked.
Keywords:
Hydrosilylation
Polyhedral oligomeric silsesquioxane (POSS)
Polysiloxanes
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
through covalent bonding to improve their thermal/mechanical
properties [20–29].
Polysiloxane derivatives have long been used widely in many
applications including elastomers, silicone oils, sealing agents,
adhesives, low dielectric materials, and biomedical materials,
because of their unique properties such as heat resistance, high
oxygen permeability, low toxicity, poor wettability, extremely low
surface tension, outstanding electrical isolating properties, chemical
resistance, and biocompatibility [1–6]. Because linear polysiloxane
derivatives have poor thermal/mechanical properties and low
dimensional stability, cross-linked polysiloxanes or copolymers
having siloxane units were prepare to improve their thermal/
mechanical properties [7–11]. As an alternative approach to improve
their thermal/mechanical properties, polysiloxane composites can
be prepared by mixing linear polysiloxanes with metal oxide fillers
such as zinc oxide, aluminum oxide, and ferric oxide, etc. or poly-
hedral oligomeric silsequioxane (POSS) nanoparticles, but phase
separation of these inorganic materials in the polymer matrix was
found to be major problem to improve their physical properties
[12–19]. POSS/polymer blend systems were also studied to improve
the polymers’ properties. However, POSS molecules were also phase-
separated and aggregated in polymer matrix. So the polymer
membranes are not fully transparent and have optical defects even at
low POSS loading levels [14]. Therefore bulky groups such as phenyl,
biphenyl, triphenyl, naphthalene, carbohydrate, cholesteric, and
cyclohexyl groups also have been attached to the polysiloxanes
In this work, POSS nanoparticles were covalently bonded to
polysiloxanes to obtain organic/inorganic hybrid polymeric mate-
rials with improved thermal/mechanical properties. These mate-
rials were synthesized from hydrosilylation reactions of silane
groups (Si–H) in poly(ethylhydrosiloxane) (PEHS) with allyl–
cyclohexylPOSS (allyl–CyPOSS)/1-octene using platinum(0)-divinyl
tetramethyldisiloxane as a catalyst. The detailed synthetic proce-
dures and characterization of these polymers using instrumental
methods such as nuclear magnetic resonance, Fourier transform
infrared, gel-permeation chromatography, wide-angle X-ray scat-
ting, differential scanning calorimetry, thermogravimetry analysis,
rheometry, dynamic mechanical thermal analysis, and universal
testing machine are discussed in this paper.
2. Experimental
2.1. Materials
1,3,5,7,9,11,14-Heptacyclohexyltricyclo[7.3.3.15,11]heptasiloxane-
endo-3,7,14-triol (trisilanol-CyPOSS) was obtained from Hybrid
Plastics Inc. and used without further purification. Allyltri-
chlorosilane, 1-octene, and Platinum(0)-divinyl tetramethyldisi-
loxane, Pt2[(CH2]CH)SiMe2OSiMe2(CH]CH2)]3, (Pt2(dvs)3, 2 wt%
Pt solution in xylene) were used as received from Aldrich. Toluene
and tetrahydrofuran were distilled from Na/benzophenone and
stored under dry nitrogen. All other reagents and solvents were
* Corresponding author. Tel.: þ82 2 880 7070; fax: þ82 2 888 1604.
E-mail address: jongchan@snu.ac.kr (J.-C. Lee).
0032-3861/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2010.01.066

H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
2297
used as received from commercial sources, unless otherwise
mentioned. Trimethyl-terminated poly(ethylhydrosiloxane) (PEHS,
viscosity at 25 ꢀC is 100 cSt on average.) was purchased from Gelest.
The degree of polymerization of PEHS was estimated from the 1H
NMR spectrum of PEHS by comparing the intensities of peaks from
18 protons of the methyl groups at the chain end and that from one
proton of the hydrosilane (Si–H) group in the backbone. The degree
of polymerization was about 370.
with a typical value of 1.0 mm. Dynamic mechanical thermal anal-
ysis (DMTA) was carried out in a nitrogen atmosphere with
a Rheometric Scientific analyzer (MARK IV). Measurement of the
tensile storage modulus (E0) and loss modulus (E00) of samples were
performed at a frequency of 1 Hz and working in single cantilever in
the temperature range from ꢁ70 to 100 ꢀC. The heating rate was
3
ꢀC/min. The testing was performed using rectangular samples
measuring approximately 35 ꢂ 5.0 ꢂ 0.2 mm3. The exact dimen-
sions of each sample were measured before the scan. The tensile
strength of the polymer film was measured with dumbbell speci-
mens according to the ASTM D638 Type V. The value for each sample
was taken as the median value of five specimens. These tests were
carried out at room temperature on a universal tensile testing
machine (Lloyd instruments LR 10K) with a crosshead speed of
50 mm/min.
2.2. Measurements
1H nuclear magnetic resonance (1H NMR) spectra were recorded
using 2 wt% samples in CDCl3 on a Jeol (JNM-LA 300) (300 MHz). The
Fourier transform infrared (FT-IR) measurements were performed
on a Perkin Elmer Spectrum 2000 FT-IR spectrometer in combina-
tion with a deuterated triglycine sulfate (DTGS) detector using KBr
pellets. In all cases, 16 scans at a resolution of 4 cmꢁ1 were used to
record the spectra. Weight-average molecular weight (Mw),
number-average molecular weight (Mn), and molecular weight
distributions were measured by conventional gel-permeation
chromatography (GPC) system equipped with a Waters 1515 Iso-
cratic HPLC pump, a Waters 2414 refractive index detector, and a set
of Waters Styragel columns (HR2 and HR4, 7.8 mm ꢂ 300 mm). GPC
measurements were carried out at 30 ꢀC using THF as eluent with
a flow rate of 1.0 mL/min. The systemwas calibrated with against ten
known polystyrene standards. The absolute molecular weights of
some polymers were determined by GPC equipped with a multi-
angle light scattering detector (GPC/MALS). THF was used as
eluent with a flow rate of 1.0 mL/min. Detectors: Wyatt OPTILAB DSP
interferometric refractive index detector and Wyatt miniDAWN
light scattering detector with a 20 mW semiconductor laser oper-
ating at 690 nm. The glass transition temperature (Tg) and melting
temperature (Tm) were determined using a TA Instruments differ-
ential scanning calorimeter (DSC 2920) equipped with a DSC Cool-
ing Can, which is allowed to cool down using liquid nitrogen, under
a continuous nitrogen purge (60 mL/min). The heating rate was
10 ꢀC/min. Pure indium was used to calibrate the instrument. The
glass transition temperatures were taken as the inflection point in
the change in heat capacity with temperature in the DSC curves. The
thermal stability of the polymers was analyzed by thermogravim-
etry analysis (TGA) using a TA Instruments TGA 2050 under
a continuous nitrogen purge of 60 mL/min. The samples were
heated from room temperature to 700 ꢀC with a uniform heating
rate of 10 ꢀC/min. The residual char yield was taken as the weight
percentage remaining at T ¼ 700 ꢀC. Wide-angle X-ray scattering
was used to analyze the chemical structures of polymer samples.
The measurements were carried out using the bending magnet
beam line 3C2 at the Pohang Light Source, Korea. X-rays of wave-
2.3. Synthesis of 1-allyl-3,5,7,9,11,13,15-
heptacyclopentylpentacyclo[9.5.1.13,9.15,15.17,13]-octasiloxane
(allyl–CyPOSS)
Allyl–CyPOSS was prepared using the method described by
Lichtenhan et al. as shown in Fig. 1 [30]. 1,3,5,7,9,11,14-Heptacyclo-
hexyltricyclo[7.3.3.15,11]heptasiloxane-endo-3,7,14-triol (trisilanol-
CyPOSS) (24.34 g, 25 mmol) and triethylamine (8.348 g, 82.5 mmol)
were dissolved in distilled THF (200 mL) and cooled in an ice bath. A
solution of allyl trichlorosilane (4.875 g, 27.5 mmol) in THF (25 mL)
was added using an addition funnel to the cooled solution. The
reaction mixture was allowed to warm to room temperature and
stirred for 24 h. Afterwards the reaction mixture was filtered to
remove the Et3N$HCl byproduct. Volatiles were removed under
reduced pressure at ambient temperature, and obtained yellow
solid was subsequently dissolved in a minimum amount of benzene
and precipitated into acetonitrile (5-fold excess) to eliminate the
remaining byproduct completely. After filtration and drying under
vacuum at room temperature, the yield of the obtained white
powder was 97%.
1H NMR (CDCl3, ppm): 5.78 (m, 1H, SiCH2CH]CH2), 5.00–4.90
(s, 2H, SiCH2CH]CH2), 1.75 (m, 35H, cyclohexyl–CH2), 1.61 (d, 2H,
SiCH2CH]CH2), 1.25 (m, 35H, cyclohexyl–CH2), 0.76 (m, 7H,
cyclohexyl–CH). 13C NMR (CDCl3, ppm): 132.6 (SiCH2CH]CH2),
114.7 (SiCH2CH]CH2), 27.5–26.6 (cyclohexyl–CH2), 23.2–23.1
(cyclohexyl–CH), 19.7 (SiCH2CH]CH2). FT-IR (KBr, cmꢁ1): 3077
(y
, allyl–CH), 3000–2800 (
y
, CH2), 1646 (y, C]C), 1447 (d, CH2), 1108
(y
, T-type Si–O–Si), : stretching mode;
y
d: bending mode.
2.4. Synthesis of poly(ethylsiloxane)s containing CyPOSS and
n-octyl side groups (PES#: # ¼ 0, 1, 4, 7, 10, 20, 25)
length
l
¼ 1.5406 Å monochromatized by a Si(111) double-crystal
The abbreviation of poly(ethysiloxane)s containing CyPOSS and n-
octyl in the side groups is PES#, where # is the mol-% of allyl–CyPOSS
groups with respect to the Si–H groups in PEHS used in the synthesis.
Fig.1 shows the synthetic routes for the preparation of PES#s and the
synthetic procedure is exemplified in the case of PES20 as follows.
Allyl–CyPOSS (1.14 g, 1.096 mmol, 20 mol% versus the Si–H groups in
PEHS) and PEHS (0.406 g, 5.480 mmol) were dissolved in freshly
distilled toluene (5.5 mL). The reaction mixture was stirred at room
temperature for 5 min to obtain a homogeneous solution, and then
0.05 mL solution of Pt2(dvs)3 in xylene (2 wt% Pt solution) was added
to the reaction mixture. After 1 h of stirring, 1-octene (0.590 g,
5.261 mmol,1.2 equiv. versus the remaining Si–H groups) was added
to the reaction mixture and the reaction solution was stirred at room
temperature for 5 days. The reaction was monitored by 1H NMR and
monochromator were focused at the sample position by a toroidal
premirror. For high resolution – transmission electron microscopy
(HR-TEM) measurement, the 0.5 wt% solution of polymers in THF
was dropped onto carbon-coated copper grid. Just after vacuum
evaporation of the solvent, thin polymer films formed between
copper grid lines were observed from the transmittance for 300 keV
using TEM (JEM-3010, JEOL). The linear viscoelastic properties were
measured with an ARES rheometer (TA Instruments). All of the
measurements were performed within the linear viscoelastic range,
using rotation rates of 0.1–500 rad sꢁ1, where the dynamic shear
storage (G0) and shear loss modulus (G00) are independent of strain.
Apparent shear viscosity [h
(Pa s)], storage modulus [G0 (Pa)], and
loss modulus [G00 (Pa)] were measured as a function of frequency in
the dynamic oscillatory mode. All of the rheological characteriza-
tions were performed using a cone and plate with 25 mm diameter
and with the gap between cone and plate being controlled precisely,
FT-IR. The disappearance of chemical shift (
d
¼ 4.7 ppm) and
stretching vibration (v ¼ 2160 cmꢁ1) contributed by the hydrosilane
(Si–H) group confirmed 100% conversion of Si–H groups through the

2298
H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
Fig. 1. Synthetic routes for allyl–CyPOSS and polysiloxanes having POSS side groups (PES#s), where the # is molar ratio of CyPOSS fed. (# ¼ 0, 1, 4, 7, 10, 20, and 25).
hydrosilylation reactions. The resulting polymer was obtained on
precipitation in methanol three times and washed several times with
methanol. The final product was obtained after drying in a vacuum
oven at room temperature for 2 days. Other PES#s (# ¼ 0, 1, 4, 7, 10,
25) were obtained with the same procedure, except for the amount of
allyl–CyPOSS and 1-octene used (Table 1). The yields were always
above 70% and 100% conversions from PEHSto PES#s were confirmed
by 1H NMR and FT-IR. The octyl and POSS content in polymers were
calculated by comparing the intensity of the peak at 1.88 ppm from
the cyclohexyl group attached to the POSS cage (k in Fig. 2) and thatof
the peak at 0.55 ppm from the two methylene groups covalently
bonded at the silicone atom substituted with 1-octene (a and c in
Fig. 2). The POSS content in the polymer was similar to the allyl–
CyPOSS used in the hydrosilylation reactions (Table 1). The physical
appearance of the isolated products changed from viscous liquid at
0–7 mol% of POSS to powders at 20 mol% or greater POSS content. A
(Table 1), high enough to ignore the effect of molecular weight on the
thermal properties of the polymers.
1H NMR (CDCl3, ppm): 1.75 (cyclohexyl–CH2), 1.46
(SiCH2CH2CH2SiO3), 1.25 (SiCHCH3CH2, SiCHCH3(CH2)5CH3,
SiCH2(CH2)6CH3, cyclohexyl–CH2), 0.94 (SiCH2CH3, SiCHCH3,
CH2CH2CH3), 0.72 (cyclohexyl–CH), 0.65 (SiCH2CH2CH2SiO3), 0.51
(CH3CH2SiO2CH2CH2). 13C NMR (CDCl3, ppm): 34.0 (SiCHCH3CH2),
32.3 and 29.8 (SiCHCH3CH2(CH2)3CH2, SiCH2(CH2)5CH2), 27.7–
26.8 (cyclohexyl–CH2), 23.5–23.0 (CH2CH2CH3, cyclohexyl–CH),
17.1 (Si(CH2)3SiO3), 16.3 (SiCH2(CH2)6CH3), 14.4 (CH2CH2CH3),
8.4–8.1 (SiCH2CH3, SiCHCH3CH2), 6.8 (SiCH2CH3). FT-IR (KBr,
cmꢁ1): 3000–2800 (
Si), : stretching mode;
y
, CH2), 1462 (
d, CH2), 1108 (y, T-type Si–O–
y
d: bending mode.
3. Results and discussion
flexible transparent film about 200 mm thickness could be obtained
3.1. Synthesis
by casting from the THF solution of PES20 (Fig. 3(a)). The weight-
average molecular weight (Mw) of the polymers measured from
GPC using polystyrene standards are in the range of 111000–157000
A series of polysiloxanes containing n-octyl and POSS side
groups (PES#) were prepared from hydrosilylation reactions of
Table 1
Compositions of poly(ethylhydrosiloxane)s containing POSS and n-octyl side groups.
POSS (mol%)
Feeding
n-octyl content (mol%)a
Mnb (ꢂ10ꢁ3
)
(Mw/Mn)c (ꢂ10ꢁ3
)
PDI
In polymera
Anti-Markovnikov addition
Markovnikov addition
PEHS
PES0
PES1
PES4
PES7
PES10
PES20
PES25
27.4
69.0
73.1
83.7
94.7
106.7
144.8
160.0
35/8.7
140/23
157/24
144/25
134/22
154/24
151/31
110/26
4.00
5.97
5.88
5.69
6.03
6.50
4.94
4.19
0
1
4
0
1.2
4.3
7.5
11.0
22.1
26.4
94.2 (94.2d)
91.6 (92.7d)
90.8 (94.9d)
87.0 (94.1d)
81.9 (92.0d)
70.3 (90.2d)
66.8 (90.8d)
5.8 (5.8d)
7.2 (7.3d)
4.9 (5.1d)
5.5 (5.9d)
7.1 (8.0d)
7.6 (9.8d)
6.8 (9.2d)
7
10
20
25
a
Calculated from 1H NMR.
Theoretical molecular weights from PEHS having 370 degree of polymerization.
Obtained from GPC using THF as solvent with respect to monodisperse polystyrene as standards.
The numbers in parentheses are mol% among the octyl side groups.
b
c
d

H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
2299
Fig. 2. 1H NMR spectra of starting compounds and polysiloxanes having POSS side groups; (a) allyl–CyPOSS, (b) PEHS, (c) PES0, and (d) PES20.
PEHS with different amounts of 1-octene and allyl–CyPOSS
(Table 1) using the platinum catalyst, Pt2(dvs)3, as shown in Fig. 1.
Poly(ethylhydrosiloxane) (PEHS) was used as the parent polymer
instead of poly(methylhydrosiloxnae) (PMHS), which is more
commonly used to obtain polysiloxane derivatives from hydro-
silylation reactions [21,23,25,31]. Because the maximum degree of
polymerization (DP) of commercially available PMHS is about 40,
polysiloxanes with POSS content under 10 mol% could not be easily
prepared. For example if 4 mol% of allyl–CyPOSS is reacted with
PMHS with DP 40, the average number of POSS groups per each
polymer chain would be about 1.6, then there should be quite large
content of polymer products that do not have any POSS side groups
at all. Those products would not be pure polysiloxanes containing
POSS side groups, but a mixture of polysiloxanes containing only
n-octyl side groups and polysiloxanes containing both n-octyl and
POSS side groups. Therefore, the chemical and physical properties
of these products would reflect these mixtures. When PMHS was
reacted with a small amount of allyl–CyPOSS, less than 4 mol%, the
POSS content in the product was always much larger than the
amount of allyl–CyPOSS used in the reaction. Some of the polymer
products without POSS side groups could have been removed
during the purification procedures because of different physical
Fig. 3. Photo images of free standing polymer films, (a) PES20 and (b) PES25, obtained by casting from THF solution.

2300
H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
properties, including polymer viscosity. Since the DP of PEHS used
in this study is larger, about 370, all the polymer chains may contain
POSS side groups, even with a small amount of allyl–CyPOSS in the
hydrosilylation reactions. Even PES1 and PES4, which have POSS
mol% of 1 and 4, respectively, prepared from the hydrosilylation
reactions show reproducible chemical and physical properties.
1-Octene was intentionally reacted with the remaining Si–H
groups in the polysiloxane backbone after the hydrosilylation reac-
tions of PEHS and allyl–CyPOSS to prevent any possible reactions of
Si–H in the polymer backbone with hydroxyl groups in alcohols
(or water), known as O-silylation [32,33]. Alkoxysilane (Si–OR) or
hydroxysilane (Si–OH) groups in the polymer backbone produced
from the O-silylation reactions can undergo condensation reactions
to produce cross-linked polymers. We obtained insoluble cross-
linked polymers when Si–H groups were not fully substituted with
n-octyl group reactions. 1-Octene not other 1-alkenes was inten-
tionally used to produce n-octyl side groups for the reactions
because (1) it is not easy to handle shorter gaseous alkenes such as
1-butene, (2) using longer 1-alkenes, such as 1-dodecene, could
produce side chain crystalline phases where both the alkyl side
chains and the POSS groups work as filler to increase the thermal
properties [34]. Because the Si–H groups in PEHS are fully
substituted by allyl–CyPOSS and 1-octene groups, the increase in
POSS content is caused by decreased n-octyl content. The effect of n-
octyl groups on thermal properties is discussed below.
Fig. 4. FT-IR spectra of starting compounds (PEHS and allyl–CyPOSS) and polysiloxanes
having POSS side groups (PES0 and PES20).
respectively (Table 1). Therefore, anti-Markovnikov addition is the
major reaction, with a probability of 92% on average.
Karstedt’s catalyst, Pt2(dvs)3, was used as a catalyst in the
hydrosilylation, although acidic Speier’s catalyst, a solution of plat-
inum hydrochloric acid (H2PtCl6$ꢂH2O) in isopropyl alcohol, was
used for hydrosilylation reactions for poly(methylhydrosiloxane) by
others [31,35]. Speier’s catalyst produced O-silylation reactions of
Si–H groups with isopropanol in the solution of the catalyst and with
other alcohols and water used during the purification process, then
the cross-linked polymers (insoluble polymer gels) were produced
through condensation reactions between the alkoxysilane
(or hydroxylsilane) groups produced when we dried the products in
a vacuum oven. On the contrary when we used Karstedt’s catalyst,
Pt2(dvs)3 in xylene, as the catalyst known for a hydrosilylation
reaction, linear soluble polymers could be obtained; since iso-
propanol is not included in Karstedt’s catalyst, O-silyation additions
should not occur. The conversion of hydride groups to POSS and
n-octyl groups was confirmed from 1H NMR and FT-IR spectrum. For
example, an absorption band at 2160 cmꢁ1 from the Si–H group is
not observed in the FT-IR spectrum of PES0 and PES20 (Fig. 4), and
the proton peak at 4.7 ppm from Si–H is not observed in the 1H NMR
spectrum of PES0 and PES20 (Fig. 2).
The weight-average molecular weight (Mw) of the polymers
measured from GPC are in the range of 111000–157000 (Table 1).
There was not a strong correlation between the molecular weight
and the POSS content, although the molecular weight of PES25 with
the largest POSS content seems to be smaller than those of other
PES#s. For comparison, we performed a SEC analysis with a light
scattering detector. The molecular weight (Mw/Mn) values of PES0,
PES7, and PES25 obtained from the light scattering detector are
264100/66500, 568700/115900, and 295500/88800, respectively;
the molecular weight of PES25 is slightly larger than that of PES0.
Since these polymers were prepared by substitution reactions of
Anti-Markovnikov addition is a major reaction in the hydro-
silylation of poly(alkylhydrosiloxane) using platinum catalysts
[36–38], and both Markovnikov and anti-Markovnikov products
can be produced with subtle changes of reaction conditions such as
the types of alkenes and catalyst [33,39,40]. In our case, only
Markovnikov products were obtained from the hydrosilylation
reactions of allyl–CyPOSS, while both anti-Markovnikov and Mar-
kovnikov additions of Si–H groups into the double bond of 1-octene
were observed. 2-Dimensional heterogeneous NMR spectroscopy
was used to identify the proton peak positions of anti-Markovnikov
and Markovnikov products from 1-octene (Fig. 5), and the inte-
gration of the peaks at 0.55 ppm from anti-Markovnikov addition
(a and c in Fig. 2(d)) and that of the peak at 1.25 ppm from Mar-
kovnikov addition (d, f, and l in Fig. 2(d)) were compared to
calculate the composition. The compositions of POSS and octyl
groups in polymers were calculated from the integration of the
NMR peaks in Fig. 2(d). The details for the calculation are listed in
the Supporting Information. The mol% of n-octyl groups produced
from the anti-Markovnikov (–Si–CH2–CH2–CH2–) and Markovnikov
additions (–Si–CHCH3–CH2–) is between 90.2 – 94.9 and 5.1–9.8,
Fig. 5. 2-Dimensional heterogeneous NMR spectra of PES20.

H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
2301
PEHS, the molecular weight measured from light scattering
detector should increase with the increase of the POSS content in
the polymer because these polymers were made by substitution
reactions of PEHS. Therefore, slight backbone cleavage may occur
during the substitution reactions. We still don’t know well the
mechanism of the backbone cleavage because we have only limited
information from NMR.
upon the incorporation of POSS groups into the polymer chains were
also observed from other polymers such as polyfluorenes,
poly(4-methylstyrene)s, poly(norbornyl)s, and poly(acrylate)s [47–
51]. In addition, the PES0 and PES1 have very flexible backbones
because of low POSS content and therefore show good enthalpy
relaxation behavior. For these polymers, increasing POSS content
increased Tg, indicating that the flexible domain of polymers
decreases with increasing POSS content [46,52–54]. Although we
could not measure Tg values of PES20 and PES25, these values would
probably be lower than 0 ꢀC based on other PES#s in this study. Still,
non-sticky free standing films could be prepared as shown in Fig. 3(a)
3.2. Thermal analysis
Fig. 6 shows DSC curves obtained from the second heating scans at
a heating rate of 10 ꢀC/min. Only glass transition temperatures (Tg)
were obtained when the samples were heated from ꢁ150 ꢀC. Because
the side chain melting transition of alkyl side groups occurs below
100 ꢀC, our polymers do not have any side chain crystalline phases
from the n-octyl side group that can affect the physical properties as
fillers in the polymer chain [41]. Still, the n-octyl side group increased
the Tg of the polysiloxane; when the silane group in PEHS was 100%
substituted with a n-octyl side group, Tg increased from ꢁ127 ꢀC for
PEHS to ꢁ97.3 ꢀC for PES0. The medium-length n-alkyl side groups
such asn-octyl decrease Tg because theyact as a plasticizer to increase
the free volume of the polymers by pushing the polymer backbones
apart. Therefore, the Tg values of poly(n-alkyl methacrylate)s, poly(n-
alkyl styrene)s, and poly(n-alkyl vinyl ether)s are lower than those
polymers without corresponding n-alkyl side groups [42,43].
However, in our case, the n-octyl side group decreases the flexibility
of the polysiloxane backbone by increasing the steric hindrance to the
segmental motion of the flexible siloxane backbones. Therefore, the
n-octyl side group can decrease Tg when attached to relatively stiff
backbone such as poly(acrylate), poly(styrene), and poly(vinyl ether),
but increases Tg for the flexible polysiloxanes as ours. Similarly, the Tg
of polysiloxane derivatives with long n-alkyl groups are higher than
that of poly(dimethyl siloxane) with small side groups [44,45]. The
POSS side groups increased Tg further; as the content of POSS side
groups increased, Tg increased. For example, Tg values of PES1, PES4,
PES7 and PES10 were ꢁ95.5, ꢁ92.4, ꢁ87.3, and ꢁ77.3 ꢀC, respectively.
We could not observe a glass transition from DSC curves of PES20 and
PES25 with larger POSS content, evenwhen we quenched the sample
from 150 ꢀC to ꢁ150 ꢀC using liquid nitrogen, possibly due to the very
high POSS content; although the mol% of POSS side groups in PES20
and PES25 are 20 and 25, respectively, the weight percents of these
polymers are 55.3% and 61.4%, respectively. Therefore it is not easy to
detect the changes in heat capacity upon heating from the small
portions of polymer chains in these polymers [46]. The increase of Tg
for PES20 at 200
thickness.
mm thickness and in Fig. 3(b) for PES25 at 340 mm
Fig. 7 shows the TGA thermograms of POSS-containing poly-
siloxanes (PES#s), the starting polysiloxane (PEHS), and pure allyl–
CyPOSS. The decomposition temperatures of 10% weight loss (Td,10%
)
and the char yield data obtained from the TGA measurements are
listed in Table 2. As shown in Fig. 7, all the polymers with POSS side
groups exhibit good thermal stability above 400 ꢀC. PEHS was
thermally decomposed below 300 ꢀC similar to PDMS by depoly-
merization [55], whereas thermal decomposition process of PES0
with only octyl side groups was different [see DTGA curves in the
Supporting Information]. The n-octyl side groups in PES#s may
interfere with the back-biting process for depolymerization [55,56].
Within the experimental temperature range, the TGA curves of all
the polymers showed one-step decomposition behavior, suggesting
that the incorporation of the CyPOSS nanoparicles changed the
decomposition mechanism of the polymers, since the allyl–CyPOSS
nanoparticle shows two-step decomposition behavior. The first
derivative peak of CyPOSS nanoparicles is attributable to decompo-
sition of the organic cyclohexyl ring at the POSS vertices, and the
second peak is attributed to the inorganic POSS framework [57,58].
The incorporation of POSS groups into the side chain of polysiloxane
increases the thermal stability, resulting in a retarded weight loss
rate and char yield. For example, the incorporation of 1 mol% POSS
groups increased the decomposition temperature by 47 ꢀC. The
enhanced thermal stability of POSS-containing polymers also occurs
in other polymers [59–65].
3.3. Morphological characterization
Fig. 8 shows the wide-angle X-ray scattering (WAXS) patterns of
the polymers and allyl–CyPOSS. The d-spacings of the samples were
calculated using the Bragg equation. The WAXS pattern of PES0
Fig. 6. DSC curves of polymers.
Fig. 7. TGA thermograms of PEHS, allyl–CyPOSS, and polymers.

2302
H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
Table 2
POSS groups should appear, as previously reported [67–73].
Thermal and mechanical properties of the polymers.
Therefore, peak broadness and/or disappearance in PES#s indicates
that the POSS groups are well-dispersed in the polymer matrix and/
or ethyl n-octyl siloxane repeating units and ethyl POSS siloxane
repeating units are randomly distributed in the polymer backbone.
The films of PES25 have sharper peaks and are quite brittle, possibly
due to the aggregation of POSS groups in the polymer due to high
POSS content of 61.4 wt%.
Tg (ꢀC) Td, 10% Char
G0/G00
viscosity E0/E00
s
(ꢀC)a
yield
(%)b
(Pa)c
(Pa s)d
(MPa)e (MPa)f
CyPOSS
PEHS
PES0
PES1
PES4
–
346
8.91
1.21
24.1
27.0
28.8
30.6
31.1
32.8
34.8
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
ꢁ127g 314
1.0h
8.3
14.2
ꢁ97.3
ꢁ95.8
ꢁ92.4
ꢁ87.3
ꢁ77.3
359
407
427
428
432
439
436
2.2/74.5
9.2/140.6
22.2/270.5 26.0
79.0/582.4 70.9
–
–
–
PES7
3.4. Rheological and mechanical properties
PES10
PES20
PES25
–
–
–
i
–
–
5.6/1.7 2.1
–
i
Small-amplitude dynamic oscillatory measurements were con-
ducted to investigate the effects of the POSS groups on rheological
properties. These measurements were all performed in the linear
viscoelastic regions of the polymer melts at 25 ꢀC. Rheological
results were obtained on the four polymers from PES0 to PES7, but
PES10, PES20, and PES25 were too viscous at 25 ꢀC to measure the
rheological properties. Fig. 9 shows the shear storage modulus (G0)
and loss modulus (G00) curves obtained by frequency sweep tests at
25 ꢀC. Both G0 and G00 increase with increased POSS content, and G00 is
dominant over G0, with no crossover, indicating the polymers show
–
a
The decomposition temperature (Td,10%) is defined as 10 wt% loss.
b
c
d
e
f
The char yield at 700 ꢀC.
Shear moduli obtained from rheometer at 10 rad sꢁ1
.
Zero-shear viscosity was obtained at lowest shear rate, 0.1 sꢁ1
Tensile moduli obtained from DMTA at 25 ꢀC.
.
Tensile strength obtained from UTM at 25 ꢀC.
g
h
i
Literature value of poly(dimethylsiloxane) in Polymer Handbook 4th edition.
The value of PEHS from supplier, Gelest.
Tgs were not detected.
predominantly viscous flow [67]. Fig. 10 shows shear viscosities (h)
versus shear rates (sꢁ1) of the polymers. The zero-shear viscosity
values of the polymers are quite low, but increase with higher POSS
content, from 8.3, 12.2, 26.0, and 70.9 Pa s for PES0, PES1, PES4, and
PES7, respectively, at 0.1 shear rate. Since these polymers were
prepared by polymer modification reactions through hydro-
silylation under very mild reaction conditions, backbone cleavage to
decrease the degree of polymerization should not occur, and back-
bone length should be consistent [74]. Since the molecular weight
values measured from the GPC of these polymers are not very
different (Mn values of PES0, PES1, PES4, and PES7 are in the range of
69,000 to 94,700), the large differences of the shear viscosity value
should not be caused by the hydrodynamic volume differences of
these polymers [75]. Although the solubility of polymers in THF may
be different with different POSS content, the differences of the shear
viscosity values seems to be much larger than the molecular weight
differences. Higher POSS content could increase the shear viscosity
by bulkiness, which can increase the hydrodynamic volume of the
polymers in the melt state and/or increase the entanglements
between side chains [76–80]. The polymer with higher POSS
content shows shear-thinning behavior at a lower shear rate; for
shows two amorphous halos: one is centered at a d-spacing of 4.3 Å
and the other one is at 13.8 Å. These two broad peaks are associated
with correlations between the polysiloxane backbones (d ¼ 13.8 Å)
and between n-octyl side groups (d ¼ 4.3 Å), respectively. When the
POSS content increases in PES#s, the d-spacing of the backbone
peak decreases, while that of the side chain peak increases. The
incorporation of the POSS groups in the polymer may increase the
distance between the n-octyl side groups, which in turn decreases
the distance between the polymer backbones (see Supporting
Information). The pure allyl–CyPOSS is a crystalline substance and
features several diffraction peaks. The diffractogram of allyl–
CyPOSS exhibits strong reflections at d-spacing of 11.3
Å
(2
q
¼ 7.85ꢀ), 8.4 Å (10.5ꢀ), and 4.8 Å (18.3ꢀ), suggestive of a rhom-
bohedral unit cell and nearly identical to the pattern for an octa-
cyclohexyl–POSS studied by J. Schwab and co-workers [66]. In
particular, there is a very intense and sharp diffraction peak at
11.3 Å (2
q
¼ 7.85ꢀ) associated with the 101 hkl reflection of a lattice.
These sharp peaks of allyl–CyPOSS become broad when incorpo-
rated in the side chain of polysiloxane, although they become
a little sharper as the POSS content increases. If the POSS groups
aggregate in the polymer matrix, relatively sharp peaks from the
Fig. 9. Dynamic storage modulus (G0) and loss modulus (G00) curves for polymers
(PES#s, # ¼ 0, 1, 4, 7); (-) PES0, (C) PES1, (:) PES4, and (;) PES7. Filled and open
symbols stand for G0 and G00, respectively.
Fig. 8. Line profiles of WAXS diffractograms of polysiloxanes having POSS side groups.

H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
2303
the polymer backbone with hydroxyl groups. As the POSS content
increases, glass transition temperature, decomposition tempera-
ture, char yield, modulus, and viscosity increase. For example when
the content of POSS-containing monomeric units increased from
0 to 7 mol%, the zero-shear viscosity values at 0.1 shear rate
increase from 8.3 to 70.9 Pa s. Even free standing films could be
prepared from the polysiloxanes with POSS monomeric units of
more than 10 mol%.
Acknowledgments
This research was supported by Samsung Electronics, Dongjin
Semichem, and a Grant from Construction Technology Innovation
Program (CTIP) funded by Ministry of Land, Transportation, and
Maritime Affairs (MLTM) of Korean government. Experiments at
the Pohang Accelerator Laboratory were supported in part by MOST
and POSTECH.
Fig. 10. Apparent shear viscosity curves for the polymers (PES#s, # ¼ 0, 1, 4, 7); (,)
PES0, (B) PES1, ( ) PES4, and (7) PES7.
6
Appendix. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.polymer.2010.01.066.
example, the decrease in shear viscosity with the increase of the
shear rate (the shear-thinning behavior) starts at about 40 sꢁ1 and
400 sꢁ1 for PES7 and PES0, respectively. Similarly, more concen-
trated polymer solutions and polymer melts with higher molecular
weight show increased shear viscosity and shear-thinning behavior
at lower shear rates due to increased entanglements of the polymer
chains [81–83].
References
[1] Kessler D, Roth PJ, Theato P. Langmuir 2009;25:10068–76.
[2] Gao ZM, Nahrup JS, Mark JE, Sakr A. J Appl Polym Sci 2003;90:658–66.
[3] Kessler D, Teutsch C, Theato P. Macromol Chem Phys 2008;209:1437–46.
[4] Paul KE, Prentiss M, Whitesides GM. Adv Funct Mater 2003;13:259–63.
[5] Kessler D, Lechmann MC, Noh S, Berger R, Lee C, Gutmann JS, et al. Macromol
Rapid Commun 2009;30:1238–42.
[6] Kessler D, Theato P. Macromolecules 2008;41:5237–44.
[7] Madhavan K, Reddy BSR. J Polym Sci Part A Polym Chem 2006;44:2980–9.
[8] Srividhya A, Reddy BSR. J Polym Sci Part A Polym Chem 2007;45:1707–26.
[9] Srividhya M, Madhavan K, Reddy BSR. Eur Polym J 2006;42:2743–54.
[10] Yamada Y, Furukawa N. Polym J 1997;29:923–30.
[11] Teramoto N, Unosawa M, Matsushima S, Shibata M. Polym J 2007;39:975–81.
[12] Dickstein WH, Siemens RL, Hadziioannou E. Thermochim Acta 1990;166:
137–45.
[13] Sim LC, Ramanan SR, Ismail H, Seetharamu KN, Goh TJ. Thermochim Acta
2005;430:155–65.
[14] Lewicki JP, Patel M, Morrell P, Liggat J, Murphy J, Pethrick R. Sci Technol Adv
Mater 2008;9:024403–10.
[15] Rajan GS, Sur GS, Mark JE, Schaefer DW, Beaucage G. J Polym Sci Part B Polym
Phys 2003;41:1897–901.
[16] Osman MA, Atallah A, Kahr G, Suter UW. J Appl Polym Sci 2002;83:2175–83.
[17] Yuan QW, Mark JE. Macromol Chem Phys 1999;200:206–20.
[18] Osman MA, Atallah A, Muller M, Suter UW. Polymer 2001;42:6545–56.
[19] Pan GR, Mark JE, Schaefer DW. J Polym Sci Part B Polym Phys 2003;41:
3314–23.
[20] Finkelmann H, Rehage G. Makromol Chem Rapid Commun 1980;1:31–4.
[21] Gray GW, Lacey D, Nestor G, White MS. Makromol Chem Rapid Commun
1986;7:71–6.
[22] Zhang B-Y, Meng F-B, Li Q-Y, Tian M. Langmuir 2007;23:6385–90.
[23] Hsu CS, Leu YJ. Mol Cryst Liq Cryst 1997;300:83–95.
[24] Cooray NF, Kakimoto M, Imai Y, Suzuki Y. Polym J 1993;25:863–72.
[25] Juang TM, Leu YJ, Hsu CS. Mol Cryst Liq Cryst 1993;237:223–34.
[26] Hsiue GH, Hsieh CJ, Yu NH. Liq Cryst 1992;12:705–14.
[27] Janini GM, Laub RJ, Shaw TJ. Makromol Chem Rapid Commun 1985;6:57–63.
[28] Kreuder W, Ringsdorf H. Makromol Chem Rapid Commun 1983;4:807–15.
[29] Jonas Gerd, Stadler Reimund. Makromol Chem Rapid Commun 1991;12:625–32.
[30] Lichtenhan JD, Otonari YA, Carr MJ. Macromolecules 1995;28:8435–7.
[31] Mukbaniani O, Tatrishvili T, Titvinidze G, Mukbaniani N. J Appl Polym Sci
2006;101:388–94.
[32] Chung DW, Kim TG. J Ind Eng Chem 2007;13:979–84.
[33] Zhang CX, Laine RM. J Am Chem Soc 2000;122:6979–88.
[34] Rim PB, Rasoul HAA, Hurley SM, Orler EB, Scholsky KM. Macromolecules
1987;20:208–11.
[35] Mukbaniani O, Titvinidze G, Dundua A, Doroshenko M, Tatrishvili T. J Appl
Polym Sci 2008;107:2567–71.
Dynamic mechanical thermal analysis (DMTA) was performed
to measure tensile storage modulus (E0) and loss modulus (E00) of
the polymer films. Although free standing films could be prepared
from PES10, PES20, and PES25, the measurement was only possible
from PES20 because PES10 is sticky and PES25 is too brittle to make
rectangular samples for the measurement. The tensile storage
modulus and loss modulus values of PES20 are 5.6 MPa and 1.7 MPa
at 25 ꢀC, respectively, and its tensile strength was in the range of
1.9–2.1 MPa, which is in the lower range of cross-linked silicone
polymers for practical applications; tensile strength values of cross-
linked silicone polymers are 0.5–12.0 MPa [84–86]. This lower
tensile strength value of PES20 compared with those of cross-
linked silicone polymers implies that the increased physical inter-
actions through the incorporation of POSS groups into the silicone
polymers are still lower than those obtained through cross-linking
reactions. However, linear polysiloxanes with POSS side groups
synthesized in this study are soluble in many organic solvents, so
they can be used for silicon coating materials with good chemical
and mechanical stability. Transparent coating materials having high
chemical and mechanical properties could be prepared if some
cross-linkable moieties such as acryl groups are included in the
polysiloxanes with POSS groups during the post-curing process
using heating or UV irradiation. Such work is in progress.
4. Conclusion
A series of polysiloxane derivatives with different amounts of
POSS groups in the side chain were prepared for the first time.
Poly(ethylhydrosiloxane) was used as starting polymer because it
has high enough molecular weight to prepare reproducible poly-
mer products with small amounts of POSS groups. Platinum(0)-
divinyl tetramethyldisiloxane was used as a catalyst for the
hydrosilylation reaction of poly(ethylhydrosiloxane) to prevent the
formation of hydroxyl and alkoxy side groups which can cause the
cross-linking of the polymers. 1-Octene was used to protect the
Si–H groups in the side chains to prevent the reactions of Si–H in
[36] Pyun SY, Kim WG. Macromol Res 2003;11:202–5.
[37] Furukawa Y, Kotera M. J Polym Sci Polym Chem 2002;40:3120–8.
[38] Boutevin B, Guida-Pietrasanta F, Ratsimihety A. J Polym Sci Polym Chem
2000;38:3722–8.
[39] Brook MA. Silicon in organic, organometallic, and polymer chemistry. 1st ed.
New York: Wiley-Interscience; 2000 [chapter 12].

2304
H.-S. Ryu et al. / Polymer 51 (2010) 2296–2304
[40] Marciniec B, Gulinski J, Kopylova L, Maciejewski H, Grundwald-Wyspianska M,
Lewandowski M. Appl Organomet Chem 1997;11:843–9.
[41] Plate NA, Shibaev VP. J Polym Sci Macromol Rev 1974;8:117–253.
[42] Rogers S, Mandelkern L. J Phys Chem 1957;61:985–91.
[43] Overberger CG, Frazier C, Mandelman J, Smith HF. J Am Chem Soc 1953;75:
3326–30.
[44] Stern SA, Shah VM, Hardy BJ. J Polym Sci Part B Polym Phys 1987;25:1263–98.
[45] Dvornic PR, Lenz RW. Macromolecules 1992;25:3769–78.
[46] Hodge IM. J Non-Cryst Solids 1994;169:211–66.
[47] Lee J, Cho HJ, Jung BJ, Cho NS, Shim HK. Macromolecules 2004;37:8523–9.
[48] Chou CH, Hsu SL, Dinakaran K, Chiu MY, Wei KH. Macromolecules
2005;38:745–51.
[49] Jeon HG, Mather PT, Haddad TS. Polym Int 2000;49:453–7.
[50] Mather PT, Jeon HG, Romo-Uribe A, Haddad TS, Lichtenhan JD. Macromole-
cules 1999;32:1194–203.
[65] Hoque MA, Kakihana Y, Shinke S, Kawakami Y. Macromolecules
2009;42:3309–15.
[66] Fu BX, Hsiao BS, Pagola S, Stephens P, White H, Rafailovich M, et al. Polymer
2001;42:599–611.
[67] Wang JL, Ye ZB, Joly H. Macromolecules 2007;40:6150–63.
[68] Zheng L, Waddon AJ, Farris RJ, Coughlin EB. Macromolecules 2002;35:2375–9.
[69] Zheng L, Hong S, Cardoen G, Burgaz E, Gido SP, Coughlin EB. Macromolecules
2004;37:8606–11.
[70] Lee KM, Knight PT, Chung T, Mather PT. Macromolecules 2008;41:4730–8.
[71] Fu BX, Hsiao BS, White H, Rafailovich M, Mather PT, Jeon HG, et al. Polym Int;
2000:437–40.
[72] Kim BS, Mather PT. Macromolecules 2006;39:9253–60.
[73] Romo-Uribe A, Mather PT, Haddad TS, Lichtenhan JD. J Polym Sci Part B Polym
Phys 1998;36:1857–72.
[74] Kim BG, Moon JK, Sohn EH, Lee JC, Yeo JK. Macromol Res 2008;16:36–44.
[75] Sperling LH. Introduction to physical polymer science. 3rd ed. New York:
Wiley-Interscience; 2001 [chapter 3].
[51] Haddad TS, Lichtenhan JD. Macromolecules 1996;29:7302–4.
[52] Andreozzi L, Faetti M, Giordano M, Zulli F. Macromolecules 2005;38:6056–67.
[53] McGonigle EA, Cowie JMG, Arrighi V, Pethrick RA. J Mater Sci 2005;40:
1869–81.
[76] Gupta RK. Polymer and composite rheology. 2nd ed. New York: Marcel Dek-
ker; 2000.
[54] Tanaka Y. Polym J 2007;39:1030–9.
[77] Graessley WW. Acc Chem Res 1977;10:332–9.
[78] Parent JS, Bodsworth A, Sengupta SS, Kontopoulou M, Chaudhary BI, Poche D,
et al. Polymer 2009;50:85–94.
[79] Sugimoto M, Suzuki Y, Hyun K, Ahn KH, Ushioda T, Nishioka A, et al. Rheol
Acta 2006;46:33–44.
[80] Piel C, Stadler FJ, Kaschta J, Rulhoff S, Munstedt H, Kaminsky W. Macromol
Chem Phys 2006;207:26–38.
[81] Graessley WW. Adv Polym Sci 1974;16:1–179.
[82] Lyons PF, Tobolsky AV. Polym Eng Sci 1970;10:1–3.
[83] Ortiz M, Kee DD, Carreaub PJ. J Rheol 1994;38:519–39.
[84] Park ES. J Appl Polym Sci 2008;110:1723–9.
[55] Grassie N, Macfarlane IG. Eur Polym J 1978;14:875–84.
[56] Camino G, Lomakin SM, Lazzari M. Polymer 2001;42:2395–402.
[57] Liu YR, Huang YD, Liu L. J Appl Polym Sci 2008;110:2989–95.
[58] Kim BS, Mather PT. Macromolecules 2002;35:8378–84.
[59] Liu H, Zheng S, Nie K. Macromolecules 2005;38:5088–97.
[60] Strachota A, Kroutilova I, Kovarova J, Matejka L. Macromolecules 2004;37:
9457–64.
[61] Zheng L, Farris RJ, Coughlin EB. Macromolecules 2001;34:8034–9.
[62] Tsuchida A, Bolln C, Sernetz FG, Frey H, Mulhaupt R. Macromolecules
1997;30:2818–24.
[63] Liu YR, Huang YD, Liu L. Polym Degrad Stab 2006;91:2731–8.
[64] Liu YR, Huang YD, Liu L. J Mat Sci 2007;42:5544–50.
[85] Xu CH, Feng SY. J Appl Polym Sci 2000;76:1554–7.
[86] Jia LY, Du ZJ, Zhang C, Li CJ, Li HQ. Polym Eng Sci 2008;48:74–9.