New approach in the synthesis of hybrid polymers grafted with polyhedral oligomeric silsesquioxane and their physical and viscoelastic properties

Article abstract of DOI:10.1021/ma047892y

Synthesis, chain characteristics, and time-dependent viscoelastic response of polyhedral oligomeric silsesquioxanes (POSS)-containing hybrid polymers were investigated. Unlike many other reported POSS hybrid copolymers works, the POSS-grafted copolymers u

Full text of DOI:10.1021/ma047892y

438
Macromolecules 2005, 38, 438-444
New Approach in the Synthesis of Hybrid Polymers Grafted with
Polyhedral Oligomeric Silsesquioxane and Their Physical and
Viscoelastic Properties
Andre Lee*
Department of Chemical Engineering & Materials Science, Michigan State University,
East Lansing, Michigan 48824
Jun Xiao and Frank J. Feher‡
†
Department of Chemistry, University of California, Irvine, California 92697
Received October 11, 2004; Revised Manuscript Received November 3, 2004
ABSTRACT: Synthesis, chain characteristics, and time-dependent viscoelastic response of polyhedral
oligomeric silsesquioxanes (POSS)-containing hybrid polymers were investigated. Unlike many other
reported POSS hybrid copolymers works, the POSS-grafted copolymers used in this study, although having
different amounts of POSS attachments, have the same degrees of polymerization and molecular weight
distribution. This was accomplished by grafting different amounts of aluminosilsesquioxane onto a
previously synthesized, random copolymer of styrene and vinyl-diphenylphosphine oxide, PSP. The
coordination bonding between aluminum and phosphine oxide is quantitative, which enables the
investigation of physical characteristics and viscoelastic response of polymers as influenced only by the
POSS attachments. Similar to other hybrid polymers with covalently bonded POSS, we observed a systemic
increase in the characteristic relaxation time of polymers at the terminal zone due to the POSS attachment.
g
Linear viscoelastic response at different temperatures above T was studied using the small-strain
amplitude oscillatory shear technique. It was found that these POSS-grafted copolymers obey the time-
temperature superposition principle. In addition, for isothermal experiments, the linear viscoelastic
response obtained for copolymer with varying amounts of POSS attachment were able to be superposed,
thus demonstrating that these hybrid copolymers obey the time-POSS content superposition principle.
We also examine the effect of POSS-POSS interactions on the long-term viscoelastic response of these
copolymers. It was found that copolymers with high POSS attachments exhibit a slow gelation response
g
when held isothermally at temperatures above T for extended periods of time, while copolymers with
low POSS attachments remain unchanged for the same periods of time. Therefore, this observed gel-like
behavior is due to the intrachain POSS-POSS interaction.
Introduction
terization results of these copolymers for the opportu-
nity to develop structure-property relationships. These
polymers are synthesized by grafting of Al-containing
POSS cages onto a random copolymer of styrene and
vinyl-diphenylphosphine oxide. Since the grafting of
POSS cages occurs after polymerization, the effects of
POSS incorporation can be investigated using polymers
of common backbones. The physical characteristics of a
single polymer chain as affected by the number of POSS
attachments were studied using standard dilute solution
techniques such as intrinsic viscosity and dynamic light
scattering. In addition, the melt rheological behavior of
these copolymers at different temperatures was inves-
tigated using the small-amplitude oscillatory shear
method. Series of isothermal, frequency sweep experi-
ments were used to examine the validity of the time-
temperature superposition principle on these POSS-
containing copolymers. We also examined time-tem-
perature shift factors for these hybrid copolymers as
affected by the amount of POSS attachments. Since
there is only a limited miscibility between POSS mol-
ecules and the organic polymeric host, it was suggested
that above a critical number of POSS attachments, a
strong POSS-POSS interaction will influence the poly-
mer dynamics. In this study we explore this interaction
by examining the isothermal, viscoelastic response at
different times at a temperature above the glass-
transition temperature of the copolymers.
Polyhedral oligomic silsesquioxanes (POSS) have at-
tracted widespread interest as precursors to high-per-
formance materials. Of particular interest have been
POSS-containing inorganic-organic polymeric materi-
als, which often exhibit enhanced performance charac-
teristics versus stoichiometrically similar non-POSS-
1
-3
containing polymers.
Although it is generally recog-
nized that POSS incorporation into organic polymers
can have profound effects on polymer properties, rela-
tively little is known with certainty about how the pre-
sence of POSS alters polymeric properties because it is
difficult to prepare the necessary series of POSS-con-
taining polymers that selectively isolate the POSS ef-
fects. Most of the work reported on POSS-containing
copolymers has been one-pot syntheses, where POSS
monomers and organic monomers were mixed together
prior to the polymerization process.4-10 It is known that
the presence of POSS monomers can have dramatic
1
1
effects on the rates of polymerization. Therefore, any
changes in POSS monomer structure and/or initial feed
ratios can lead to nonrandom and difficult-to-quantify
changes in the resulting polymer. Hence, it is difficult
to attribute the observed differences purely due to the
addition of POSS.
In this paper we report a new approach to the syn-
thesis of POSS-grafted copolymers and provide charac-
*
To whom correspondence should be addressed. E-mail:
leea@egr.msu.edu.
Experimental Section
†
Currently at GE-China. E-mail: Jun.Xiao@ge.com.
Currently at Goodyear Chemical, Arkon, OH. E-mail:
‡
Styrene (St, Aldrich) was washed three times with aqueous
1 M NaOH to remove inhibitors, then washed three times with
frank_feher@goodyear.com.
1
0.1021/ma047892y CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/21/2004

Macromolecules, Vol. 38, No. 2, 2005
Hybrid Polymers Grafted with POSS 439
water, dried with anhydrous sodium sulfate overnight, distilled
under nitrogen at reduced pressure, and kept refrigerated.
Azobisisobutylnitrile (AIBN, Aldrich) was recrystallized from
methanol before use. The solvents, toluene and dichlo-
romethane, were water free and oxygen free. Ethanol was used
as received. Vinyl-diphenylphosphine and Oxone (potassium
peroxymonosulfate) were purchased from Aldrich Chemical Co.
and used without further purification.
prepared by copolymerization of styrene and vinyl-diphen-
ylphosphine oxide. As described below, the reaction chemistry
works equally well and provides a convenient method for
preparing well-defined POSS-containing polymers for structural/
property studies.
Synthesis of Vinyl-Diphenylphosphine Oxide (VD-
PhPO). In a 1 L one-necked flask, vinyl-diphenylphosphine
(5.0 mL, 0.025 mol), Oxone (30.7 g, 0.050 mol), and dichlo-
romethane (120 mL) were stirred at room temperature for 5
days. The solution was filtered to remove insoluble materials,
and the filtrate was washed with water three times, dried over
anhydrous sodium sulfate overnight, filtered, and evaporated
to give a white solid. The white solid was purified by column
chromatography (hexane/ether); the solvent was evaporated
Dimerization of POSS Cages. Trisilanols with structure
1
are versatile precursors to POSS monomers. These trisilanols
typically are reacted with trifunctional organosilicon reagents
e.g., R′SiX ) to produce POSS monomers with the ubiquitous
Si 12 framework.
(
3
8
O
to give vinyl-diphenylphosphine oxide (VDPO, 3.3 g, yield
1
5
6
7
8%) as a white microcrystalline solid. H NMR (CDCl
3
): δ
.22-6.40 (m, 2H), 6.61-6.75 (m, 1H), 7.44-7.51 (m, 4H),
3
1
.51-7.58 (m, 2H), 7.67-7.74 (m, 4H). P NMR (CDCl
PO as external standard): δ 24.83.
3
, 85%
H
3
4
NMR Tube Reaction of 2 with VDPhPO: Synthesis of
The reaction chemistry of trisilanol 1a (R ) c-C
6
H
11) is
3a. VDPhPO (10 mg, 0.04 mmol) and 2 (43.7 mg, 0.02 mmol)
especially well developed because it was the first POSS
trisilanol available in synthetically useful quantities. In ad-
were dissolved in CDCl (0.5 mL) in a 5 mm NMR tube. The
3
formation of 3a was quantitative within 5 min at room
dition to “corner capping” reactions with RSiX
3
, 1a reacts with
1
3
temperature. H NMR (CDCl ): δ 0.55-0.79 (m, 7H), 1.10-
a wide range of main-group, transition-metal, and lanthanide
1.35 (m, 35H), 1.58-1.85 (m, 35H), 6.45-6.78 (m, 3H), 7.51-
elements to afford fully condensed metal-silsesquioxanes. For
7.58 (m, 4H), 7.65-7.70 (m, 2H), 7.78-7.86 (m, 4H). 31P NMR
example, the reaction of 1a with Me
3
Al affords 2, which reacts
(CDCl , 85% H PO as external standard): δ 38.45. 13C NMR
3
3
4
quickly and quantitatively with phosphine oxides (e.g., Ph
3
-
(CDCl ): δ 23.53, 23.77, 24.36, 26.70, 26.97, 27.07, 27.13, 27.22,
3
PO) to produce cubeoctameric AlSi
7
O12 frameworks (e.g., 3a)
27.38, 27.54, 27.69, 27.75, 27.96, 124.95, 125.28, 125.78,
that are isostructural with many common POSS monomers.
126.18, 129.12, 129.23, 131.95, 132.05, 133.89, 133.91, 139.79.
The details of the chemistry for the synthesis of 2 and 3a have
29Si NMR (CDCl ): δ -69.79, -68.67, -65.61 (3:1:3).
3
1
2
been described elsewhere.
Synthesis of Styrene and VDPhPO Copolymers. In a
50 mL Schlenk flask, vinyl-diphenylphosphine oxides (1.9980
1
g), styrene (4.537 g), 2,2′-azobisisobutronitrile (AIBN, 0.3% of
the total weight of monomers), and toluene (15.024 g) were
added. The reactor was evacuated and backfilled with argon
three times, and then it was tightly sealed and immersed in
an oil bath at 60 °C. The solution was stirred for 3 days under
argon. After cooling the solution was poured into a large excess
of ethanol to precipitate the copolymer, which was collected
by filtration. The polymer was then twice purified by dissolving
in toluene, filtering, and then reprecipitating with ethanol. The
resulting poly(styrene-co-vinyl diphenylphosphine oxide), PSP,
was dried at 60 °C under vacuum to give a white powder (4.9
3
1
g, yield 75%); the composition was determined by P NMR
spectroscopy using monomer VDPhPO as the reference; the
3
1
ratio of integrated P NMR intensities for free VDPhPO and
PSP copolymer was 1:1.6, which corresponds to a P content of
0
.7 mmol of P per gram of copolymer. By changing the amount
of AIBN used we were able to synthesize two different
molecular weights of PSP; however, the molar ratio of dPhPO
to styrene in the PSO copolymer is the same for both PSP.
Here we extend the previous work with aluminum-contain-
ing silsesquioxanes to a polymer-bound phosphine oxide

440 Lee et al.
Macromolecules, Vol. 38, No. 2, 2005
Synthesis of POSS-Containing Polymers: Reactions
Table 1.
of PSP with 2. Samples containing 0.1, 0.2, 0.5, and 0.9 Al
per polymer-bound PdO were prepared. In a typical procedure
low molecular weight
samples 8.0 mol % dPhPO
high molecular weight
samples 7.4 mol % dPhPO
PSP (ca. 1 g) and 2 were dissolved in dry CH
glass vial equipped with a small magnetic stirring bar. The
total concentration of PSP and 2 in CH Cl was 10 wt % for
2 2
Cl in a 20 mL
[
η]
R
H
(nm)
[η]
R
H
(nm)
3
3
sample
PSP 1-0 (0%)a
(cm /g)
K
H
sample
(cm /g)
K
H
2
2
PSP 2-0 (0%)
17.82 0.67 5.7
all reactions. Within 2 min of mixing the solution became clear
and colorless; the reaction was continued for another 10 min.
Evaporation of the solvent gave a white solid, which was dried
overnight in a vacuum oven at 110 °C and 0.1 Torr. Provided
that the number of PdO groups from PSP exceeds the number
of Al sites provided by 2, the grafting reaction proceeds to
completion with respect to 2 and the yield is quantitative. The
mole fraction of POSS attached to PdO on the 1:1.6 VDPhPO/
PSP 1-1 (10%) 12.16 0.51 4.6 PSP 2-1 (10.9%) 17.56 0.51 5.8
PSP 1-2 (20%) 12.15 0.48 4.6 PSP 2-2 (21.7%) 17.52 0.47 6.1
PSP 1-3 (50%) 11.89 0.42 4.8 PSP 2-3 (54.5%) 17.12 0.42 6.5
PSP 1-4 (90%) 11.78 0.38 5.0
a
(
x%) indicates the fraction dPhPO sites that is attached with
POSS molecules.
1
PSP copolymer was confirmed by H NMR spectroscopy in
to be negligible. Four different concentrations were used for
each sample, and the average efflux time used to determine
intrinsic viscosity was determined using average of 10 mea-
surements with an error of 0.5%.
3
CDCl using the total integrated intensities for all cyclohexyl
groups on the POSS framework (δ 0.60-2.80 (m br)) and all
aromatic protons (δ 5.90-7.90 (m br)). NMR spectral data for
1
the copolymer containing 0.5 Al/PdO are as follows. H NMR
Rheological Experiments. All rheological experiments
were performed using a Paar Physica rheometer, Universal
Dynamical Spectrometer UDS200 operated with direct-strain
control mode. The rheometer is equipped with a convection-
heated air oven for temperature control within (0.1 °C of the
set temperature. A parallel plate geometry with a diameter
of 8 mm and a gap of 0.5 mm was used for all measurements
obtained in this study. The fixture was preheated to the
highest testing temperature, 200 °C, and lowered to obtain
the zero-gap reference point. The molded disc-shaped specimen
was then placed in the rheometer, and the top fixture was
slowly lowered to the preset gap of 0.5 mm with minimal
normal force. The sample was then subjected to a series of
isothermal, small-strain oscillatory shear tests with oscillatory
frequencies ranging from 20 to 0.01 rad/s and strain amplitude
of 2%. To examine the validity of the time-temperature
superposition principle and the effect of POSS addition on the
values of the time-temperature shift factor, each sample was
tested at temperatures ranging from 200 to 120 °C with a 10
3
1
(
(
3
(
2
1
(
CDCl
CDCl
3
): δ 0.60-2.80 (m br), 5.90-7.90 (m br). P NMR
3
3 4
, 85% H PO as external standard): δ 32.8-34.0 (br),
5.0-38.1 (br) (free ‘PO’), 49.0-50.1 (br), 50.5-53.2 (br)
13
coordinated ‘PO’). C NMR (CDCl
3
): δ 23.52 (br), 23.79 (br),
4.34 (br) 26.4-28.2 (m, br), 40.29 (br), 43.77 (br), 125.59 (br),
2
9
27.58 (br), 127.88 (br), 131.14 (br), 145.13 (br). Si NMR
) δ 69.61 (br), -68.45 (br), -65.58 (br) (3:1:3).
CDCl
3
°
C interval. To examine the effect of long-term POSS-POSS
interactions on the dynamics of polymer, the sample was first
placed in the rheometer set at 210 °C and held isothermally
for 15 min to equilibrate the sample temperature. The tem-
perature of the testing chamber was then reduced to 170 °C
while holding the gap constant. The sample was then subjected
to a series of isothermal, small-strain oscillatory shear tests
with oscillatory frequencies ranging from 20 to 0.01 rad/s and
a strain amplitude of 2% for different elapsed times. The
elapsed time begins when the sample first reaches 170 °C.
Since only 5 min is needed for each frequency-sweep cycle, it
is assumed that the linear viscoelastic experiment performed
is a direct reflection of the state of polymer at the given elapsed
time and the mechanical stimuli used to probe the state of
polymer is too small to alter the molecular morphology of
polymer melts.
Dynamic Light Scattering. A DynaPro Dynamic Light
Scattering Instrument with Dynamic V6 (Protein Solution Inc.)
software was used to determine the hydrodynamic radius of
POSS-containing PSP in the diluted solution with a concentra-
tion of 12 mg/mL of anhydrous toluene. Anhydrous toluene
was purchased from Aldrich and used as received. The
instrument was equipped with a temperature-controlled mi-
crosampler (MSXTC-12), a host (DynaPro-99-E-6), and a
temperature controller unit. The standard procedure is as
follows: polymer was first dissolved in anhydrous toluene
overnight with a concentration of 12 mg/mL, 80 µL of the
polymer solution was extracted using a microsyringe, and then
injected into the microcell through a 0.45 µm filter. The
microcell was then inserted into the temperature-controlled
microsampler, an elapsed time of 10 min was used for the
temperature to equilibrate at 30 °C, and each measurement
was then obtained using an acquisition time of 10 s with 80%
laser power. The same sample solution was measured 30 times.
The fluctuations in the intensity of the scattered light are
related to the diffusion coefficient. The hydrodynamic radius,
Results and Discussion
Hydrodynamic Radius and Intrinsic Viscosity.
In Table 1 we depict the values intrinsic viscosity, [η],
and Huggins coefficient, KH, as obtained using the
standard dilute solution capillary viscometry technique
and hydrodynamic radius, RH, as obtained from the
diffusion coefficient as measured by the dynamic light
scattering measurement with spherical scatter assump-
tion, respectively.
As observed, it is reasonable to suggest that as more
POSS are attached to the host PSP polymer chain, the
value of RH for the corresponding copolymer was ex-
pected to increase. However, it is interesting to point
out here that the value of the intrinsic viscosity, [η], of
the copolymer decreases with increasing amounts of
POSS attachments. From the dilute solution theory it
is suggested that the intrinsic viscosity relates to the
Brownian dynamics of polymeric coils in solution. In the
H
R , is then calculated from the diffusion coefficient using the
Stokes-Einstein relation of a spherical scatter.
Dilute Solution Viscometry. CANNON CT-1000 constant
temperature bath and cannon-manning semi-micro viscom-
eters with size 25 were used to measure the efflux time of
copolymers at different concentrations. The measurements
were taking at 30 ( 0.01 °C. The initial concentration of
samples was 75 mg/mL in anhydrous toluene. Anhydrous
toluene was purchased from Aldrich and used as received. The
filtration of polymer solution was carried out using a syringe
with 0.2 µm filters. The amount of polymer lost was assumed

Macromolecules, Vol. 38, No. 2, 2005
Hybrid Polymers Grafted with POSS 441
Figure 1. Isothermal loss modulus, G′′(ω), versus oscillatory
angular frequency, ω, at different temperatures as indicated.
The POSS-grafted copolymer used was a lower molecular
weight copolymer of poly(styene-co-vinyl diphenylphosphine
oxide), PSP, of 8 mol % diphenylphosphine oxide. A 50 mol %
of diphenylphosphine oxide was attached with POSS.
dilute limit, for the hard spheres model and Rouse
approximation, the value of [η] can be expressed as13,14
3
R N
φ
4
3
H
A
[
η] ) 2.5 ) 2.5 π
(1)
F
M
where φ is the volume fraction occupied by the spheres
in the solution and F is the density of spheres. With the
hard-sphere assumption, [η] is re-expressed with the
hydrodynamics radius, RH, molecular weight of polymer,
Figure 2. (a) Time-temperature loss modulus master curve,
G′′(a ω), versus reduced angular frequency, a ω, of Figure 1.
T
T
3
M, and NA as Avogadro’s number. If the increase in R H
(b) Master curves of G′ and G′′ versus a
T
ω, shifted using the
same time-temperature factor of the copolymer shown in
Figure 2a. The reference temperature used for the master
curve was 160 °C.
is slower than the increase in the molar mass of the
copolymer as the amount of POSS attachments in-
creases, then it is reasonable for the value of [η] to
decrease as the number of POSS attachments per chain
in the copolymer increases. Another interesting obser-
vation from the results of dilute solution viscometry is
the changes of Huggin’s coefficient with varying amounts
of POSS per chain. From Table 1 it was observed that
the value of the Huggin’s coefficient, the slope of the
specific viscosity versus concentration, decreases with
an increasing number of POSS attachments per chain;
the decrease in the Huggin’s coefficient means that the
specific viscosity of the copolymer is becoming less
dependent on polymer concentration as more POSS are
attached to the copolymer. In other words, as more
POSS molecules are attached to the polymer chain, the
interaction between the polymeric segment and solvent
becomes weaker. Hence, the polarity and solubility of
these POSS-containing copolymers in the solvent are
changed with increasing number of POSS attachments
per chain and thus affect the value of intrinsic viscosity
for these hybrid copolymers.
reduced G′(ω) and G′′(ω) as shown in Figure 2a and b,
thus validating the applicability of the time-tempera-
ture superposition principle for this POSS-containing
copolymer. Examination of the applicability of time-
temperature superposition was extended to all copoly-
mers, regardless of the number of POSS attachments
per chain. As shown in Figures 3a,b and 4 a,b, all POSS-
grafted copolymers investigated in this study obey the
time-temperature superposition principle.
In Figure 5 a and b we plotted the logarithm of the
temperature shift factors of copolymers containing dif-
ferent amounts of POSS attachments versus tempera-
ture. From these figures it was observed that the
amount of shift needed to form the master curve
increases as the POSS content increases if one uses the
same reference temperature for all copolymers tested.
Since these copolymers have the same common back-
bone, this observation suggests that the POSS attach-
ment delays the dynamics of the polymer chain in the
melt state. This retardation of polymer dynamics is
attributed to the nanoscopic size of the POSS cages
attached to the host polymer chain. This may be related
to the increase in the glass-transition temperature as
the number of POSS attachments per chain increases.
The uniqueness of our synthesis approach enables us
to examine the effects of increasing numbers of POSS
per chain without any influence from the change in the
degree of polymerization and/or molecular weight dis-
tribution, since the same host polymer is used for this
Viscoelastic Response
Effect of Temperature and POSS Content.
The results of oscillatory experiments at different
temperatures were obtained by plotting the storage
modulus, G′(ω), and loss modulus, G′′(ω), as a function
of oscillatory frequency, ω. In Figure 1 we show the
G′′( ω) versus ω at temperatures ranging from 120 to
2
00 °C. Using a reference temperature of 160 °C we
were able to shift the G′′(ω) and G′(ω) curves obtained
at different temperatures to form master curves of the

442 Lee et al.
Macromolecules, Vol. 38, No. 2, 2005
Figure 3. (a) Time-temperature storage modulus master
curves, G′(a ω), versus reduced angular frequency, a ω, of
T
T
Figure 4. (a) Time-temperature loss modulus master curves,
POSS copolymers with different extent of POSS grafting on
the high molecular weight copolymer of poly(styene-co-vinyl
diphenylphosphine oxide), PSP, of 7.4 mol % diphenylphos-
phine oxide. (b) Time-temperature storage modulus master
T T
G′′(a ω), versus reduced angular frequency, a ω, of POSS
copolymers with different extent of POSS grafting on the high
molecular weight copolymer of poly(styene-co-vinyl diphen-
ylphosphine oxide), PSP, of 7.4 mol % diphenylphosphine
oxide. (b) Time-temperature loss modulus master curves, G′′-
T T
curves, G′(a ω), versus reduced angular frequency, a ω, of
POSS copolymers with different extent of POSS grafting on
the low molecular weight copolymer of poly(styene-co-vinyl
diphenylphosphine oxide), PSP, of 8 mol % diphenylphosphine
oxide. The percent indicates the mole fraction of diphenylphos-
phine oxide sites with POSS attachment. The reference
temperature used was 160 °C.
T T
(a ω), versus reduced angular frequency, a ω, of POSS co-
polymers with different extent of POSS grafting on the low
molecular weight copolymer of poly(styene-co-vinyl diphe-
nylphosphine oxide), PSP, of 8 mol % diphenylphosphine oxide.
The percent indicates the mole fraction of diphenylphosphine
oxide sites with POSS attachment. The reference temperature
used was 160 °C.
experiment. In Figure 6a we plotted loss modulus, G′′-
(ω), and complex viscosity, η*(ω), versus angular fre-
as indicated for the copolymer with the highest POSS
content used in this study (about 7 mol % of polymer is
grafted with POSS molecules). At a specified angular
frequency the value of G*(ω) increases as the elapsed
time increases. This observation alone does not imply
any morphological change in the polymer. We then
plotted the damping factor, tan δ ) G′′(ω)/G′(ω), versus
ω, as depicted in Figure 7b. As the elapsed time
increases, the values of tan δ become less dependent
on the angular frequency. For the two longest elapsed
times values of tan δ are almost independent of ω. This
observation indicates that the viscoelastic response of
this POSS-containing copolymer is changed from a
liquidlike to gel-like behavior. The gel-like behavior may
be due to a strong POSS-POSS interaction within the
polymer chain, which acts like a physical cross-link and
can have a significant effect on the mechanical proper-
ties of polymeric melts when annealed for extended
periods of time at elevated temperatures. If the gel-like
behavior is attributed to the intrachain POSS-POSS
quency, ω, for copolymers grafted with different amounts
of POSS at a given temperature, T ) 170 °C. Using the
superposition principle we are able to shift these curves
to form master curves of reduced G′′(ω) and η*(ω), as
shown in Figure 6b. Therefore, for a given temperature
the linear viscoelastic response of POSS-grafted hybrid
copolymers obeys the time-POSS content superposition
principle. The validity of the time-POSS content su-
perposition principle suggests that although the grafting
of POSS molecules on the host polymer chain reduces
the overall mobility of the polymer chain, the activation
energy for motion of the polymer chain is not affected.
In other words, although the POSS-POSS interaction
in the polymer melts increases the characteristic relax-
ation time, the overall relaxation spectrum of the chain
is not affected by the addition of POSS attachments.1
Effect of Time on POSS-POSS Interaction.
4-16
In Figure 7a we plotted the complex modulus, G*(ω),
versus angular frequency, ω, at different elapsed times

Macromolecules, Vol. 38, No. 2, 2005
Hybrid Polymers Grafted with POSS 443
T
Figure 5. (a) Time-temperature shift factor, a , versus
Figure 6. (a) Isothermal loss modulus, G′′(ω), and complex
viscosity, η*(ω), versus angular frequency, ω, at 170 °C of
POSS copolymers with different extent of POSS grafting on
the high molecular weight copolymer of poly(styene-co-vinyl
diphenylphosphine oxide), PSP, of 7.4 mol % diphenylphos-
temperature, T, of POSS copolymers with different extent of
POSS grafting on the high molecular weight copolymer of poly-
(
styene-co-vinyl diphenylphosphine oxide), PSP, of 7.4 mol %
diphenylphosphine oxide. (b) Time-temperature shift factor,
T
a , versus temperature, T, of POSS copolymers with different
phine oxide. (b) Master curves of G′′(a
p P
ω) and η*(a ω) versus
extent of POSS grafting on the low molecular weight copolymer
reduced frequency a ω. Solid symbols represent loss modulus,
p
of poly(styene-co-vinyl diphenylphosphine oxide), PSP, of 8 mol
and empty symbols represent complex viscosity: (O) no POSS
attachment, (0) 10.9 mol % of dPhPO sites was attached with
POSS, (]) 21.7 mol % of dPhPO sites was attached with POSS,
%
diphenylphosphine oxide. The percent indicates the mole
fraction of diphenylphosphine oxide sites with POSS attach-
ment. The reference temperature used was 200 °C.
(
3) 54.5 mol % of dPhPO sites was attached with POSS).
interactions, then this transition of liquidlike to gel-like
behavior requires that a sufficient amount of POSS is
attached to the host polymer chain. To test this hypoth-
esis the copolymer with the lowest POSS attachments
per chain was used. In Figure 8 we plotted the complex
modulus, G*(ω), versus angular frequency, ω, at differ-
ent elapsed times as indicated using the copolymer with
the lowest POSS content in this study (about 0.8 mol %
of polymer is grafted with POSS molecules). We ob-
served no change of its viscoelastic responses for the
entire duration of the experiment. We believe for
copolymers with only a few POSS attachments, as in
the case of one POSS attachment per 120 repeat units,
the POSS molecule is surrounded mostly by other
segments of the polymer chain, which will screen out
the possibility of any POSS-POSS interaction. How-
ever, as in the case of a copolymer with an average of
one POSS attachment per 14 repeat units, the length
scale between successive POSS attachments is compa-
rable to the length scale of a single POSS molecule.
Therefore, strong POSS-POSS interactions were to be
developed in time. The development of these POSS-
POSS interactions is also supported by the observed
decrease in the intrinsic viscosity and the Huggin’s
coefficient in the dilute solution viscosity measurements.
Conclusions
Two series of copolymers grafted with different
amounts of POSS cages were synthesized. These POSS
copolymers were synthesized by grafting Al-containing
POSS cages onto a random copolymer of styrene and
vinyl-diphenylphosphine oxide. By changing the con-
centration of initiator while maintaining identical molar
ratios of styrene to vinyl-diphenylphosphine oxide we
were able to synthesize different molecular weights of
poly(styrene-diphenylphosphine oxide) with the same
molar ratio of styrene and vinyl-diphenylphosphine for
the host copolymers using free-radical polymerization.
The Al-containing POSS is then attached to PdO via
coordination bonding. Since grafting of POSS cages
occurs after polymerization, this enables us to examine
the effects of POSS incorporation in polymers of the
same backbone.
Using dynamical light scattering of dilute solutions,
the hydrodynamic radius of POSS-grafted copolymers

444 Lee et al.
Macromolecules, Vol. 38, No. 2, 2005
details of the polymer chain structure in dilute solution
were examined using dilute solution viscometry. To our
surprise, the value of the intrinsic viscosity for POSS-
grafted copolymers decreases with increasing POSS
attachment per chain. Two possible explanations were
suggested for this observation. On the basis of the hard-
sphere model, it is possible to reduce the value of the
intrinsic viscosity if the increase in hydrodynamic
volume is slower than the increase in molar mass of the
copolymer as the amount of POSS attachments in-
creases. However, coupled with the observed decrease
in intrinsic viscosity is the decrease in the Huggin’s
coefficient, the slope of the specific viscosity versus
concentration, as the POSS attachments per chain
increases. The combination of these two observations
suggests that the POSS-POSS interactions are stronger
than POSS-monomer, POSS-solvent, monomer-mono-
mer, and monomer-solvent interactions.
The melt rheology of POSS copolymers was examined
using small-strain amplitude oscillatory shear with
different angular frequencies. It was observed that the
linear viscoelastic response of the POSS copolymers
obeys the time-temperature superposition principle. In
addition, for a given isothermal condition, we were able
to demonstrate the applicability of a time-POSS con-
tent superposition principle to these POSS-containing
copolymers provided the average length between suc-
cessive POSS cages is long enough and it is freshly
quenched to the testing temperature from a higher
temperature. Otherwise, if the average length between
successive POSS cages is comparable to the size of a
POSS cage, the intrachain POSS-POSS interaction will
gradually dominate the dynamics of POSS copolymer.
The linear viscoelastic response of the POSS-grafted
copolymers will then change from a viscoelastic liquid-
like response to a viscoelastic gel-like behavior.
Figure 7. (a) Isothermal complex modulus, G*(ω), versus
angular frequency, ω, of the low molecular weight PSP (8 mol
%
dPhPO) with 90 mol % of dPhPO sites attached with POSS.
References and Notes
(
b) Isothermal damping factor, tan δ(ω), versus angular
(
(
(
(
(
(
(
(
(
1) Haddad, T. S.; Tomczak, S. J.; Phillips, S. H. Curr. Opin.
Solid State Mater. Sci. 2004, 8, 21.
frequency, ω, of low molecular weight PSP (8 mol % dPhPO)
with 90 mol % of dPhPO sites was attached with POSS. The
sample was annealed at 170 °C for amounts of time annealed
prior to the small-strain oscillatory shear experiment.
2) Pittman, C. U.; Ni, H.; Wang, L.; Li, G. J. Inorg. Organomet.
Polym. 2001, 11, 123.
3) Lichtenhan, J. D. In Polymeric Materials Encyclopedia;
Salamone, J. C., Ed.; CRC Press: New York, 1996; Vol. 10.
4) Haddad, T. S.; Lichtenhan, J. D. Macromolecules 1996, 29,
7302.
5) Romo-Uribe, A.; Mather, P. T.; Haddad, T. S.; Lichtenhan,
J. D. J. Polym. Sci., Polym. Phys. Ed. 1998, 36, 1857.
6) Mather, T. P.; Jeon, H. G.; Romo-Uribe, A. Macromolecules
1999, 32, 1194.
7) Tsuchida, A.; Bolln, C.; Sernetz, F. G.; Frey, H.; Mulhaupt,
R. Macromolecules 1997, 30, 2818.
8) Coughlin, E. B.; Farris, R. J.; Zheng, L. J. Polym. Sci., Polym.
Chem. 2001, 39, 2920.
9) Coughlin, E. B.; Waddon, A. J.; Zheng, L.; Farris, R. J. Nano.
Lett. 2002, 2, 1149.
(
10) Gonzalez, R. I. Synthesis and in-situ Atomic Oxygen Erosion
Studies of Space-Survivable Hybrid Organic-Inorganic POSS
Polymers. Ph.D. Dissertation, University of Flordia, 2002.
11) Haddad, T. S.; Viers, B. D.; Phillips, S. H. J. Inorg. Orga-
nomet. Polym. 2001, 11, 155.
(
(
(
(
12) Feher, F. J.; Weller, K. J. Organometallics 1990, 9, 2638.
13) Rouse, P. E. J. Chem. Phys. 1953, 21, 1272.
14) Larson, R. G. The Structure and Rheology of Complex Fluids;
Oxford University Press: New York, 1999; Chapters 3 and
Figure 8. Isothermal complex modulus, G*(ω), versus angular
frequency, ω, of the low molecular weight PSP (8 mol %
dPhPO) with only 10 mol % of dPhPO sites attached with
POSS. The sample was annealed at 170 °C for amounts of time
annealed prior to the small-strain oscillatory shear experiment.
4.
(
15) Ferry, J. D. Viscoelastic Properties of Polymers, 3rd ed.;
Wiley: New York, 1980.
16) Donth, E.-J. Relaxation and Thermodynamics in Polymers;
Akademie Verlag: Berlin, 1992; Chapter 2.
(
was determined. As expected, the hydrodynamic radius
increases as the POSS content increases. However, more
MA047892Y