DEVELOPMENT OF MISCIBLE BLENDS FROM INTRAPOLYMER
REPULSIVE INTERACTIONS: POLYCARBONATE/POLYSTYRENE IONOMERS
R. XIE, R. A. WEISS
Polymer Science Program and Department of Chemical Engineering, University of Connecticut
Storrs, CT 06269-3136
ABSTRACT
The phase behavior of blends of zinc sulfonated polystyrene (ZnSPS) and bisphenol-A
polycarbonate (PC) was studied as a function of the sulfonation level and the molecular weight
of the ZnSPS ionomer. The system exhibits upper critical solution temperature (UCST) behavior.
The cloud point temperatures increased with increasing ZnSPS molecular weight and decreased
with increasing sulfonation level. No strong interactions between ZnSPS and PC were detected
by FTIR. The composition dependence of the glass transition temperatures of the miscible
blends exhibited negative deviation from linear additivity, which is consistent with no or weak
interactions between ZnSPS and PC. Miscibility is believed to arise from strong repulsive
interactions between the charged and uncharged species on the ionomer chain.
INTRODUCTION
Many studies have shown that the introduction of attractive intermolecular interactions in
a polymer blend, such as hydrogen bonding,' ion-dipole interactions,2 acid-base interactions, 3 or
transition metal complexation4 , is effective at enhancing the miscibility of otherwise immiscible
polymers. Although it has been shown that the miscibility of random copolymers and
homopolymers may occur as a result of intramolecular repulsive interactions within the
copolymer5' 8 , relatively few reports of such systems are available in the literature. None of those
discuss blends of homopolymers with ionomers, for which one would expect exceptionally
strong repulsive interactions between the non-ionic and ionic repeat units.
lonomers appear to be exceptional candidates for developing miscible blends with other
polymers or for compatibilizing two-phase polymer blends. On the one hand, the ionic group
provides a rich chemistry for attaining specific interactions with polymers that contain
complementary polar groups, such as amine or amide groups. On the other hand, the strong
repulsive interactions that occur between the ionic and non-ionic species of the ionomer, which is
essentially a random copolymer, may be an especially effective mechanism for attaining
miscibility between polymers where no specific intermolecular interactions occur. The
contribution of the repulsive interactions to the free energy of mixing is expected to depend on
the concentration of the ionic groups and the strength of the dipole, which is dependent on the
choice of the anion and cation pair.
In this paper, we report the phase behavior of blends of bisphenol-A polycarbonate and
the zinc salt of lightly sulfonated polystyrene determined by optical microscopy and light
scattering. The absence of specific interpolymer interactions was confirmed using temperatureresolved Fourier transform infrared spectroscopy (FTIR).
EXPERIMENT
Materials. Bisphenol-A polycarbonate (PC) with Mn = 48,000 g/mole was obtained from
13
Mat. Res. Soc. Symp. Proc. Vol. 461 01997 Materials Research Society
General Electric Co. and was used as received. Lightly sulfonated polystyrene (SPS) was
9
prepared according to the procedure of Makowski et al. The reaction substitutes a sulfonic acid
group at the para-position of the phenyl ring and sulfonation is random along the chain. Three
different molecular weight polystyrenes (PS) were used as the starting polymer. The sulfonation
level was determined by titration of the sulfonic acid derivative of the SPS product in a mixed
solvent of toluene/methanol (90/10 v/v) with methanolic sodium hydroxide. The zinc salt
(ZnSPS) was prepared by neutralizing the acid derivative in toluene/methanol (90/10) with a
methanol solution of zinc acetate dihydrate. The nomenclature used for the ionomers is x.yZnSPS, where x.y denotes the sulfonation level in mol% of sulfonated styrene. The ionomer
samples used in this study are summarized in Table I.
Sample
8.7ZnSPS
10.3ZnSPS
13.7ZnSPS
5.88ZnSPS
5.89ZnSPS
Table 1.Characteristics of the Zinc Sulfonated Polystyrene
MW/M.
M. of starting PS
Sulfonation of the
(g/mole)
lonomer (mol%)
1.06
4000
8.7
1.06
4000
10.3
1.06
4000
13.7
1.13
135 000
5.88
1.09
280000
5.89
Sample Preparation. Blends were prepared by solution mixing the two polymers in a common
solvent. PC was first dissolved in tetrahydrofuran (THF) to form a 1% (g/100ml) solution, after
which a small amount of methanol was added to achieve a ratio of THEF and methanol of 10/1
v/v. The PC solution was then added dropwise to a stirred 1%solution of ZnSPS in a mixture of
THF and methanol from 10/1 to 5/1 (v/v) depending on the molecular weight of the ZnSPS.
Blend samples for DSC measurements were solution cast at 30'C. The solutions were allowed to
evaporate over 3 days, and were further dried in a vacuum oven at 80'C for another 3 days. Thin
films of the blends for FTIR spectroscopy were cast from solution onto KBr windows, and those
for optical microscopy and light scattering measurements were cast from solution onto glass
plates. All the films were then further dried in vacuum at 80'C for at least 3 days.
Measurements The morphologies and the phase transitions of the blends were observed with a
0
Nikon microscope equipped with a hot stage whose temperature was controlled within ±0.5 C.
0
measures
that
device'
built
Cloud points were also determined by light scattering using a custom
the backscattered light from a blend contained within a sealed glass vial. The scattering intensity
of the sample was recorded as the temperature was raised at a rate of 1°C/min.
Glass transition temperatures, T., were measured with a Perkin Elmer DSC-7 using a dry
nitrogen atmosphere and a heating rate of 20°C/min. The glass transition temperature was
defined as the midpoint of the change in the heat capacity at the transition. Miscibility of the
blends at various temperatures was assessed by annealing the blend at the desired temperature
under nitrogen for 30-60 minutes, quenching the sample rapidly to 0°C and then running a DSC
heating experiment. The cooling rate was assumed to be fast enough so that the subsequent
heating thermogram represented the state of the blend at the annealing temperature. The blend
was considered to be miscible at the annealing temperature if the subsequent heating thermogram
exhibited a single Tg and phase-separated if the thermogram showed two Tgs.
FTIR spectroscopy was used to assess any interactions involving either the carbonate or
the sulfonate groups. FTIR spectra were measured with a Mattson Cygnus 100 FTIR
14
spectrometer using a resolution of 1 cm-'. A total of 320 scans was signal-averaged for each
spectrum. A hot stage with temperature control of +1 °C over the range of 25°C -300'C was used
for measuring spectra at elevated temperatures.
RESULTS
All the as-cast films exhibited a two-phase morphology at room temperature. Fig. 1 shows
optical micrographs for a (20/80) 13.7ZnSPS/PC blend for three different temperature histories.
Fig. 1(a) shows a two-phase morphology for a film cast from solution at room temperature and
then heated to 200'C. When the temperature was raised to 225°C, only a single, homogeneous
phase was observed, Fig. 1(b). That result indicates that the two polymers were miscible at the
higher temperature. When the temperature was then slowly lowered from 225°C to 200'C and
annealed at 200'C for 20 minutes, phase separation was again observed, Fig. 1(c). The result
clearly shows the reversibility of the phase behavior of ZnSPS/PC blends and demonstrates
upper critical solution temperature (UCST) type phase behavior, which is unusual for high
molecular weight binary polymer blends.
100k.m
(c)
(b)
(a)
Fig.1 Optical micrographs of the morphology of the (20/80) 13.7ZnSPS/PC blend at different
temperatures: (a) 200°C; (b) 225 0 C; (c) cooled from 225°C to 200'C and annealed for 20
minutes.
1800
Further evidence of the UCST
behavior was obtained from light scattering,
as shown in Fig.2. Below the cloud point,
the film was cloudy and the scattering
intensity was nearly constant as the
temperature was increased. Above the cloud
point temperature, the film became
transparent and the backscattered light
intensity decreased sharply. The cloud point
temperature, Tcp, was calculated from the
intercept of the baseline below the cloud
point and the linear section of the transient
in the scattered intensity above the cloud
point. The Tcp data are plotted in Fig 3. In
8.7ZnSPS/PC 20Y80
1600
1O.3ZrnSPS/PC 2CY80
1400
U,
1200
800
w
180
190
200
210
220
230
T(°C)
Fig. 2. Backscatterd light scattering intensity of
ZnSPS/PC blends as a function of temperature.
15
Fig. 3 (a), the molecular weight of the ionomer used in the three blends was held constant, but
the sulfonation level was varied from 8.7 to 13.7 mol%. Although the SPS molecular weight
used for the blends in Fig. 3(a) was relatively low, M. = 4000 g/mole, blends of PC and the
starting PS were completely immiscible over a comparable temperature range. All the blends
exhibited UCST-type phase behavior, and the cloud point temperature decreased with increasing
sulfonation. This observation is in stark contrast with the effect of strong intermolecular
interactions on the phase behavior of blends, such as SPS/polyamide blends, which exhibit
1 1
LCST-type phase behavior and a cloud point that increases with increasing sulfonation level.' , 2
The critical point for the blends in Fig. 3(a) is skewed towards the higher molecular weight
component, i.e., PC, which is what is usually expected. The critical composition, ca. 60 wt% PC,
was relatively insensitive to the sulfonation level.
Id.
. I -i.. I -- 1'0
220
0 00
210 -
0
0.
0
0
0
00
290(b0
b
o
fl
009
A
A
0
A
270
200
0
.-
190
250]
A
180
•
0
I
20
,
I
40
,
I
60
,
I
80
240
•,
100
Vx
I
0
20
,
I
40
,
I
60
•
•
80
100
PC (wt %)
PC (wt%)
Fig.3 Tcp vs. composition for blends of ZnSPS and PC. (a) Effect of sulfonation: (o) 8.7ZnSPS, (J)
10.3ZnSPS, (A) 13.7ZnSPS. (b) effect of ionomer molecular weight: (O)Mw =280 000, (V)Mw = 135
000.
Blends of ZnSPS/PC based on a higher molecular weight ionomer than was used to
construct the phase diagrams in Fig. 3(a) showed similar UCST-phase behavior. Fig. 3(b) shows
the effect of the SPS molecular weight on the phase behavior of ZnSPS/PC blends. The
sulfonation level of the PS was held constant at ca. 5.9 mol%, while the molecular weight of the
ionomer was varied from M,- 135,000 g/mole to 280,000 g/mole. As is normally expected for
polymer blends with UCST-type phase behavior, increasing the molecular weight of either
component reduces the miscibility, and TcP increased when the molecular weight of the SPS was
doubled. A comparison of Figs. 3(a) and (b) indicates that the critical composition shifted from
PC-rich to ionomer-rich, i.e., from 60% PC to 40% PC, when the ionomer molecular weight was
increased from Mw = 4000 to 135,000 g/mole, which again is the expected result for a polymer
blend when the higher molecular weight component switches from one polymer to the other.
The TO obtained after annealing at a temperature within the single phase region (see Fig.
3) are plotted against composition in Fig. 4 for the blends based on the 10.3 mol% ionomer (M,
= 4000 g/mole and 5.88 mol% ionomer (M, = 135,000). The T9 vs. composition curve for these
blends, as well as for the other miscible blends investigated in this study, showed negative
deviation from a simple, weighted average of the two component polymer Tgs. The GordonTaylor equation"3 , which is indicative of very weak or no specific intermolecular interactions
between the component polymers, i.e., X - 0, fit the data in Fig. 4 very well. This indicates that
any attractive interaction between ZnSPS and PC is weak at best, and certainly much weaker
than the intermolecular interactions that occur in ZnSPS/polyamide blends that exhibit positive
deviation of T. from a linear weighted average of the component Tgs.
16
Xie et al.' 4 observed that although
weak interactions between PS and
poly(vinyl methyl ether) occurred in a
miscible blend, no interaction was detected
when the blend was phase-separated. To
ascertain whether a similar result might be
obtained for the Zn-SPS/PC blends, we
measured the
FTIR spectra of a
(50/50)10.3ZnSPS/PC blend at 200'C,
235°C and 250'C. The lower temperature
was below Tcp and the two higher
temperatures were above Tcp. Fig.5 shows
the FTIR spectra at those temperatures for
the carbonyl and sulfonate stretching
regions. The spectra were recorded after the
film was annealed at the indicated
temperatures for at least 30 minutes, which
170
160
150
140
05 130
120(
110
100(
90
0
20
40
60
80
100
PC (wt%)
Fig.4 Tg vs. composition for (0) PC/l0.3ZnSPS
(M, = 4000 g/mole) and (U) 5.88 ZnSPS/PC (M, = 135,000) . The solid lines
are fits of the Gordon-Taylor equation.
was long enough to reach its equilibrium morphology according to the DSC data (After a two-Tg
sample annealed at one-phase region for 30 minutes, it just showed one Tg). Although there is
some broadening of the absorbance peaks at the higher temperatures, the maximum in the
carbonate carbonyl absorption remains constant over the temperature range studied, Fig. 5(a).
Similarly, the absorption peak for the symmetric S-0 stretching vibration of the sulfonate anion
at 1044 cm- remains constant with increasing temperature, even though there appears to be some
changes to a lower frequency neighboring vibration at 1034 cm-', which is due to the S-0 stretch
of an isolated SO3.However, this change, together with the intensity changes of the band at 1130
cm' due to the in-plane skeleton vibration of the disubstituted benzene ring, and those of the
weak absorbances between 850 - 880 cm', was found to have nothing to do with the addition of
PC."5 The constancy of the frequency of the carbonyl and S-0 stretching vibration from room
temperature to 250'C indicates that little change in the environment of both groups occurred over
that temperature interval, which represented two-phase and single-phase blends. That result
further supports the conclusion that no specific interaction occurred in the ZnSPS/PC system.
(a)
(b)
\
T("C)
250
235
200
1900
1850
1800
1750
1700
1650
1150 1100 1050 1000 950 900 850
wavenurrter(crrr 1)
wavenurrter(arr')
800
Fig. 5 FT-IR spectra for a (50/50) 10.3ZnSPS/PC blend in (a) the carbonyl stretching region and (b) the
sulfonate stretching region.
17
CONCLUSION
Blends of polycarbonate and the zinc salt of light sulfonated polystyrene ionomers are
partially miscible and exhibit UCST-type phase behavior. Tcp decreased with increasing
sulfonation of the ionomer over the range of 8.7-13.7mol%, and for a constant sulfonation level,
Tcp increased with increasing molecular weight of the ionomer. The Tg-composition behavior of
miscible ZnSPS/PC blends exhibited negative deviation from a linear weighted average of the
component polymer TgS, which is consistent with a miscible blend with weak or no interpolymer
interactions. FTIR analyses also demonstrated that no specific interaction occurred involving
either the carbonate carbonyl group or the metal sulfonate group. These results are consistent
with previously published results for blends of PC with SPS ionomers 6, and they strongly suggest
that the miscibility between these two polymers arises from intrapolymer repulsive interactions
between the ionic and non-ionic species within the ionomer.
ACKNOWLEDGEMENT
We gratefully acknowledge support of this research by the Polymer Program of the
National Science Foundation (Grant DMR 9400862).
REFERENCES
I
E. M. Pearce, T. K. Kwei, T.K.; B. Y. Min, J. Macromot. Sci. Chem. 1984, A21, 1181.
2
3
A. Eisenberg, M. Hara, Polym. Eng. Sci. 1984, 24, 1306
Z. L. Zhow, A. Eisenberg, J. Polym. Sci. Polym. Phys. Ed. 1983, 21, 223.
4.
X. Y. Lu, R. A. Weiss, Macromolecules 1992, 25, 6185
5
X. Y. Lu, R. A. Weiss, Macromolecules 1996, 29, 1216
6
R. P. Kambour, J. T. Bendler, R. C. Bopp, Macromolecules 1983, 16, 753.
7.
8.
9.
10.
11.
12
13.
14.
15.
D. R. Paul, J. W. Barlow, Polymer 1984, 25, 487.
G. Ten Brinke, G.; F. E. Karasz, W. J. Macknight, Macromolecules 1987,28,957
H. S. Makowski, R. D. Lunderg, G. H. Singhal,. US. Patent 3,870,841, 1975
D. J. Moonay, R. J. Wu, M. T. Shaw, Proc. Soc. Plastics.Eng., ANTEC, 1994, 2038.
R. A. Weiss, X. Y. Lu, Polymer 1994, 35, 1963.
Y. Feng, R. A. Weiss; Macromolecules, 1996, 29, 3925.
M. Gordon, J. S. Taylor, J. Appl. Chem. 1952, 2, 495.
R. Xie, B. X Yang and B. Z. Jiang, B.Z. J Polym. Sci. Polym. Phys. 1995, 33, 25.
R. Xie, R. A. Weiss, to be published results.
18