chapter 3 effect of annealing on the reaction induced miscibility …shodhganga.inflibnet.ac.in ›...
TRANSCRIPT
Chapter 3
Effect of Annealing on the Reaction Induced Miscibility and Phase Behaviour
In this chapter, the effect of annealing on the reaction induced miscibility and
phase behavior of Sorona® {poly (trimethylene terephthalate), PTT} and
bisphenol-A polycarbonate (PC) blends was discussed. The unannealed
PTT/PC blends exhibited heterogeneous phase-separated morphology and two
well-spaced glass transition temperatures indicating immiscibility. The PTT/PC
blends were thermally annealed at 260 °C for different times to induce various
extents of transreactions between the two polymers. After annealing at high
temperature the original two Tgs of blends were found to merge into one single
Tg, exhibiting a homogeneous morphology. It is interesting to note that upon
extended annealing the original semicrystalline morphology transformed into an
amorphous nature. This is attributed to chemical transreactions between the
PTT and PC chain segments as evidenced with , SEM, FTIR, DSC, DMA, 1H
NMR, WAXD, PVT and Rheology measurements. Thermal stability of the blends
was also analyzed. A random copolymer formed as a result of the transreactions
between PTT and PC, serves as a compatibiliser at the beginning, and upon
extended annealing this became the main species of the system which is finally
transformed to a homogeneous and amorphous phase.
A part of the results of this chapter has been published in Journal of Physical Chemistry - B
104 Chapter 3
3.1. Phase morphology analysis of PTT/PC unannealed blends.
3.1.1. Introduction
Because of their potential to exhibit tailor-made properties, polymer blends
continue to attract much attention in academia and industry. Nowadays polymer
blending is a versatile and widely used method for optimizing the cost-
performance balance and increasing the range of potential application. The
properties and performance of polymer blends are critically dependent on blend
morphology. Morphology control plays a key role in optimising the performance
of multi component polymer blends. Morphology development is the path of
morphological change, in which the material undergoes its transformation from
large to small domains. The evolution of blend morphology from pellet or
powder sized particles to the sub micrometer droplets depends on several
processing parameters including the rheology, interfacial properties and
composition of the blend [1-6]. The competing processes of drop break-up and
coalescence during processing of polymer blends determine the final
morphology of these mixtures as explained in a growing body of literature on this
subject [6-15]. The interface has a crucial role in controlling the morphology and
final properties of an immiscible polymer blend. The interfacial tension is the
most basic parameter, which characterises the interface between polymers [16-
18]. Owing to the high molecular weights of the component polymers and
negligible combinatorial entropy during mixing, most of the blends are
characterised by coarse, unstable morphology and poor interfacial adhesion
between the phases. Hence the major challenge in blending involves the
manipulation of blend morphology via judicious control of mixing parameters and
the interfacial interactions.
The fundamental reasons responsible for the unstable morphology are the
unfavourable interactions at the interface between the components which create
a high interfacial energy and low interfacial thickness, which would, in turn lead
to poor interfacial adhesion between the phases that may result in premature
failure of the interface upon stress transfer. Another aspect that deserves
Miscibility and phase behaviour of PTT/PC blends 105
attention is the coalescence of the dispersed phase, which makes the dispersed
particles larger and non-uniform, leading to an unstable morphology.
As discussed earlier, development and stability of the morphology of multiphase
polymer melts is a complex function of blend composition, interfacial
characteristics, rheological properties and shear conditions. In the early 1930s,
Taylor developed a theory for the break-up of individual droplets for Newtonian
fluids [19, 20]. A relationship was established between the capillary number, Ca,
a ratio of shear to interfacial forces and the viscosity ratio ηr = ηd/ηm
ma
γη DC
2Γ= [3.1]
Where γ is the shear rate, D is the diameter of the droplet, Γ is the interfacial
tension, ηd is the dispersed phase viscosity and ηm is the matrix phase viscosity.
The predicted drop size for a simple shear field is proportional to the interfacial
tension and inversely proportional to shear rate and matrix phase viscosity. If Ca
is small, the interfacial forces dominate and a steady drop shape develops.
When Ca exceeds a critical value, Ca crit the drop deforms and subsequently
breaks down under the influence of interfacial tension. According to Tokita [5] when
coalescence and break down balance, the equilibrium particle size (de) can be
expressed as,
de ≅ 24PrΓ/ πτ12 {φd + [4PrEdk/πτ12] φd2 } [3.2]
where τ12 is the shear stress, Γ is the interfacial tension, Edk is bulk breaking
energy, φd is the volume fraction of the dispersed phase and Pr the probability for
a collision to result in coalescence. Tokita’s expression incorporates the
composition variable and predicts that particle size at equilibrium diminishes as
the magnitude of the stress field increases between the component phases and
volume fraction of the dispersed phase result in an enhancement of particle size.
Callan et al. [21] extensively studied the dependence of morphology on
composition of the blends. Danesi and Porter [22] showed that under same
processing conditions, the blend ratio and melt viscosity differences of the
106 Chapter 3
components determine the morphology. When the components have similar melt
viscosities the resultant morphology shows a distribution of minor component in
the major one. When the components have different melt viscosities the
morphology of the resultant blends depends on whether the minor component
has a lower or higher viscosity than the major one. The minor component will be
finely dispersed, when it has got lower viscosity. The minor component will be
dispersed as spherical domains if its viscosity is higher than the major
component. Recently different studies were reported on morphology of polymer
blends [23-46]. Among the blends studied, aromatic polyesters represent a
major class of engineering plastics having excellent properties with large variety
of applications and the possible transesterification reactions that they can
possibly undergo in certain environments. As a consequence it can strongly
enhance the applications of homopolyesters. Studies on various aspects for
blends of PET and PBT [47–50] and of PET and PTT [51, 52] are available in
literature. Recently more research publications are coming in the field of
polyester blends especially for PET [36, 53-55], PBT [56-59] and PTT [60-65].
There is a growing urgency to develop biobased materials as replacements/
substitutes of fossil fuel based materials. A new aromatic polyester,
poly(trimethylene terephthalate) (PTT), has been commercialized by DuPont
under the trade name Sorona® which is prepared by the melt condensation
polymerization of 1,3-propanediol (derived from renewable corn sugar) with
either terephthalic acid or dimethyl terephthalate. Its mechanical properties are
comparable to those of PET and PBT and its crystal structure and thermal
properties have been studied and some studies on PTT-based blends were
conducted [66-72]. Poly (ether imide) (PEI) and poly (ethylene-co-cyclohexane
1, 4-dimethanol terephthalate) (PETG) are miscible counterparts reported for
PTT [73-75]. The blends of PTT with immiscible counterparts such as
polystyrene (PS) and polyamide-12 (PA12) were investigated [76-77].
It is known that annealing at high temperature can have a thermodynamic effect
on phase behavior of blends due to the variation of free energy of mixing. For
polyesters blends and in some polyamides temperature effects on phase
Miscibility and phase behaviour of PTT/PC blends 107
behavior can be more complex than just variation of free energy. Transreactions
can induce variation in chemical structures of the polymer constituents in blends.
By the careful control of the extent of the interchange (transesterification)
reactions, miscible blends as well as tailored block and/or random copolyesters
can be produced with desirable properties. Several examples of such
transreactions in polyester blends have been reported, including, e.g., poly
(trimethylene terephthalte) (PTT) [78-79], poly(butylene terephthalate)
(PBT)[80,81], poly(ethylene terephthalate) (PET), polycarbonate (PC), polyarylate
(PAR) and poly(ethylene 2,6-naphthalate) (PEN)[82, 91]. Devaux et al. [92] have
pointed out that transesterification could take place in the temperature range used
for melt-blending, and the observed Tg changes could be accounted for by
copolymer formation rather than purely thermodynamic modifications.
Solution-cast blends of PTT with bisphenol-A polycarbonate (PC) were studied
[93] recently which is found to be inherently immiscible and after annealing at
260 °C, they become miscible due to the transesterification reaction. According
to Yavari et al. [94] PTT/PC blends are partially miscible and after annealing at
300 °C for 10 min the blends changed to a miscible state through a
transesterification reaction. From these investigations, it can be concluded that
transesterification plays an important role in controlling the properties of PTT/PC
blends. Therefore, the effect of annealing on the transreactions and various
properties of PTT/PC blends are of paramount importance and should be
investigated.
In the present study, melt-mixed PTT/PC blends with different compositions
were prepared through melt mixing technique and the effect of annealing on the
extents of transreactions and the apparent changes in miscibility, phase
morphology and thermal properties of the blends were evaluated. For this
purpose, the PTT/PC blends were annealed at 260 °C for different times (from 0
to180 min) to induce various extents of transreactions between the two
polymers. All the annealing experiments were done inside the vacuum oven at
260oC in which the sample is placed between two parallel metallic plates which
is also in the same temperature of the oven. Therefore the sample attains the
108 Chapter 3
annealing temperature very fast from the surface of the metal plates. Also the
measurements started only after giving an incubation period of ~ 1 min for each
sample to attain the temperature.
This part of the chapter is devoted to the investigations on phase morphology of
unannealed PTT/PC blends. The effect of blend composition on the phase
morphology development in the unannealed blends has been analysed.
3.1.2. Results and discussion
The samples for the morphology measurements were prepared by cryogenically
fracturing the samples in liquid nitrogen. Dispersed PC phase is preferentially
extracted from the blend using methylene chloride. The size of the dispersed
phase was analysed by image analysis technique. About 300 particles were
considered for the diameter measurements. The number average (Dn) and
weight average diameters (Dw), polydispersity index (pdi), interfacial area per
unit volume (Ai) were determined using the following equations;
The number average diameter:
i in
i
N DDN
Σ=
Σ [3.3]
The weight average diameter:
2i i
wi i
N DDN D
Σ=
Σ [3.4]
Poly dispersity index:
w
n
Dpdi D= [3.5]
Interfacial area per unit volume:
3iA R
φ= [3.6]
Interparticle distance
Miscibility and phase behaviour of PTT/PC blends 109
1 3
16
IPD D πφ
= −
[3.7]
where φ is the volume fraction and R the average radius of the dispersed
particles.
It should be noted that the viscosity difference between polymers has significant
impact on the phase morphology of the blends. If the minor component has
lower viscosity compared to the major one, it will be finely and uniformly
dispersed in the major continuous phase owing to the diffusional restrictions
imposed by the matrix [3] and otherwise coarsely dispersed. It is believed that
viscosity ratio should be approximately unity when designing the polymer blends
for superior properties. Wu’s equation (Equation 3.8) suggests that minimum
particle size is achieved when the viscosities of the two phases are closely matched
and as the viscosity moves away form unity in either direction, the dispersed
particles become larger [95].
0.844
m
D λγη
±Γ=
& [3.8]
where D is the droplet diameter, mη is the viscosity of the matrix, λ is the
viscosity ratio of the droplet phase to the matrix, γׂ is the shear rate and Γ is
the interfacial tension.
We observed a similar result from the SEM micrographs of the PTT/PC
unannealed blends presented in Fig. 3.1 These SEM micrographs demonstrate
the phase morphology of cryogenically fractured surfaces of PTT/PC blends
which clearly disclose two-phase morphology. One can distinguish two types of
morphology from the figure: (a) dispersed droplet type morphology in blends up
to 30 wt% of PC (where PC forms the dispersed phase) and up to 40 wt% of
PTT (where PTT forms the dispersed phase) and (b) co-continuous phase
structure in blends with 40 and 50 wt% of PC. From the dispersed droplet type
morphology average domain diameter, the polydispersity index/distribution of
110 Chapter 3
dispersed particles, interparticle distance interfacial area/unit volume etc. can
be calculated.
PTT90 PTT80
PTT30 PTT50
PTT40 PTT20
Figure 3.1: Scanning electron micrographs of unannealed PTT/PC blends
Miscibility and phase behaviour of PTT/PC blends 111
The average domain diameters (_
nD and_
wD ) of the dispersed particles of
unannealed PTT/PC blends as a function of blend ratio is presented in Fig. 3. 2.
From the figure, one can see that, as the weight % of PC in PTT matrix
increases, particle size increases and beyond 60 wt% both PTT and PC form bi-
continuous phase structure at which a phase inversion occurs and after this
point (60 wt%), PC forms the matrix in which PTT phase is distributed as
dispersed particles. This is a typical morphology of an incompatible binary blend.
The difference in particle size of dispersed PC and PTT phases for a given
dispersed phase concentration (eg. 90/10 and 10/90) can be explained by
considering the relative difference in their viscosities in the blend (see Fig. 3.3).
It should be noted that the less viscous component (PTT) forms finely dispersed
particles in more viscous matrix (PC) due to comparatively restricted diffusion
effects on coalescence of particles and increased shear stress resulting from the
more viscous matrix phase. The fundamental reasons responsible for the
unstable morphology are the unfavourable interactions at the interface between
the components which create a high interfacial energy and low interfacial
thickness, which would, in turn lead to poor interfacial adhesion between the
phases that may result in premature failure of the interface upon stress transfer.
Another aspect that deserves attention is the coalescence of the dispersed
phase, which makes the dispersed particles larger and non-uniform, leading to
an unstable morphology.
112 Chapter 3
0 20 40 60 80 1000.0
0.5
1.0
1.5
2.0
2.5
3.0 Dn Dw
Dom
ain
Dia
met
ers(
µ m
)
Weight percentage of PC
Cocontinuous
Region
Figure 3.2: Effect of blend ratio on the average domain diameter of unannealed PTT/PC Blends
0.1 1 10 100
102
103 PTT PC
Com
plex
vis
coci
ty η
∗ (P
as)
Frequency (rad/sec)
Figure. 3. 3: Complex viscosities of PTT and PC as a function of frequency at 260°C
Miscibility and phase behaviour of PTT/PC blends 113
The development of continuity as described by percolation theory can be
summarised as follows: Initially at low concentrations, there is a dispersion of
particles in the matrix. As the concentration of the minor phase increases,
particles become close enough to behave as if they were connected. Further
addition of minor phase material extends the continuity of network until the minor
phase is continuous throughout the sample [96-98].
The continuity of the dispersed phase is calculated by solvent dissolution
method. When PTT forms the matrix, the minor phase PC was extracted using
dichloromethane solvent. The continuity of the component is defined as the ratio
of the difference of the weight of the component present initially and the
calculated weight of the residual component after extraction to the weight of the
component present initially.
Continuity of A (%) = component) the of weight (Initial A)component of fraction (Wt.
)extraction after(Weight component) the of weight (Initial×
× [3.9]
The results are summarized in Fig. 3.4. When the percentage continuity of both
the components equals 100 %, the morphology of the blend is considered to be
cocontinuous. From the Fig.3.4 it is evident that the continuity of the PC phase is
close to 90% in PTT40 and above 90 % in PTT50 blends. This suggests that
PTT40 and PTT50 exhibit co-continuous morphology. For the other three blend
compositions (PTT90, PTT80 and PTT70) the continuity is less than 65 %,
suggesting matrix/droplet morphology. The sample after extraction didn’t break
down (disintegrate) between 0-50 wt % of PC, and this indicates that the PTT
phase is continuous in that range.
114 Chapter 3
0 20 40 60 80 1000
20
40
60
80
100
% c
ontin
uity
of P
C p
hase
Weight percentage of PC Figure. 3.4: Effect of blend ratio on cocontinuity of unannealed blends.
Here the morphological parameters showed that all blends are associated with
two-phase unstable morphology owing to the high interfacial tension and greater
coalescence effects in the absence of favourable interactions at interface
between the phases. As the concentration of one phase in the blends increases,
the incompatibility intensifies. This is a typical morphology of an incompatible
heterogeneous binary blend in which the less viscous component is more finely
dispersed in highly viscous matrix (PC) due to comparatively restricted diffusion
effects on coalescence of particles and increased shear stress resulting from the
more viscous matrix phase. This is evident from the polydispersity index values
shown in Fig.3.5. It is obvious from the figure that blends containing dispersed
PTT phase show the narrowest while dispersed PC show the broadest
distribution of particles. When the concentration of dispersed phase increases,
due to the enhanced unfavourable cross-correlations of the component polymers
at the interface (derived from the surface tensional forces along with the
coalescence process) the morphology become more coarse and unstable.
Further it can be observed that when the concentration of dispersed phase
increases, due to the enhanced unfavourable cross-correlations of the
component polymers at the interface between them (derived from the surface
tensional forces), the morphology becomes more coarse and unstable. It is
Miscibility and phase behaviour of PTT/PC blends 115
evident that blends with dispersed PTT phase possess more uniform
morphology compared to those with dispersed PC phase. This behaviour is due
to the fact that the relatively more viscous PC matrix suppresses the
coalescence of PTT phase which facilitates the formation of more uniform
dispersed morphology.
0 20 40 60 80 100
1.02
1.04
1.06
1.08
1.10
1.12
1.14
1.16
1.18
1.20
Poly
Dis
pers
ity In
dex
Weight percent of PC
Cocontinuous
Region
Figure 3.5: Effect of Blend ratio on the polydispersity index of PTT/PC blends
The effect of blend ratio on the domain distribution of dispersed phase in PTT/PC
blends is shown in Fig. 3.6. It can be seen that blends containing 10 wt% of minor
component (both PTT90 and PTT10) show the narrowest while PTT70 and PTT40 show
the broadest distributions of particles. The distribution of domains in all the other
blends remains in between these two limits. This is well expected and can be directly
related to the relative stability of phase structure.
116 Chapter 3
0 2 4
10
20
30 PTT90 PTT80 PTT70 PTT40 PTT30 PTT20
Dom
ain
dist
ribut
ion
(%)
Domain diameter (µm)
Figure 3.6: Effect of Blend ratio on the domain distribution of unannealed blends
Fig. 3.7 displays the effect of blend ratio on the interfacial area per unit volume
(Ai) and interparticle distance (IPD) of PTT/PC uncompatibilised blends. It is
evident from the figure that Ai diminishes with increasing concentration of
minor component in the blend. Blends with dispersed PTT phase possess
greater interfacial area compared to the corresponding blends with PC
dispersed phase. This is because Ai depends on the average domain size of
dispersed particles. On the other hand, on the basis of Ai values, one can
claim that blends with lower Ai exhibit maximum unfavourable interactions
(derived from maximum interfacial tension) at the interface and thus associated
with more coarse, non-uniform and unstable morphology. The higher value of
IPD indicates the tendency of a material to be failed brittley upon mechanical
loading. It is obvious from the figure that with increasing concentration of PTT
dispersed phase in the blend, IPD increases in all blends except for PTT40
suggesting that the blends are prone to brittle failure with increasing
concentration of PTT in PC component. In short, the morphological parameters
showed that all blends are associated with two-phase morphology owing to
greater coalescence effects in the absence of favourable interactions at interface
between the phases. As the concentration of one phase in the blends increases,
Miscibility and phase behaviour of PTT/PC blends 117
the incompatibility intensifies. It should be noted that the melt blended samples
were used directly for SEM analysis before annealing, and therefore there is no
possibility for transreactions to take place between the blend components.
0 20 40 60 80 1000.6
0.9
1.2
1.5
1.8 interfacial area
Weight % of PC
Inte
rfac
ial a
rea
per u
nit v
olum
e (µ
m-1)
co-continuous phase
0.2
0.3
0.4
0.5
0.6
0.7
0.8
IPD
Inter particle distance(µm)
Figure 3.7: Effect of Blend ratio on the Interfacial area /unit volume (Ai) and
interparticle distance of unannealed PTT/PC blends
3. 2. FTIR spectroscopy
Fourier transform infrared spectroscopy (FTIR) and high-resolution solid state
nuclear magnetic resonance (NMR) spectroscopy have been proven to be the
most powerful techniques for investigating the intermolecular specific
interactions and the phase behaviour of polymer blends. Miscibility between the
component polymers in the blends often perturbs the environment of their
molecular chains which causes the variation of intensities and/or shifts their
characteristic absorption in IR spectra, which promises IR spectra as a good tool
in determining miscibility in polymer blends [80-85, 93]. In FTIR approach, the
information on the intermolecular interactions in blends can be detected through
the variation of relative spectroscopic vibration bands.Blends of different
composition mixed in the Haake mixer, unannealed and annealed at 260°C for
different times (0-180 min) were analyzed and spectral regions of interest were
chosen, zoomed and evaluated.
118 Chapter 3
3.2.1. Unannealed blends
Figure 3.8 shows the complete mid-IR spectra of neat polymers, PTT (Sorona®)
and PC, and Fig. 3.9 shows the spectra of PTT/PC blends of different composition.
As expected, new bands indicating copolymer structures of transesterification
products were not found in the spectra of unannealed blends. But an influence of
both polymer components on phase behaviour of the blends could be detected
using special spectral data treatment: Since PTT itself is semicrystalline, carbonyl
groups exist in the well-ordered crystalline as well as in the disordered amorphous
phase. They absorb at slightly different wavenumbers due to the influence of the
supermolecular interactions on C=O stretching vibrations. In the crystalline phase
strong interactions lead to absorption at lower wavenumbers; in the amorphous
phase there are more “free” or less interacting carbonyls, which absorb at higher
wavenumbers. The polymer chains of the added amorphous PC can disturb
especially the ordered crystalline PTT in the blend; that means that some of
ordered polyester chains were transformed into more disordered ones which can
be followed by detailed analysis of the carbonyl spectral region.
1800 1600 1400 1200 1000 800 600
0.00
0.05
0.10
0.15
0.20
0.25
0.30 PTT Polycarbonate (PC)
ATR
Uni
ts
Wavenumber cm-1
1768
.6
1707
.2
1502
.7
1159
.011
87.4
1218
.812
44.8
Figure 3.8: The complete mid-FTIR spectra of neat PTT and PC.
Miscibility and phase behaviour of PTT/PC blends 119
1800 1600 1400 1200 1000 800 600
0.00
0.05
0.10
0.15
0.20 1 PTT/PC20/80 2 PTT/PC30/70 3 PTT/PC50/50 4 PTT/PC70/30 5 PTT/PC80/20 6 PTT/PC90/10
ATR
Uni
ts
Wavenumber cm-1
1
5
2
3
46
Figure 3.9: The complete mid-FTIR spectra of PTT/PC blends of different
composition. Evaluation of the normalized carbonyl region 1850 – 1600 cm-1 is represented in Fig.
3.10 and its second derivative in Fig. 3.11. Dependence of the position of PC
carbonyl (C=O) stretch on PTT content is shown in Fig. 3.11. It can be seen that
there is a shift from 1768 cm-1 (pure PC) to 1773 cm-1 (90% PTT), indicating an
increasing interruption of the PC-PC interaction of the PC chains in the blends due
to the added PTT chains (dilution effect), and as a result more and more “free” PC
carbonyls were formed. Also the position of PTT carbonyl (C=O) stretch shows a
shift in dependence on PC content from 1707 cm-1 (pure PTT) to 1712 cm-1 (80%
PC) indicating an increasing interruption of the PTT-PTT interaction due to the PC
chains. Detailed analysis of the structured asymmetric shape of the PTT carbonyl
stretching band (1745-1650 cm-1) to assign carbonyls in well-ordered (e.g. in
crystalline phase) and less ordered state (e.g. in amorphous phase or “free” carbonyl
groups) was done using the OPUS curve fit program by peak-fitting with Lorentz-
Gauss curves. As result three PTT individual band components could be separated:
(1) C=O stretch around 1674 cm-1 (ester groups with strong molecular
interaction in ordered regions of PTT rich blends)
(2) C=O stretch around 1710 cm-1 (ester groups with lower (medium) molecular
interaction in PC rich blends)
120 Chapter 3
(3) C=O stretch around 1725 cm-1 (ester groups with low molecular interaction in
unordered regions).
1850 1800 1750 1700 1650 1600
0.0
0.5
1.0
1.5
2.0
2.5 1 PC 2 PTT 3 PTT/PC 20/80 4 PTT/PC 30/70 5 PTT/PC 50/50 6 PTT/PC 70/30 7 PTT/PC 80/20 8 PTT/PC 90/10
ATR
Uni
ts
Wavenumber cm-1
2
13
5
4
678
Figure 3.10: Evaluation of the normalized carbonyl region (1850 – 1600 cm-1)
of PTT/PC blend.
1850 1800 1750 1700 1650 160
-0.03
-0.02
-0.01
0.00
0.01
Wavenumber cm-1
2
1
87
6
354
Figure 3.11: The second derivative of normalised carbonyl region which indicate
the dependence of PTT content on the position of PC carbonyl (C=O) stretch.
Miscibility and phase behaviour of PTT/PC blends 121
In order to understand whether the band shift in the carbonyl region is really the
only contribution of a change in crystallinity, we prepared a highly amorphous
sample by melting of a dry pure PTT sample (thin foil, heated for 30 sec. at
260°C, which means above Tm ~ 228°C) and quenching the melt in liquid
nitrogen. The FTIR- ATR spectrum of the quenched sample was reordered
immediately and compared with that of the semi crystalline PTT material as
shown in Fig. 3.12. Indeed, we found the C=O stretching vibration of the less
ordered /less interacting groups in the spectrum of the quenched, highly
amorphous PTT sample at higher wave numbers (only one band maximum at
1709 cm-1) in comparison to the initial semi crystalline PTT ( band maximum at
1706 cm-1 with a shoulder near 1674 cm-1). That means changes in crystallinity
or, more generally, changes in intermolecular interactions are responsible for the
band shift and change in band shape. i.e., the band shifts observed in the
spectra of unannealed blends are only ascribed to modification in the level of the
intermolecular interactions in dependence on composition, and it will be seen
that annealing yield more complex spectral effects.
1800 1760 1720 1680 1640 1600
0.0
0.5
1.0
1.5
2.0
2.51706.1 cm
-1
PTT amorphous PTT semicrystalline
ATR
uni
ts
Wave number cm-1
1709.2 cm-1
Figure 3.12: Evaluation of the normalized carbonyl region (1800 – 1600 cm-1)
of semi crystalline and highly amorphous PTT samples.
Miscibility and phase behaviour of PTT/PC blends 123
Annealing time of 5 min gives no detectable transesterification that means nospectral changes when compared to the spectra of unannealed blend samples. So, these spectra serve as initial ones for the evaluation of the subsequent
annealing steps. The most significant spectral changes can be seen at annealing
time of 3 hours. The complete mid-IR spectra of annealed (3 h) and unannealed
PTT/PC 70/30 and 50/50 blends are shown in Figs. 3.13 and 3.14.
4000 3000 2000 1500 1000
0.00
0.05
0.10
0.15
0.20 unannealed 3h annealed
ATR
Uni
ts
Wavenumber cm-1
1714
.417
72.1
1243
.6
1069
.0
PTT/PC 70/30
Figure 3.13: Complete mid-IR spectra of unannealed and 3 h annealed
PTT/PC 70/30 blends.
. 4000 3000 2000 1500 1000
0.00
0.05
0.10
0.15
0.20 unannealed 3h annealed
ATR
Uni
ts
Wavenumber cm-1
1769
.517
16.3
1069
.7
PTT/PC- 50/50
124 Chapter 3
Figure 3.14: Complete mid-IR spectra of unannealed and 3 h annealed PTT/PC 50/50 blends
The new band present at 1070 cm-1 (C-O-C stretching vibration) for the annealed
blends beside the PTT and PC bands indicates the formation of the fully aromatic
ester structure of the transesterification product, i.e. COO linked to two phenyl
groups on each side as shown below (see also PTB and BTB in Scheme 1). Figs.
3.15 and 3.16 show the formation of this band with increasing annealing time.
CO
O
At the same time the intensity of the PC band at 1080 cm–1 decrease because of
the consumption of aromatic carbonate groups. It seems that there is no
remarkable transesterification up to 60 min annealing time. Such an “induction
period” was also found in NMR analysis which is discussed in the later part of
this chapter (section 3.3).
1100 1090 1080 1070 1060 10500.02
0.04
0.06
0.08
0.10
0.12 6 PC 7 PTT 0 unannealed 1 15 min annealed 2 30 min annealed 3 1h annealed 4 2h annealed 5 3h annealed
ATR
Uni
ts
Wavenumber cm-1
PTT-PC 70/30
6
7
5
4
3
2
1
0
Figure 3.15: The formation of new band at ~1070 cm-1 (C-O-C stretching
vibration) with increasing annealing time for 70/30 blends indicates the presence of fully aromatic ester structure of the transesterification product.
Miscibility and phase behaviour of PTT/PC blends 125
1100 1090 1080 1070 1060 10500.02
0.04
0.06
0.08
0.10
0.12 6 PC 7 PTT 0 unannealed 1 15 min annealed 2 30 min annealed 3 1h annealed 4 2h annealed 5 3h annealed
ATR
Uni
ts
Wavenumber cm-1
PTT/PC 50/50
6
7
5
4
32 01
Figure 3.16: The formation of new band at ~1070 cm-1 (C-O-C stretching
vibration) with increasing annealing time for 50/50 blends indicates the presence of fully aromatic ester structure of the transesterification product.
Again, the important carbonyl spectral range was evaluated in more detail. For
that, difference spectra were calculated as follows:
Difference spectrum = (spectrum of blend sample annealed for 3h) – (spectrum
of blend sample annealed for 5 min).
In Figs. 3.17 (70/30 blend) and 3.18 (50/50 blend) these difference spectra are
shown together with the two blend spectra used for subtraction procedure. The
annealing effects (transreactions) are discussed for both blend compositions, but
they are more pronounced in the 70/30 blend.
There is a shift of the ester carbonyl band (C=O stretch) from 1709 cm-1 to 1714
cm-1 (in 70/30 blend) or 1712 cm-1 to 1716 cm-1 (in 50/50 blend) connected with
an intensity increase, the small band at 1674 cm-1 disappeared. That generates
a positive band in the difference spectra at 1718 cm-1 and negative ones at 1705
cm-1 and 1674 cm-1. These features demonstrate a strong increase of the
number of less ordered ester segments in the annealed sample (1718 cm-1)
126 Chapter 3
which is related to a corresponding dramatic decrease of ordered ester
segments (1705 cm-1 and 1674 cm-1).
1850 1800 1750 1700 1650 1600 1550
0.00
0.05
0.10
0.15
ATR
Uni
ts
Wavenumber cm-1
PTT-PC 70/30
3h annealed (2)
5 min annealed (1)
difference (2) - (1)
Figure 3.17: Difference spectra indicating the annealing effects on carbonyl
spectral range of PTT/PC 70/30 blend.
1850 1800 1750 1700 1650 1600 1550
0.00
0.05
0.10
ATR
Uni
ts
Wavenumber cm-1
3 h annealed (2)
5 min annealed (1)
difference (2) - (1)
PTT-PC 50/50
Figure 3.18: Difference spectra indicating the annealing effects on carbonyl spectral range of PTT/PC 50/50 blend.
Miscibility and phase behaviour of PTT/PC blends 127
A new high frequency shoulder at 1739 cm-1 band of the annealed samples is
better seen in the difference spectrum, which indicates the formation of the new
C=O stretch in the ester linkage of the fully aromatic ester structure of
transesterification product. The shoulder at 1760 cm-1 in the difference spectra is
coming from the C=O stretch of the carbonyl groups in the carbonate linkage of
aliphatic carbonate structures of transesterification product. Nevertheless, the
decrease of the carbonyl band at 1775 cm-1 of the carbonate unit of the PC
component in the 3 h annealed blends (gives negative band in the difference
spectra) demonstrates the consumption of initial PC aromatic carbonate groups
due to the transesterification.
3.3. 1H NMR analysis
Solid state NMR techniques are highly effective in the measurement of the
dimensional aspects of structural inhomogeneties down to molecular scale. Solid
state NMR also offers a promising tool for high-resolution measurements of
interface [92a-d]. Furthermore, because of its generality, it can be used to study
a much wider range of polymer materials [93]. Another significant advantage of
this method is that it doesn’t require any further sample modification. All these
methods are based on the principle of proton spin diffusion.
The progress of transesterification reactions can be well followed and quantified
by NMR spectroscopy. In principle, the exchange reactions in PTT/PC blends
are the same like in PBT/PC blends intensively studied by Devaux et al. [92a-d]
except that the aliphatic component is propylene instead of butylene. Starting
from PTT with the components propylene (P) and terephthalate (T) in a (PT)n
chain and PC with the components bisphenol-A (B) and carbonate (C) in a
(BC)m chain the exchange reactions will generate a four-component
polycondensate containing the components in a polymer structure with a certain
degree of randomness as shown in Scheme 2.
128 Chapter 3
O C
O
OC
O C
O
C
O
OT
C
CH3
CH3
B
P 3CH2
Scheme 2: Components in the exchange reaction between PTT and PC
This process results in characteristic changes in the 1H NMR spectra [94] (Figs.
3.19a and 3.19b ) because, e.g., terephthalate originally bonded to two
propylene units in a PTP triad appears after transreaction with PC under
exchange of one propylene unit by a bisphenol-A unit in a BTP triad and after a
second exchange in a BTB triad. These three triads can be well distinguished in
the 1H NMR spectrum (Fig. 3.19b) and the progress of transesterification is
reflected in their ratio. The second insert in Fig. 3.19b shows the signals of the
central methylene group of the P unit which can be located in TPT, CPT or CPC
triads after transesterification. Because of higher accuracy, the integral values of
terephthalate signals were used to describe the segmental sequence structures
of the copolyesters produced by transesterification applying the statistical model
developed by Devaux et al. [92a].
O
OCH2
O
OCH2CH2n
P T
1 2
3
C
CH3
CH3
OC
O
O
m
CB
5 64
a)
130 Chapter 3
120 72.5 25.0 2.5 85.0 15.0 0.33 6.7
180 58.5 36.5 5.0 77.0 23.0 0.51 4.3 c) 30.47 49.46 20.07 55.2 44.8 1.0 2.23
PTT/PC 70/30: FP = FT = 0.742 and FB = FC = 0.258
5 100 0.0 0.0 100 0.0 0.0 -
15 99.0 1.0 0.0 99.5 0.5 0.02 200
30 98.0 2.0 0.0 99.0 1.0 0.04 100
60 91.5 8.0 0.5 95.5 4.5 0.17 22.2
120 66.5 30.5 3.0 82.0 18.0 0.70 5.60
180 58.5 36.5 5.0 77.0 23.0 0.89 4.30 c) 55.06 38.28 6.66 74.2 25.8 1.0 3.88
a) estimated absolute errors: ∆f = ± 1 %; ∆B = ± 0.02 b) fPT = fPTP + 0.5 * fBTP; fBT = fBTB + 0.5 * fBTP c) theoretical values for the statistical four-component polyester
Table 3.1: Relative triads and diads contents f determined from the 1H NMR spectra of two PTT/PC blends after different annealing times at 260°C and calculated degrees of randomness B and number-average length of PT sequences LPT.
Comparing the compositions at 180 min annealing time with the calculated
composition of the corresponding random four-component condensates it is
obvious that the progress in transesterification is higher for the PTT/PC 70/30
blend due to the higher molar excess of PTT. This is in agreement with the IR
results (see Figs. 3.15-3.18). Devaux et al [92b] used a degree of randomness B
which is associated with the distribution of monomer units in the copolyester. A
value of B = 0 corresponds to the mixture of the two polycondensates whereas B
= 1 is characteristic of a random polymer. Here, values between both limits
indicate increasing transesterification. B can be calculated from the diad mole
fraction, e. g. FBT = fBT * FT, according to equation (3.10)
B = FBT / (FP * FB) [3.10]
Miscibility and phase behaviour of PTT/PC blends 131
with FP (= FT) and FB (= FC) are the mol fractions of propylene and bisphenol A
units (FP + FB = 1).
A further parameter, the number-average sequence length X, gives an impression
about the shortening of the initial homopolyester chains, (PT)n and (BC)m, by the
random transreactions with the second one. From the 1H NMR spectra, only the
shortening of (PT)n can be calculated according to Equation (3.11)
LPT = FPT / FBT + 1 [3.11]
Both B and LPT are given in Table 3.1 for the annealing series. Additionally,
Table 3.2 and Fig. 3.20 illustrate the time-dependence randomization.
PTT/PC 70/30 PTT/PC 50/50 Annealing Time (min) B F(B)ln
F(B)-f(BT)
B F(B)lnF(B)-f(BT)
0 0.00 0.00 0.00 0.00
15 0.02 0.02 0.01 0.01
30 0.04 0.04 0.02 0.02
60 0.17 0.19 0.04 0.05
120 0.70 1.20 0.33 0.41
180 0.89 2.22 0.51 0.72
(random) 1.00 1.00
Table 3. 2: The time-dependence randomization of 70/30 and 50/50 blends.
It is obvious that the transesterification starts after an induction period of about
30- 45 min. This becomes more clear from a plot of ln[FB/(FB - fBT)] vs. annealing
time (Fig. 3.20), which gives the apparent transesterification constant k2 as slope
[92d]. Again, an induction period appears for both blends followed by the
expected linear dependence which gives a higher k2 value for the PTT/PC 70/30
blend. Because k2 depends on the concentration of catalysts used in the
polyester synthesis [92d] it can be assumed that the higher PTT content also
causes the higher transesterification rate for the 70/30 blend.
132 Chapter 3
0 20 40 60 80 100 120 140 160 180 200
0.0
0.2
0.4
0.6
0.8
1.0
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0 B 70/30 B 50/50
Deg
ree
of ra
ndom
ness
, B
time (min)
In{ F(B)/F(B)-f(BT)} 70/30 In{ F(B)/F(B)-f(BT)} 50/50
In{ F
(B)/F
(B)-
f(B
T)}
Figure 3.20: The time-dependence randomization of 70/30 and 50/50 blends
3.4. Wide angle X-ray diffraction.
Wide angle X-ray diffraction (WAXD) has been widely used for evaluating
polymer crystallinity. It is proved to be more successful method for the determination
of some structural changes occurring as a result of blending [99]. The crystallinity
with respect to the crystallite size and perfection can be determined by wide
angle X-ray scattering (WAXD).
Miscibility and phase behaviour of PTT/PC blends 133
3.4.1. Unannealed Blends
In WAXD measurements, PTT/PC blends of different composition were analysed
with a scanning angle ranged from 2θ = 3° to 41°, with a step scanning of 2° for
1 min. Fig. 3.21 shows the WAXD patterns of PTT/PC blends. The characteristic
X-ray peaks for PTT were observed at the scattering angles 2θ of about 15.80,
17.50, 20.10, 22.10, 24.1, 25.20, and 27.40, corresponding to the reflection planes
of (0 1 0), (0 1 2), (0 1 2), (1 0 2 ), (1 0 2), (1 1 3), and (1 0 4 ), respectively
indicating that PTT has triclinic crystalline structure [99]. But PC gives only an
amorphous halo in the WAXD spectrum indicating that it is amorphous in nature.
It can be seen that the intensity of the crystalline diffraction peaks of PTT is
decreased with increase in PC content in the blends.
The amorphous halo of PTT was found out for crystallinity calculation. The
background was adapted considering the air scattering in the scattering region
around ~ 8° in 2theta. Therefore the relative ordering parameter αX ("crystallinity
index") calculated using equation αX = Icr / (Icr + Ia) based on peak-area method
(ratio of relevant crystalline scattering to total scattering, integral method in the
range 2Θ = 5°...36°) with applying an amorphous scattering curve. The results
obtained here are the overall crystallinity assuming Icr= Icr (PTT) and Ia=Ia (PTT)+Ia
(PC). The crystallinity values of the PTT/PC blends are shown in Table 3.3. It is
evident from the table that the crystallinity of blends decreases with increase in
PC content. But the crystalline structure of PTT is unaffected by the second
component in the blend.
134 Chapter 3
1 0 2 0 3 0 4 0
P T T P T T 9 0 P T T 8 0 P T T 5 0 P T T 4 0 P T T 2 0 P C
2 th e ta (d e g )
inte
nsity
/cps
Figure 3.21: WAXD patterns of PTT/PC blends
Sample PTT PTT/PC 90/10
PTT/PC 70/30
PTT/PC 50/50
PTT/PC 40/60
PTT/PC 20/80 PC
Crystallinity
Index (αX) 0.630 0.513 0.300 0.192 0.117 0.041 0
Table 3.3: The variation of crystallinity with blend ratio of PTT/PC blends
3.4.2. Annealed blends
The WAXD patterns for the blends PTT/PC 70/30 and 50/50 annealed at 260°C
for different times (0- 180 min) were shown in Fig. 3.22 and Fig. 3.23,
respectively. Both systems showed the semi crystalline behavior corresponding
to 3 sub-phases: crystalline PTT, amorphous PTT and amorphous PC,
respectively, due to the immiscibility of the blend partner. In the case of
unannealed 70/30 blends, the intensity of crystalline diffraction peaks is nearly
the same as that of the neat PTT. As the annealing time increased in steps from
5 min to 180 minutes the peak intensity decreases gradually, indicating reduction
Miscibility and phase behaviour of PTT/PC blends 135
in crystallinity of the PTT to different degrees. PTT/PC 70/30 blend annealed for
1 h shows the minimum PTT crystallinity and when it was annealed for 2 and 3
h, gives only an amorphous halo, indicating the complete absence of
crystallinity. Thus, WAXD data showed reduced peak intensities upon annealing
at 260 °C from 5 – 180 min. The calculated crystallinity data for 70/30 and 50/50
blends are shown in Table 3.4. As expected, at the beginning of the annealing
procedure, i.e. at 5 min, a small, but not negligible improvement of the crystalline
structure and/or increasing of the crystallinity αX was found, caused by healing
effects of the ordered phase.
5 10 15 20 25 30 35
S/PC 70/30 unannealed S/PC 70/30 annealed for 5 min S/PC 70/30 annealed for 15 min S/PC 70/30 annealed for 30 min S/PC 70/30 annealed for 1 h S/PC 70/30 annealed for 2 h S/PC 70/30 annealed for 3 h
Inte
nsity
(a.u
.)
2Theta (deg)
Figure 3.22: The WAXD patterns for the PTT/PC 70/30 blends annealed at 260°C for different times (0- 180 min)
WAXD analysis of PTT/PC 50/50 blends also showed the same behavior. PTT
diffraction peaks are clearly seen in the WAXD pattern of unannealed blends.
For annealed blends, as the annealing time increased in steps from 5 to 180
minutes, the PTT crystallinity is decreased to different degrees. The PTT/PC
50/50 blend annealed for 1 h shows the minimum PTT crystallinity and when it is
136 Chapter 3
annealed for 2 and 3 h gives only amorphous halo, which indicates the complete
disappearance of crystallinity.
5 10 15 20 25 30 35
S/PC 50/50 unannealed S/PC 50/50 annealed for 5 min S/PC 50/50 annealed for 15 min S/PC 50/50 annealed for 30 min S/PC 50/50 annealed for 1 h S/PC 50/50 annealed for 2 h S/PC 50/50 annealed for 3 h
Inte
nsity
(a.u
.)
2Theta (deg)
Figure 3.23: The WAXD patterns for the PTT/PC 50/50 blends annealed at 260°C for different times (0- 180 min)
Concerning the discussion of the induction period of the transesterification, the
randomization became more significant after annealing times in the range
above 30 min. In the same range a perceptible decrease of the crystallinity in
both blend systems are found. On the other hand, this degrease of crystallinity
corresponds very well with the shortening of the PT sequences connected
each other in a segment, which is notable to crystallise if the length fall below a
certain length.
Miscibility and phase behaviour of PTT/PC blends 137
∆αX ~ ≥ ± 0.015
Table 3.4: The variation of crystallinity with annealing time of PTT/PC 70/30 and 50/50 blends
3.5. Differential scanning calorimetric and morphological Analysis
Knowledge of non-isothermal crystallisation behaviour of polymer is necessary
for optimising its processing conditions for designing a product. It has been
reported that blending of polymers has significant impact on the crystallisation
properties of individual polymer. The degree of crystallinity is the one of the most
important parameters for characterising crystalline and semicrystalline polymers.
The incorporation of a second component to a crystallisable polymer may lead to
the following modifications in its crystallisation behaviour: (a) no effect on
crystallisation rate or morphology, (b) retardation of crystallisation with or without
change in morphology, (c) prevention of crystallisation at high loadings and (d)
acceleration of normally non-crystallising polymer as a result of induced mobility
[37, 93, 100,101]. The miscibility, melting and crystallisation behaviours of
polymer blends can be analysed by differential scanning calorimeter (DSC).
DSC analysis gives the heat flow rate associated with a thermal event as function of
time and temperature to obtain quantitative information about melting and phase
transition of polymeric materials
3.5.1. Unannealed blends
When a polymer crystallises in immiscible matrices such as in a polymer blend,
various crystallisation behaviours are possible depending on the component
polymers, their compositions, the interfacial adhesion, the processing
Annealing time(min) 0 min 5 min 15
min 30 min
60 min
120 min
180 min
PTT/PC 70/30
Crystallinity (αX) 0.272 0.300 0.274 0.267 0.15 0.0 0.0
PTT/PC 50/50
Crystallinity (αX)
0.188
0.192
0.180
0.144
0.05
0.0
0.0
138 Chapter 3
parameters, etc. The crystallisation behaviour of poly (trimethylene
terephthalate) in PTT/ PC (polycarbonate) blends is investigated. It is to be
noted that PTT is semicrystalline and PC is an amorphous polymer. The effect of
blend ratio on the melting and crystallisation parameters of PTT in PTT/PC
blends is depicted in Table 3.5. The crystallisation temperatures Tc and Tm of
PTT 170 °C and 228 °C, respectively. Figure 3.24 shows the DSC cooling scans
of PTT/PC blends at 10 oC/ min. When PTT/PC blends were cooled from the
melt the crystallisation exotherms of PTT were observed at ~ 170 –140 o C due
to the PTT phase crystallisations. Although the two polymers are immiscible, the
presence of one component appeared to influence the onset and peak
crystallisation temperatures of the other component depending on the blend
compositions. Changes in the Tc with blend composition showed that
crystallisation in the PTT phase was affected by the presence of the other
component, implying that there is some interaction between the components
which affect the crystallisation process. The differential scanning calorimetry
(DSC) results showed that the crystallisation behaviour of PTT/PC blends were
very sensitive to PC content. The onset (Tci) and the peak (Tc) crystallisation
temperatures shifted to lower temperatures whereas the area of the exotherm
decreased quickly as the PC content was increased. This suggests that the
crystallisation process of PTT was suppressed by the presence of PC. The
crystallisation temperature Tc of the PTT phase shifted from 170 to 141 oC on
adding 30 wt% PC and above which the crystallisation peaks disappeared.
When PTT content was greater than 50 wt.%, in addition to crystallisation
exotherms, the cooling curves exhibited the glass transitions of the PTT-rich phase
at ~56–75oC (arrow marked), which shifted to higher temperatures as PC content
was increased. However, when the PC content was greater than 70 wt. %, no
crystallisation exotherms were seen, but the glass transitions of the PC-rich
phase (arrow marked) were exhibited, which shifted to lower temperatures as
PTT content was increased. From the variations of the two glass transitions with
composition, it is concluded that the miscibility of PTT/PC blends is correlated
with blend composition. When the weight percent of PC is greater than 20 wt. %,
the crystallisation exotherms became very broad and indistinct. From 20 to 50
Miscibility and phase behaviour of PTT/PC blends 139
wt. % of PC contents, the broad crystallisation exotherms which appeared to end
at the glass transition temperatures of the PTT-rich phase. This suggests that
PC severely restrained the mobility of PTT molecules or segments, which led to
much longer and more varied relaxation times. As a result, the crystallisation
process takes place over a wider temperature range. Once the temperature
decreased to the glass transition temperature, the segments were frozen
instantaneously at various crystallisation stages. When PC content is greater
than 70 wt. %, the crystallisation of the PTT-rich phase appeared to be
completely restricted.
Tc (°C) Tm (°C) ∆Hc,n (J/g) ∆Hf,n (J/g) % Crystallinity Blends
PTT PTT PTT PTT PTT
PTT 170 228 52.8 53.9 37.02
PTT90 158 220 50.4 50.2 34.48
PTT80 142 216 44.5 45 30.94
PTT70 141 209 4 37.8 25.96
PTT50 215 - 35.8 24.58
PTT30 223 - 32 21.97
PTT20 227 - 25.5 17.5
Table 3.5: Crystallisation and melting behaviour of PTT/PC blends
140 Chapter 3
50 100 150 200 250
PCPTT 20PTT 30PTT 50
PTT 80PTT 70
PTT 90
PTT
0.1W/g
Nor
mal
ised
hea
t flo
w (
W/g
)en
do
Temperature (oC)
Figure 3.24: DSC cooling scans of PTT/PC blends
Figure 3.25 shows the second heating scans of PTT/PC blends. The melting
temperature(Tm) of PTT is also shifted from 228 to 209 oC as the PC content is
increased to 30 wt%. Above 30 wt% the Tm gradually increases. The shift in Tg of
PTT to higher temperature with increasing PC content is also clear from the
heating curves. The melting endothers also decreased with increase in PC
content. The Tg of PC is shifted to lower temperatures with increase in PTT. In
the heating curves (melting) the blends exhibited reorganization
(recrystallisation) exotherms as indicated by arrow before melting peak. These
observations also indicate the crystallisation and melting of PTT is affected by
the amount PC in the blends.
Miscibility and phase behaviour of PTT/PC blends 141
0 50 100 150 200 250
PCPTT20
PTT30PTT50
PTT70
PTT80
PTT90
PTT
PC
0.5 W/g
norm
aliz
ed h
eat f
low
(W/g
) e
ndo
Temperature (°C)
Figure 3.25: DSC second heating scans of PTT/PC blends.
Percentage crystallinity of PTT in the blend is obtained from the expression
( )0% 100f fcrystallinity H H= ∆ ∆ × [3.12]
where fH∆ is the enthalpy of fusion obtained calorimetrically and 0fH∆ is the
enthalpy of fusion of the 100 % crystalline PTT .
The percentage crystallinity values calculated using the equation 3.12 is
presented in Table 3.5. The percentage crystallinity values decreased with
increase in PC content which shows that the interaction between the
components decreased the crystallinity. Fig. 3.26 indicates the variation of
percentage crystallinity with PC content
142 Chapter 3
0 20 40 60 80 10010
20
30
40
Cry
stal
linity
(%)
Weight % of PC Figure 3.26: Effect of blend ratio on the percentage crystallinity of PTT in
PTT/PC blends.
3.5.2. Annealed blends: DSC and phase morphology
Figure 3.27 shows the DSC thermograms of the PTT/PC 70/30 blend annealed
at 260oC for increasingly longer times from 0 to 180 min. Two well-defined glass
transition temperatures (Tgs) can be seen in the DSC curves of the unannealed
blends and indicate that the system is immiscible. Upon annealing, the glass
transition temperatures of the amorphous PTT and PC rich phases shift to higher
and lower temperatures, respectively. After annealing at 260°C for more than 30
min, the original two Tgs merged in the blends to a single and sharp Tg. In
addition, the melting temperature (Tm) decreases with increase of the annealing
time imposed on the blends. Eventually, at extended annealing times (e.g., 120
min or longer), Tm of the blends disappear which indicate the transition from
semicrystalline to an amorphous state. Furthermore, the annealed blends
seemed to reach a final state where one glass transition was observed. This
behavior is due to the compatibilising effect of the copolyester formed as a result
of transesterification. Similar homogenization of the blend upon annealing at
Miscibility and phase behaviour of PTT/PC blends 143
high temperatures was reported elsewhere [93,100]. With the increase in
annealing time the heat of fusion and the peak temperature decreased [101].
The Tms of the blends eventually disappeared and the Tg stayed constant for the
samples annealed at 260 °C for 120 min or longer.
Figure 3.27: The DSC thermograms of the PTT/PC 70/30 blend annealed at 260 oC (0 - 180 min)
Fig. 3.28 shows the SEM micrographs of annealed and unannealed blends of
PTT/PC 70/30 and 50/50. These pictures show that the melt compounded
PTT/PC 70/30 blend exhibits phase separated domains, while the same blend
that had been heated for 120 and 180 min is apparently free from such phase
separated domains, which further indicates that the annealed blends readily
underwent a homogenization process. The 50/50 blend initially having co-
continuous phase morphology was also transformed to a homogeneous one
after extended annealing. Therefore it can be concluded that on progressively
longer annealing, the original phase-separated morphology eventually
disappear, and the morphology of the annealed PTT/PC blends turned
homogeneous.
Tg1Tg2
144 Chapter 3
(a) 70/30 unannealed (b) 50/50 unannealed
(c) 70/30 after annealed for 3 h (d) 50/50 after annealed for 3 h
Figure 3.28: SEM pictures of annealed and unannealed blends of PTT/PC 70/30 and 50/50. (a) 70/30 unannealed (b) 50/50 unannealed (c) 70/30 annealed for 3 h (d) 50/50 annealed for 3 h
It is well known that a physical state is readily reversible, but a chemically
changed state is irreversible. Figure 3.29 shows the second heating scans of the
PTT/PC 70/30 blends after first heat treatment. The SEM and DSC results
(Figs. 3.28 and 3.29) show that the morphology of the annealed PTT/PC blends
was homogeneous and is different from the original phase-separated
morphology of the as prepared blends and a single glass transition is apparent.
In other words, the changes upon annealing of the blends were irreversible (Fig.
3.29).
Miscibility and phase behaviour of PTT/PC blends 145
0 50 100 150 200 250
unannealed 5 min 15 min 30 min 60 min 120 min 180 min
0.1 W/g
norm
aliz
ed h
eat f
low
(W/g
)---
-> e
ndo
Temperature (°C)
Figure 3.29: The second heating scans of the PTT/PC 70/30 blends after first heat treatment.
The 70/30 blend exhibited significant decrease in the endothermic (crystalline
melting, Tm,) peak temperatures and the degree of crystallinity (as indicated by
the peak areas) upon annealing at 260°C. This behavior is attributed to an
increase in the degree of transesterification between PTT and PC, which
produce significant quantities of statistical and short random-block copolymers,
which inhibit crystallisation. Figure 3.30 shows the DSC cooling scans of the
70/30 blend.
146 Chapter 3
0 50 100 150 200 250
180 min 120 min 60 min 30 min 15 min 5 min virgin
0.05 W/g
norm
aliz
ed h
eat f
low
(W/g
)<-
--- e
xo
temperature (°C) Figure 3.30: The DSC cooling scans of the 70/30 blend.
From the cooling scans of unannealed blends we have seen that when the
weight percent of PC is greater than 20, the crystallisation exotherms became
very broad and indistinct. From 20 to 50 wt. % of PC contents, the broad
crystallisation exotherms which appeared to end at the glass transition
temperatures of the PTT-rich phase. This suggests that PC severely restrained
the mobility of PTT molecules or segments, which led to much longer and more
varied relaxation times. As a result, the crystallisation process takes place over a
wider temperature range. For annealed blends the crystallisation exotherms
decreases with increase in annealing time and the Tgs of individual components
which are shown initially come closer and finally a single Tg is observed which
indicate the occurrence of transreactions between PTT and PC.
3.6. Dynamic mechanical analysis
Dynamic mechanical thermal analysis (DMTA) is another powerful technique to
investigate the performance of polymer blends as it measures response of a
material to cyclic stress. The investigation of dynamic modulus and damping
behaviour over a wide range of temperatures and frequencies has proven to be
very useful in studying the structural features of polymer blends and the variation
of properties with respect to end use applications [102, 103]. These rely on
Miscibility and phase behaviour of PTT/PC blends 147
structure, crystallinity, extent of cross-linking etc., which in turn depends on the
phase morphology of the blends. The dynamic mechanical properties are
sensitive not only to different molecular motions but also to various transitions,
relaxation processes, structural heterogeneity and the morphology of multiphase
systems. Further, the dynamic mechanical properties of polymers give mirror
image of their molecular and morphological features.
3.6.1. Unannealed blends
The dynamic mechanical spectroscopy has become a classical method for the
determination of blend miscibility because the height and position of the mechanical
damping peaks are remarkably affected by miscibility, inter molecular interaction,
interface feature and morphology. The dynamic mechanical properties of the blends
are also affected by the composition with particular emphasis on the amount of the
minor phase composition. The dynamic mechanical properties such as storage
modulus (E’), loss modulus (E”) and the damping (tanδ) were evaluated from 30 oC
to 180 oC. The dynamic mechanical properties such as storage modulus, loss
modulus and tan δ of PTT/PC blends are presented in Figs. 3.31-3.33. Each blend
showed two separated glass transition relaxations corresponding to a PTT-rich
phase and a PC-rich phase, respectively [102, 103].
Figure 3.31 shows the variation of storage modulus as a function of temperature for
PTT and PC homoploymers and their blends. Polycarbonate has higher value of
storage modulus than PTT in all temperature ranges except at higher temperature
(above the Tg of PC) and the blends have values in between. As in the case of blend
components, the storage modulus of the PTT/PC blends also decreases with
increase in temperature. PTT shows a very sharp drop in storage modulus in the
temperature range from 45 to 85 oC and PC shows a sharp drop from 145 to 169 oC
as shown in the Fig. 3.31. For the blends a sharp drop in E’ is observed when the
temperature is increased from 500 C to 90 0 C due to the presence of PTT, followed
by another sharp drop in storage modulus in the temperature range 140 to 164 oC
due to the PC content. Since PC is amorphous polymer, it tends to decrease the
crystallinity of the blend system due to small interactions with PTT. Figure 3.32
shows the variation of loss modulus as a function of temperature for various blend
148 Chapter 3
compositions. These curves show two maxima corresponding to the glass
transitions temperature of PTT and PC.
60 120 180
0
500
1000 PTT PTT90 PTT80 PTT50 PTT40 PTT20 PC
Stor
age
Mod
ulus
( M
Pa )
Temperature (0C )
Figure 3.31: The variation of storage modulus of PTT, PC and their blends as a function of temperature.
40 60 80 100 120 140 160
0
50
100
150
200 PTT PTT90 PTT80 PTT50 PTT40 PTT20 PC
Los
s Mod
ulus
(MPa
)
Temperature ( 0C ) Figure 3.32: Effect of blend ratio on the variation of loss modulus as a function
of temperature
Miscibility and phase behaviour of PTT/PC blends 149
Fig. 3.33 shows the variation of tan δ as a function of temperature for PTT/PC
blends. Tan δ curve of PTT shows peak at ~ 70 0 C due to the α-transition
arising from the onset of segmental motion. This corresponds to the glass
transition temperature of PTT. Polycarbonate shows tanδ peak at ~ 160 0 C
which corresponds to its glass transition temperature. The reports say that
generally for an incompatible blend, the tanδ Vs temperature curve shows the
presence of two tanδ or damping peaks corresponding to the glass transition
temperatures of the component polymers [104-107]. For a highly compatible
blend the curves shows only a single peak in between the transition
temperatures of the component polymers [104], where as broadening of
transition peak occurs in the case of partially compatible system [107]. In the
case of compatible and partially compatible blends the Tgs are shifted to higher
or lower temperatures as a function of composition. The PTT/PC blends show
two transitions peaks corresponding to the glass transition temperature of PTT
and PC. On adding PC into PTT there is a slight shifting of tan δmax of PTT and
PC towards each other indicating partial miscibility due to the transreactions
taking place in the system due to the annealing effect caused by the reaction
conditions(sample preparation conditions (melt blending compression moulding,
etc.), even though the samples used are not annealed separately. This shift is
more pronounced in PTT90 and PTT80 blends where the PC content is low
thereby the transreactions rate is high. This is due to the fact that
transesterification reactions are more pronounced in blends with higher ester
content than with lesser ester content. The variation of Tg of PTT and PC
obtained from tan δ curve with blend composition is shown in Table 3.6.
150 Chapter 3
50 100 150
0.0
0.7
1.4
2.1
60 80 100 1200.00
0.05
0.10
0.15
PTT PTT90 PTT80 PTT50 PTT40 PTT20 PC
Tan
δ
Temperature (0C) Figure 3.33: Effect of blend ratio on the variation of tan δ as a function of
temperature
Blends Tg of PTT Tg of PC
PTT 70 -
PTT90 79 149
PTT80 80 150
PTT50 79 153
PTT40 80 156
PTT20 80 158
PC - 160
Table 3.6: variation of Tg of PTT and PC with blend composition
Miscibility and phase behaviour of PTT/PC blends 151
3.6.1.1. Theoretical modeling of Viscoelastic properties
Various composite models such as parallel model, series model, Corans and
Takayanagi model have been applied to predict the viscoelastic behaviour of
binary blends.
The highest upper bound parallel model is given by the rule of mixtures as
follows
1 1 2 2uE E Eφ φ= + [3.13]
This model is applicable to the materials in which the components are connected
parallel to one another so that the applied stress lengthens each component to
the same extent. In the lowest-lower bound series model, the blend components
are arranged in series (Reuss prediction) perpendicular to the direction of the
applied force. The modulus prediction is given by the inverse rule of mixtures as:
1 2
1 2
1
LE E Eφ φ
= + [3.14]
In these models Eu is any mechanical property of the blend in the upper bound
parallel model and EL the moduli of the blend in the series model. E1 and E2 are
the mechanical properties of components 1 and 2, respectively; φI and φ2 are their
corresponding volume fractions. For both these models, no morphology is
required, but strain or stress can be continuous across the interface, and
Poisson’s ratio is the same for both phases.
According to Coran's equation [108, 109]
( )U L LM f M M M= − + [3.15]
where f can vary between zero and unity. The value of f is given by
( )1nH Sf V nV= + [3.16]
where n contains the aspects of phase morphology, and VH and VS are the
volume fractions of the hard phase and soft phase respectively.
152 Chapter 3
Takayanagi proposed a series-parallel model [110, 111] in which, the concept of
percolation is introduced. It is a phenomenological model consisting of mixing rule
between two simple models involving connection in series (Reuss prediction) or in
parallel (Voigt prediction) of the components. According to this model,
[ ] 1211 )()1()1( −+−+−= EEEE φφλ [3.17]
E1 is the property of the matrix phase, E2 is the property of the dispersed phase,
and φ is the volume fraction of the dispersed phase and is related to the degree
of series-parallel coupling. The degree of parallel coupling of the model can be
expressed by
% parallel = [φ (1- λ) / (1- φ λ)] x 100 [3.18]
The effect of blend ratio on the experimental and theoretical storage moduli of
PTT/PC blends at a temperature, i.e. Tg of PTT (~ 70 0 C) is shown in Fig. 3.34.
0 20 40 60 80 100
400
600
800
1000
Experimental Parallel Series Coran's Takayanagi
Stor
age
Mod
ulus
(MPa
)
Weight % of PC Figure 3.34: Comparison of the experimental and theoretical data of storage
modulus of PTT/PC blends at the Tg of PTT
Miscibility and phase behaviour of PTT/PC blends 153
It is clear from the figure that both Coran and Parallel models are close to the
experimental data except at 70 wt %.of the PC. However, it is important to note
that the best fit curve is obtained in the case of Coran’s model since it contains
the aspects of phase morphology (‘n’ is a parameter related to the phase
morphology of the blends). The value of ‘n’ in the present case is low and can
be explained by taking into account of the fact that Coran’s model was originally
proposed for thermoplastic elastomers with a hard matrix and an elastomeric
minor phase. However, in the present case, the minor phase is not elastomeric
and there is no distinction between the natures of two phases, consequently ‘n’
values are very low. Further, it is seen that Takayanagi model shows deviation in all
the cases. It is mainly due to the fact that there is no big difference between the
storage moduli of PTT and PC. These observations may be due to the fact that
the blends are incompatible to certain extent so that the deterioration of
properties is shown as expected.
3.6.1.2. Determination of apparent weight fractions of components.
In PTT/PC blends two shifted glass transition regions are observed due to the
interaction between component phases and from the glass transition
temperatures of PTT and PC we can estimate the apparent weight fractions of
PTT and PC dissolved in the PC rich phase and PTT rich phases, respectively.
The apparent weight fractions of PC in the PTT rich phase and PC rich phase
were determined by the following empirical equation, which is often used to
describe the dependence of Tg on composition in random copolymers and
plasticized systems [112, 113].
1 1 2 2g g gT wT w T= + (3.19)
where Tg is the observed Tg of the copolymer w1 is the weight fraction of the
homo polymer 1 having Tg1 and w2 is the weight fraction of homo polymer 2
having Tg2. Equation 3.19 may be rearranged to
1, 2'1
1 2
g b g
g g
T Tw
T T−
=− (3.20)
154 Chapter 3
where w1’ is the apparent weight fraction of polymer 1 in the polymer 1 rich
phase, Tg1b is the observed Tg of polymer 1 in the blends, and Tg1 and Tg2 are
the Tgs of homo polymers 1 and 2, respectively [112].
Applying the DMA results of Tg (PTT) and Tg (PC) in PTT/PC blends we have
calculated the apparent weight fractions of PTT and PC in the PTT rich phase
and PC rich phase, which are shown in Table 3. 7.
PTT rich phase PC rich phase Wt. fraction of PTT Tg1 Tg2
w1’ w2’ w1” w2”
1 64 1.0000 0.0000 - -
0.9 68 145 0.9575 0.0425 0.1383 0.8617
0.8 70.5 146 0.9308 0.0692 0.1277 0.8723
0.5 73 147 0.9043 0.0957 0.1170 0.8830
0.4 77 148 0.8617 0.1382 0.1064 0.8936
0.2 79 150 0.8404 0.1596 0.0851 0.9149
0 158 - - 0.0000 1.0000
Table 3.7: Apparent weight fractions (w) of PTT and PC in the PTT rich phase and PC rich phase.
It can be seen from the table 3.7 that the apparent weight fractions of dissolved
PTT in PTT rich phase and in PC rich phase decreased with increase in PC
content. Similarly the apparent weight fractions of PC in PC rich phase and in
PTT rich phase also decreased with increase in PTT content. This shift in Tg
values and the corresponding decrease in apparent weight fractions are
attributed to the transreaction induced miscibility of PTT and PC under the
experimental conditions.
3.6.2. Annealed Blends
It is well known that miscible binary polymer blends exhibit a single Tg registered
between the Tgs of the neat components. If the polymers are immiscible, the two
Miscibility and phase behaviour of PTT/PC blends 155
Tgs (or the corresponding α relaxation processes) appear well separated [83].
Therefore, spectroscopic techniques sensitive to glass transition relaxations may
provide useful information concerning local concentration fluctuations and
miscibility in polymer blends [114]. PTT/PC blends (70/30 and 50/50) were
annealed at 260 °C for different times. Fig. 3.35 shows the DMA spectra of
PTT/PC 70/30 blends annealed at 260 °C from 0 to 120 min. It is evident that
there is a substantial difference in the tan δ maximum with annealing time. After
annealing at 260 oC for more than 15 min, the two distinct tan δ peaks present in
the unannealed blends (two Tgs) move closer to each other and when the
annealing time exceeds 60 min the two peaks are merged to a single tan δ peak
(Tg). The single tan δ peak at extended annealing times, e.g. 120 min or longer
indicates that the system is homogenous. It is proved that progressively longer
annealing, the original phase-separated domains eventually disappear, and the
morphology in the annealed PTT/PC 70/30 blends turned homogeneous.
40 60 80 100 120 140 160 180
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
unannealed 30 min annealed 2 h annealed
tan
δ
Temperature( o C)
Figure 3.35: DMA spectra showing the tan δ of PTT/PC 70/30 blends annealed at 260 °C from 0 to 120 min.
156 Chapter 3
PTT/PC 50/50 unannealed blends also show two distinct tan δ peaks (two Tgs)
as indicated in Fig. 3.36. Upon annealing at 260 °C for 15 min these tan δ peaks
come closer and when the annealing time is more than 120 min, they merged
into a single peak (Tg). The single tan δ peak at extended annealing times, e.g.
120 min or longer indicates that the system is homogenous. Therefore the
morphology of PTT/PC 50/50 blend also turned homogeneous upon
progressively longer annealing. This is also clear from the morphological
observations indicated in Fig.3.28 where the phase separated morphology
become homogenous on progressive annealing.
40 60 80 100 120 140 160 180
0.0
0.1
0.2
0.3
0.4
0.5 unannealed 15 min annealed 2 h annealed
tan
δ
Temperature ( o C) Figure 3.36: DMA spectra showing the tan δ of PTT/PC 50/50 blends annealed
at 260 °C from 0 to 120 min.
3.7. Pressure-Volume-Temperature (PVT) Measurements of Annealed and Unannealed blends
Pressure-Volume-Temperature (PVT) measurements provide the specific
volume of a material or density as a function of pressure and temperature [115-
117]. The specific volume of a material changes during reactions including
physical changes, phase changes, degradation reactions and the data is of
direct importance to engineering applications of materials such as
compressibility, bulk modulus, thermal expansivity etc. Generally, PVT data are
Miscibility and phase behaviour of PTT/PC blends 157
useful in the prediction of service life and performance of polymeric materials
based on the free volume concepts [118] and also in the prediction of miscibility
between polymers [119]. Chemical reactions, which are accompanied by volume
effects, can also be followed [120].
The permeation properties as well as the viscosity, viscoelasticity, the glass
transition, volume recovery and mechanical properties are related to the free
volume present in amorphous polymers. The mobility of small molecules in
polymers is closely connected to the hole-free volume. [121–124] This special
type of free volume appears in amorphous polymers in excess to the interstitial
free volume, which is observed in polymer crystals, due to their static or dynamic
disorder. Recently Fernández et. al [ 125] studied the relation between
Pressure–Volume–Temperature (PVT) and rheological behaviour of several
polymers including polypropylene, poly(methyl methacrylate), polyamide,
polycarbonate, polystyrene and polystyrene/polycarbonate blends. Pressure–
viscosity coefficient was calculated by means of a revisited
version of the Miller equation that accounts for pressure and temperature effect
on Newtonian viscosity through the activation energy of flow and PVT
parameters.
Sato et al [126] studied the PVT properties of polyethylene copolymer melts
Dependence of properties such as specific volume, thermal expansion
coefficient, isothermal compressibility, and characteristic parameter of equations
of state on the length of the polymer branched chains were examined. It was
found that the length of the branched chain did not affect the thermal expansion
coefficient and isothermal compressibility. The specific volume of copolymers
having longer branched chains were slightly larger than those copolymers with
short branched chains.
In this part the PVT measurements of neat PTT, neat PC and PTT/PC 70/30 and
50/50 blends were performed with varying annealing time. PTT is a
semicrystalline polymer while PC is amorphous. Figure 3.37 indicates the
variation of specific volume (Vsp) with temperature (T) of neat PTT and neat PC.
For PTT the first heat starts from a semicrystalline state and during heating cold
158 Chapter 3
crystallisation occurs at a temperature greater than 185 °C. At about 208 °C the
melting starts which is finished at 238°C. While cooling PTT shows crystallisation
between 202 °C and 177 °C. A soft glass transition of PTT is detectable in the 1st
and 2nd heating run between 50 °C and 70 °C. The second heat shows a similar
fast melting like in the first heat. The cold crystallisation behaviour is clearly
visible in second heating. Since PC is amorphous, it shows only one slope
change at its Tg region, between138 and 150 oC. There isn’t any melting or
crystallisation behaviour.
50 100 150 200 250 300
0.76
0.80
0.84
0.88
0.92
Tm
Tg
PTT 1st
Heating PTT cooling
PTT 2nd
Heating
PC 1st
Heating PC cooling
PTT 2nd
Heating
V sp(c
m3 /g
)
Temperature oC
Tg
Figure 3.37: PVT data showing the variation of Vsp with T of neat PTT and PC
Our goal is to follow the degree of the thermally initiated transesterification in
PTT/PC blends by changes in the density, glass transitions, and / or crystallinity.
As long as the two components are pure they are phase separated showing the
typical melting/ crystallisation behaviour and glass transitions like the pure
components. As soon as the transesterification starts, the crystallinity should
reduce resulting in changed densities and the formation of mixed phases should
form separate Tg’s or shifts in the Tg. It is assumed that the transesterification
occurs at 260 °C. Therefore, the mixtures were heated with a defined rate to this
temperature, annealed for a defined time, and then cooled down with the same
Miscibility and phase behaviour of PTT/PC blends 159
rate. Below ca. 100 °C the heat exchange is too small for keeping the cooling
rate and the cooling is reduced. This process was repeated with increasing hold
times at 260°C. The heating and cooling is necessary since at 260°C the
mixtures are completely amorphous and changes in the thermal behaviour
caused by transesterification can only be analysed during the heating/cooling
runs. Thus, we also have to consider that during this time the transesterification
continues, so that the real annealing time is underestimated when adding all the
times at which the samples are annealed at 260 °C. Furthermore all IBA runs are
done under a pressure of 10 MPa that shifts the transition temperatures to
slightly higher values compared to environmental pressure and that may also
influence the transesterification rate constants. When comparing the PVT data
with DSC and other methods we must also take into account that the methods
are sensitive to different physical responses caused by changed temperature
and therefore the Tg and Tm values determined by different methods differ
sometimes to more that 10K.
The first ITS runs are performed just to determine the accurate filling state of the
cell and to calibrate the run to the specific volume which was determined
separately by means of a Helium Pycnometer under standard conditions (1
atmosphere, 25 °C). The specific volumes calculated from the component values
agree rather well with the values of the mixtures determined by the Helium
Pycnometer which are shown in Table 3.8.
Vsp
PTT
(cm³/g)
Vsp
PC
(cm³/g)
Vsp (70/30)
experimental
(cm³/g)
Vsp (70/30)
calculated
(cm³/g)
Vsp (50/50) experimental
(cm³/g)
Vsp (50/50)
calculated
(cm³/g)
0.7626 0.8361 0.7748 0.7833 0.7980 0.7977
Table 3.8: specific volumes, Vsp (cm³/g) of the components and mixtures
The monitoring of the transreactions in PTT/PC 70/30 and 50/50 blends due to
annealing through PVT measurements was carried out and the variation of
specific volume with temperature was shown in Fig.3.38 and 3.39.
160 Chapter 3
0 50 100 150 200 250
0.78
0.80
0.82
0.84
0.86
0.88
0.90 heat 1 cool 1 heat 2; 5 min 260°C cool 2 heat 3; 15min 260°C cool 3 heat 4; 35min 260°C cool 4 heat 5, 65min 260° cool 5 heat 6; 125min 260°C cool 6
V sp (c
m3 /g
)
T (°C)
PTT/PC = 70/30
Figure 3.38: PVT data showing the variation of Vsp with T of PTT/PC 70/30
blends annealed at 260 °C from 0 to 120 min
0 50 100 150 200 250
0.78
0.80
0.82
0.84
0.86
0.88
0.90
0.92
heat 1 cool 1 heat 2; 5 min 260°C cool 2 heat 3; 15 min 260°C cool 3 heat 4; 35 min 260°C cool 4 heat 5; 65 min 260°C cool 5 heat 6; 125 min 269°C cool 6
Vsp
(cm
3 /g)
T (°C)
PTT/PC = 50/50
Figure 3.39: PVT data showing the variation of Vsp with T of PTT/PC 50/50
blends annealed at 260 °C from 0 to 120 min
Miscibility and phase behaviour of PTT/PC blends 161
From the Figs. 3.38 and 3.39 it is clear that the blends keep its semi crystalline
nature up to 30 min annealing i.e. the Vsp vs. T curves show a change of slope at
two temperature ranges, a lower temperature range representing the Tg region
and higher one at Tm region of the semi crystalline component. On extended
annealing, due to the transesterification reactions the blends transformed from
crystalline to amorphous nature and the Vsp vs. T curves shows the behavior of a
typical amorphous polymer. i.e. slope of the curve gradually changes near the
glass transition temperature region and then increases steeply.
Let us consider the PTT/PC 70/30 (Fig. 3.38) in detail. The first heat starts from
a semicrystalline state, which was obtained after the preparation. During heating
cold crystallisation occurs which is pronounced at T >185 °C. At about 212 °C
the melting starts which is finished at 237°C. The first cooling run shows
crystallisation between 205 °C and 187 °C. A soft glass transition is detectable in
the 2nd and 3rd heating run between 50 °C and 70 °C. In the second and third
cooling run also, at the low temperature limit, beginning of glass transition is
detectable. The second heating shows a similar fast melting like in the first heat.
However, in the 3rd heat just before the melting, first signs of cold crystallisation
are detectable (205-215°C) and then a fast melting shifted slightly (~ 4 K) to
lower temperature. The 2nd cool is shifted to lower temperature while in the 3rd
cool no crystallisation occurred. So heat 4 starts from the amorphous state
exhibiting a nice glass transition at 54 °C, a cold crystallisation between 115 and
155°C , and a melting between 175 and 215 °C. All following runs show no
crystallisation. The glass transition is rather well pronounced in the 5th (~ 49 °C)
and 6th (~ 44 °C) run, showing that there is still a high and with annealing time
increasing degree of phase separation, which is not necessarily expected. It is to
be noted that in the 3rd cooling and in the low temperature range of the 4th
heating run the specific volume is slightly lower than in the following runs,
possibly due to soft crystallisation. With increasing annealing time in general the
specific volume slightly increases. The reasons may be manifold: (i) still some
very small degree of crystallisation; (ii) changes due to ongoing
transesterification, and (iii) beginning of degradation. For the last counts a slight
162 Chapter 3
pressure built up during the measurements caused by low molecular weight
compounds which were noticed when opening the cell.
PTT/PC 50/50 blend also showed the same behaviour. The semi crystalline
nature of the blend is maintained till 30 min annealing. (Fig. 3.39) which show
change of slopes at a lower temperature range representing the Tg region and at
a higher temperature range at the Tm region of the semi crystalline component.
On extended annealing, due to the transesterification reactions the blends
transformed from crystalline to amorphous nature and the Vsp vs T curves shows
the behavior of a typical amorphous polymer.
PVT experiments allow measuring the specific volume, Vsp, at a defined
temperature and pressure as a function of time. Typical Volume versus Pressure
(V vs. P) and volume versus temperature (V vs. T) plots called ITS files
(Isothermal Standard runs at constant temperature and repeating at different
temperature intervals) as a function of temperature and pressure of PTT/PC
70/30 and 50/50 blends are represented in Fig.3.40a and b and in Fig.3.41a and
b respectively. The last ITS runs (first heating from RT to 260°C and than
cooling) again show a tendency to increased specific volume with annealing time
(the difference between the values of the heating and then of the cooling run are
larger than commonly (due to the limited heat exchange rate) observed). The
glass transition of the PTT and its pressure dependency are clearly detectable,
while that of the PC is very broad.
Miscibility and phase behaviour of PTT/PC blends 163
.50 100 150 200 250 300
0.74
0.76
0.78
0.80
0.82
0.84
0.86
0.88
0.90
0.92
12 3 4 5 6 7 8 9
10 11 12 1314
1516
1718
19
2021
2223
2425
262728
2930
3132
3334
3536
3738
3940
4142
4344
4546
4748495051
AB C D E F G H I
J K L MN
OP
QR
ST
UV
WX
Y
ZAAAB
ACAD
AEAF
AGAH
AIAJ
AKAL
AMAN
AOAP
AQAR
ASAT
AUAVAWAXAY
ab c d e f g h i
j k l mn
op
qr
st
uv
wx
y
zaaab
acad
aeaf
agah
aiaj
akal
aman
aoap
aqar
asatauavawaxay
V sp (
cm³/g
)
T / °C
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
1 160A 170a 180
190 200
PTT/PC 70/30 (a)
0 50 100 150 2000.750.760.770.780.790.800.810.820.830.840.850.860.870.880.890.900.910.920.93
12
34
56
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
AB
CD
EF
GH I J K L M N O P Q R S T U
ab
cd
ef
gh i j k l m n o p q r s t u
12
34
56
78
910 11 12 13 14 15 16 17 18 19 20 21
AB
CD
EF
GH
I J K L M N O P Q R S T U
ab
cd
ef
gh i j k l m n o p q r s t u
23.835 22.695 27.61 32.435 37.27 41.913 46.91999 51.528 56.251 26.396 32.47 47.275 61.65501 76.09999 90.72398 105.79
1 120.72A 135.765a 151
166.063 181.43 196.875 212.35 227.815 242.63 258.35 257.964 247.87 237.505 227.33 216.97 206.49 196.315 185.995 175.725
1 165.65A 155.42a 145.635
135.06 125.112 115.165 105.085 95.42999 85.77601 76.304 66.52901 56.835 47.23 42.52 35.19 30.49
V sp (
cm3 /g
)
P / MPa
PTT/PC = 70/30 (b)
Figure 3.40: (a) V-T plot of 70/30 blend at specified Pressures (b) V-P plot of 70/30 blend at specified Temperatures
164 Chapter 3
0 50 100 150 200 250 3000.76
0.78
0.80
0.82
0.84
0.86
0.88
0.90
0.92
12 3 4 5 6 7 8 910 1112 13 14
1516
17
18
19
20
21
22
23
24
25
262728
29
30
31
32
33
34
35
36
3738
3940414243
AB C D E F G H IJ K L M NO
PQ
R
S
T
U
V
W
X
Y
ZAAAB
AC
AD
AE
AF
AG
AH
AI
AJ
AKAL
AMANAOAPAQ
ab c d e f g h ij k l m n op
qr
s
t
u
vw
x
y
zaaab
ac
ad
ae
af
ag
ah
ai
aj
akal
amanaoapaq
Vsp
(cm
³/g)
T / °C
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
1 160A 170a 180
190 200 P21 P22 P23 P24 P25 P26 P27 P28 P29 P30 P31 P32 P33 P34
1 P35A P36a P37
P38 P39 P40 P41 P42
PTT/PC 50/50 (a)
0 50 100 150 2000.760.770.780.790.800.810.820.830.840.850.860.870.880.890.900.910.92
12
34
56
78
9 10 11 12 13 14 15 16 17 18 19 20 21
AB
CD
EF
GH
IJ K L M N O P Q R S T U
ab
cd
ef
gh
ij
k l m n o p q r s t u
12
34
56
78
9 10 11 12 13 14 15 16 17 18 19 20 21
AB
CD
EF
GH I J K L M N O P Q R S T U
ab
cd
ef g h i j k l m n o p q r s t u
20.528
22.73
27.53701
32.31001
37.158
41.93
46.665
51.404
56.26
24.642
32.47
47.04
61.439
75.98
90.54201
105.431 120.41A 135.415a 150.692
165.8
181.105
196.555
211.865
227.362
243.115
258.745
258.395
242.722
227.36
212.045
196.315
181.16
165.555
150.565
135.4951 120.67A 105.905a 90.98
76.32
61.99
%(41) 41.965 34.93
Vsp
/ cm
3 /g
P / MPa
PTT/PC 50/50 (b)
Figure 3.41: (a) V-T plot of PTT/PC 50/50 blend at specified Pressures (b) V-P
plot of PTT/PC 50/50 blend at specified Temperatures
Miscibility and phase behaviour of PTT/PC blends 165
3.8. Dynamic rheological measurements of unannealed blends.
Blend morphology is affected by rheological characteristics of the base
polymers, and shear stress applied during the mixing of the components
[77,127,128]. In order to design proper molds, appropriate process equipment
selection and for the assessment of the optimum process conditions, it is very
important to know the relationship among the melt viscosity, elasticity, shear rate,
pressure and processing temperature. Rheology is one of the most frequently
used methods for characterizing interfacial properties such as interfacial tension
and strength that are necessary for predicting the mechanical properties of
immiscible polymer blends [129,130]. Nowadays dynamic rheological
measurements have received much attention as an extremely powerful rheological
technique, which offers several advantages over the conventional steady shear
rheometry because of its unique ability to assess and provide important
informations on the frequency dependence of rheological properties and on the
physical and microstructure of materials without disturbing the conformation of the
material appreciably. In this technique, a sinusoidally varying strain is imposed on
the polymeric material and the resulting stress may be separated into pure elastic
and viscous responses from which useful informations on the melt rheology and
processing characteristics can be obtained.
The rheological properties of molten components in immiscible polymer blends
affect the processing/morphology/property relationships [131-135]. The classic
theory of rheology of emulsions focuses on dilute emulsions of spherical,
Newtonian drops [136, 137]. Palierne [138] reported a cell theory for more
concentrated emulsions that applies to dynamic shear with very small drop
deformation from a spherical shape. Computational results on concentrated
emulsion rheology were repoted by Loewenberg and Hinch [139] for shear flows
with appreciable departures from a spherical shape for the dispersed phase. The
Palierne theory has an added distinction of being formulated for viscoelastic
constituents. Oldroyd [140] and Choi and Schowalter [141] models are the other
two models of emulsion rheology that are applied widely to polymer blends in the
dilute and semi dilute regimes to explain their rheological behaviour.
166 Chapter 3
The phase morphology and rheology of multiphase polymer blends are
strongly affected by interfacial characteristics. Asthana and Jayaraman [142]
and Shi et al. [143] reported that the interfacial tension in polymer blends can be
estimated from particle size distribution using Palierne model. Also Friedrich et
al. [144] have shown that if the interfacial tension is known, the particle size
distribution can be derived from measured data. Micro-mechanical models, such
as that of Coran and Patel [145] reflects the morphology, together with the
common series and parallel mixing rule approaches, have been used to describe
the observed rheological response [146].
But under the reaction conditions of melt rheological measurements (5 min
compression moulding for making samples, 2 min annealing time before
rheological measurements and 5 min for melt rhological analysis) there is
sufficient time for the transreactions to occur between the blend components.
From the literature it can be seen that solution-cast PTT/PC blends are inherently
immiscible [93] and after annealing at 260 °C, the blends could become miscible
due to the transesterification reaction. According to Yavari et al. [94] PTT/PC
blends are partially miscible and after annealing at 300 °C for 10 min the blends
changed to a miscible state through a transesterification reaction. From these
investigations, it can be concluded that transesterification plays an important role
in controlling the properties of PTT/PC blends. The most important feature of this
study is that we calculated interfacial tension between the polymers from the
storage modulus of the blends using two well-known models, viz. Palierne and
Choi-Schowalter.
Effect of blend ratio on the complex viscosity of unannealed PTT/PC blends is given
in Fig. 3.42. As the frequency increases, the complex viscosity decreases.
Further, with increase in frequency, the relaxation time decreases or in other
words, the shear rate increases. Thus an increase in frequency has the same
effect as that of increase in shear rate. Thus in all cases, pseudo plastic
behaviour is seen. It can be seen from Fig. 3.42 that PTT has the minimum while
PC has the maximum complex viscosity in the whole range of frequency. The
complex viscosity of all the blends is found to be intermediate between the neat
Miscibility and phase behaviour of PTT/PC blends 167
polymers in such a way that addition of PC into PTT increases the complex
viscosity due to the interaction between blend components because of trans
reaction taking place between the components.
0.1 1 10 100
400
800 PTT PTT90 PTT80 PTT70 PTT60 PTT50 PTT40 PTT30 PTT20 PC
Com
plex
vis
coci
ty η
∗
Frequency (rad/sec) Figure 3.42: Effect of blend ratio on the complex viscosity of PTT/PC blends
The most important rheological parameters determining the morphology
development of immiscible blends are the viscosity and elasticity ratios of the
components. The values of viscosity and elasticity ratio of the components,
necessary to calculate their ratios, have been derived from the dynamic moduli
by applying the Cox–Merz rule [3] for the pure polymers: with η* is the complex
viscosity (Pa s), ηss is the steady shear viscosity (Pas), steady shear rate, and
ω frequency. The ratios are calculated from the dynamic data at 100 rad/s: The
results are summarised in Table 3.9.
viscosity ratio : p = η*PC / η*PTT or η*PTT/PC ( at 100 rad/sec) [3.21]
elasticity ratio : e = G’PC/ G’PTT or G’PTT/PC ( at 100 rad/sec) [3.22]
168 Chapter 3
Sample Complex viscosity η*(Pas)
Viscosity Ratio (p) η*droplets/η*matrix
Elasticity(Pa) G’(Pa)
Elasticity ratio (e)
G’droplets/G’matrix
PTT 125.241 3.474 845.05 12.74
PTT90 128.815 1.029 1163.67 1.377
PTT80 135.160 1.080 1391.89 1.647
PTT70 168.188 1.343 1750.99 2.072
PTT40 279.153 0.450 5099.05 0.145
PTT30 297.454 0.421 5809.40 0.166
PTT20 332.934 0.376 7647.96 0.111
PC 435.064 0.288 10769.7 0.078
Table 3.9: Rheological data of blend components used including viscosity ratio and elasticity ratio.
From the table it is evident that when the viscosity ratio and elasticity ratios are
high the droplet coalescence occurs and the dispersed phase size will be high. If
the ratios are lower there the highly viscous matrix offers suppression to
dispersed phase coalescence and there by smaller dispersed particles. In this
case PC matrix viscosity is more and thereby results in finely dispersed PTT in
PC matrix than PC dispersed in PTT. These results are in good agreement with
the morphological observations.
3.8.1. Theoretical modelling of rheological properties
In polymer blends, in addition to the characteristics of the component polymers
the viscosity depends on the interfacial adhesion. This is because in polymer
blends, there is interlayer slip along with the orientation and disentanglement
upon the application of shear stress. When shear stress is applied to a blend, it
undergoes an elongational flow. If the interface is strong, the deformation of the
dispersed phase is effectively transferred to the continuous phase. However, in
the case of a weak interface, interlayer slip occurs and as a result, the viscosity
Miscibility and phase behaviour of PTT/PC blends 169
of the system decreases. In order to compare the rheological behaviour of binary
blends, the theoretically predicted viscosities of the uncompatibilised PTT/PC
blends for the entire composition range, were calculated using different
rheological models (equations 3.23 to 3.26).
The viscosity of a biphasic system can be calculated using a series of mixing rules.
According to the rule of additivity [147]
2211 φηφηη += [3.23]
For heterogeneous materials, Hashin’s upper and lower limit models [148]
suggests that
2
2
21
12
2)(1
ηφ
ηη
φηη+
−
+=mix [3.24]
1
1
12
21
2)(1
ηφ
ηη
φηη+
−
+=mix [3.25]
where, ηmix is the viscosity of the blend, η1, η2 and φ 1, φ 2 are the viscosities
and volume fractions of the pure components respectively.
The log additivity rule assumes that the viscosity of the blend depends on their
logarithmic addition [147].
iimix c ηη ∑= loglog [3.26]
where, Ci is the weight fraction and ηi is the viscosity of the components.
Figure 3.43 presents the comparison of the experimental value of complex viscosities
of the unannealed blends with those predicted by the theoretical rheological models at
a frequency of 1Hz. It can be seen from Fig. 3.43 that the blends exhibit a positive-
negative deviation behavior from the additivity line. Similar behavior was also reported
earlier by Utracki [132, 147] for polybutadiene-polyisoprene blends. Here when the PC
content in the blend is increased from 10-60 wt % the experimental viscosity of the
blends show a negative deviation from additivity line model values. But above 60 wt. %
of PC content the viscosity of the blends follows a slight positive deviation from the
170 Chapter 3
additivity line. When PC is the dispersed phase in PTT matrix, the experimental values
seem to lie close to those of the Hashin’s lower limit and log additivity values which
show a positive deviation from 50 wt. % PC more towards the values as predicted by
the additivity rule. An immiscible blend can exhibit three types of behavior: (a) positive
deviation as in a homogeneous blend in which there is a large interaction between the
phases; (b) negative deviation when the interaction is small; and (c) a positive–
negative deviation, when there is a concentration-dependent change of structure.
Therefore the positive–negative deviation observed for the PTT/PC system is
expected to be the result of change in the phase morphology and interfacial
interactions due the transesterification reactions occurred. Fig. 3.44 shows the
variation of storage modulus of PTT/PC blends. The storage modulus increases with
increase in frequency. This is due to the fact that with increase in frequency, polymer
chains relax slowly (less relaxation time) and consequently moduli increased. Storage
modulus of PTT has minimum while PC has the maximum in the whole range of
frequency. Addition of PC into PTT increases the storage modulus and the blends
show storage modulus values intermediate between the neat polymers. This is due to
the transesterification reaction taking place between PTT and PC under the
experimental conditions.
0 20 40 60 80 100100
200
300
400
500
Com
plex
vis
cosi
ty(P
as)
weight percent of PC
experimental additivity rule Hashin's upper limit Hashin's lower limit log additivity
Figure 3.43: Theoretical modelling of complex viscosity of PTT/PC blends.
Miscibility and phase behaviour of PTT/PC blends 171
0.1 1 10 100
1
10
100
1000
10000 PTT PTT90 PTT80 PTT70 PTT60 PTT50 PTT40 PTT30 PTT20 PC
Stor
age
Mod
ulus
(Pa)
Frequency (rad/sec) Figure 3.44: Effect of blend ratio on the storage modulus of PTT/PC blends
3.8.2. Interfacial tension measurements
Palierne model has been shown to be very useful for predicting the rheological
behaviour of the immiscible blends which describes the linear viscoelastic
behaviour of emulsions of viscoelastic fluids [149-152]. Also, Palierne model was
used to determine the interfacial tension between the components [149, 153] to
determine the volume average radius of the dispersed particles [154], to calculate
the sphere-size distribution from rheological data [155] and to analyse the
deformation of droplets under elongational flow [156]. With an electric formalism,
Palierne derived an equation for predicting the complex modulus of molten
(emulsion type) blends (Gb*), which is a function of the complex moduli of both
phases Gm* (for the matrix) and Gb* (for the inclusions or dispersed phase) taking
into account of several important features of a multiphase system. The
viscoelasticity of phases, the hydrodynamics interactions, the droplet size and size
distribution and the interfacial tension are indeed included in this formulation.
Jacobs et al. [157] developed an extended form of the Palierne model, written as,
172 Chapter 3
( )( ) ( )
( )( ) ( )
0* *
0
,1 3
, =
,1 2
,
b m
E RR dR
D RG G
E RR dR
D R
ων
ωω
νω
∞
∞
+
−
∫
∫
[3.27]
in which
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
* * * *
* * * *
* *2 2
4, = 19 16
X 5 2 23 16
2 2413 8 16
d m d m
l
d m d m
ll ll lll
d m
E R G G G GR
G G G GR
G GR R R
αω ω ω ω ω
β ωω ω ω ω
β ω β ω α α β ωω ω β ω
− + +
− + −
+ + − + +
[3.28]
and
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
* * * *
* * * *
* *2 2
, = 2 3 19 16
240 23 32
4 4813 12 32
d m d m
l
d m d m
ll l lll
d m
D R G G G G
G G G GR R
G GR R R
ω ω ω ω ω
β ωα ω ω ω ω
β ω β ω α α β ωω ω β ω
+ +
+ + + +
+ + + + +
[3.29]
where, ( )*bG ω , ( )*
mG ω and ( )*dG ω represent complex modulus of blend,
matrix and dispersed phase, respectively. ( )lβ ω and ( )llβ ω are the complex
interfacial dilation and shear moduli, respectively. ( )Rν denotes the particle
size distribution function while , R andα ω are particle radius, interfacial
tension, and strain frequency, respectively. When the deformation of dispersed
phase is small enough so that viscoelastic properties remain linear, we can set
both ( )lβ ω and ( )llβ ω to zero. Graebling et al. [149] by assuming the particle
size distribution to be narrow ( )2v nR R ≤ ) and interfacial tension to be
independent of shear and interfacial area variation, simplified equation as:
Miscibility and phase behaviour of PTT/PC blends 173
( )( )
* * 1 3 =
1 2i i i
b mi i i
HG G
Hϕ ωϕ ω
+ Σ− Σ
[3.30]
where
( )( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )( )( ) ( ) ( )( ) ( ) ( )( ) ( ) ( )( )
* * * * * *
* * * * * *
4 2 5 16 19 =
40 2 3 16 19i m d d m m d
ii m d d m m d
R G G G G G GH
R G G G G G G
α ω ω ω ω ω ωω
α ω ω ω ω ω ω
+ + − +
+ + − + (3.31)
in which, and i iR φ denote the ith particle fraction radius and the ith volume
fraction of dispersed phase, respectively. The interfacial tension can then be
estimated by fitting the experimental data to the Palierne model. Using ( )α as
fitting parameter, the best fit gives the interfacial tension.
We calculated the interfacial tension based on the weighted relaxation spectrum
(τH(τ)) with the relaxation time (τ) for PTT/PC blends. In order to get the weighted
relaxation spectrum the following equations were used:
( ) ( )( ) ( ) ( )' 2 2 2 21 lnG H dω τ
ω τ ω τ τ∞
−∞ = + ∫ [3.32]
( ) ( )( ) ( ) ( )" 2 21 lnG H dω τ
ωτ ω τ τ∞
−∞ = + ∫ [3.33]
the relaxation spectrum can be determined using Tschoegle approximation [158]
as given in following equation:
( )
( ) ( )( )( ) ( )( )
( ) ( ) ( )
'
2' '
22 '
1 2
log log
0.5 log log
1/ 4.606 [ log log ]
d G d
H G d G d
d G d
τ
ω τ
ω
ω
ω=
− − =
[3.34]
where ω is the frequency and τ is the relaxation time. It should be noted for
neat polymer one will get one relaxation time where as for blends two relaxation
times 1τ and 2τ corresponding to the component polymers. The difference in
the values ( 1 2τ τ− ) was used to calculate the interfacial tension between the
174 Chapter 3
polymers in the presence and absence of compatibilisers. The interfacial tension
(α ) was calculated using two methods:
(i) Palierne [138] (equation 3.35)
( )( )( ) ( )
19 16 2 3 2 ( 1)4 10( 1) 2 (5 2)v m K K KR
K Kφηα
τ φ + + − − = + − +
[3.35]
and (ii) Choi-Schowalter [141] (equation 3.36).
( )( )19 16 2 3 5(19 16)140( 1) 4( 1)(2 3)
v m K KR KK K K
ηα φτ
+ + + = + + + + [3.36]
where mη is the viscosity of the matrix, φ is the volume fraction of the dispersed
phase, K is the viscosity ratio and is given as d mK η η= ( dη is the viscosity of
the dispersed phase). Both these equations are similar to Taylor’s equation.
The interfacial tension values of PTT/PC blends calculated from these equations
are given in Table 3.10. In both methods, the blends show different interfacial
tension values, even though the difference is small. It is very interesting to note
that the blends show very low interfacial tension values, which means that there
is some interaction between the blend components PTT and PC. Here it is the
transesterification reaction occurred under the reaction conditions and the
random copolyester formed as a result of the transesterification reaction
between PTT and PC is the main factor for the change in miscibility. This
random copolymer formed as a result of these exchange reactions acted as a
compatibiliser in the initial stages of reactions. It should be noted that when PC
forms dispersed phase the interfacial tension increases with increase in PC
content since the amount of transreaction is more at high PTT content. But when
PTT forms the dispersed phase the interfacial tension decreases with increase in
PC content. Palierne model gives lower interfacial tension values than Choi-
Schowalter model. However, for a polymer blend system, it is believed that
irrespective of the blend composition, α should be the same. This slight
difference between the α values arises from the parameter Rv, which is derived
Miscibility and phase behaviour of PTT/PC blends 175
from the phase morphology. Note that since the blend is not a dilute system, the
average particle size (Rv) contains contributions from interfacial tension as well
as coalescence effect. Thus the difference arises from the coalescence effect
associated with Rv.
Interfacial tension (mN/m) Blend
Palierne Choi-Schowalter
PTT90 0.050 0.060
PTT80 0.110 0.134
PTT70 0.120 0.150
PTT30 0.042 0.052
PTT20 0.032 0.040
PTT10 0.010 0.015
Table 3.10: Interfacial tension values of PTT/PC blends
3.9. Mechanical property measurements of unannealed Blends An important aspect is that polymers are viscoelastic materials, which have
some of the characteristics of both viscous liquids and elastic solids. Elastic
materials have a capacity to store mechanical energy with no dissipation of
energy; on the other hand, a viscous fluid in a non-hydrostatic stress state has a
capacity for dissipating energy, but none for storing it. When polymeric materials
are deformed, part of the energy is stored as potential energy and part is
dissipated as heat. The energy dissipated as heat manifests itself as mechanical
damping or internal friction. Therefore the interpretations of these properties at
molecular level are of great scientific and practical importance in understanding
the mechanical behaviour of the polymers [159-162].
For incompatible blends containing at least one semi-crystalline component, the
final tensile properties are determined by two competing factors: the increase in
crystallinity due to the presence of more crystalline component and the extent of
176 Chapter 3
compatibility between the two component polymers. The former is the property
determining factor at low strain level and the latter determines the properties at
high strain level. The phase morphology and the interfacial adhesion between
the component polymers also influence the mechanical properties of polymer
blends. Two-phase morphology with lack of adhesion between the component
polymers leads to premature failure and thus to lower tensile strength [163]. The
stress-strain behaviour of PTT/PC blends is demonstrated in Fig. 3.45.
0 2 4 60
20
40
60
PC PTT PTT90 PTT80 PTT70 PTT50 PTT40 PTT20
Stre
ss(M
Pa)
Strain % Figure 3.45: Stress-strain behaviour of unannealed PTT/PC blends
From the stress-strain curves, we estimated maximum tensile strength (σm),
elongation at break (Eb) and Young’s modulus (E) etc and these tensile properties
are summarised in Table 3.11. The results indicate that the addition of PC phase
decreases the tensile strength and modulus. The effect of PC content on the
ultimate tensile strength (σm) of PTT/PC blends are shown in Fig. 3.46. Since PTT
and PC have almost near tensile strength values and there isn’t much interaction
(even though small amount of transesterification reaction taking place under
reaction conditions) between the two component polymers, the tensile strength
shows negative deviation from the additivity line. Addition of PC to PTT decreases
the Young’s modulus (E) of the blend system as indicated in Fig. 3.47. The values
Miscibility and phase behaviour of PTT/PC blends 177
of these mechanical properties of blends with PTT dispersed phase are closer to
the additivity line than dispersed PC (Compare the values of PTT90, PTT80 with
PTT 20), since PTT forms finely dispersed particles in PC matrix.
The impact toughness is often the deciding factor in material selection
because impact test measures the ability of a polymer to withstand the load
imposed upon being struck by an object at high velocity. Thus, it is a
measure of the energy required to propagate a crack across the specimen.
Therefore, the impact properties of these blends are especially important.
The impact strength of PC is very much higher than that of PTT. From Fig.
3.48, it is obvious that the impact strength of blends also registered a
negative deviation from the additivity line. But the impact strength increases
with increase in PC content in the blends. The impact strength results are
also summarised in Table 3.11.
sample
Ultimate Tensile strength
(MPa)
Deformation at break (%)
Young’s Modulus (MPa)
Impact strength
J/m
PTT 59. 5 +/- 2.3 5.1 +/- 0.24 2592 +/- 39 34.5 +/- 2
PTT90 52.1 +/- 2.7 1.7 +/- 0.04 2481 +/- 41 56.2 +/- 4
PTT80 48.1 +/- 2.4 1.8 +/- 0.03 2483 +/- 29 86.8 +/- 3
PTT70 46.2 +/- 1.8 1.9 +/- 0.04 2417 +/- 33 108.5 +/- 5
PTT50 48.4 +/- 1.3 2.3 +/- 0.03 2140 +/- 46 155.4 +/- 7
PTT40 51.3 +/- 2.5 2.5 +/- 0.03 2208 +/- 30 250.2 +/- 9
PTT20 57.3 +/- 1.4 3.2 +/- 0.04 2260 +/- 44 452.3 +/- 6
PC 60.1 +/- 1.2 4.5 +/- 0.2 2189 +/- 28 659.5 +/- 7
Table 3. 11: Mechanical properties of PTT/PC blends.
178 Chapter 3
0 20 40 60 80 10010
20
30
40
50
60
70
80
Ulti
mat
e te
nsile
str
engt
h (M
Pa)
Weight percent of PC Figure 3.46: Effect of blend ratio on the ultimate tensile strength of
uncompatibilised PTT/PC blends
0 20 40 60 80 1002000
2500
Youn
g's
Mod
ulus
(MP
a)
Weight percent of PC Figure 3.47: Effect of blend ratio on the Young’s modulus of PTT/PC blends
Miscibility and phase behaviour of PTT/PC blends 179
0 20 40 60 80 1000
100
200
300
400
500
600
700
Impa
ct s
tren
gth
(J/m
)
Weight percent of PC
Figure 3.48: Effect of blend ratio on the impact strength of PTT/PC blends
3.9.1. Theoretical analysis of mechanical properties.
In order to understand Young’s modulus behaviour, applicability of various
composite models such as Parallel, Series, Coran and Takayanagi are
examined.
The highest upper bound parallel model is given by the rule of mixtures as
follows
1 1 2 2uE E Eφ φ= + [3.37]
This model is applicable to the materials in which the components are connected
parallel to one another so that the applied stress lengthens each component to
the same extent. In the lowest-lower bound series model, the blend components
are arranged in series (Reuss prediction) perpendicular to the direction of the
applied force. The modulus prediction is given by the inverse rule of mixtures as:
1 2
1 2
1
LE E Eφ φ
= + [3.38]
180 Chapter 3
In these models Eu is any mechanical property of the blend in the upper bound
parallel model and EL the moduli of the blend in the series model. E1 and E2 are
the mechanical properties of components 1 and 2, respectively; φI and φ2 are their
corresponding volume fractions. For both these models, no morphology is
required, but strain or stress can be continuous across the interface, and
Poisson’s ratio is the same for both phases.
According to Coran's equation [164, 165]
( )U L LM f M M M= − + [3.39]
where f can vary between zero and unity. The value of f is given by
( )1nH Sf V nV= + [3.40]
where n contains the aspects of phase morphology, and VH and VS are the
volume fractions of the hard phase and soft phase respectively.
Takayanagi proposed a series-parallel model [166, 167] in which, the concept of
percolation is introduced. It is a phenomenological model consisting of mixture
rule between two simple models involving connection in series (Reuss prediction) or
in parallel (Voigt prediction) of the components. According to this model,
[ ] 1211 )()1()1( −+−+−= EEEE φφλ [3.41]
E1 is the property of the matrix phase, E2 is the property of the dispersed phase,
and φ is the volume fraction of the dispersed phase and is related to the degree
of series-parallel coupling. The degree of parallel coupling of the model can be
expressed by
% parallel = [φ (1- λ) / (1- φ λ)] x 100 [3.42]
We have generated data according to these models and these results are
presented in Fig. 3.49. It can be seen that all the theoretical models are near to
each other especially when PTT forms the dispersed phase. The experimental
data show some agreement with series, Coran and Takayanagi models except
for 60 wt% PC. These models takes into account of the morphological aspects of
Miscibility and phase behaviour of PTT/PC blends 181
the blend and this may be the reason why the experimental value shows
agreement with these models. This shows that there is a small interaction
between the blend components under the experimental conditions due to the
small amount of transreaction between PTT and PC under the reaction
conditions i.e. melt blending, compression molding etc. but the effect is not
much pronounced.
0 40 80
2500
Experimental Parallel Series Coran Takanaygi
Youn
g's
mod
ulus
(MPa
)
Weight % of PC
Figure 3.49: Plots of experimental and theoretical Young’s moduli as a function of PC content
3.10. Positron Annihilation Lifetime Spectroscopy (PALS) measurements of unannealed Blends
Nowadays, Positron Annihilation Lifetime Spectroscopy (PALS) analysis is an
established technique for studying the local free volume in polymers [168-173].
Positrons emitted from radioactive sources, like 22Na, into the polymeric solid
become thermalised and may annihilate with an electron or form positronium
(Ps)- a hydrogen- like bound state [173]. Ps is the result of the combination of a
thermalized positron with one of the available free electrons. The Ps formation is
affected by a large variety of processes, such as the mobility of electrons and
positrons and the appearance of scavengers for electron or positrons [173]. One
182 Chapter 3
quarter of the formed Ps appear as para-positronium (p-Ps). The electron and
positron spins are anti-parallel in p-Ps, which decays quickly via self annihilation
with a life time of ~150 ps. Three quarters of the formed Ps appear as ortho-
positronium (o-Ps), with the electron and positron spins parallel. In vacuum, o-Ps
has a life time of 142 ns [172]. In matter, the positron in o-Ps may annihilate during
collision with molecules with an electron of opposite spin. This process, called
pick-off annihilation, reduces the o-Ps lifetime in polymers to ~1-5 ns. From this life
time, the size of the local free volumes in which the o-Ps is trapped can be
calculated. These holes have typical dimensions of between 0.3 and 1 nm and
appear as a consequence of the structural disorder in amorphous polymers.
All PALS measurements were performed at room temperature and two or three
positron lifetime spectra (with more than a million counts in each spectrum) were
recorded. The consistently reproducible spectra were analyzed into three lifetime
components with the help of the PATFIT-88 computer program with proper
source and background corrections. The source correction term and resolution
function were estimated from the lifetime of well-annealed aluminium using the
RESOLUTION [174] program. The three Gaussian resolution functions were
used in this analysis of positron lifetime spectra for the blend and pure samples.
The positron lifetime spectra from PALS measurements were resolved into three
lifetime components τ1, τ2 and τ3 with intensities I1, I2 and I3, respectively. The
shortest lifetime component τ1 with intensity I1 is attributed to p-Ps and free
positron annihilations. The intermediate lifetime component τ2 with intensity I2 is
usually considered to be due to annihilation of positrons trapped at the defects
present in the crystalline regions or trapped at the crystalline–amorphous
interface. The longest-lived component τ3 with intensity I3 is due to pick-off
annihilation of the o-Ps in the free-volume sites present mainly in the amorphous
regions of the polymer matrix. The o-Ps lifetime τ3 is related to the free-volume
hole size by a simple relation developed by Nakanishi et al [175] which is based
on the quantum mechanical models of Tao [176] and Eldrup et al [177]. In this
model, Positronium (Ps) atom is assumed to be localized in a spherical potential
well having an infinite potential barrier of radius Ro with an electron layer in the
Miscibility and phase behaviour of PTT/PC blends 183
region R < r < Ro. Accordingly, the relation between τ3 and the radius R of the
free volume hole or cavity is given by:
[3.43]
where Ro = R + δR and δR is an adjustable fitting parameter which represents
the thickness of the electron layer or the probability of the overlap of the Ps
wave function and electron wave function. The free-volume radius R was
calculated from Eqn (3.43) and the average size of the free-volume holes Vf
was evaluated as
Vf = (4/3)πR3. [3.44]
The relative fractional free volume or the free-volume content (Fvr) of the sample
could then be estimated as
Fvr = Vf I3 [3.45]
To determine the free-volume parameters of the blends we consider only the o-
Ps lifetime τ3 and its intensity I3. The positron data of PTT/PC blends are shown
in Figs. 3.50 and 3.51. From these figures is clear that the average free volume
size (Vf3) and its intensity (I3) increases slightly with increasing concentration of
PC in the blend. This variation of the free-volume hole sizes of the blends was
tested with the linear additivity rule and found to have a slight positive deviation
from this rule [178]. The continuous but small increase in the free-volume hole
size of the blend with increase of PC content is possibly due to coalescence of
the free volume of PC. Under normal blending conditions the physical and
chemical interactions across the phase boundaries of PTT and PC will be small.
This leads to a weak interface. As a result, there is the possibility of void
formation at the interface. But here, the samples used for PALS measurements
are compression moulded for 5 min after melt blending which is sufficient to
induce transesterification reactions between the blend components to start.
Therefore, the void formation become small, i.e. free volume values become
small, and in PC rich blends the amount of transesterification reaction will be
184 Chapter 3
low, which means that an increase in free-volume size with increase in PC
content is on the expected lines.
0 20 40 60 80 1001.5
1.6
1.7
1.8
1.9
2.0
Weight percent of PC
Free
vol
ume
hole
siz
e, V
f (Å
3)
o-Ps
life
time(
ns)
τ3
60
70
80
90
100
Vf
Figure 3.50: Effect of blend ratio on the o-Ps life time and free volume hole size
of PTT/PC blend
0 20 40 60 80 10015
20
25
30
I3
Fvr
10
15
20
25
Rel
ativ
e fr
actio
nal f
ree
volu
me,
Fvr
(%)
o-Ps
inte
nsity
, I3 (%
)
Weight percent of PC
Figure 3.51: Effect of blend ratio on o-Ps intensity and relative fractional free volume in PTT/PC Blends
Miscibility and phase behaviour of PTT/PC blends 185
3.11. Thermal degradation properties of PTT/PC unannealed blends. Thermal degradation behaviour of polymers and polymer blends are very
relevant to the potential use of these materials in many demanding applications.
In order to develop durable industrial products it is necessary to investigate the
thermal stability of these blends. Thermal properties are important due to the
fact that the stability of polymeric materials towards thermal degradation is one
of the important criteria for designing these materials for specific applications.
Polymeric materials are subjected to various types of degradation ranging from
thermal degradation to biodegradation. Polymer degradation is generally
undesirable as far as their application is concerned since it leads to deterioration
of properties. Further, a change in heat flow and stability of polymers will give
some idea on the extent of chemical interaction occurring between the
components, their bond strength, activation energy, melting temperature and
degradation kinetics. Fabrication and design of a variety of articles with improved
mechanical properties as well as their end uses need a detailed understanding
of the thermal degradation of polymers, because the threshold temperature for
decomposition determines the upper limit of the fabrication temperature. One of
the most accepted methods for studying the thermal properties of polymeric
materials is thermogravimetry. The integral (TGA) and derivative (DTG)
thermogravimetric curves provide information about the nature and extent of
degradation of the polymeric materials.
Several research publications are available to prove the great impact of the
thermal stability of polymers by blending [179-193]. The thermal stability of the
blends depends strongly on the compatibility of the polymers [37, 183,189].
Varughese et.al. [179] reported that the blending of epoxidised natural rubber
(ENR) with poly vinyl chloride (PVC) reduced the rate of HCl elimination in the
first degradation step of PVC. As far as the thermal degradation is concerned,
blending of a polymer with other polymers has stabilizing as well as destabilizing
effects. Effects of blend ratio and compatibiliser concentration on the thermal
degradation properties of the PA12/PP blends were analysed by Jose et.al [37].
They found that the blend ratio as well as the presence of compatibiliser has
significant effect on the thermal stability of the blends. Phase morphology was
186 Chapter 3
found to be one of the decisive factors that affected the thermal stability of both
uncompatibilised and compatibilised blends. Guo et.al. [189] reported the
degradation behaviours and thermal properties of polystyrene (PS)/polyolefin
elastomer (POE) blends. Thermo gravimetric analysis (TGA) was adopted to
reveal the effects of in situ grafting reaction and degradation of blending
compounds on the thermal properties of PS/POE blends. It was found that the
changes in both catalyst content and blend composition influenced the
competition between in situ grafting reaction and degradation, resulting in the
complexity of the thermal properties of PS/POE/AlCl3 blends.
The effect of blend ratio on the thermal degradation properties of the PTT/PC
blends are presented in this part. The activation energy for degradation in the
unannealed blends computed using Horowitz-Metzger equation is reported. The
thermograms (TGA) and derivative thermograms (DTG) of PTT/PC blends are
given in Fig. 3.52a and b. Detailed evaluations of the thermograms are
presented in Tables 3.12 to 3.14.
Table 3.12 gives an idea about the effect of blend ratio on the temperature
corresponding to different weight losses (viz. Ton - onset of degradation, T10 -
temperature corresponding to 10wt% degradation, and so on). It is seen from the
table that PTT is more susceptible to degradation where as PC shows maximum
thermal stability. The Ton of PTT (330°C) is much lower than that of PC (432°C).
The thermal stability of the blends is in between these limits. As the amount of
PC in PTT increases, thermal stability of the blends increases. For example, the
Ton of PTT increases from 330 to 333°C by the addition of 10wt% of PC (PTT80)
and to 338°C by 80% addition of PC into PTT (PTT20), i.e. Ton increased by ~
8°C. The same trend is seen in case of T10, T20, etc. This means that phase
morphology has a definite role in determining the thermal stability of the blends.
It should be noted that in PTT80 blend, PTT is the matrix and in PTT50 blend,
both PTT and PC form continuous phases. Thus in PTT80 and PTT50 blends,
PTT phase are more susceptible to thermal degradation. On the other hand, in
PTT20 blends, PC forms matrix where as PTT is the dispersed phase. As result,
Miscibility and phase behaviour of PTT/PC blends 187
the thermal degradation of PTT is suppressed since PC matrix offers protection
to the dispersed PTT domains.
200 400 600 8000
40
80
PTT PTT90 PTT80 PTT70 PTT50 PTT30 PTT20 PC
Wei
ght %
Temperatureo C
(a)
100 200 300 400 500 600 700 800
20
15
10
5
0
PTT PTT90 PTT80 PTT70 PTT50 PTT30 PTT20 PC
dW/d
T
TemperatureoC
(b)
Figure 3.52: Effect of blend ratio on the thermograms of unannealed PTT/PC blends (a) TG (b) DTG
188 Chapter 3
Blends Ton(°C) T10 (°C) T20(°C) T30(°C) T40(°C) T50(°C)
PTT 330 362 371 377 383 387
PTT90 335 365 378 386 392 397
PTT80 333 364 376 386 392 397
PTT70 332 363 376 387 393 400
PTT50 332 355 370 385 395 409
PTT30 336 364 376 388 401 430
PTT20 338 377 395 412 439 467
PC 432 473 486 496 502 508
Table 3.12: Effect of blend ratio on the temperatures corresponding to different percentage weight losses in unannealed PTT/PC blends
Table 3.13 shows the effect of blend ratio on the weight percentage of the
sample at seven selected temperatures, gives a clear idea about how much
improvement in the thermal stability has been achieved by the addition of PC
into PTT
Weight % of the sample remained at selected temperatures Blends
320(°C) 360 (°C) 400(°C) 440 (°C) 480 (°C) 520(°C)
PTT 99.5 65.15 22.44 9.3 8.4 7.9
PTT90 99 92.1 45.1 16.3 13.4 10.3
PTT80 99 88 43 24.1 17.6 13
PTT70 99 87 49 31 23 16
PTT50 99 85 55 43 30 20
PTT30 99.5 92 61 48 33 22
PTT20 99.6 96 77 61 43 23
PC 99.9 99.8 99.5 98 86 34
Table 3.13: Effect of blend ratio on the weight remained at selected temperatures in unannealed PTT/PC blends
Miscibility and phase behaviour of PTT/PC blends 189
Table 3.14 displays the effect of blend ratio on the Tmax of the blends. The Tmax of
PTT and PC were found to be 391 and 507°C, respectively. As the amount of
PC in the blend increases from 0- 50 wt%, the Tmax of PTT increases marginally
above which decreases slightly.
Blends Tmax (°C)
PTT 391
PTT90 397, 479
PTT80 392, 472
PTT70 391, 476
PTT50 393, 475
PTT30 387, 476
PTT20 388, 487
PC 507
Table 3.14: Effect of blend ratio on the Tmax of unannealed PTT/PC blends
3.11.1. Activation energy for thermal decomposition
Activation energy for the decomposition of PTT and PC in unannealed PTT/PC
blends was measured using Horowitz and Metzger (HM) method [194].
In HM, activation energy was calculated using the equation:
( ) 1 2maxln ln 1 a aE E RTα θ− − = [3.46]
where α is the decomposed fraction and is given as α = Ci-C/Ci-Cf, where C the
weight at temperature chosen, Ci the weight at initial temperature and Cf is the
weight at final temperature, Ea is the activation energy for decomposition, Tmax
the temperature at maximum rate of weight loss, R the universal gas constant
and θ is given by T-Tmax. Kinetic plots were made with ( ) 1ln ln 1 α − − versusθ.
From the slope of the plots Ea was calculated. Fig. 3.53 shows the Arrhenius plots
190 Chapter 3
for the activation energy (Ea) for the decomposition of PTT and PC in unannealed
PTT/PC blends.
-80 -60 -40 -20 0 20-4
-3
-2
-1
0
1
PTT PTT80 PTT50 PTT30 PCln
[ln(1
-α)-1
]
T - Tmax(°C)
Figure 3.53: Arrhenius plots for calculating the activation energy for degradation of PTT, PC and their blends
The effect of blend ratio on the activation energy (Ea) values of PTT and PC are listed
in Table 3.15. Activation energy of PTT and PC was found to be 148.3 and 190.1
kJ/mol, respectively. It is seen that addition of PC into PTT increases the Ea of the
blends. The blends are two-phase heterogeneous systems; we obtained two Ea values
for each blend due to the degradation peaks of PTT and PC. It is also important to
note that an increase in Ea indicates that more energy is required for the major
degradation step, which in turn implies an improvement in thermal stability of the
blends. Thus it can be concluded that as the amount of PC in the blend increases, the
thermal stability also increases.
Miscibility and phase behaviour of PTT/PC blends 191
Blends Ea (kJ/mol)
PTT 148.3
PTT90 134.2 245.3
PTT70 145.5 263. 2
PTT50 139.5 256. 2
PTT20 155.5 270. 2
PC 190.1
Table 3.15: Effect of blend ratio on the activation energy of degradation of PTT/PC blends
3.12. Conclusions
PTT/PC melt blends were characterized by their transreaction, morphology,
thermal and crystallisation behavior upon annealing using SEM, FTIR, WAXD, 1H NMR, DSC, PVT, DMA, Dynamic Rheology, PALS and TGA measurements.
Effect of annealing on the reaction induced miscibility and phase behaviour of
these blends were analysed.
These investigations showed that the unannealed blends are having
heterogeneous phase morphology, i.e., PTT and PC are inherently immiscible
and the copolyester content is exactly zero. SEM analysis showed that upon
annealing at 260 °C, for more than 120 min the original two phase morphology is
converted into a homogeneous one.
The copolyester content increases with increase in annealing time and the PTT
content. FT-IR spectra revealed the occurrence of a transesterification reaction
between PTT and PC chain segments upon annealing at 260 °C, and this
exchange reaction produced the new copolymer, which acted as the
compatibiliser in the PTT/PC blends in the earlier stage of the reaction. A new
absorption peak present at ~ 1070 cm-1 in the spectra of annealed blends is
characteristic of the new aromatic polyester structure. 1H NMR studies confirm
the dependence of transesterification rate on the annealing time and PTT
192 Chapter 3
content. The sequence structures of the produced co-polyesters were
determined by a triad analysis, which showed that the degree of randomness
increased proportionally with time of annealing. It was found that the
randomness on the PTT/PC 70/30 and 50/50 blends increased with annealing
time at 260 °C. Up to 180 min of reaction, the degrees of randomness B are 0.89
(for 70/30 blend) and 0.51 (for 50/50 blend) i.e., degree of randomness B
approaching 1.0 upon extended annealing for the 70/30 blend, where PTT
content is more compared to the 50/50 blend, indicating that fully random
copolyesters (with B ~1) are finally formed after extensive reactions at 260 °C.
Two well defined glass transition temperatures present in the DSC curves of the
unannealed blends are indicative of an immiscible system. Miscibility of PTT/PC
blends is correlated with blend composition. When the weight percent of PC is
greater than 20 wt. %, the crystallisation exotherms became very broad and
indistinct and the broad crystallisation exotherms appeared to end at the glass
transition temperatures of the PTT-rich phase. This suggests that PC severely
restrained the mobility of PTT molecules or segments, which led to much longer
and more varied relaxation times. As a result, the crystallisation process takes
place over a wider temperature range. Once the temperature decreased to the
glass transition temperature, the segments were frozen instantaneously at various
crystallisation stages. When PC content in the blend is greater than 70 wt. %, the
crystallisation of the PTT-rich phase appeared to be completely restricted. After
annealing at progressively longer times (0 -180 min), the original two Tgs in the
blends come closer and finally merged to a single Tg. The melting temperature
decreases with the increase in annealing time and when it is 120 min or longer, Tm
of the blends disappears, indicating transition into an amorphous state. Also DSC
measurements on the heated PTT/PC 70/30 blend showed that transesterification
reaction between PTT and PC at 260 °C is irreversible. SEM analysis showed that
upon annealing at 260 °C, for more than 120 min the original two phase
morphology is converted into a homogeneous one.
From WAXD analyses of unannealed blends, it can be concluded that PTT has
triclinic crystalline structure. But PC gives only an amorphous halo in the WAXD
Miscibility and phase behaviour of PTT/PC blends 193
spectrum indicating that it is amorphous in nature. It can be seen that the
intensity of the crystalline diffraction peaks of PTT is decreased with increase in
PC content in the blends. When annealed at 260°C for more than one hour, the
PTT/PC blends turned out to be amorphous, hence, indicating formation of
random copolyesters as a result of transesterification reactions.
The Dynamic mechanical Analysis of unannealed blends revealed that for each
blend there are two well separated glass transition relaxations corresponding to
a PTT-rich phase and a PC-rich phase, respectively. On adding PC into PTT
there is a slight shifting of tanδmax of PTT and PC towards each other indicating
partial miscibility due to the transreactions taking place in the system due to the
annealing effect caused by the reaction conditions (sample preparation
conditions i.e. melt blending compression moulding etc.), even though the
samples used are not annealed separately. This shift is more pronounced in
PTT90 and PTT80 blends where the PC content is low there by the transreactions
rate is high. But in the case of annealed blends, the single tan δ peak at
extended annealing times indicates transition into a homogeneous system. Thus,
it is clear that on extended annealing, the original phase-separated domains
eventually disappear, and the morphology of the annealed PTT/PC blends
turned homogeneous.
Pressure Volume Temperature (PVT) measurements showed that semi
crystalline nature of the blends continued till 30 min of annealing i.e. the Vsp vs. T
curves show a change of slope at two temperature ranges, a lower temperature
range representing the Tg region and higher one at Tm region of the semi
crystalline component. On extended annealing, due to the transesterification
reactions the blends transformed from crystalline to amorphous nature and the
Vsp vs. T curves shows the behavior of a typical amorphous polymer.
Also very low interfacial tension values calculated from rheological
measurements indicate that there is interaction between the blend components,
PTT and PC, under the reaction conditions. Mechanical measurements of
unannealed blends also showed that the unannealed blends are having phase
194 Chapter 3
separated morphology and there is only slight interaction due to transreactions
under the reaction conditions. PALS results
The PALS results revealed that the average free volume size (Vf3) and its intensity
(I3) increases slightly with increasing concentration of PC in the blend. The
continuous but small increase in the free-volume hole size of the blend with
increase of PC content is possibly due to coalescence of the free volume of PC.
Here, the samples used for PALS measurements are compression moulded after
melt blending which is sufficient to induce small amount of transesterification
reactions between the blend components. Therefore, the void formation become
small, i.e. free volume values become small, and in PC rich blends the amount of
transesterification reaction will be low, which means that an increase in free-
volume size with increase in PC content is on the expected lines.
The thermal degradation studies of unannealed blends revealed that PTT is
more susceptible to thermal degradation where as PC showed maximum
thermal stability, that means, the amount of PC in PTT increases, thermal
stability of the blends increases. We can see that the phase morphology has a
definite role in determining the thermal stability of the blends. The thermal
stability of blends with dispersed PTT phase is greater than those with dispersed
PC phase. This is due to the fact that the thermal degradation of PTT is
suppressed since PC matrix offers protection to the dispersed PTT domains.
According to the experimental results PTT/PC unannealed and annealed blends
it can be concluded that the random copolyester formed as a result of the
transesterification reaction between PTT and PC is the main factor for the
change in miscibility. This random copolymer formed as a result of these
exchange reactions acted as a compatibiliser in the initial stages of reactions.
After annealing at 260 °C for more than one hour, this random copolymer
became the main species of the product exhibiting a homogeneous phase.
Miscibility and phase behaviour of PTT/PC blends 195
3.13. References
1. B.D. Favis, J.P. Chalifoux, Polymer, 29, 1761,1988.
2. K. Min, J.L. White, J.F. Fellers, Polym. Eng. Sci., 24, 1327, 1984.
3. S. Wu, Polym. Eng. Sci., 27, 335, 1987.
4. J.L. White, K .Min, Adv. Polym. Technol., 5, 225, 1985.
5. N.Tokita, Rubber Chem. Technol., 50, 292, 1977.
6. G. Wildes, H. Keskkula, D.R. Paul, Polymer, 40, 5609, 1999.
7. B. Majumdar, D.R. Paul, A.J. Oshinski, Polymer, 38, 1787, 1997.
8. U. Sundararaj, C.W, Macosko, Macromolecules, 28, 2647, 1995.
9. I. Fortelny, A. Zivny, Polymer, 36, 4113, 1995.
10. J.J. Elmendorp, A.K. van der Vegt, L.A. Utracki, Eds. Two-phase polymer
systems, Munich: Carl Hanser, 1991.
11. C.W. Macosko, P. Guegan, A. Khandpur, A. Nakayama, P. Marechal, T.
Inoue, Macromolecules, 29, 5590, 1996.
12. L.A. Utracki, Polymer alloys and blends. Munich: Carl Hanser/ Dusseldorf:
VDI Verlag, 1989.
13. I. Fortelny, J. Kovar, M.J. Stephan, Elast Plast., 28, 106, 1996.
14. J.M.H. Janssen, H.E.H. Meijer, Polym. Eng.. Sci., 35, 1766, 1987.
15. C.C. Chen, E. Fontan, K. Min, J.L. White. Polym. Eng. Sci., 28, 69, 1988.
16. L. Minkova, Hr. Yordanov, S. Filippi, N. Grizzuti, Polymer, 44, 7925, 2003.
17. S.E. Harton, F.A. Stevie, H. Ade, Macromolecules, 38, 9, 2005.
18. H. Pernot, M. Baumert, F. Court, L. Leibler, Nat. Mater., 1, 54, 2002.
19. G.I. Taylor, Proc. R. Soc., A 138, 41, 1932.
20. G.I. Taylor, Proc. R. Soc., A 146, 501, 1934.
196 Chapter 3
21. J.E. Callan, W.M. Hessand C.E. Scott; Rubber Chem. Technol., 44, 814,
1971.
22. S. Danesi and R.S. Porter Polymer, 18, 69, 1977.
23. W. Chen, J.A. Sauer, A. Hara, Polymer, 45, 7219, 2004.
24. J. Yonezawa, S.M. Martin, C.W. Macosko, M.D. Ward, Macromolecules, 37,
6424, 2004.
25. P. Wang, J.T. Koberstein, Macromolecules, 37, 5671, 2004.
26. X. Li, Y.C. Han, L.J. An, Polymer, 44, 8155, 2003.
27. H.A. Khonakdar, S.H. Jafari, A. Yavari, A. Asadinezhad, U. Wagenknecht,
Polymer Bulletin, 54, 75, 2005.
28. P. Tsotra, O. Gryshchuk, K. Friedrich, Macromolecular Chemistry and
Physics, 206, 787, 2005.
29. P.S. Calvao, M.Yee, N.R. Demarquette, Polymer, 46, 2610, 2005.
30. S. Jose, S. Thomas, E. Lievana, J. KargerKocsis, J. Appl. Polym. Sci., 95,
1376, 2005.
31. S. Mbarek, M. Jaziri, C. Carrot, Polym. Eng. Sci., 46, 1378, 2006.
32. H.Y. Bai, Y. Zhang, Y.X Zhang, X.F Zhang, W. Zhou, 101, 54, 2006.
33. C.C. Su, C.K .Shih, J. Appl. Polyrn. Sci., 100, 3840, 2006.
34. M. Anthamatten, J. Polym. Sci. Polym.Phy. 45,3285,2007.
35. S. Maeda, E.Kamei, T. Takigawa, T.Masuda, Kobunshi Ronbunshu, 64,
791,2007.
36. W.X. Shi, Y.Y. Li, J. Xu, G.Q. Ma, J. Sheng, J. Macro. Sci. .-Phy., 46,
1115, 2007.
37. S. Jose, S. Thomas, P.K. Biju, P. Koshy, J. Karger-Kocsis, Polym. Degrad.
Stab. 93, 1176, 2008.
38. C. Carotenuto, M. Grosso, P.L. Maffettone, Macromolecules, 41, 4492, 2008.
Miscibility and phase behaviour of PTT/PC blends 197
39. J. Kim, R.W. Sandoval, C.M. Dettmer, S.T. Nguyen, J.M Torkelson,
Polymer, 49, 2686, 2008.
40. B. Andricic, T. Kovacic, S. Perinovic, I. Klaric, J. Appl. Polyrn. Sci., 1002,
109, 2008.
41. L. Barangi, H. Nazockdast, F. A . Taromi, J. Appl. Polyrn. Sci., 108, 2558,
2008.
42. T.F. Wu, G.S. Yang, J. Appl. Polyrn. Sci., 108, 1044, 2008.
43. C. Qin, J. F.Shen, Y.Z. Hu, W.S. Huang, M.X. Ye, Polym. Eng. Sci., 48,
556, 2008.
44. I. Fortelny, M. Lapcikova, F. Lednicky, Z. Stary, Z. Krulis, Polym. Eng.
Sci., 48, 564, 2008.
45. D. Cangialosi, A. Alegria, J.Colmenero, Macromolecules, 41,1565,2008
46. J.D.Martin, S.S. Velankar, AICHE J. 54, 791, 2008.
47. P. Supaphol , N. Dangseeyun, P. Srimoaon, Polym. Testing 23, 175, 2004.
48. B. Jacques, J. Devaux, R. Legras, E. Nield, Polymer, 38, 5367, 1997.
49. S.C.E. Backsona, R.W. Richards, S.M. King, Polymer, 40, 4205, 1999
50. K.J. Song, J.L. White, Polym. Sci. Engng. 40, 902, 2000
51. Y. Yang, S. Li, H. Brown, P. Casey, AATCC Rev. 2, 54, 2002.
52. C.F. Ou, W.C. Li, Y.H. Chen, J. Appl. Polym. Sci. 86, 1599, 2002.
53. Y. Peng, W.H. Guo, P. Zhu, C.F. Wu, J. Appl. Polym. Sci , 109,483, 2008.
54. S. Mubarek, M. Jaziri, C. Carrot, Polym. Eng. Sci., 46, 1378, 2006.
55. A.N. Wilkinson, E.M.I. Nita, M.L. Clemens, E .Jobstl, J.P.A.. Fairclough, J.
Macro. Sci.- Phy. 44, 1087, 2005.
56. C.H. Lei, D.H. Chen, J. Appl. Polym. Sci, 109, 1099, 2008.
57. M.C. Righetti, M.L. Di Lorenzo, M. Angiuli, E. Tombari, P. La Pletra, Euro.
Polym. J., 43,4726,2007.
198 Chapter 3
58. M.L. Di Lorenzo, P. La Pietra, M.E. Errico, M.C. Righetti, M. Angiuli,
Polym. Eng. Sci., 47, 323, 2007.
59. K.E. Fagelman, J.T. Guthrie, Surf. Coatings Int. Part B-Coatings Trans.,
89, 1, 2006.
60. F. C.Chiu, H.Y. Lee, Y.H. Wang, J. Appl. Polym. Sci., 107, 3831, 2008.
61. M.S. Montaudo, e-Polym., 1618, 51, 2008.
62. F.C.Chiu, M.H. Ting, , Polym. Testing, 26, 338, 2007.
63. H.Y. Bai, Y. Zhang, Y.X Zhang, X.F Zhang, W. Zhou, 101, 54, 2006.
64. C.C. Su, C.K .Shih, J. Appl. Polyrn. Sci., 100, 3840, 2006.
65. A .Sanchez-Solis, A. Garcia-Rejon, A. Martinez-Richa, F. Calderas, O.
Manero, J. Polym. Eng. 25,553, 2005.
66. B.Wang, C.Y. Li, J.Hanzlicek, S.Z.D. Cheng, P.H.Geil, J. Grebowicz, R.M.
Ho,. Polymer, 42, 7171, 2001.
67. M. Pyda, A. Boller, J. Grebowicz, H. Chuah, B.V.Lebedev, B. Wunderlich,
J. Polym. Sci., Part B: Polym. Phys. 36, 2499, 1998.
68. M. Pyda, B. Wunderlich, J. Polym. Sci., Part B: Polym. Phys., 38, 622, 2000.
69. J.M. Huang, F.C.Chang, J. Polym. Sci., Part B: Polym. Phys. 38, 934, 2000.
70. P.D. Hong, W.T. Chung, C.F. Hsu, Polymer, 43, 3335, 2002.
71. W.T. Chung, W.J. Yeh, P.D Hong, J. Appl. Polym. Sci., 83, 2426, 2002.
72. X.S. Wang, D. Yan, G.H. Tian, X.G. Li, Polym. Eng. Sci., 41, 1655, 2001.
73. Y.H Kuo, E.M. Woo, T.Y. Kuo, Polym. J. 33, 920, 2001.
74. J.M. Huang, F.C. Chang, J. Appl. Polym. Sci., 84, 850, 2002.
75. F.C. Chiu, K.H. Huang, J.C Yang,. J. Polym. Sci., Part B: Polym. Phys.,
41, 2264, 2003.
76. J.M. Huang, J. Appl. Polym. Sci. 88, 2247, 2003.
Miscibility and phase behaviour of PTT/PC blends 199
77. A. Asadinezhad, A. Yavari, S.H.Jafari, H.A.Khonakdar, F. Boehme, R.
Haessler, Polym. Bull., 54, 205, 2005.
78. F.C. Chiu.; M.H. Ting. Polym. Test., 26, 338, 2007.
79. D. H. Huang.; E. M. Woo,; L.T. Lee. Colliod. Polym. Sci., 284, 843, 2006.
80. I.Hopfe, G.Pompe, K.-J Eichhorn,. Polymer, 38, 2321, 1997.
81. P.Godard, J.M. Dekoninck, V.Devlesaver, J. Devaux, J. Polym. Sci., Part
A: Polym. Chem. 24, 3301, 1986.
82. R.S Porter, L.H. Wang, Polymer, 33, 2019, 1992.
83. R.S. Porter, J.M. Jonza, M.Kimura, C.R. Desper, E.R. George, Polym.
Eng. Sci., 29, 55, 1989.
84. T. S. Oh, J. H.Ryou, Y. S.Chun, W. N. Kim, Polym. Eng. Sci., 37, 838, 1997
85. M. Guo, W.J. Brittain, Macromolecules, 31, 7166, 1998.
86. H.Yang, J. He, B.Liang, J. Polym. Sci., Part B: Polym. Phys., 39, 2607,
2001.
87. A.M. Kenwright, S.K.Peace, R.W.Richards, A.Bunn,; W.A. MacDonald,
Polymer, 40, 5851, 1999.
88. S.C. Lee, K.H. Yoon, I.H. Park, H.C. Kim, T.W. Son, Polymer, 38, 4831,
1997.
89. L.M. Robeson, J. Appl. Polym. Sci., 30, 4081, 1985.
90. S. Liu, W. Li, B.Liang, Y. Yu, Eur. Polym. J., 36, 2159, 2000.
91. H. Yoon, Y. Feng, Y.Qiu, C.C. Han, J. Polym. Sci., Part B: Polym. Phys.,
32, 1485, 1994.
92. a) J. Devaux, P.Godard, J. P.Mercier, J. Polym. Sci.: Polym. Phys., 20,
1875,1982. b) J. Devaux, P. Godard, J. P. Mercier, J. Polym. Sci.: Polym.
Phys. 20, 1881, 1982. c) J. Devaux, P. Godard, J. P. Mercier, J. Polym.
Sci.: Polym. Phys., 20, 1895, 1982. d) J. Devaux, P. Godard, J. P. Mercier,
J. Polym. Sci., Polym. Phys., 20, 1901, 1982.
200 Chapter 3
93. L.T. Lee, E.M. Woo, Colloid. Polym. Sci., 282, 1308, 2004.
94. A. Yavari, A.Asadinezhad, S. H. Jafari, H. A. Khonakdar, F. Boehme, R.
Haessler, Eur. Polym. J. 41, 2880, 2005.
95. S. Wu, Polymer, 28, 1144, 1987.
96. G.M. Jordhamo, J.A. Manson, L.H. Sperling, Polym. Eng Sci., 26, 517, 1986.
97. T.H. Chen, A.C. Su, Polymer, 34, 4826, 1993.
98. V. Everaet, L. Aerts, G. Groeninckx, Polymer, 43, 6627, 1999.
99. P. Krutphun, P. Supaphol, Euro. Polym. J., 41 1561, 2005.
100. E.M. Woo, S.S. Hou. D. H. Huang.; L. T. Lee, Polymer, 46, 7425, 2005.
101. G. Pompe, L. Haeussler, J. Polym. Sci., Part B: Polym. Phy., 35, 2161, 1997.
102. C.C.Gonza´lez,; J.M Peren˜a,.; A. Bello, J. Polym. Sci., Part B, Polym.
Phys., 26, 1397,1988.
103. D.C. Wahrmund, D.R. Paul, J.W. Barlow, J. Appl. Polym. Sci., 22, 2155, 1978.
104. K. T. Varughese, G.B. Nando, P.P.De and S.K.De. J.mater. Sci, 23, 3894, 1988.
105. S. Thomas and A. George, Eur .Polym. J., 28, 145, 1992.
106. D.J.Walsh and J.S Higgns, Polymer, 23, 336, 1982.
107. S. Thomas, B.R.Gupta and S.K.De, J. Vinyl Technol, 9,71,1887.
108. A.Y. Coran, Handbook of Elastomers, New developments and Technology
Eds. A.K. Bhowmick, H.L. Stephens, Marcel Dekker, New York, 249 ,1988.
109. A.Y Coran, in “Thermoplastic Elastomers”, N.R.Legge, G.Holden, H.E
Schroeder, eds; Hanser Publishers, Distributed by Macmillan Publishing
Co; N.Y. 1987.
110. R.A. Dickie, J. Appl. Polym. Sci., 17, 45, 1973
111. R.M. Holsto-Miettiner, J.Y. Seppala, O.T. Ikkala, I. T. Reima, Polym. Eng. Sci.,
34, 395, 1994
Miscibility and phase behaviour of PTT/PC blends 201
112. W.N Kim, C.M Burns,Macromolecules, 20, 1876, 1987.
113. W.N Kim, C.M Burns, J.Appl. Polym. Sci., 34, 945, 1987.
114. J. F. Mano, Z. Denchev, A.Nogales, M. Bruix, T. A. Ezquerra, Macromol.
Mater. Eng., 288, 778, 2003.
115. J. Pionteck, S. Richter, S. Zschoche, K. Sahre , K.-F. Arndt, Acta
Polymer., 49, 192, 1998.
116. L. A. Utracki and R. Simha, Macromolecules, 37, 10123, 2004.
117. G. Dlubek, J. Pionteck, Y. Yu, S. Thranert, M. Elsayed, E. Badawi, R. K.
Rehberg, Macromol. Chem. Phys., 209, 1920, 2008.
118. W. Brostow in: Brostow W and Corneliussen R (eds) Failure of Plastics,
Hanser, Munich Chapter 10. (1986)
119. I. Sanchez in: Paul D and Newman S Polymer Blends Acad press, New
York, Chapter 3. (1978)
120. J.M. Berry, W. Brostow, M. Hess, E.G. Jacobs, Polymer, 39, 4081, 1998.
121. I. A. Ronova, E. M. Rozhkov, A. Yu. Alentiev, Yu. P. Yamploskii,
Macromol. Theory. Simul., 12, 425, 2003.
122. C. Nagel, K. Gu¨nter-Schade, D. Fritsch, T. Strunsky, F. Faupel,
Macromolecules, , 35, 2071, 2002
123. G. Dlubek, D. Kilburn, M. A. Alam, Electrochem. Acta, 49, 5241, 2004.
124. D. Bamford, A. Reiche, G. Dlubek, F. Alloin, J.-Y. Sanchez, M. A. Alam, J.
Chem. Phys., 118, 9420, 2003
125. M. Fernández, M. E. Muñoz, A. Santamaría, S. Syrjälä, J. Aho, Polym.
Test., 28, 109, 2009.
126. Y. Sato, H. Hashiguchi, K. Inohara, S.Takishima, H. Masuoka , Fluid
Phase Equilibria,257, 124, 2007.
127. L.A. Utracki, Polymer Blends Handbook, Kluwer Academic Publishers,
Netherlands (2003).
202 Chapter 3
128. B.D. Favis, in Polymer Blends, Volume 1, D.R. Paul, C.B. Bucknall, Eds.,
Wiley Interscience, New York, (2000).
129. S. Wu. Polymer interfaces and adhesion. New York and Basel: Marcel
Dekker Inc.; (1982).
130. I.C. Sanchez, L.E. Fitzpatrick, editors. Physics of polymer surfaces and
interfaces. Boston: Butterwort-Heinemann; (1992).
131. T. Matsumoto, N. Hori, M. Takahashi, Langmuir, 12, 5563, 1996.
132. L.A. Utracki, Commercial Polymer Blends; Chapman & Hall, London, (1998).
133. T. Araki, Q. Tran-Cong, M. Shibayama, Structure and Properties of
Multiphase Polymeric Materials; Marcel Dekker Inc., New York, (1998).
134. S. Walheim, M. Ramstein, U. Steiner, Langmuir, 15, 4828, 1999.
135. H.J. Kim, Y. Seo, Langmuir, 19, 2696, 2003.
136. N.A. Frankel, A. Acrivos, Chem. Eng. Sci., 22, 847, 1967.
137. W.R. Schowalter, C.E. Chaffey, H. Brenner, J. Colloid. Interface Sci., 26, 152, 1968.
138. J.F. Palierne, Rheol. Acta, 29, 204, 1990.
139. M. Loewenberg, E.J. Hinch, J. Fluid Mech., 321, 395, 1996.
140. J.G. Oldroyd, Proc. Royal Soc., London, A218, 122, 1953.
141. S.J. Choi, W.R. Schowalter, Phys. Fluids, 18, 420, 1975.
142. H. Asthana, K. Jayaraman, Macromolecules, 32, 3412, 1999.
143. D. Shi, Z. Ke, J. Yang, Y. Gao, J. Wu, J. Yin, Macromolecules, 35, 8005,
2002.
144. Chr. Friedrich, W. Gleinser, E. Korat, D. Maier, J. Wees, J. Rheol., 39,
1411, 1995.
145. A.Y. Coran, R. Patel, J. Appl. Polym. Sci., 20, 3005, 1976.
146. W.G.F. Sengers, P. Sengupta, J.W.M. Noordermeer, S.J. Picken, A.D. Gotsis,
Polymer, 45, 8881(2004).
Miscibility and phase behaviour of PTT/PC blends 203
147. L.A. Utracki and M.R. Kamal, Polym. Engg. Sci., 22, 96, 1982.
148. Z. Hashin, Second Order Effects in Elasticity, Plasticity, and Fluid
Dynamics; M. Reiner, S. Abir., Eds.; Macmillan: New York, (1964)
149. D. Graebling, R. Muller, J.P. Palierne, Macromolecules, 26, 320, 1993.
150. D. Graebling, R. Muller, J. Rhol., 34, 193, 1990.
151. M. Bousmina, R. Muller, J. Rhol., 37, 663, 1993.
152. P.J. Carreau, M. Bousmina, A. Ajji, in K.P. Ghiggin, Ed., Progress in Pacific
Science-3, Springer-Verlag, New York, (1994).
153. C. Lacroix, M. Bousmina, P.J. Carreau, B.D. Favis, A. Michel, Polymer, 37,
2939,1996.
154. I. Vinckier, P. Moldenaers, J. Mewis, J. Rheol., 40, 613, 1996.
155. T. Jansseune, J. Mewis, P. Moldenaers, M. Minale, P.L. Maffettone, J.
Non Newtonian Fluid Mech., 93, 153, 2000.
156. I. Delaby, B. Ernst, Y. Germain, R. Muller, J. Rheol., 38, 1705, 1994.
157. U. Jacobs, M. Fahrlander, J. Winterhalter, C. Friedrich, J. Rheol., 43,
1495, 1999.
158. N.W. Tschoegle, The Phenomenological Theory of Linear Viscoelastic
Behaviour, Berlin, Springer, (1989).
159. K.A. Moly, Ph.D Thesis, Mahatma Gandhi University, 2005.
160. S. Jose, Ph.D Thesis, Mahatma Gandhi University, 2006.
161. S. V. Nair, Ph.D Thesis, Mahatma Gandhi University, 2006.
162. C. Komalan, Ph.D Thesis, Mahatma Gandhi University, 2007.
163. H. Yang, M. Lai, W. Liu, C. Sun and J. Liu. J. Appl. Polym. Sci., 85, 2600, 2002.
164. A.Y. Coran, Handbook of Elastomers, New developments and Technology
Eds. A.K. Bhowmick, H.L. Stephens, Marcel Dekker, New York, 249
(1988)
204 Chapter 3
165. A.Y Coran, in “Thermoplastic Elastomers”, N.R.Legge, G.Holden, H.E
Schroeder, eds; Hanser Publishers, Distributed by Macmillan Publishing
Co; N.Y. (1987)
166. R.A. Dickie, J. Appl. Polym. Sci., 17, 45, 1973.
167. R.M. Holsto-Miettiner, J.Y. Seppala, O.T. Ikkala, I. T. Reima, Polym. Eng. Sci.,
34, 395, 1994.
168. H. R. Radfarnia, G. M. Kontogeorgis, C. Ghotbi, V. Taghikhani, 257, 63, 2007.
169. L. Terlemezyan, P. Mokreva, D. Tsocheva, S. Peneva, K. Berovsky, T.
Troev, Rad. Phy. . Chem., 77, 591, 2008.
170. Z. Minfeng, S. Xudong, W. Yun, Y. Xiandong, X. Huiquan, W. Baoyi, Q.
Chenze, Rad. Phy. . Chem., 77, 1062, 2008.
171. M. Heuchel, D. Fritsch, P. M. Budd, N. B. McKeown, D. Hofmann, J. Mem.
Sci. 318, 84, 2008.
172. R.A. Pethrick, Prog. Polym. Sci., 22, 1, 1997.
173. O.E. Mogensen, Positron Annihilation in Chemistry; Springerverlag: Berlin,
Heidelberg, 1995.
174. P. Kirkegaard, N.J. Pedersen, and M. Eldrup, Riso Nat Lab Reports,
Denmark M- 2740, 1989.
175. H. Nakanishi, S.J.Wang, Y.C. Jean. S.C. Sharma, editor. Positron
annihilation in fluids. Singapore: World Scientific; 1988.
176. S J. Tao, J. Chem. Phys., 56, 5499, 1972.
177. M. Eldrup, D Lightbody, N Sherwood, J. Chem. Phys., 63, 1981.
178. H. B. Ravikumar, C. Ranganathaiah, G. N. Kumaraswamy, M. V. Deepa
Urs, J. H Jagannath, A. S. Bawa, Sabu Thomas J. Appl. Polym. Sci., 100,
740, 2006.
179. K.T. Varughese, K. G. Kunststoffe, 41, 114, 1988.
180. S. Stack, O. O’Donogbue, C Birkinshaw, Polym. Degrad. Stab. 79, 29, 2003.
Miscibility and phase behaviour of PTT/PC blends 205
181. N.S. Vrandečić, I. Klarić, T. Kovačić, Polym. Degrad. Stab., 84, 23, 2004.
182. S. S. Pesetskii, Y.A. Olkhov, S.P. Bagdanovich, V.N. Koval, Euro. Polym.
J. 41, 1380, 2005.
183. J. Roeder, R.U.B. Oliveira, D. Becker, M.W. Goncalves, V. Soldi, A.T.N.
Pires, Polym. Degrad. Stab., 90, 481, 2006.
184. K. Aouachria, N.B. Bensemra, Polym. Degrad. Stab., 91, 504, 2006.
185. R. Stephen, S. Jose, K. Joseph, S. Thomas, Z. Oommen, Polym. Degrad.
Stab., 91,1717,2006.
186. S. Jose, P. S. Thomas, S. Thomas, J. Karger-Kocsis, Polymer, 47, 6328, 2006.
187. P. Malik, M. Castro, C. Carrot, Polym. Degrad. Stab., 91, 634, 2006.
188. J. H. Kim, B. R. Min, Y. S. Kang, Macromolecules,39, 1297,2006.
189. Z.Guo, Z. Fang, L. Tong, Z. Xu, Polym.Degrad.Stab. 92,545,2007.
190. S. Saikrasun, T Amomsakchai, J. Appl. Polym. Sci.103, 917, 2007.
191. S. Tiptipakorn, S. Damrongsakkul, S. Ando, K. Hemvichian, S. Rimdusit,
Polym. Degrad. Stab., 92, 1265,2007.
192. R. Paul, U. Karabiyik, M. C. Swift, A. R. Esker, 24, Langmuir, 5079, 2008.
193. F. Signori, M.B. Coltelli, S. Bronco, Polym. Degrad. Stab., 94,74, 2009.
194. H.H. Horowitz, G. Metzger, Anal. Chem., 35, 1464, 1963.