Chapter 3 Effect of Annealing on the Reaction Induced Miscibility …shodhganga.inflibnet.ac.in ›...

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


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


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


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


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


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,


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.


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.


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.


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.


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.


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]


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).


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


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.


∆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


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


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.


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


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).


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


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.


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


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


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


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


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


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


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


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.


.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


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


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


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.


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:


( )( )

* * 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


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


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


0 20 40 60 80 1000

100

200

300

400

500

600

700

Impa

ct s

tren

gth

(J/m

)


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


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


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


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 (%

)


Figure 3.51: Effect of blend ratio on o-Ps intensity and relative fractional free volume in PTT/PC Blends


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,


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


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.


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


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.


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.


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.


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


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).


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.


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.

Chapter 3 Effect of Annealing on the Reaction Induced Miscibility …shodhganga.inflibnet.ac.in ›...

Documents

Transcript of Chapter 3 Effect of Annealing on the Reaction Induced Miscibility …shodhganga.inflibnet.ac.in ›...