1. Intercalation of the polymer or pre-polymer from solution. This method uses a solvent system in which the polymer or pre-polymer is soluble and the silicate layers are swellable. The clay is first swollen in a suitable solvent, such as water, chloroform, or toluene. When the polymer and clay solutions are mixed, the polymer chains intercalate and displace the solvent within the interlayers of the silicate. On solvent removal, the intercalated structure remains, resulting in a nanocomposite.

2. In situ intercalative polymerization. In this method, the clay is swollen within the liquid monomer or a monomer solution so polymer formation can occur between the intercalated sheets. Polymerization can be initiated either by heat or radiation, by the diffusion of a suitable initiator, or by an organic initiator or catalyst fixed through cation exchange inside the interlayer before the swelling step.

3. Melt intercalation. This method involves annealing, statically or under shear, a mixture of the polymer and organically modified clay at the softening point of the polymer. This method has great advantages over either in situ intercalative polymerization or polymer solution intercalation. First, this method is environmentally benign due to the absence of organic solvents. Second, it is compatible with current industrial processes, such as extrusion and injection molding. Melt intercalation can use polymers that are not suitable for in situ polymerization or solution intercalation.

10.3.1 Intercalation of PET from solution

Although a solvent casting method has been used to prepare a PET/clay nanocomposite, the most popular ways to prepare PET/clay nanocomposites are in situ intercalation and melt blending. Ou et al.¹² modified MMT with cetylpyridinium chloride (CPC) (CPC/clay = 2/1 by equivalent) and then prepared a PET/clay nanocomposite using a phenol/chloroform mixture as the solvent. The paper contains more details of the method used to produce the nanocomposites from PET.

10.3.2 In situ intercalative polymerization

In situ polymerization to prepare PET/clay nanocomposite can be carried out in several ways. The first method is the intercalation and ring-opening polymerization of ethylene terephthalate cyclic oligomers (ETC) in pre-swelled OMMT galleries. Usually, OMMT is swelled in dichloromethane (DCM). The ETC is dissolved in DCM separately. Then the OMMT solution is added to the ETC solution under vigorous stirring followed by solvent extraction and drying resulting in an intercalated nanocomposite.⁷ Several sessions of extensive nitrogen purging and high vacuum treatment are necessary during polymerization. A PET/clay nanocomposite has been synthesized via the ester exchange reaction of ethylene glycol (EG) and dimethyl terephthalate (DMT) in the clay gallery in the presence of a zinc acetate catalyst followed by, finally, polycondensation in the presence of an antimony (III) oxide catalyst.² Here, the clay was ultrasonicated with ethylene glycol and zinc acetate before the ester interchange reaction was initiated due to the addition of DMT to this mixture.

The clay can also be added in the polycondensation step after transesterification of EG and DMT in the presence of a manganese acetate catalyst, but the reaction temperature has to be increased.³ Instead of DMT, the combination of purified terephthalic acid (PTA), antimony (III) oxide as a catalyst, and additives, such as triethyl phosphate (TEP) as a thermal stabilizer and cobalt (II) acetate tetrahydrate (CoAc) as a color inhibitor, can also be used.¹³

Jung et al.¹⁴ prepared a PET/organically modified mica hybrid by transesterification of EG and DMT in the presence of isopropyl titanate. They cooled the sample to room temperature as soon as the ester interchange reaction was finished, repeatedly washed it with water and, finally, vacuum dried it in order to obtain the PET hybrid. After compression molding of the sample they extruded it through the die of a capillary rheometer.¹³

Since polyamide 6 (PA6) has better compatibility with clay, Li et al.¹⁵ expected that, if PA6 chains were introduced into the PET, the compatibility between PET and the clay could be improved. So they synthesized a PET/PA6 copolymer/MMT nanocomposite by transesterification of DMT and EG in the presence of a zinc acetate catalyst at 180 °C, then dissolved the PA6/MMT master batch sample with an antimony trioxide catalyst in the reaction system and increased the reaction temperature to 280 °C. The results of ¹H nuclear magnetic resonance (NMR) proved that the ester amide exchange reaction had taken place and an average 3.75 repeat units of PA6 were dispersed randomly in the PET molecular chains.¹⁵

With the aim of exfoliating the clay platelets in a polyester/clay nanocomposite, Tsai et al.¹⁶ proposed a new preparative technique. They prepared three agents: Agent A contained antimony acetate dissolved in EG, Agent B contained purified clay swollen in EG, and Agent C contained sodium cocoamphohydroxypropylsulfonate dissolved in deionized water. After mixing Agents A and B, calcination resulted in modified clay intercalated with antimony oxide (Sb₂O₃). The Sb₂O₃-modified clay was added to Agent C to give Sb₂O₃-SB modified clay. Then the modified clays were mixed with terephthalic acid bis(2-hydroxyethylester) (BHET) at its melting point for 2 h to prepare the nanocomposite named PET/PK805/Sb-SB.¹⁶

In order to understand the effect of different in situ intercalation techniques on the dispersion of clay platelets in a PET matrix, transmission electron micrographs (TEM images) were used. Figure 10.1 shows TEM micrographs of (a) PET nanocomposite containing 2 wt% organically modified mica,¹⁴ (b) PET/PA6 copolymers/organically modified MMT nanocomposite containing 2 wt% clay loading,¹⁵ and (c) a PET nanocomposite based on organically modified clay (PET/PK805/Sb-SB composite) with antimony acetate as a catalyst.¹⁶ From these images, we can see that the dispersion of clay platelets in the PET/PA6 copolymers/OMMT nanocomposite (Fig. 10.1(b)) is better than the composite prepared via conventional in situ intercalation (Fig. 10.1(a)). The antimony-acetate-treated clay-based composite has better delaminated clay layers (Fig. 10.1(c)) compared with the PET/PA6 copolymers/OMMT nanocomposite. Note that different authors used different types of OMMT. However, the improvement in the dispersion of clay layers in the PET/PK805/Sb-SB composite may be attributed to the catalyst-treated clay, since antimony acetate catalyzes the polymerization reaction inside the clay galleries.

10.3.3 Melt intercalation

Since melt processing is compatible with current industrial processes, such as extrusion and injection molding, Costache et al.¹¹ modified natural clays, such as MMT, hectorite, and magadiite, with the highly thermally stable surfactants hexadecyl-quinolinium (Q16) and vinylbenzyl-ammonium chloride-larnyl-acrylate copolymer (L-surfactant). They prepared PET/clay nanocomposites by melt blending in a Brabender plasticorder for 7 min at 280 °C. The authors found that, although the modified clay has better thermal stability, the nanocomposites did not show any improvement in stability, though in some cases it decreased. This can be explained from the thermal stability results of clays at the processing temperature for the processing time. The current authors believe that during processing at high temperature the clay starts to degrade and hence the thermal stability decreases.¹¹

Processing time can be reduced significantly by preparing the nanocomposite sample in an extruder. The PET and the clay must be sufficiently dry before extrusion in order to reduce decomposition due to the moisture in the sample. So far various types of commercially available clay, such as Cloisite®Na⁺ (CNa),¹⁷ Cloisite®10A (C10A),¹⁷ Cloisite®15A (C15A),^17,18 Cloisite®20A (C20A),¹⁹ and Cloisite®25A (C25A),²⁰ have been used to prepare PET/clay nanocomposites via melt extrusion. The processing temperature (the temperature of different zones of the extruder to the die) of a PET/clay nanocomposite can vary between 240 and 285 °C. Giraldi et al.²⁰ reported that a low screw speed during processing results in composites with a high Young’s modulus, low elongation at break, strength, and impact. This is because the polymer matrix is degraded in the long residence time due to the slow speed.²⁰ Further, a nitrogen environment must be maintained in the hopper and the feeder to keep the materials dry¹⁷ and, of course, to avoid possible degradation at such a high processing temperature. Another important parameter in extrusion through a die is the draw ratio. A large draw ratio results in debonding of the organoclay and the matrix polymer and creates many nano-sized voids due to excess stretching of the fibers. These voids (bubbles) hinder energy dissipation and stress transfer under a certain strain during the measurement of mechanical properties. As a result, the modulus and strength of the material decrease. It can also affect the flexibility of the material, measured by elongation at break. N,N-dimethyl-N,N-dioctadecyl ammonium modified clay is not suitable for the preparation of PET/clay nanocomposites since degradation of this modifier makes the nanocomposite brittle. On the other hand, 1,2-dimethyl-3-N-hexadecyl imidazolium tetrafluoroborate exhibits good dispersion in the PET matrix.²⁰

Barber et al.¹⁷ melted an extruded PET ionomer and organically modified MMT directly to prepare a nanocomposite. The preparation scheme is shown in Fig. 10.2(a). The concentration of the C10A clay in the composite was 5 wt%. The ionomer used was sulfonated poly(ethylene terephthalate).^17,18

Ammala et al.²¹ tried to improve the compatibility between the clay and the PET matrix by incorporating the PET ionomer (AQ55S) in a different way. They compared the effect of the PET ionomer on the dispersion of C10A, a synthetic fluoromica clay (Somasif ME100) before modification, and quaternary ammonium modified synthetic fluoromica clay (Somasif MEE). In order to prepare the nanocomposite, they dispersed the clay in water with the PET ionomer, coated the suspension onto the solid PET, followed by removal of water, and then extruded. The preparation scheme is shown in Fig. 10.2(b). The concentration of clay in the composites was 5 wt%. The composite with the modified clay showed some discoloration compared with the composite with Somasif ME100. This may be due to degradation of the quaternary ammonium modifier during extrusion. Although the optical transparency improved in the PET/Somasif MEE nanocomposite compared with the PET/Somasif ME100 nanocomposite, the optical transparency attained its maximum value with the PET/C10A nanocomposite. This is due to the smaller particle size of C10A in comparison to the mica clays. Due to its small size, C10A will scatter less light and, hence, the optical transparency increased.²¹

Barber et al.¹¹ and Ammala et al.²¹ both used 5 wt% C10A clay and a PET ionomer to prepare a nanocomposite with PET. The Barber group used conventional melt blending (Fig. 10.2(a)), whereas the Ammala group used melt blending in a different way (Fig. 10.2(b)). Now the question is which method shows better dispersion of the clay layers in the PET matrix, since the properties of the nanocomposites are directly related to the dispersion characteristics in the PET matrix. To compare the dispersion of the clay layers in the PET matrix, TEM images of the nanocomposites prepared via two different melt blending techniques are shown in Fig. 10.3. According to this figure, the dispersion of clay platelets in the PET matrix prepared by modified melt blending is more homogeneous compared with that by the conventional melt blending method.

The dispersion of clay in a PET matrix can also be improved by equibiaxial stretching when drawing the composite through the die of the extruder. The stretching may increase the aspect ratio due to slippage of the clay platelets during stretching and preferential orientation of the clay platelets. Hence, the mechanical and barrier properties of the nanocomposites improve, which in turn is advantageous for packaging.^22,23 For example, for the same stretch ratio, the oxygen permeability coefficient of neat PET (~ 2.35 cm³-mm/(m²-day-bar)) reduces after nanocomposite preparation (~ 1.82 cm³-mm/(m²-day-bar)).²²

Therefore, when preparing a PET/clay nanocomposite one should consider the higher thermal stability, large d-spacing, and compatibility of the clay surface with the PET matrix. Compatibility can be determined by the polar solubility parameter using group contribution methods, such as the Fedors approach.²⁴ The incorporation of a PET ionomer during melt blending is also effective in improving the dispersion of clay platelets in the nanocomposite.²¹ With in situ intercalation, it is better to treat the clay with a catalyst and then allow polymerization.¹⁶

10.5.1 Thermal properties and kinetics of crystallization

The most common technique for evaluating thermal properties, such as phase changes, is differential scanning calorimetry (DSC). DSC determines the amount of heat change (either absorbed or released) when a substance undergoes physical or chemical changes. Such changes alter the internal energy (U) of the substance. At a constant pressure, the internal energy is known as enthalpy (H). Any thermal transition, e.g., a phase change, can be described in terms of the change of enthalpy between the two phases. When the internal energy of a solid is increased by the application of an external energy source, the molecular vibrations in the substance increase. As a result, the substance becomes less and less ordered. Hence the enthalpy and entropy (S) of the material increase. Melting occurs when the Gibbs free energy (G = H–TS) of the liquid is lower than that of the solid since every system seeks to attain a minimum free energy state. Crystallization is the opposite of melting. During crystallization, entropy decreases because the ordering of molecules within the system is overcompensated by the thermal randomization of the surroundings. Due to the release of heat of fusion, the entropy of the universe increases. Hence, crystallization is an exothermic process. Usually crystallization occurs at a lower temperature than melting, known as supercooling. This means a crystal can be destroyed more easily than it is formed. Crystallization consists of two phenomena: nucleation and growth. Nucleation is the onset of a phase transition in a small region. The presence of foreign materials (e.g., dispersed nanoparticles in nanocomposites) can facilitate nucleation. Crystal growth is the subsequent growth of nuclei that have reached a critical cluster size.

Hamzehlou and Katbab²⁹ described the properties of clay-based nanocomposites with bottle grade PET, and the flash PET that is left at the gate of an injection mold during preform manufacturing of PET bottles from neat PET by injection molding (which is known as recycled PET). The clay used for this purpose was MMT organically modified by methyl tallow bis(2-hydroxyethyl) ammonium salt. According to these authors, the recycled-PET-based nanocomposites showed an improvement in thermal properties. Although there was no change in melting temperature, the glass transition temperature (T_g) and crystallization temperature (T_c) shifted towards higher temperatures because they are quite thermally stable systems.²⁹ Even for this particular case, the intrinsic viscosity remained unaltered in the nanocomposites when compared with the recycled PET. However, a change in the organic modifier of the clay to dimethyl dehydrogenated tallow ammonium salt resulted in a decrease in T_g due to clay agglomeration and may be due to degradation of the polymer matrix during processing.³¹ T_g for the nanocomposite may remain the same as that of the neat polymer if the composite has a homogeneous intercalated structure. In such cases, the degree of crystallinity decreases in the nanocomposite relative to the neat polymer.¹⁸ For any industrial application, it is essential to achieve high nucleation efficiency and a low degree of crystallinity in the polymer nanocomposite compared with the matrix polymer because higher nucleation efficiency allows preform injection molding, blow molding, etc., at higher temperatures. On the other hand, a reduction in the degree of crystallinity makes the material tougher.

The melting and crystallization behaviors of neat PET and PET/C20A nanocomposite samples were investigated by both conventional and temperature modulated DSC.¹³ The results indicate that a clay particle acts as a nucleating agent in the crystallization of the PET. To find the effect of clay on the cold crystallization of neat PET, conventional DSC and temperature-modulated DSC (TMDSC) of melt-quenched samples were also carried out. With nanocomposites, cold crystallization was accompanied by a small degree of fusion and subsequent crystallization in the reversal component during TMDSC. This may be due to the presence of short polymer chains formed by the degradation of the matrix at a high temperature in the presence of clay. These small chains undergo a small amount of fusion during crystallization of the bulk. The shift of the cold crystallization peak temperature (T_cc) of PET toward lower temperatures in nanocomposites suggested the intercalated silicate layers act as nucleating agents, so that crystallization of the matrix starts sooner. However, the decrease in enthalpy of cold crystallization (ΔH_cc) of the nanocomposites confirmed that, although the clay particles act as nucleating agents, the nanocomposites lose some crystallizable moiety due to the intercalation of the polymer chains in the clay galleries. A further increase in temperature resulted in melting with re-crystallization. In the quenched state, the initial percentages of crystallinity in the nanocomposites were higher than in the neat PET.

Crystallization can simultaneously involve varying rates of nucleation, growth, aggregation breakage, re-dissolution of unstable crystal clusters, and filler concentration. The result is a complex process that is irreversible and often difficult to predict accurately – especially during scale-up or in technology transfer to alternative equipment. For polymers, however, kinetic control is the dominant factor. Polymer crystallization only occurs at a reasonable rate at temperatures well below the equilibrium melting temperature. The reasons for this behavior will emerge from the discussion of crystallization kinetics. There are several different kinds of experimental method that are commonly used to observe the time development of crystallinity (crystallization kinetics) in polymers. One method assesses the cooling rate at which the total amount of crystallinity develops from the supercooled liquid, known as non-isothermal crystallization. Another method assigns a time for crystal growth at a certain temperature near or above the supercooled phase, known as isothermal crystallization. The study of non-isothermal crystallization is more important because most current processing techniques, such as injection molding, blow molding, etc., with polymeric materials follow non-isothermal crystallization. The Avrami model³² was developed for isothermal crystallization kinetics; Ozawa³³ extended the Avrami equation to non-isothermal crystallization.³⁴ The Liu model, which is applicable for non-isothermal crystallization kinetics, is a combination of the Avrami and Ozawa models.³⁵ Prior to the study of crystallization kinetics using these models, it is necessary to determine the relative degree of crystallinity as a function of temperature and time.

The relative degree of crystallinity, X_T can be defined as:

[10.2]

where T_o (T_{c, on}) and T_α (T_{c, fin}) are, respectively, the initial and final temperatures of crystallization. For example, X_T as a function of temperature for neat PET and nanocomposites with different clay loadings at different cooling rates during non-isothermal crystallization are plotted in Fig. 10.5. To see the change of the relative degree of crystallinity as a function of time (X_t), the temperature scale was transformed into a time scale using Eq. 10.3 and the equivalent time dependences of the fractional crystallinity are plotted in Fig. 10.6.

[10.3]

where T_o is the onset temperature of crystallization and T is the same temperature used to determine X_T. As a function of time or temperature, the relative degree of crystallinity remains the same; the suffix only denotes the abscissa.

By truncating non-isothermal crystallization into an infinitesimally small isothermal process, Ozawa extended the Avrami model for isothermal crystallization to analyze non-isothermal crystallization kinetics. According to this model, X_T can be written as a function of the cooling rate:

[10.4]

where K(T) is the Ozawa crystallization rate constant and m is the Ozawa exponent depending on the dimension of crystal growth. Taking a double logarithm on both sides of Eq. 10.4:

[10.5]

Therefore, a plot of ln[–ln(1–X_T)] versus ln ϕ should be a straight line if this model is valid. K(T) and m can be estimated from the antilogarithms of the y-intercept and slope, respectively.

Although the Ozawa model is designed to analyze non-isothermal crystallization kinetics, this model is not valid for PET/C20A nanocomposites since this model does not take into account the secondary crystallization process (impingement of crystals). Therefore, the Ozawa model fails to describe non-isothermal crystallization kinetics with secondary crystallization. As an alternative approach, the Avrami model can be used to explain crystallization kinetics for such systems.^34,36

According to the Avrami model, the equivalent time-dependent crystallinity can be expressed as:

[10.6]

where Z_t is a composite rate constant involving both nucleation and growth rate parameters and the Avrami exponent, n, is a constant depending on the type of nucleation and the growth process. Taking a double logarithm on both sides of Eq. 10.6:

[10.7]

Equation 10.7 should be a straight line if this model is valid, and Z_t and n can be determined from the antilogarithms of the y-intercept and slope, respectively. Note that, during analysis of non-isothermal crystallization by this model, Z_t and n do not possess the same physical meaning as in the original Avrami analysis for isothermal crystallization because the temperature changes instantaneously in the non-isothermal process. Here they are adjustable parameters to fit the experimental results and they help to analyze the crystallization kinetics. However, the variation of the Avrami exponent with cooling rate implies that crystallization of the sample has occurred in various growth forms.³⁷ According to this model, the primary and secondary crystallization of PET produce almost the same size of crystallites. Hence, there is almost no deviation from straight lines in the Avrami plot as shown in Fig. 10.7(a). In nanocomposites, although crystallization starts by nucleation in the presence of foreign materials, the dispersed clay layers hinder impingement and, as a result, the growth of small crystallites occurs due to secondary crystallization. Actually, the different size distribution of crystallites is responsible for the deviation of the linear portion of the Avrami plot as shown in Fig. 10.7(b) and Fig. 10.7(c).

Jeziorny suggested the parameter Z_t should be modified when Avrami analysis is applied to non-isothermal crystallization kinetics. Assuming a constant or almost constant cooling rate, the final form of this parameter as suggested by Jeziorny is:³⁸

[10.8]

The combined Avrami and Ozawa model, i.e., the Liu model, is given by Eqs 10.9 and 10.10, and is also valid for the current system as shown in Fig. 10.8.

[10.9]

By rearranging Eq. 10.9, the final form of the Liu model is:^35,39

[10.10]

where F(T) = [K(T)/Zt]^1/m is for the cooling rate to reach a defined degree of crystallinity and a is the ratio of the Avrami exponent to the Ozawa exponent, i.e., a = n/m. For a given degree of crystallinity, F(T) and a can be determined from the y-intercept and slope of the straight lines defined by Eq. 10.10.

The different kinetic parameters determined from these models proved that, in nanocomposites, organoclay is efficient at initiating crystallization earlier by nucleation, but crystal growth will decrease in nanocomposites due to intercalation of the polymer chains in the silicate galleries. Wang et al.⁴⁰ also observed similar non-isothermal crystallization kinetics as for PET/C20A nanocomposites in their systems. If there is a strong interfacial interaction between the clay and the matrix PET, the polymer segmental motion reduces and hence the crystal growth rate decreases either because the crystals form more slowly or because there is less overall crystallization, i.e., the n value reduces.²⁷

Similar crystal growth with secondary crystallization has also been observed during isothermal crystallization.³⁰ For neat PET, the average value of the Avrami exponent, n, is 3. This implies that crystal growth is three dimensional (spherulitic). This value changes in nanocomposites depending on secondary crystallization.^41,42 Fourier transform infrared spectroscopy (FTIR) can be used to determine crystalline structure and perfection. In some nanocomposites, the crystal lamellar thickness and perfection of the crystal is reduced, since the ethylene glycol segments are more likely to sit in the polymer/clay interface.⁴¹

From a technical point of view, the study of non-isothermal cold crystallization kinetics is important since it is frequently encountered in processing methods, such as the reheat stretch blow molding of bottles, heat setting, production of films and fibers, etc. The physical and mechanical properties of such products are directly or indirectly controlled by the crystallization processes.

In order to study non-isothermal cold crystallization kinetics, it is necessary to quench samples from the molten state to a temperature below T_g, and then heat them at different rates.⁴³

Crystal growth can be directly visualized using polarized optical microscopy (POM) images taken at the same isothermal or non-isothermal crystallization conditions as used in DSC.^{34,41,42,44,45}

The activation energy (ΔE) for non-isothermal crystal growth can be determined using the Augis–Bennett, Kissinger, and Takhor methods represented by the following equations.

Augis–Bennett method:⁴⁶

[10.11]

Kissinger method:⁴⁷

[10.12]

Takhor method:⁴⁸

[10.13]

The reduction of the activation energy in nanocomposites compared with the neat polymer leads to the conclusion that the nanoparticles in the nanocomposite facilitate crystal growth.³⁴

The determination of the activation energy by the Kissinger method is more reliable than that determined by the Augis–Bennett method since the Kissinger method is model free. On the other hand, there are two assumptions in the Augis–Bennett method: first, ΔE > > RT and second, (T_1/2)₁. (T_1/2)₂ ≈ T_c²;⁴⁹ where (T_1/2)₁ and (T_1/2)₂ are the temperatures at the half maximum before and after the crystallization peak, respectively. Determination of the activation energy by the Augis–Bennett method is really fruitful if the reaction obeys the Avrami law.⁴⁹

10.5.2 Thermal stability

Although some surfactants used to modify pristine clays are very thermally stable, there is a chance of degradation of the surfactant during nanocomposite preparation due to the high processing temperature of PET.^19,28 Therefore, to reduce the possibility of organoclay degradation, it is better to dry the clay overnight under vacuum before processing and to minimize the processing time. The processing time can be reduced significantly by preparing the sample in an extruder. The processing time can be estimated from a study of isothermal degradation of the organoclay at the processing temperature.¹⁹ Further, under oxidative conditions a PET/C20A nanocomposite sample exhibits a two-step decomposition. In the first half of the degradation process, the nanocomposite exhibits less onset thermal stability than neat PET. This is due to the degradation of the surfactant used for the modification of MMT,^50,51 as alkyl ammonium modifiers are known to undergo Hoffman degradation at around 200 °C.⁵² As for the oxidative condition, under an inert atmosphere the nanocomposite samples also exhibit less onset thermal stability (at 10% weight loss) than neat PET. However, the main degradation temperature for the nanocomposite samples increased in air compared with a nitrogen atmosphere. It is possible that the different types of char formation mechanism under an oxidative environment actually slow down oxygen diffusion, thus hindering oxidation under thermo-oxidative conditions. This observation suggests that there is improved flame retardancy for nanocomposites.¹³ The same behavior was also obtained for PET/C30B nanocomposites.²⁸ On the other hand, phosphonium-modified MMT with low phosphonium content exhibits an improvement in thermal stability of the PET/clay nanocomposite when compared with the neat PET.^28,53,54 Different phosphonium salts, such as (4-carboxybutyl) triphenylphosphonium bromide⁵³ and dodecyltriphenyl phosphonium chloride,^54,55 can be used to modify the natural MMT. The thermal stability of the PET/ammonium-salt-modified MMT nanocomposites can be improved by using recycled PET instead of neat PET.²⁹ An imidazolium-surfactant-modified MMT also enhances the thermal stability of PET/clay nanocomposites.^56,57 Aminosilane- and imidosilane-modified palygorskite clay-based PET nanocomposites also show an improvement in thermal stability compared with the neat PET resin.⁵⁸ However, it has already been proved, at the same level of loading, that clay-based PET composites exhibit much more thermal stability compared with composites based on silica nanoparticles.⁵⁹ Because of its high aspect ratio, the clay has a much higher surface area for polymer–filler interactions, which in turn allows better dispersion of the clay layers in the polymer matrix than spherical nanoparticles. Due to this better dispersion, the tortuous path decelerates the permeation of oxygen gas and enhances the thermal stability of the PET/clay nanocomposite.

The main factors for polymer degradation in clay-based nanocomposites are the number of hydroxyl groups on the edge of the clay platelets and the ammonium linkage on the clay.⁶⁰ For example, acid-treated sodium MMT (H-MMT) has a reduced thermal stability of the PET matrix due to the larger number of Brensted acid sites generated by the acid treatment (Fig. 10.9). On the other hand, silane-modified MMT (S-MMT) shows less degradation of the PET matrix during preparation of the nanocomposite. After silane modification, the gallery spacing of the clay remains unaltered since the silane coupling agent was grafted onto the sides of the clay layers as shown in Fig. 10.9. Therefore, ammonium modification is necessary. However, the preparation of ammonium-modified clay (A-MMT) further accelerates the degradation of the PET since the ammonium modifiers take part in the Hoffman elimination reaction, which produces more Brønsted acid sites. The effect of ammonium modification on the degradation of PET can be lowered by washing the modified clay with ethanol (W-A-MMT, Fig. 10.9) or adding a silane grafting agent (S-A-MMT in Fig. 10.9).

We have discussed the thermal stability of clay-based PET nanocomposites. What happens if the clay has fibrous rod-like structures instead of layered ones? Yuan et al.⁶¹ described the thermal stability of organically modified fibrous silicate (palygorskite) (modified by water-soluble polyvinylpyrrolidone)/PET nanocomposites. According to these researchers, the thermal stability of a nanocomposite containing 3 wt% of clay exhibited an improvement in the onset degradation temperature for all heating rates under air, as shown in Table 10.1. On the other hand, under a nitrogen atmosphere the nanocomposite exhibited a reduction of the onset degradation temperature for slower heating rates; however, stability increased at higher heating rates. Due to the coexistence of the barrier effect and catalytic decomposition, such nanocomposites exhibit a complex degradation mechanism. For both atmospheres, the catalytic decomposition reaction relied on the metal cations dissociating from the chemical constitution of the clay crystal structure, which mainly exists in the inner part of the matrix. Hence, the thermal degradation reaction should principally originate from the inner part of the matrix. Regardless of the atmospheres examined, the well-dispersed silicate layers in the PET matrix hindered thermal transport from the surface to the inner matrix and weakened the thermal transduction rate. Furthermore, the barrier effect of the nanocomposite was also ascribed as being due to blocking by the thermal degradation products volatized from the inner matrix to the external environment. However, compared with nitrogen, in air there was an additional path for oxygen from the outside to the interior of the nanocomposite, as shown in Fig. 10.10 by the black solid arrows. The delaminated clay also acted as an obstacle in the transfer of oxygen gas, which enhanced the barrier effect of the nanocomposite compared with the nitrogen environment. The additive barrier effect of the nanocomposite is the primary reason for the increasing thermal oxidative stability of the nanocomposite at the first stage for all heating rates in air compared with nitrogen.

In order to avoid decomposition of the organic modifier of organoclay at the processing temperature of PET, Wang et al.⁶² first prepared a master batch of PET and sodium MMT (PET:MMT = 50:50) by solid-state shear milling. Then they prepared the nanocomposites by extruding the neat PET with a varying proportion of the master batch. Although the nanocomposites had intercalated structures, the thermal stability of the nanocomposites improved compared with the neat PET resin.⁶² They also compared the thermal stability of PET/sodium-MMT nanocomposites prepared by simple extrusion (PETCNx where x is the clay content) with nanocomposites prepared by extrusion of the PET/MMT master batch (PETSNx). PETSNx showed better thermal stability compared with PETCN with the same amount of clay loading.

Therefore, to prepare a PET/organically modified MMT nanocomposite one must choose a surfactant (used for organic modification) with a high thermal stability. Otherwise, degradation of the surfactant will facilitate the collapse of the clay layers. As a result, intercalation of the polymer chains in the clay gallery will be difficult. This will result in an intercalated nanocomposite, possibly even a phase-separated composite material. The collapse of the clay structures results in much better heat transfer between the polymer and the clay. This accelerates the degradation of the matrix polymer during processing.

10.5.3 Mechanical properties

Generally, the mechanical properties of nanocomposites depend on the extent of delamination of the nanoparticles (particularly nanoclay) in the polymer matrix, which actually depends on the processing conditions and the chemical treatment and organic modification of the clay surface.⁶³ Better dispersion and orientation of the nanoclay in the direction of application of the mechanical force promotes better stress transfer from one end to the other of the sample.⁶⁴ As a result, the nanocomposites exhibit enhanced mechanical properties when compared with the neat polymer.

The tensile property of a phosphonium-modified MMT-containing PET nanocomposite exhibited a significant improvement in modulus, but on the other hand the strength and elongation at break decreased a lot. Usually, clay has a very high modulus compared with polymers. Therefore, it is expected that the incorporation of clay in a polymer matrix will result in an improved modulus for the nanocomposite. If the clay layers are not delaminated homogeneously, the energy-dissipation mechanism will be hindered, which in turn will be responsible for a reduction of the elongation at break.

The tensile strength can be improved by preparing the nanocomposite with recycled PET instead of neat PET (Fig. 10.11(a)).²⁹ In this figure neat PET is labeled as FPET and recycled PET as RPET. The tensile strength is the ultimate capacity of the material to resist a tensile load regardless of deflection. Figure 10.11(a) shows that the highest value of the tensile strength can be achieved with 3 wt% clay loading. The RPET-based nanocomposite containing 3 wt% clay loading has the maximum tensile strength. According to the current authors, the probable reason for this improvement can be explained as follows. During injection molding or melt stretch molding, the polymer chains become disentangled or partially oriented. This disentangled structure is responsible for a better dispersion of the clay platelets in the recycled PET matrix. However, according to XRD results, the nanocomposites still possess intercalated structures. As a result, the energy dissipation mechanism improves slightly and hence the tensile strength. However, as shown in Fig. 10.11(b), the tensile modulus is lower for a RPET-based composite compared with a FPET-based composite when the clay loading is below 5 wt%. The tensile modulus is a measure of the stiffness of an isotropic elastic material. It is defined as the ratio of the uniaxial stress over the uniaxial strain. It is determined from the slope of the stress-strain curve traced during tensile tests conducted on a sample of the material. Now it is interesting to note that the tensile strength decreases above 3 wt% clay loading but the modulus of the composites still increases above this value. Although the authors only reported the TEM image of the composite containing the 3 wt% clay loading, the dispersion of the clay in the composite containing a higher clay loading can be inferred from the results of tensile property measurements. The improvement of tensile strength up to 3 wt% OMMT loading indicates that the intercalation of PET chains in the clay gallery reaches a maximum for this specific composite. After that the stacking of clay layers increases. This hinders the energy dissipation mechanism during tensile testing. As a result, the tensile strength decreases with higher clay loadings beyond 3 wt%. On the other hand, the modulus increases as the clay loading increases above 3 wt% due to the high modulus of clay itself.

PET/silane-modified organoclay composites exhibit improved tensile properties, which can be useful for fiber and film applications.⁶⁵ Nanocomposites prepared via solid-state shear milling (PETSNx, as mentioned above) can exhibit increased strength compared with nanocomposites prepared via simple extrusion (PETCNx) (Fig. 10.12(a)). However, tensile strength increases systematically with an increase in clay loading in both samples. Since, according to the TEM images, the dispersion of PETSNx is better than PETCNx, it is expected that the modulus and strength of the PETSNx will be higher in this case due to the high modulus of the dispersed silicate layers. However, there is no significant change in elongation at break (Fig. 10.12(b)) between the composites prepared via the two different techniques.⁵⁴ For the PET/C20A nanocomposite such improvements can be achieved with 1 wt% clay loading. The reason is that the enhanced amorphous orientation of the intercalated clay layers leads to improvements in modulus and strength.⁶⁶

As well as the tensile properties, the thermomechanical properties of clay-based PET nanocomposites were studied by dynamic mechanical analysis (DMA). The variation of storage and loss moduli as a function of temperature explains how the rigidity of the nanocomposite material varies with clay content and how clay concentration affects the glass transition temperature of the matrix.⁶⁷

10.5.4 Rheological properties

Rheological studies on PET nanocomposites are scarce, but exhibit very interesting features.⁵⁵ A rheological analysis of a polymer/clay nanocomposite is a powerful and complementary technique for characterizing the state of clay dispersion due to the sensitivity of the rheological response to the structure and surface characteristics of the dispersed phase.⁶⁰ The advantages of rheological characterization over electron microscopy and X-ray scattering reside in the fact that measurements are performed of the molten state and that rheological methods can be used under both linear and non-linear deformation. A disadvantage of rheological methods is that they only provide indirect information on the structures of nanocomposites.⁶⁸

Vidotti et al.⁶⁸ studied the rheological properties of C20A (o-MMT)-based PET nanocomposites in the presence of a polyester ionomer (PETi). The melt-state rheological properties were measured by a stress-controlled rheometer with parallel plate geometry (25 mm diameter) at a temperature of 280 °C. Figure 10.13 show the complex viscosity (η*) as a function of angular frequency (ω) for the neat and extruded PET and PET/PETi/o-MMT nanocomposites containing 1, 3, and 5 wt% of o-MMT. Figure 10.13(a) shows that both the neat and extruded PET exhibited a Newtonian behavior (η* being nearly constant) over the whole frequency range studied. The complex viscosity of all the PET/PETi/o-MMT nanocomposites containing 1 wt% of clay was smaller than that of the neat PET, but similar to or higher than that of the extruded PET. Moreover, the complex viscosity increased when increasing the amount of PETi in the nanocomposite. This observation indicates that an increase in the amount of PETi in a nanocomposite improves the state of dispersion of the clay particles in the PET matrix. A decrease in viscosity of the extruded PET compared with that of the neat PET indicates that the PET had undergone degradation during extrusion.^63,68 Parts (b) and (c) of Fig. 10.13 show that at very low frequencies the complex viscosity of the PET nanocomposites was higher than that of the neat PET and much higher than that of the extruded PET. As shown in Fig. 10.14, the elastic modulus (G′) at low frequencies for the nanocomposites is almost independent of frequency. This is typical of solid-like behavior and indicates the formation of a percolated network, which in the case of nanocomposites implies good dispersion of the organoclay.

Giraldi et al.²⁰ described how the intrinsic viscosity of recycled PET changes with the screw speed in the extruder and also in the presence or absence of an antioxidant (Table 10.2). Higher screw speeds lead to higher viscosity. Moreover, the viscosity increases in the presence of an antioxidant for both screw speeds. However, they did not report the same result for nanocomposites.²⁰

Usually a homopolymer exhibits Newtonian behavior in a complex viscosity versus frequency plot. In comparison, the complex viscosity of nanocomposites increases significantly in the low frequency region. This shear thinning phenomenon observed for nanocomposites in the low frequency region is initially due to the disruption of the network structure and then in a later stage by the orientation of filler particles in the flow direction.⁶⁵ For a silane-modified commercially available organoclay, the complex viscosity either increases or decreases depending on the silane coupling agent. However, if the complex viscosity of a nanocomposite exceeds the complex viscosity of the neat polymer, then it is expected that this is because the degradation of the matrix was significantly reduced during processing and rheological property measurements.⁶⁵ An inert (e.g., nitrogen) atmosphere also plays an important role in achieving this. However, during silanization, particularly in the case of C10A and C30B, the retention of water within silicate layers via a chemical reaction between the organic groups of the organoclay and the silane modifier can reduce the melt strength of the composite material.⁶⁵

It has been reported that clay nanoparticles suppress crystallization and orientation of the PET in high-speed melt spinning. On the other hand, polyhedral oligomeric silsesquioxane (POSS) nanoparticles slightly promote the crystallization of PET even in high-speed melt spinning. Hence, Lee et al.⁶⁹ studied the shear-induced crystallization behavior of PET/clay nanocomposites and PET/POSS nanocomposites. They melted samples at 280 °C for 5 min in the 1 mm gap of parallel plates (diameter of the plates was 25 mm) and then performed time sweep experiments at 220, 210, and 200 °C over an angular frequency range 0.5 to 1 rad/s. A comparison of the shear-induced crystallization behavior of PET/clay nanocomposites and PET/POSS nanocomposites is shown in Fig. 10.15. According to this figure, the increase of G′ at the early stage of crystallization is almost negligible. An abrupt increase of G′ follows in a few minutes because the homogeneous melt becomes a heterogeneous system with the formation and growth of crystallites. Hence G′ increases with time. In addition, the clay nanoparticles are more effective in the crystallization process than POSS nanoparticles.⁶⁹

10.5.5 Flame retardancy

The wide application of PET is sometimes hindered due to its combustibility, and the increasing uses of PET also place emphasis on improving flame retardancy. Incorporating a chemically reactive phosphorus-containing monomer into the polymer chain is one of the most efficient methods of improving the flame retardancy of polyesters.⁷⁰

On the other hand, the study of polymer/clay nanocomposites has become even more attractive due to the demonstrations of their flame-retardant properties, which mainly demonstrate a significant decrease in the peak heat release rate (PHRR), a change in char structure, and a decrease in the rate of mass loss during combustion in a cone calorimeter.⁷⁰ Not only was reduced flammability obtained at very low organoclay contents (2–5 wt%) without increasing carbon monoxide or smoke yields, but also the physical properties of the polymers improved simultaneously. Moreover, the incorporation of nanoclay particles with other flame retardants allows a significant portion of a conventional flame retardant to be removed from the formulated polymer products while maintaining or improving flammability performance and enhancing the physical properties. So, the preparation of nanocomposites is another tool for chemists who are engaged in the research and development of flame-retardant materials.⁷⁰

A novel phosphorus-containing flame-retardant copolyester/MMT nanocomposite (PET-co-HPPPA/O-MMT) was synthesized by in situ intercalation polycondensation of terephthalic acid, ethylene glycol, and 2-carboxyethyl(phenyl phosphinic) acid (HPPPA) with OMMT.⁷⁰ The flame retardancy of the samples was characterized by the limiting oxygen index test (LOI) and the UL-94 vertical test. The LOI is defined as the minimum fraction of oxygen (O₂) in a mixture of O₂ and nitrogen (N₂) that will just support flaming combustion. It is an important and representative parameter for describing the flame-retardant properties of a material. UL-94 ratings are used to describe the ease with which a polymer can be burned or extinguished. LOI and UL-94 test results are shown in Fig. 10.16 and Table 10.3, respectively. It can be seen that PET-co-HPPPA/O-MMT nanocomposites exhibit much better flame retardancy than PET-co-HPPPA. It is interesting that there is a considerable increase in LOI (from 31.4 to 34.0) at very low OMMT content (1 wt%), while for a further increase in OMMT content (from 1 wt% to 3 wt%) no further increase in LOI was observed. Furthermore, the UL-94 vertical test results show that the after-flame time of each sample was 0 s. When the OMMT content is lower than 2 wt%, the samples cannot reach a rating of V-0 because the surgical cotton ignites. However, when the amount of OMMT increases to 2 wt%, the sample (PET5-2) can achieve the rating of V-0 in the UL-94 test, which also indicates that the incorporation of nanoclay particles with HPPPA improves flame retardancy. The higher loadings of MMT are not necessary for improving the LOI, but, by increasing the viscosity of the burning polymer, they reduce its tendency to drip with fire. This phenomenon can be explained by the following flame-retardant mechanism: the carbonaceous–silicate char builds up on the surface during burning, and insulates the underlying material, reducing the mass loss rate of the decomposition products.⁷⁰

10.5.6 Gas barrier properties

PET is a widely used packaging material for beverages and drugs. However, its application could be broadened if packaging materials could be made more sensitive to oxygen and carbon dioxide by improving the barrier property. One approach for achieving this is in the preparation of PET/clay nanocomposites. If the clay layers are dispersed nicely in the polymer matrix, a tortuous path will be created by the homogeneously dispersed clay layers in the nanocomposite, as shown in Fig. 10.17,¹ which actually reduces the permeability of oxygen and carbon dioxide. For example, the Nanolin DK₂ clay-based PET nanocomposite exhibits an improved gas barrier property compared with a C15A-based nanocomposite.¹⁸ Tsai et al.¹⁶ showed that the barrier property against carbon dioxide gas can be improved by in situ polymerization of a monomer/oligomer in the clay gallery in the presence of an antimony acetate catalyst. According to Fig. 10.18, it is clear that the addition of a few wt% OMMT in the PET matrix can effectively reduce the oxygen (O₂) gas permeability of a PET film. When the content of the OMMT reached 3 wt%, the permeation of O₂ was reduced to half of the neat PET film.⁷¹ Equibiaxial stretching sometimes promotes the gas barrier property since the effective length of the clay tactoids increases due to stretching.²²

On the other hand, recycled PET-based nanocomposites show some property improvement, but the oxygen gas permeability decreases significantly for such composites.²⁹

1. Sinha Ray, S., Okamoto, M. Polymer/clay nanocomposites: A review from preparation to processing. Prog Polym Sci. 2003; 28:1539–1641.

2. Sinha Ray, S., Bousmina, M. Biodegradable polymers/layered silicate nanocomposites: in greening the 21st century materials science. Prog Polym Sci. 2003; 50:962–1035.

3. Bogue, R. Nanocomposites: A review of technology and application. Assembly Automation. 2011; 31:106–112.

4. Camargo, P.H.C., Satyanarayana, K.G., Wypych, F. Nanocomposites: Synthesis, structure, properties and new application opportunities. Mat Res. 12, 2009. [Print version ISSN 1516–1439].

5. Yin, M., Li, C., Guan, G., Yuan, X., Zhang, D., Xiao, Y. In-situ synthesis of poly(ethylene terephthalate)/clay nanocomposites using TiO₂/SiO₂ sol-intercalated montmorillonite as polycondensation catalyst. Polym Eng Sci. 2009; 1562–1572.

6. www.worldofplastic.net/PolyethyleneTerephthalate.htm, Types of Plastic – Worldofplastic.net, downloaded on 24 October 2011.

7. Lee, S.-S., Ma, Y.T., Rhee, H.-W., Kim, J. Exfoliation of clay facilitated by ring-opening reaction of cyclic oligomers in PET-clay nanocomposites. Polymer. 2005; 46:2201–2210.

8. www.petmachine.in/advantages_of_pet.htm’ Advantages of PET – Plastic PET, downloaded on 24 October 2011.

9. Hwang, S.Y., Lee, W.D., Lim, J.S., Park, K.H., Im, S.S. Dispersibility of clay and crystallization kinetics for in situ polymerized PET/pristine and modified montmorillonite nanocomposites. J Polym Sci: Part B: Polym Phys. 2008; 46:1022–1035.

10. Yuan, X., Li, C., Guan, G., Liu, X., Xiao, Y., Zhang, D. Synthesis and characterization of poly(ethylene terephthalate)/attapulgite nanocomposites. J Appl Polym Sci. 2007; 103:1279–1286.

11. Costache, M.C., Heidecker, M.J., Manias, E., Wilkie, C.A. Preparation and characterization of poly(ethylene terephthalate)/clay nanocomposites by melt blending using thermally stable surfactants. Polym Adv Techno. 2006; 17:764–771.

12. Ou, C.F., Ho, M.T., Lin, J.R. Synthesis and characterization of poly(ethylene terephthalate) nanocomposites with organoclay. J Appl Polym Sci. 2004; 91:140–145.

13. Kim, S.-H., Park, S.-H., Kim, S.-C. Novel clay treatment and preparation of poly(ethylene terephthalate)/clay nanocomposite by in situ polymerization. Polym Bull. 2005; 53:285–292.

14. Jung, M.-H., Chang, J.-H., Kim, J.-C. Poly(ethylene terephthalate) nanocomposite fibers with organomica via in situ intercalation. Polym Eng Sci. 2007; 1820–1826.

15. Li, C., Xiao, Y., Guan, G., Liu, X., Zhang, D. Preparation and properties of PET/PA6 copolymer/montmorillonite hybrid nanocomposite. J Appl Polym Sci. 2006; 101:2512–2517.

16. Tsai, T.-Y., Li, C.-H., Chang, C.-H., Cheng, W.-H., Hwang, C.-L., Wu, R.-J. Preparation of exfoliated polyester/clay nanocomposites. Adv Mater. 2005; 17:1769–1773.

17. Barber, G.D., Calhoun, B.H., Moore, R.B. Polyethylene terephthalate) ionomer based clay nanocomposites produced via melt extrusion. Polymer. 2005; 46:6706–6714.

18. Frounchi, M., Dourbash, A. Oxygen barrier properties of poly(ethylene terephthalate) nanocomposite films. Macromol Mater Eng. 2009; 294:68–74.

19. Bandypadhyay, J., Sinha Ray, S., Bousmina, M. Thermal and thermo-mechanical properties of poly(ethylene terephthalate) nanocomposites. J Ind Eng Chem. 2007; 13:614–623.

20. Giraldi ALF de, M., Bizarria, M.T.M., Silva, A.A., Velasco, J.I., d’Ávila, M.A., Mei, L.H.I. Effects of extrusion conditions on the properties of recycled poly(ethylene terephthalate)/nanoclay nanocomposites prepared by a twin-screw extruder. J Appl Polym Sci. 2008; 108:2252–2259.

21. Ammala, A., Bell, C., Dean, K. Poly(ethylene terephthalate) clay nanocomposites: improved dispersion based on an aqueous ionomer. Compo Sci Tech. 2008; 68:1328–1337.

22. Rajeev, R.S., Soon, K., McNally, T., Menary, G., Armstrong, C.G., Martin, P.J. Studies on the effect of equi-biaxial stretching on the exfoliation of nanoclays in poly(ethylene terephthalate). Euro Polym J. 2009; 45:332–340.

23. Rajeev, R.S., Harkin-Jones, E., Soon, K., McNally, T., Menary, G., et al. A method to study the dispersion and orientation of nanoclay tactoids in PET matrix-focused ion beam milling combined with electron microscopy. Mater Lett. 2008; 62:4118–4120.

24. Krevelen, D.W.V.Properties of Polymers. Amsterdam, the Netherlands: Elsevier, 1990.

25. Bandypadhyay, J., Maity, A., Khatua, B.B., Sinha Ray, S. Thermal and rheological properties of biodegradable poly[(butylene succinate)-co-adipate] nanocomposites. J Nanosci Nanotechnol. 2010; 10:4184–4195.

26. Kim, S.H., Kim, S.C. Synthesis and properties of poly(ethylene terephthalate)/clay nanocomposites by in situ polymerization. J Appl Polym Sci. 2007; 103:1262–1271.

27. Monemian, S.A., Goodarzi, V., Zahedi, P., Angaji, M.T. PET/imidazolium based nanocomposites via in situ polymerization: Morphological, thermal, and non-isothermal crystallization studies. Adv Polym Technol. 2007; 26:247–257.

28. Patro, T.U., Khakhar, D.V., Misra, A. Phosphonium-based clay-poly(ethylene terephthalate) nanocomposites: Stability, thermal and mechanical properties. J Appl Polym Sci. 2009; 113:1720–1732.

29. Hamzehlou, Sh., Katbab, A.A. Bottle-to-bottle recycling of PET via nanostructure formation by melt intercalation in twin screw compounder: Improved thermal, barrier, and microbiological properties. J Appl Polym Sci. 2007; 106:1375–1382.

30. Ke, Y.-C., Yang, Z.-B., Zhu, C.-F. Investigation of properties, nanostructure, and distribution in controlled polyester polymerization with clay. J Appl Polym Sci. 2002; 85:2677–2691.

31. Bizarria, M.T.M., Giraldi ALF de, M., de Carvalho, C.M., Velasco, J.I. Morphology and thermomechanical properties of recycled PET-organoclay nanocomposites. J Appl Polym Sci. 2007; 104:1839–1844.

32. Avrami, M.J. Kinetics of phase change. II Transformation-time relations for random distribution of nuclei. Chem Phys. 1940; 8:212–225.

33. Ozawa, T. Kinetics of non-isothermal crystallization. Polym. 1971; 12:150–158.

34. Bandypadhyay, J., Sinha Ray, S., Bousmina, M. Nonisothermal crystallization kinetics of polyethylene terephthalate nanocomposites. J Nanosci Nanotechnol. 2007; 8:1–11.

35. Liu, T.X., Mo, Z.S., Wang, S.E., Zhang, H.F. Nonisothermal melt and cold crystallization kinetics of poly(Ary1 Ether Ether Ketone Ketone). Poly Eng Sci. 1997; 37:568–575.

36. Chen, G.-X., Yoon, J.-S. Nonisothermal crystallization kinetics of poly(butylene succinate) composites with a twice functionalized organoclay. J Polym Sci: Part B: Polym Phys. 2005; 43:817–826.

37. Durmus, A., Ercan, N., Soyubol, G., Deligöz, H., Kaşgöz, A., Nonisothermal crystallization kinetics of poly(ethylene terephthalate)/clay nanocomposites prepared by melt processing. Polym compos 2010; 1056–1066, doi: 10.1002/pc.20892.

38. Jeziorny, A. Parameters characterizing the kinetics of the nonisothermal crystallization of poly(ethylene terephthalate) determined by d.s.c. Polymer. 1978; 19:1142–1144.

39. Yao, X., Tian, X., Zheng, K., Zhang, X., Zheng, J., Wang, R., Liu, C., Li, Y., Cui, P. Non-isothermal crystallization kinetics of poly(butylene terephthalate)/silica nanocomposites. J Macromol Sci. 2009; 48:537–549.

40. Wang, Y., Shen, C., Li, H., Chen, J. Nonisothermal melt crystallization kinetics of poly(ethylene terephthalate)/clay nanocomposites. J Appl Polym Sci. 2004; 91:308–314.

41. Wan, T., Chen, L., Chua, Y.C., Lu, X. Crystalline morphology and isothermal crystallization kinetics of poly(ethylene terephthalate)/clay nanocomposites. J Appl Polym Sci. 2004; 94:1381–1388.

42. Guan, G., Li, C., Yuan, X., Xiao, Y., Liu, X., Zhang, D. New insight into the crystallization behavior of poly(ethylene terephthalate)/clay nanocomposites. J Polym Sci: Part B: Polym Phys. 2008; 46:2380–2394.

43. Wang, Y., Shen, C., Chen, J. Nonisothermal cold crystallization kinetics of poly(ethylene terephthalate)/clay nanocomposites. Polym J. 2003; 35:884–889.

44. Calcagno, C.I.W., Mariani, C.M., Teixeira, S.R., Mauler, R.S. The effect of organic modifier of the clay on morphology and crystallization properties of PET nanocomposites. Polymer. 2007; 48:966–974.

45. Stoeffler, K., Lafleur, P.G., Denault, J. Thermal decomposition of various alkyl onium organoclays: effect on poly(ethylene terephthalate) nanocomposites’ properties. Polym Degrad Stabil. 2008; 93:132–1350.

46. Augis, J.A., Bennett, J.E. Calculation of the Avrami parameters for heterogeneous solid state reactions using a modification of the Kissinger method. J Therm Anal Cal. 1978; 13:283–292.

47. Kissinger, H.E. Reaction kinetics in differential thermal analysis. Anal Chem. 1957; 29:1702–1706.

48. Takhor, R.L., Advances in Nucleation and Crystallization of Glasses. American Chemical Society, Columbus, OH, 1971:166–172.

49. Bandypadhyay, J., Sinha Ray, S., Bousmina, M. Effect of nanoclay incorporation on the thermal properties of poly(ethylene terephthalate)/liquid crystal polymer blends. Macromol Mater Eng. 2010; 295:822–837.

50. Xie, W., Gao, Z., Pan, W.-P., Hunter, D., Singh, A., Vaia, R. Thermal degradation chemistry of alkyl quaternary ammonium montmorillonite. Chem Mater. 2001; 13:2979–2990.

51. Xie, W., Gao, Z., Liu, K.-L., Pan, W.-P., Vaia, R., et al. Thermal characterization of organically modified montmorillonite. Thermochim Acta. 2001; 367:339–350.

52. Sinha Ray, S., Bousmina, M., Okamoto, K. Structure and properties of nanocomposites based on poly(butylene succinate-co-adipate) and organically modified montmorillonite. Macromol Mater Eng. 2005; 290:759–768.

53. Lai, M.-C., Chang, K.-C., Huang, W.-C., Hsu, S.-C., Yeh, J.-M. Effect of swelling agent on the physical properties of PET-clay nanocomposite materials prepared from melt intercalation approach. J Phys Chem Solids. 2008; 69:1371–1374.

54. Chang, J.-H., Mun, M.K. Nanocomposite fibers ofpoly(ethylene terephthalate) with montmorillonite and mica: thermomechanical properties and morphology. Polym Int. 2007; 56:57–66.

55. Chang, J.-H., Mun, M.K., Lee, I.C. Poly(ethylene terephthalate) nanocomposite fibers by in situ polymerization: The thermomechanical properties and morphology. J Appl Polym Sci. 2005; 98:2009–2016.

56. Guan, G., Li, C., Zhang, D., Jin, Y. The effects of metallic derivatives released from montmorillonite on the thermal stability of poly(ethylene terephthalate)/montmorillonite nanocomposites. J Appl Polym Sci. 2006; 101:1692–1699.

57. Kráčalík, M., Studenovský, M., Mikešová, J., Sikora, A., Thomann, R., Friedrich, C. Recycled PET-organoclay nanocomposites with enhanced processing properties and thermal stability. J Appl Polym Sci. 2007; 106:2092–2100.

58. Yuan, X., Li, C., Guan, G., Xiao, Y., Zhang, D. Thermal stability of surfactants with amino and imido groups in poly(ethylene terephthalate)/clay nanocomposites. J Appl Polym Sci. 2008; 109:4112–4120.

59. Vassiliou, A.A., Chrissafis, K., Bikiaris, D.N. In situ prepared PET nanocomposites: Effect of organically modified montmorillonite and fumed silica nanoparticles on PET physical properties and thermal degradation kinetics. Thermochimica Acta. 2010; 500:21–29.

60. Xu, X., Ding, Y., Qian, Z., Wang, F., Wen, B., Zhou, H. Degradation of poly(ethylene terephthalate)/clay nanocomposites during melt extrusion: effect of clay catalysis and chain extension. Polym Degrad Stabil. 2009; 94:113–123.

61. Yuan, X., Li, C., Guan, G., Xiao, Y., Zhang, D. Thermal degradation investigation of poly(ethylene terephthalate)/fibrous silicate nanocomposites. Polym Degrad Stabil. 2008; 93:466–475.

62. Wang, G., Chen, Y., Wang, Q. Structure and properties of poly(ethylene terephthalate)/Na⁺-montmorillonite nanocomposites prepared by solid state shear milling (S³M) method. J Polym Sci: Part B: Polym Phys. 2008; 46:807–817.

63. Sánchez-Solís, A., Romero-Ibarra, I., Estrada, M.R., Calderas, F., Manero, O. Mechanical and rheological studies on poly(ethylene terephthalate)/montmorillonite nanocomposites. Polym Eng Sci. 2004; 44:1094–1102.

64. Bandypadhyay, J., Sinha Ray, S. Mechanism of enhanced tenacity in a polymer nanocomposite studied by small-angle X-ray scattering and electron microscopy. Polymer. 2010; 51:4860–4866.

65. Kráčalík, M., Studenovský, M., Mikešová, J., Sikora, A., Thomann, R., et al. Recycled PET nanocomposites improved by silanization of organoclay. J Appl Polym Sci. 2007; 106:926–937.

66. Litchfield, D.W., Baird, D.G. The role of nanoclay in the generation of poly(ethylene terephthalate) fibers with improved modulus and tenacity. Polymer. 2008; 49:5027–5036.

67. Hao, J., Lu, X., Liu, S., Lau, S.K., Chua, Y.C. Synthesis of poly(ethylene terephthalate)/clay nanocomposites using aminododecanoic acid-modified clay and a bifunctional compatibilizer. J Appl Polym Sci. 2006; 101:1057–1064.

68. Vidotti, S.E., Chinellato, A.C., Hu, G.-H., Pessan, L.A. Preparation of poly(ethylene terephthalate)/organoclay nanocomposites using a polyester ionomer as a compatibilizer. J Polym Sci: Part B: Polym Phys. 2007; 45:3084–3091.

69. Lee, S.J., Hahm, W.G., Kikutani, T., Kim, B.C., Effects of clay and POSS nanoparticles on the quiescent and shear-induced crystallization behavior of high molecular weight poly(ethylene terephthalate). Polym Eng Sci 2009; 317–323, doi: 10.1002/pen.21262.

70. Ge, X.-G., Wang, D.-Y., Wang, C., Qu, M.-H., Wang, J.-S., et al. A novel phosphorous containing copolyester/montmorillonite nanocomposites with improved flame retardancy. Euro Polym J. 2007; 43:2882–2890.

71. Ke, Z., Yongping, B. Improve the gas barrier property of the PET film with montmorillonite by in situ interlayer polymerization. Mater Lett. 2005; 59:3348–3351.

72. Yang, Y., Gu, H. Superfine structure, physical properties, and dyeability of alkaline hydro lyzed poly (ethylene terephthalate)/silica nanocomposite fibers prepared by in situ polymerization. J Appl Polym Sci. 2006; 102:3691–3697.

73. Abid, K., Dhouib, S., Sakli, F. Addition effect of nanoparticles on the mechanical properties of coated fabric. J Textile Institute. 2010; 101:443–451.

74. Minelli, M., Rocchetti, M., De Angelis, M.G., Montenero, A., Doghieri, F. Gas barrier properties of polymeric films with hybrid organic/inorganic coatings for food packing applications. AIChE Annual Meeting, San Francisco. 2006.

75. Chen, H. Smart nanotechnology in biomaterials, sensors, actuators and textiles. AIChE Annual Meeting and Fall Showcase, Cincinnati. 2005.

76. Bandyopadhyay, J., Sinha Ray, S. Determination of structural changes of dispersed clay platelets in a polymer blend during solid-state rheological property measurement by small-angle X-ray scattering. Polymer. 2011. [submitted].

77. Piegat, A., El Fray, M., Jawad, H., Qz, Chen, Boccaccini, A.R. Inhibition of calcification of polymer-ceramic composites incorporating nanocrystalline TiO₂. Adv Appl Ceram. 2008; 107:287–292.

78. Kim, M.J., Shin, D.W., Kim, J.-Y., Park, S.H., Han, I., Yoo, J.B. The production of a flexible electroluminescent device on polyethylene terephthalate films using transparent conducting carbon nanotube electrode. Carbon. 2009; 47:3461–3465.

79. Litchfield, D.W., Baird, D.G., Rim, P.B., Chen, C., Improved mechanical properties of poly(ethylene terephthalate) nanocomposite fibers. Polym Eng Sci 2010;, doi: 10.1002/pen.21758.