Printer Friendly

Studies of post drawing relaxation phenomena in poly(ethylene terephthalate) by infrared spectroscopy.


The orientation of thermoplastics affords an excellent approach to the improvement of their strength in the direction of elongation. In this context there have appeared numerous studies in the literature addressing changes in physical properties of polymeric systems as these are elongated. One of the most thoroughly examined systems is poly(ethylene terephthalate), PET, frequently used in the manufacture of bottles and fibers (1-5).

Most studies of PET orientation have made use of polarized infrared (IR) spectroscopy, which exploits the dependence of IR absorption efficiency on the direction of polarization relative to the direction of the transition dipole moment. It has been well established that orientation of PET is associated with conformational changes about the C-O-C bonds in the ethylene glycol moiety of the repeat unit, with the level of trans conformer increasing relative to that of the gauche as orientation increases (6-8). Furthermore, crystallographic studies have established that the crystalline domains in PET comprise solely trans conformers, whereas both are present in the amorphous domains (9). The trans conformer has been estimated to be higher in energy than the gauche by [approximately]28 J/g (10). This is small compared to the enthalpy of fusion, which has been estimated to be 140 J/g (11) for fully crystalline PET.

While many publications have dealt with the development of orientation in PET, there is a scarcity of information concerning the relaxation of orientation. A few publications have appeared discussing the results of shrinkage experiments on oriented PET (12, 13). These studies have demonstrated that beyond a certain degree of orientation shrinkage is hindered. This observation is explained in terms of orientational relaxation, which is inhibited because of the development of crystallinity (3, 12, 13).

In this paper we wish to present our results of orientational relaxation in drawn PET films, as obtained by polarized IR spectroscopy. Our results will show that the reported hindrance of shrinkage is indeed due to an inhibition of orientational relaxation.


Samples. Extrusion grade PET was obtained from DuPont (Selar PT 7086) without any nucleating agent. Its molecular weight characteristics as determined by GPC were [M.sub.n] = 28,800 and [M.sub.w] = 54,600. This material was dried at 100 [degrees] C for 24 h and then compressed at 280 [degrees] C in a Carver laboratory press. The resulting films were quenched into ice water and had a thickness prior to stretching of [approximately]40 [[micro]meter]. Thin strips were cut for subsequent stretching.

Elongation. Samples were stretched at 80 [degrees] C on an Instron tensile tester equipped with an environmental chamber. Stretching was performed at a cross head speed of 2 cm/min and the initial distance between the clamps was 4 cm. Draw ratios ([Lambda]) were measured by making a series of ink marks on the sample, 0.5 cm apart ([Lambda] = final distance divided by initial distance between marks). Following drawing, the chamber door was immediately opened and the temperature allowed to drop to room temperature quickly, so as to freeze in the orientation.

Infrared Spectroscopy. Transmission measurements were performed on a Nicolet 170SX FT-IR instrument at a resolution of 4 [cm.sup.-1]. A total of 32 scans were accumulated for each spectrum and background. Polarization of the beam was accomplished by means of a zinc selenide wire grid polarizer from Spectra-Tech. Spectra were recorded with the plane of polarization parallel and perpendicular to the draw direction.

Heat Treatment. The temperature of sample films was varied by wrapping them in aluminum foil and subsequently immersing them in a water bath for various lengths of time. The temperature was controlled to within [+ or -] 0.1 [degrees] C.

Crystallinity Measurements. Measurements of sample crystallinity were performed on a Perkin-Elmer DSC-7 at a scan rate of 40 [degrees] C/min under a nitrogen atmosphere. The total crystallinity was calculated by subtracting the area of the crystallization exotherm from the subsequent melting endotherm. A heat of fusion of 140 J/g was assumed for 100% crystalline PET (11). Indium was used for calibration.


The development of orientation in stretched films of PET is best described in terms of the orientation function [P.sub.2] (cos [Theta]), where, for measurements of IR dichroism (14):

<[P.sub.2](cos [Theta]) = D - 1 / D + 2 [P.sub.2](cos [[Theta].sub.m) (1)

The quantity [P.sub.2] (cos [Theta]) is the second-order Legendre polynomial:

[P.sub.2](cos [Theta]) = 3 [cos.sup.2] [Theta] - 1 / 2 (2)

The angle [Theta] is that between the chain axis and the draw direction, [[Theta].sub.m] denotes the angle between the vibrational transition dipole moment and the chain axis, D is the dichroic ratio given by D = A([parallel to])/A([perpendicular]), i.e. the ratio between the absorbance when the electric field vector is parallel to the draw direction to that when it is perpendicular. The angle brackets on the left-hand side of Eq 1 denote an ensemble average.

In our analysis, the absorbances were normalized to the area under the peak at 1410 [cm.sup.-1], which has been shown to be suitable for this purpose (15). The angle [[Theta].sub.m] has been evaluated to be 21 [degrees] for the peak at 1340 [cm.sup.-1] and 34 [degrees] for the peak at 970 [cm.sup.-1] (16). Using these values we have calculated the orientation function based on IR dichroism as a function of draw ratio [Lambda]. The results are shown in Figs. 1 and 2, together with the experimentally determined values of D. It can be seen that a sigmoidal curve is obtained, the orientation function increasing slowly at low draw ratios and then more rapidly beyond [Lambda] [approximately equal to] 2.

Numerous IR peaks of PET exhibit the phenomenon of dichroism (17, 18). However, the peaks at 1340 and 970 [cm.sup.-1] are highly dichroic and hence most suitable for monitoring changes in orientation, particularly at low draw ratios, where most other peaks have D values too close to unity to be measured accurately. These peaks also have the added advantage of being fairly weak in intensity and hence are less prone to saturation. This feature is especially important for transmission studies.

Figure 3 shows transmission spectra of a film stretched to [Lambda] = 2. A careful examination of the peaks at 1340 and 970 [cm.sup.-1] reveals a positive dichroism (i.e. the parallel peaks are more intense than the perpendicular peaks). When this film is kept at 70 [degrees] C for a period of 2 min, the dichroism is found to disappear. This is revealed in Fig. 4. The loss of molecular orientation as a function of time can be seen in Fig. 5. It can be observed that after 60 sec the level of orientation has been reduced by approximately one half.

Figure 5 also shows the loss of orientation with time at 72 [degrees] C for a film stretched to [Lambda] = 2. Once again, the relaxation is seen to be rapid. Approximately 40 sec are required for the orientation function to be reduced to half of its initial value. If the temperature is increased further to 76 [degrees] C, the relaxation is more rapid, as expected. In the same figure, it can be observed that only [approximately]10 sec are required for half the orientation to disappear.

The situation when [Lambda] is increased beyond 2 is quite different. Figure 6 shows polarized spectra for a film drawn to [Lambda] = 3. The dichroism is clearly visible at 1340 and 970 [cm.sup.-1]. If this film is kept at 70 [degrees] C for 160 sec, the spectra shown in Fig. 7 are obtained. While the orientation present in the film with [Lambda] = 2 relaxed after 120 sec, there is no such relaxation evident in Fig. 7. This observation is borne out by the data in Fig. 8. Exactly the same behavior was exhibited by a film with [Lambda] = 4, as shown in the same Figure.

The above-mentioned behavior was observed at all temperatures examined. For example, at 80 [degrees] C, the film with [Lambda] = 4 showed no obvious signs of relaxation even after 10 min, as demonstrated in Fig. 8.

Several reports have appeared in the literature in which the physical properties of PET, notably crystallization and melting, have been examined as a function of [Lambda]. Terada et al. (13), using density measurements, observed a constant low level of crystallinity up to [Lambda] = 2 and a rapid increase between [Lambda] of 2 and 3. This has been confirmed by Dargent et al. (19) using birefringence measurements. These authors suggest the onset of strain-induced crystallization occurs at [Lambda] [approximately equal to] 2.8. Padibjo and Ward (8) also used density measurements and observed significant crystallization beyond [Lambda] [approximately equal to] 2. The most exhaustive study of this phenomenon is that of Salem (20), who investigated the effects of temperature and strain rate on crystallization. For the films investigated here, the trend is as shown in Fig. 9. The onset of strain-induced crystallization was found to occur between [Lambda] values of 2.2 and 2.5.

We postulate that the significant inhibition of orientational relaxation observed for [Lambda] = 3 and higher is due to the presence of strain-induced crystallization. This accords with the results of shrinkage experiments, as reported by Gupta et al. (12) and Terada et al. (13). In both publications, shrinkage at temperatures between 80 [degrees] C and 90 [degrees] C was found to increase with [Lambda] until a draw ratio is attained at which crystallinity increases. Beyond this point shrinkage is severely hindered. These authors attributed this behavior to the development of crystallites, which restrict relaxation in the amorphous zones, thus preventing these conformational segments from returning to an isotropic orientation distribution. Such a mechanism is often termed pseudo-crosslinking, to distinguish it from chemical crosslinking, as found in thermoset polymers. Matsuo et al. (21) have demonstrated by light scattering that PET crystallites formed upon strain-induced crystallization are oriented with their c-axes in the draw direction.

There is an alternative explanation that may be considered. It is to be expected that any orientational relaxation will be confined to the trans conformers located in the amorphous regions, which according to some studies (19) constitute a mesophase. Therefore, an important issue is the distribution of trans content between the crystalline and amorphous regions. If the amorphous trans content is extremely small compared to the crystalline content, it is conceivable that any relaxation will be "masked" by the rigid crystalline trans conformers.

Several authors have examined the conformational changes which accompany orientation and crystallization in PET. Padibjo and Ward (8) have shown that orientation induces a conversion of gauche to trans. It is possible to calculate the relative amounts of the two conformers by first calculating the so-called isotropic spectrum, obtained by

[A.sup.iso] = 1/3{A([parallel to]) + 2A([perpendicular to])} (3)

where A is absorbance. This expression applies to uniaxial elongation (i.e. a sample with biaxial symmetry) and represents the spectrum one would obtain if the sample were isotropically oriented. When such a calculation is performed, followed by a calibration procedure described elsewhere (22), the curves shown in Fig. 10 are obtained. These results agree with those reported previously (23); the amount of trans for [Lambda] = 1 is slightly higher here due to a slightly higher initial crystallinity.

Combining the results in Fig. 10 with the crystallinity data, it is possible to calculate the distribution of trans conformers between the crystalline and amorphous regions. According to Padibjo and Ward (8):

[Mathematical Expression Omitted] (4)

where [X.sup.trans] is the overall fraction of trans conformer, [Mathematical Expression Omitted] is the crystalline trans contribution, and [Mathematical Expression Omitted] is the amorphous trans contribution. Since the crystalline phase is comprised exclusively of trans conformers, [Mathematical Expression Omitted] can be equated to the degree of crystallinity [X.sub.cryst]. If we consider a sample having [Lambda] = 2 and 6% crystallinity [ILLUSTRATION FOR FIGURE 9 OMITTED], [X.sup.trans] is 0.21 [ILLUSTRATION FOR FIGURE 10 OMITTED]. Thus [Mathematical Expression Omitted] is 0.15 and there are [approximately]3 times as many trans conformers in the amorphous zones as in the crystalline zones. A similar calculation for [Lambda] = 3 shows a different situation. For this draw ratio [Mathematical Expression Omitted] and [Mathematical Expression Omitted]. Hence there are now [approximately]5 times more trans conformers in the crystalline domains than in the amorphous ones. When [Lambda] is increased to 4, the amount of crystallinity has leveled off and [Mathematical Expression Omitted], [Mathematical Expression Omitted], so that [approximately]30% of trans conformers are in the amorphous zones while 70% are in the crystalline zones. It is clear from these simple calculations that the relative number of trans conformers in the amorphous zones decreases as [Lambda] increases beyond 2. Nevertheless there remains a sufficient fraction of conformers in the amorphous regions so that their relaxation should be detectable. Such relaxation is not evident from Fig. 8.

The concept of crystallinity affecting the properties of the amorphous zones in semicrystalline polymers is hardly new. For example it has been recognized for many years that crystallization can lead to an increase in the glass transition temperature [T.sub.g] in certain polymers (24-26) owing to a decrease in segmental mobility.

The effect of temperature, relative to [T.sub.g], on relaxation is illustrated in Fig. 11, which shows the decay of dichroism for the 970 [cm.sup.-1] peak at temperatures of 68 and 70 [degrees] C. The former temperature is below the beginning of the glass transition region and no relaxation is evident. An increase of temperature to 70 [degrees] C is sufficient to induce relaxation. As reported by Dargent et al. (19), the latter temperature corresponds to the beginning of the glass transition region in PET.

The orientation of PET has, in the past, been found to correspond with the deformation of a rubber-like network, for which (8, 27):

<[P.sub.2](cos [Theta])> = 1/5N ([[Lambda].sup.2] - 1/[Lambda]) (5)

where N is the number of random links between entanglements. It was thought instructive to verify such a dependence for our data, and the results are shown in Fig. 12. Despite the scatter that is evident, the fit to Eq 5 seems reasonable. The present data yield a value of N [approximately equal to] 3, in reasonable agreement with estimates by Terada et al. (13) and Padibjo and Ward (8).

The result that strain-induced crystallization inhibits orientational relaxation in the amorphous domains suggests that these zones should not be considered as discrete phases. Rather, as elongation progresses, a transformation of gauche to trans occurs (7), followed by orientation of trans segments. A certain critical orientation of the trans segments is attained (3) beyond which these conformers pack together in fibrillar-type structures and orient parallel to the draw direction (21). Concomitantly, the remaining conformational segments are trapped between these crystallites and can no longer relax to an isotropic orientation distribution.


In this paper it has been shown that when PET is drawn to [Lambda] values of 3 or higher, the orientational relaxation of the amorphous regions is hindered. This effect is ascribed to the development of strain-induced crystallites, which are believed to act as pseudo-crosslinks.


1. T. Sun, J. Pereira, and R. S. Porter, J. Polym. Sci.: Polym. Phys. Ed., 22, 1163 (1984).

2. G. C. Alfonso, M.P. Verdona, and A. Wasiak, Polymer, 19, 711 (1978).

3. F. S. Smith and R. D. Steward, Polymer, 15, 283 (1974).

4. G. LeBourvellec, L. Monnerie, and J.P. Jarry, Polymer, 27, 856 (1986).

5. P. Nicholas, A. R. Lane, T. J. Carter, and J. N. Hay, Polymer, 29, 894 (1988).

6. G. Farrow and I. M. Ward, Polymer, 1, 330 (1960).

7. A. Cunningham, I. M. Ward, H. A. Willis, and V. Zichy, Polymer, 15, 749 (1974).

8. S. R. Padibjo and I. M. Ward, Polymer, 24, 1103 (1983).

9. R. de P. Daubeny, C. W. Bunn, and C. J. Brown, Proc. Royal Soc. (A), 226, 531 (1954).

10. R. Qian, D. Shen, F. Sun, and L. Wu, Macromol. Chem. Phys., 197, 1485 (1996).

11. A. Mehta, U. Gaur, and B. Wunderlich, J. Polym. Sci.: Polym. Phys. Ed., 16, 289 (1978).

12. V. B. Gupta, J. Radhakrishnan, and S. K. Sett, Polymer, 34, 3814 (1993).

13. T. Terada, C. Sawatari, T. Chigono, and M. Matsuo, Macromolecules, 15, 998 (1982).

14. I. M. Ward, Structure and Properties of Oriented Polymers, Wiley, New York (1975).

15. D. J. Walls, Appl. Spectrosc., 45, 1193 {1991).

16. P. Spiby, M. A. O'Neill, R. A. Duckett, and I. M. Ward, Polymer, 33, 4479 (1992).

17. M. Yazdanian, I. M. Ward, and H. Brody, Polymer, 26, 1779 (1985).

18. K. C. Cole, J. Guevremont, A. Ajji, and M. M. Dumoulin, Appl. Spectrosc., 48, 1513 (1994).

19. E. Dargent, J. Grenet, and X. Auvray, J. Thermal Analysis, 41, 1409 (1994).

20. D. R. Salem, Polymer, 33, 3182 (1992).

21. M. Matsuo, M. Tamada, T. Terada, C. Sawatari, and M. Niwa, Macromolecules, 15, 988 (1982).

22. K. Edelmann and H. Wyden, Kantschuk Gummi-Kunststoffe, 25, 353 (1972).

23. A. Ajji, K. C. Cole, M. M. Dumoulin, and J. Brisson, Polymer, 36, 4023 (1995).

24. D. W. Woods, Nature, 174, 753 (1954).

25. S. Newman and W. P. Cox, J. Polym. Sci., 46, 29 (1960).

26. K. O'Reilley, F. E. Karez, and H. E. Blair, Bull. Amer. Phys. Soc., 9, 285 {1964).

27. L. R. G. Treloar, The Physics of Rubber Elasticity, 3rd Ed., Clarendon Press, Oxford, England {1975).
COPYRIGHT 1997 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:First Symposium on Oriented Polymers
Author:Pearce, R.; Cole, K.C.; Ajji, A.; Dumoulin, M.M.
Publication:Polymer Engineering and Science
Date:Nov 1, 1997
Previous Article:Constant speed drawing of amorphous films.
Next Article:Orientation of amorphous poly(ethylene terephthalate) by tensile drawing, roll-drawing, and die-drawing.

Related Articles
Aging characteristics of oriented poly(ethylene terephthalate).
The effect of block structure of an ethylene terephthalate/hydroxybenzoate copolymer on its ability to form compatible blends.
Transesterification reaction of polyarylate and copolyester (PETG) blends.
Chain orientation in sheared molten poly(ethylene oxide) fractions by measurement of infrared dichroism.
Constant speed drawing of amorphous films.
Orientation of amorphous poly(ethylene terephthalate) by tensile drawing, roll-drawing, and die-drawing.
Evolution of hot strain induced crystalline texture of poly(ethylene terephthalate) films.
Rotational isomeric relaxation in poly(ethylene glycols) studied by acoustic spectroscopy.
Synthesis and non-isothermal crystallization behavior of poly(ethylene phthalate-co-terephthalate)s.
Stress relaxation in low-shrink-force polyolefin films.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters