Hostname: page-component-74656f9884-rdgjk Total loading time: 0 Render date: 2026-11-08T00:49:56.783Z Has data issue: false hasContentIssue false
  • English
  • Français

Out-of-plane orientation and crystallinity of biaxially stretched polyethylene terephthalate

Published online by Cambridge University Press:  15 May 2014

Sudheer Bandla ,
Masoud Allahkarami  and
Jay C. Hanan
Sudheer Bandla
Affiliation:
School of Mechanical and Aerospace Engineering, Oklahoma State University, Tulsa, Oklahoma 74106
Masoud Allahkarami
Affiliation:
School of Mechanical and Aerospace Engineering, Oklahoma State University, Tulsa, Oklahoma 74106
Jay C. Hanan*
Affiliation:
School of Mechanical and Aerospace Engineering, Oklahoma State University, Tulsa, Oklahoma 74106
*
a)Author to whom correspondence should be addressed. Electronic mail: jay.hanan@okstate.edu

Abstract

The dependence of polymer properties on their processing history can be used advantageously. Polyethylene terephthalate (PET), a semi-crystalline polymer, exhibits a microstructure reliant on process and thermal history. PET undergoes strain-induced crystallization, making it sensitive to mechanical stretching. As the level of crystallinity in PET governs its mechanical behavior, it is necessary to quantify the effect of crystallinity and molecular orientation for efficient use. The present research is focused on an approach that will aid in correlating the stretch ratio of PET films to the percent crystallinity and mechanical properties. PET films with different local stretch ratios were obtained through bi-axially stretching injection-molded cylinders of increasing thickness and weight. Percent crystallinity of the PET films with different stretch ratios was measured using X-ray diffraction. Film samples were marked with respect to the stretch directions for measuring their longitudinal (primary stretch direction) and transverse mechanical properties. Local molecular orientation in the form of pole figures was mapped using the (100) plane corresponding to the PET lattice. This will help in linking the physical sample directions and processing to the molecular orientation. Associating the mechanical properties with molecular alignment helps in designing production processes that realize the material's structural potential.

Keywords

Polyethylene terephthalate (PET)Young's modulusCrystallinityDiffraction, and Orientation

Information

Type
Technical Articles
Information
Powder Diffraction , Volume 29 , Issue 2 , June 2014 , pp. 123 - 126
Copyright
Copyright © International Centre for Diffraction Data 2014 

I. INTRODUCTION

Microstructural information is essential to accurately model a material's mechanical behavior. For polymers, development of the microstructure is reliant on several elements such as chemistry, chain structure, molecular weight distribution, and processing history (Meijer and Govaert, Reference Meijer and Govaert2005). The above factors can affect the size, shape, and the extent of crystal formation in polymers, and make it more difficult to model their mechanical behavior. Polyethylene terephthalate (PET) is semi-crystalline aromatic polyester with a process-sensitive microstructure. PET can exhibit an amorphous or semi-crystalline microstructure based on the thermal treatment (Jog, Reference Jog1995). Furthermore, PET can be crystallized when subjected to strain, referred as strain-induced crystallization (SIC) (Thompson, Reference Thompson1959; Jabarin, Reference Jabarin1992). Strain-induced crystallinity is an interesting behavior of polymers, where stretching of the molecular chains results in an orderly structure. PET can undergo SIC either through biaxial stretching (bi-axial film) or uniaxial stretching (fiber drawing). Progress of SIC in PET under different conditions has been studied using techniques such as X-ray diffraction (Mahendrasingam et al., Reference Mahendrasingam, Martin, Fuller, Blundell, Oldman, Harvie, MacKerron, Riekel and Engström1999) and thermal analysis (Jabarin, Reference Jabarin1992). SIC in PET is dependent on several factors such as temperature, strain rate, and stretch/draw ratios (Jabarin, Reference Jabarin1992).

The mechanical behavior of PET is of interest for a wide range of industrial applications from textiles to packaging and engineering moldings (Bandla and Hanan, Reference Bandla and Hanan2012; Bandla et al., Reference Bandla, Winarski, Hanan, Jin, Sciammarella, Furlong and Yoshida2013). The modulus of PET is influenced by the extent of crystallinity and molecular orientation (Lyons, Reference Lyons1958; Viswanathan et al., Reference Viswanathan, Wiff and Adams1976). PET processing induces crystallization from both thermal and strain-induced processes. This led researchers to study and relate the stretch ratio with molecular orientation and the resulting crystallinity. Stretch-blow molding of PET involves both bi-axial stretching and molecular orientation (Smith et al., Reference Smith, Cooper, Winter and Everall2006). The effects of stretch ratio (Michaeli and Leopold, Reference Michaeli and Leopold2010) and crystallinity (Silberman et al., Reference Silberman, Omer, Ophir and Kenig1998) on the mechanical behavior of PET film from blow-molded containers were studied individually. As the process involves multiple variables (material and process conditions), it is difficult to develop a correlation between these studies. The current study presents an approach to relate the mechanical behavior of bi-axially stretched PET films with the amount of crystallinity and molecular orientation.

Crystallinity in PET can be measured through different methods (Johnson, Reference Johnson1959, Bashir et al., Reference Bashir, Al-Aloush, Al-Raqibah and Ibrahim2000). X-ray diffraction is widely used for measuring the crystallinity in PET, as it provides the advantage of obtaining additional microstructural details such as crystal size and orientation. PET exhibits a triclinic crystal structure, with four principle reflections from the (010), $\lpar \overline 1 10\rpar $, (100), and $\lpar \overline 1 05\rpar $ planes (17.5°, 22.5°, 25.6°, and 42.6° 2θ for CuKα X-rays) (Daubeny et al., Reference Daubeny, Bunn and Brown1954). In the present work, X-ray diffraction was used to measure the percent crystallinity of PET films with different levels of stretching and determine crystal orientation with respect to testing directions.

II. MATERIALS AND METHODS

Commercially available PET of M w 36 000 g mol−1 (0.76 dl/g I.V.) was used for the current study. PET samples of increasing wall thickness and weight (6.2, 7.5, 8.5, 9.5, and 11 g) were injection molded using a cylindrical mold. Each cylinder had the same outside diameter. To increase the thickness at each weight the “core” of the mold was changed to a smaller diameter. The 6.2 g mold was also shorter than the others. Molded PET cylinders were marked along the length at uniform intervals of 5 mm, as shown in Figure 1(a) to measure the amount of axial stretch. Marked cylinders were stretch-blow molded into hexagonal faced containers, shown in Figure 1(b). Samples were stretched at 98 °C temperature with a strain rate of 11 s−1. Axial and transverse film samples with different stretch ratios were obtained for mechanical testing.

Figure 1.

(Color online) (a) Injection-molded cylinder with markings, (b) blow-molded hexagonal containers, (c) radial sample, and (d) schematic on the circumferential and radial stretch ratios.

Stretch Ratios

The linear stretch ratio is the ratio of the final length to the initial length. Two independent stretch ratios can be calculated for the bi-axially stretched films: the axial stretch ratio (ASR) and the circumferential stretch ratio (CSR). A radial stretch ratio (RSR) can also be calculated, but is dependent on the other two, as volume is conserved. The formulas used to calculate each stretch ratio are presented in Figure 1. Using the different weight cylinders a range of ASRs from 3.3 to 5.1 and CSRs of 4.2 to 4.7 were produced. The CSR remained same for a particular container weight.

Mechanical Testing

PET film samples collected were tested following ASTM D 882 on a hydraulic driven universal testing machine (Instron 8802), at an extension rate of 5 mm min−1. A laser extensometer (LE-05 model) was used for accurate strain measurements. The Young's modulus was obtained from the slope of the initial region of the stress–strain curve. Figure 2 is a plot of the stress–strain curves of films tested in both axial and transverse directions with the amorphous PET cylinders from injection molding. A minimum of five samples for each different stretch ratio was tested. The average Young's modulus of each sample type was 2.1 ± 0.1, 4.3 ± 0.3, and 4.7 ± 0.5 GPa for the amorphous, axially stretched, and transverse-stretched directions, respectively.

Figure 2.

(Color online) Comparison of stress–strain curves for amorphous and semi-crystalline PET.

X-ray Analysis

Diffraction patterns for the bi-axially stretched PET films were collected on a Bruker D8 Discover micro-X-ray diffraction system equipped with a two-dimensional (2D) detector. CuKα radiation (λ = 1.541 84 Å) was used for this work. Crystallinity and texture measurements were performed using a 0.5 mm collimator. Film samples were fixed on to a Si single crystal (zero background slide) for data collection in reflectance. Percent crystallinity was measured based on the ratio of the crystalline fraction (A c) to the total area (A a + A c) under the diffracted curve, as represented in Eq. (1). Percentage crystallinity was determined by rotating the sample to have orientation independent measurements.

(1)$$X_c \percnt=\displaystyle{{A_c } \over {A_a+A_c }} \times 100 \percnt$$

Partial and concentrated diffraction rings from the 2D frames, indicate strong orientation of the crystallites. Although PET exhibits multiple characteristic reflections, only the (100) reflection was visible to evaluate orientation under the current laboratory coordinates. To determine the crystallite orientation, diffraction patterns were collected at fixed θ 1 = 12.5° (source) and θ 2 = 12.5° (detector) by rotating the sample along the ϕ-direction in steps of 5° at multiple χ-positions (−5° and 32°). Collected 2D diffraction frames were integrated along the χ-direction at 1° step size. Pole figures were generated based on the intensity of the (100) diffraction peak, as shown for samples from regions R1 and R6 with different stretch ratios in Figure 4. Linking the axial and transverse directions with the ϕ-position, as shown in Figure 4, will help in relating the properties with molecular orientation.

Figure 3.

(Color online) (a) Schematic representation of the sample orientation with respect to the diffractometer geometry (diffraction vector is perpendicular to the film surface, allowing to measure out-of-plane texture and crystallinity), (b) 2D diffraction frame (semi-crystalline), and (c) comparison of 1D diffraction pattern for amorphous and semi-crystalline samples.

Figure 4.

(Color online) Pole figures of the samples with different axial and circumferential stretch ratios [colors indicate intensity of the (100) reflection, χ – radial direction (−45° to −115°), ϕ – on the circumferential axis].

III. DISCUSSION

Comparison of the stress–strain curves obtained from the amorphous (injection-molded cylinders) and semi-crystalline PET samples (with ASR – 4.3 and CSR – 4.4) in Figure 2, points out the importance of sample orientation in studying their mechanical properties. Crystal development and their orientation within the amorphous matrix are important in governing the mechanical behavior. Application of the (100) plane for texture analysis, allows determining polymer chain orientation with respect to the film surface. As can be seen in Figure 4, the texture shows up as a peak along the χ-axis perpendicular to the θ-direction. The texture of the (100) plane indicated that the crystallites were oriented within ±15° χ, parallel to the film surface. The pole figure is strikingly uniform showing considerable symmetry about ϕ. In some cases, discontinuity along ϕ because of partial in-plane orientation was observed. Comparing the pole figures for different samples, indicate that the change in the molecular orientation is independent of the sample location (region) and stretch ratios. Plotting the measured crystallinity with respect to the stretch ratios (ASR and CSR) as in Figure 5, indicate that the crystallinity increases with increasing CSR. As can be seen from Figure 5, crystallinity increased till an axial stretch of 3.8 and then decreased with increasing axial stretch up to 4.4 and showed no increase with further stretching. In general, crystallinity is considered to increase with higher stretch ratios. The crystallinity in the out-of-plane direction may be influenced by the thermal history from cooling rather than simply increase as a function of mechanical stretching. As seen from Figure 6, films with higher thickness show more crystallinity. Thicker films would cool more slowly. In addition to that, the decrease in the crystallinity observed here could be due to the presence of in-plane texture in the SIC domains. As seen from Figures 4(a) and 4(b), crystalline content is measured at 39 and 47%. Along with the significant out-of-plane texture, this suggests the importance of measuring crystallinity in the in-plane direction of the film, which will be measured in future work. Unless all the film samples have their molecular orientation aligned with the testing direction, it is more difficult to make a comparison of the relationship to crystalline fraction and modulus.

Figure 5.

(Color online) Percent crystallinity from the (100) peak in PET films with respect to their axial and circumferential stretch ratios (legend – CSR).

Figure 6.

(Color online) Comparison of film thickness with the percentage crystallinity (legend – CSR).

Only relating the stretch ratios with percent crystallinity does not include the thermal influence. While in this work the thermal history before blowing was kept constant, some differences in the quench rate (during stretching) may arise from the difference in final film thickness. Accounting the influence of thermal effects on the crystallinity development may help in modeling the behavior more accurately.

IV. CONCLUSIONS

With the current approach, PET films with different stretch ratios can be obtained. X-ray diffraction allowed associating the orientation of PET crystals with sample directions (axial and transverse). Furthermore, it aided in non-destructively quantifying the extent of crystallinity present in the film, for relating with mechanical properties. The (100) crystallinity (averaged over ϕ) increased till an axial stretch of 3.8 and decreased with increasing stretch ratio till 4.4. From the data, it is evident that stretch ratios have an effect on both the crystallinity and film modulus. Significant texture was observed. All of the (100) planes were aligned with the film plane within ±15°. The impact of the increasing stretch ratio is directionally dependent, with more dependence in the transverse direction than in the axial direction. Determining the stretch ratios of the actual samples used for measuring the percent crystallinity and modulus will aid developing better models to predict the performance of PET.

ACKNOWLEDGEMENTS

This work was possible partially through OSU foundation grants to the Helmerich Research Center and industry sponsorship. The authors acknowledge Krones AG, for the stretch-blow molding of hexagonal containers at their facility.

References

Bandla, S. and Hanan, J. (2012). “Microstructure and elastic tensile behavior of polyethylene terephthalate-exfoliated graphene nanocomposites,” J. Mater. Sci. 47(2), 876–882.CrossRefGoogle Scholar
Bandla, S., Winarski, R., and Hanan, J. (2013). Nanotomography of Polymer Nanocomposite Nanofibers. Imaging Methods for Novel Materials and Challenging Applications, edited by Jin, H., Sciammarella, C., Furlong, C., and Yoshida, S. (Springer, New York), Vol. 3, pp. 193–198.Google Scholar
Bashir, Z., Al-Aloush, I., Al-Raqibah, I., and Ibrahim, M. (2000). “Evaluation of three methods for the measurement of crystallinity of pet resins, preforms, and bottles,” Polym. Eng. Sci. (USA) 40(11), 2442–2455.CrossRefGoogle Scholar
Daubeny, R. D. P., Bunn, C. W., and Brown, C. J. (1954). “The crystal structure of polyethylene terephthalate,” Proc. R. Soc. Lond. A, Math. Phys. Sci. 226(1167), 531–542.Google Scholar
Jabarin, S. A. (1992). “Strain-induced crystallization of poly(ethylene terephthalate),” Polym. Eng. Sci. 32(18), 1341–1349.CrossRefGoogle Scholar
Jog, J. P. (1995). “Crystallization of polyethyleneterephthalate,” J. Macromol. Sci. – Polym. Rev. (USA) 35(3), 531–553.CrossRefGoogle Scholar
Johnson, J. E. (1959). “X-ray diffraction studies of the crystallinity in polyethylene terephthalate,” J. Appl. Polym. Sci. 2(5), 205–209.CrossRefGoogle Scholar
Lyons, W. J. (1958). “Theoretical values of the dynamic stretch moduli of fiber-forming polymers,” J. Appl. Phys. 29(10), 1429–1433.CrossRefGoogle Scholar
Mahendrasingam, A., Martin, C., Fuller, W., Blundell, D. J., Oldman, R. J., Harvie, J. L., MacKerron, D. H., Riekel, C. and Engström, P. (1999). “Effect of draw ratio and temperature on the strain-induced crystallization of polyethylene terephthalate at fast draw rates,” Polymer 40(20), 5553–5565.CrossRefGoogle Scholar
Meijer, H. E. H. and Govaert, L. E. (2005). “Mechanical performance of polymer systems: the relation between structure and properties,” Prog. Polym. Sci. (UK) 30(8–9), 915–938.CrossRefGoogle Scholar
Michaeli, W. and Leopold, T. (2010). Modeling the Structural Performance of Stretch-Blow Moulded PET Bottles, ANTEC 2010 (Society of Plastics Engineers, Orlando).Google Scholar
Silberman, A., Omer, M., Ophir, A., and Kenig, S. (1998). The Effect of Stretch and Heat Transfer on the Thermo-Mechanical Properties of PET Bottles. ANTEC 1998 (Society of Plastics Engineers, Atlanta).Google Scholar
Smith, M. R., Cooper, S. J., Winter, D. J., and Everall, N. (2006). “Detailed mapping of biaxial orientation in polyethylene terephthalate bottles using polarised attenuated total reflection FTIR spectroscopy,” Polymer 47(15), 5691–5700.CrossRefGoogle Scholar
Thompson, A. B. (1959). “Strain-induced crystallization in polyethylene terephthalate,” J. Polym. Sci. 34(127), 741–760.CrossRefGoogle Scholar
Viswanathan, A., Wiff, D. R., and Adams, W. W. (1976). An Investigation of Structure-Property Correlation in Polyethylene Terephthalate Films, Air Force Materials Laboratory, 127.Google Scholar
Figure 0

Figure 1. (Color online) (a) Injection-molded cylinder with markings, (b) blow-molded hexagonal containers, (c) radial sample, and (d) schematic on the circumferential and radial stretch ratios.

Figure 1

Figure 2. (Color online) Comparison of stress–strain curves for amorphous and semi-crystalline PET.

Figure 2

Figure 3. (Color online) (a) Schematic representation of the sample orientation with respect to the diffractometer geometry (diffraction vector is perpendicular to the film surface, allowing to measure out-of-plane texture and crystallinity), (b) 2D diffraction frame (semi-crystalline), and (c) comparison of 1D diffraction pattern for amorphous and semi-crystalline samples.

Figure 3

Figure 4. (Color online) Pole figures of the samples with different axial and circumferential stretch ratios [colors indicate intensity of the (100) reflection, χ – radial direction (−45° to −115°), ϕ – on the circumferential axis].

Figure 4

Figure 5. (Color online) Percent crystallinity from the (100) peak in PET films with respect to their axial and circumferential stretch ratios (legend – CSR).

Figure 5

Figure 6. (Color online) Comparison of film thickness with the percentage crystallinity (legend – CSR).