Influence of Rheological Properties On the Sagging of Polypropylene and ABS Sheet for Thermoforming Applications.
The isothermal sagging resistance of different grades of conventional and a high melt strength (HMS) PP has been correlated with the rheological characteristics of the polymers, such as dynamic shear properties, melt strength, and zero shear viscosity. A thermoforming grade of acrylonitrile-butadiene-styrene (ABS) was used as a reference material. At 190[degrees]C, ABS had the highest viscosity and elastic modulus in the frequency range measured, showing that this polymer is highly elastic. HMS PP had a greater shear thinning behavior than conventional PP because of its broader molecular weight distribution. The tan [delta] of the polymers showed that conventional PP had a higher tendency to flow than HMS PP and ABS when heated above 172[degrees]C. This was confirmed with sagging experiments performed in an air circulating oven, where the rate of sagging decreased as the melt strength and the zero shear viscosity of the polymer increased.
The thermoforming process is widely used in the plastics industry to produce articles for the packaging, automotive, domestic construction (e.g., shutters, skylights) and leisure industries. Polymers such as polystyrene (PS), acrylonitrile-butadiene-styrene (ABS), polypropylene (PP), vinyl, and acrylic polymers have been used in this process.
In thermoforming, polymer, usually in sheet form, is heated until it is softened. The softened polymer sheet is then drawn into a mold by either mechanical means (e.g., solid mold, plug) or pneumatic means (e.g., air pressure, vacuum) or a combination of both to form the desired shape. Once the polymer is cooled, it retains the detail of the mold and becomes ready for removal from the mold. A good thermoforming material must have both viscous and elastic components at the forming temperature . The viscous component allows the polymer sheet to flow easily into the mold under stress and the elastic component influences the degree of sagging of the sheet under gravity prior to forming.
Amorphous polymers such as ABS and PS have been widely used for thermoforming because their rubbery elastic state is exhibited over a wide temperature range. This means these polymers have a large processing window for thermoforming . For example, ABS has a processing temperature range of 127 to 182[degrees]C, as reported by Throne . Now there is an increasing demand to use PP for thermoforming because of its lower material cost  and its desirable end use properties. However, PP cannot be easily thermoformed because of its sharp melting point and poor melt strength [5, 6]. PP changes rather abruptly from a solid state to a fluid state once the melting point of PP has been reached. Melt phase thermoforming of PP is difficult to perform because of the poor sagging resistance of the polymer. As a result, PP has a narrow processing window, reported by Throne to be 143 to 166[degrees]C . Solid-phase forming has been developed to thermoform plastic sheet below its melting point. However, this process r equires higher initial equipment cost and it leaves the finished product with molecular stress orientation . Recently, new grades of PP have been developed to improve the processability of the polymer during thermoforming [6, 8, 9].
The processability of polymers in thermoforming depends upon the material characteristics at the forming conditions. Different techniques have been used to characterize the properties of PP to predict its sagging behavior and processability in thermoforming [1, 10]. With the help of finite deformation analysis, computer simulation of sagging of a polymer sheet can be carried out by knowing the material characteristics at the processing condition . However, little work has been done to develop a correlation between the measured material parameters and the sagging and drawing behavior of polymers during thermoforming .
In this work, the rheological properties and the sagging behavior of different grades of PP have been studied to assess their suitability in thermoforming operations. The rheological parameters related to the sagging of the polymer have been identified and discussed.
The rheological properties of five different grades of PP were studied. PP-1 and PP-4 were homopolymers. PP-2 and PP-3 were block copolymers with 6% weight content of ethylene comonomer. The ethylene/propylene ratio in the rubber phase of the copolymers is approximately 50:50. PP-5 is a high melt strength (HMS) grade of PP. It contains long chain branching and is believed to have been prepared by an irradiation treatment. PP-1 to PP-4 were supplied by ICI Australia Plastics (now Orica Australia Plastics). PP-5 was supplied by Montell, Australia. Since ABS can be easily processed in thermoforming, a thermoforming grade of ABS supplied by Huntsman Chemical Company of Australia was used as a reference material in this study and its rheological properties were also measured.
The Melt Flow Index (MFI) of the polymers was measured according to the ASTM standard D1238. The material properties of the polymers are shown in Table 1.
2.2 Melt Strength Measurement
The sagging resistance of PP and ABS was studied by measuring the polymer's melt strength using a Gottfert "Rheotens" Melt Strength Tester as described previously . The melt strength measurements were performed at an extrusion temperature of 190[degrees]C with an extrusion rate of 3.2 X [10.sup.-3] kg/min. The capillary die used has an entrance angle of 90[degrees], diameter of 2 X [10.sup.-3] m and L/D = 20. The melt strength tester was used with non-fluted steel rollers. The acceleration of the rollers was 1.2 X [10.sup.-2] m/[s.sup.2]. The distance between the die exit and the rollers of the melt strength tester was 0.21 m.
2.3 Dynamic Shear Measurement
The viscoelastic shear properties of the polymers were studied by measuring the storage modulus (G'), loss modulus (G") and the complex viscosity ([eta]*) within the linear viscoelastic region of the polymers using the Dynamic Stress Rheometer, Rheometrics Model SR 200, equipped with a parallel plate geometry. The storage modulus measures the elastic response of a polymer while the loss modulus measures the energy dissipated during flow deformation. The complex viscosity depicts viscosity behavior similar to that obtained from steady shear measurements.
The polymer samples were molded into circular discs by compression molding at 190[degrees]C. The disc sample had a diameter of 25 mm and a thickness of 2 mm. All measurements were carried out in a nitrogen environment to protect the polymer from degradation. The ABS sample was dried in an air circulating oven at 70[degrees]C for two hours prior to measurement to eliminate any absorbed moisture. A frequency sweep test was performed on the sample at 190[degrees]C over a frequency range of 0.01 to 100 rad/s at a maximum strain of 6%. The maximum experimental error of this measurement was 10%.
2.3.1 Measurement of Complex Modulus as a Function of Temperature
In thermoforming processes, the sagging of a polymer sheet is an extensional deformation. Since there is no shear deformation involved, the storage and loss moduli of the polymers were measured as a function of temperature in the low frequency (low shear rate) region in order to apply to thermoforming applications. In this study, a dynamic stress sweep test was carried out at a frequency of 0.05 rad/s and in the temperature range of 172 to 190[degrees]C for PP and 150 to 190[degrees]C for ABS, which approximate the thermoforming temperature region of these polymers. The complex modulus of PP-4 was measured three times at 172[degrees]C to assess the reproducibility of the experiment and the maximum experimental error was 5%.
2.4 Thermal Property Analysis
The thermal characteristics of the polymers were studied using a Perkin-Elmer Differential Scanning Calorimeter (DSC), Model 7. Polymer samples of 3 to 5 mg were sealed in an aluminium pan and heated and cooled at a rate of 10[degrees]C/min. Initially, the polymer samples were heated from room temperature to 180[degrees]C at a heating rate of 10[degrees]C/min and annealed for 1 minute to eliminate any thermal and mechanical history in the sample. They were then cooled to obtain the crystallization peaks and heated again to obtain the melting peaks.
2.5 Isothermal Sagging Test
The sagging behavior of PP and ABS was studied in an air circulating oven that was set at a temperature of 190[degrees]C. The oven was provided with a front glass door, which allowed the sagging of the polymer sheet to be monitored. Polymer sheets were prepared by compression molding at 200[degrees]C with an average thickness of 1.4 X [10.sup.-3] and 1.7 X [10.sup.-3] [+ or -] 5 X [10.sup.-5] for PP and ABS. respectively. A wooden stand was constructed to support the polymer sheet in the oven. A sheet sample with the size of 0.115 m X 0.115 m was clamped onto the wooden stand, leaving an area of 0.09 m X 0.09 m exposed to the hot air in the oven. A steel ruler was attached to the front of the wooden stand so that the amount of sag at the center of the sheet (i.e.. where the maximum amount of sag occurred) could be monitored at any point of time. Grids with the size of 0.005 m X 0.005 m were drawn onto the polymer sheet prior to the sagging test.
In the sagging experiment, a sheet sample was put in the oven for 12 minutes to ensure a distinguishable amount of sagging could be obtained for different polymers without tearing of the polymer sheet. The sag distance at the center of the sheet was recorded as a function of time by using a video camera. In the first 4 minutes, the PP sheet absorbed heat and underwent thermal expansion. The amount of sag recorded in this period of time was not reproducible because the polymer sheet expanded either in an upward or downward direction. After thermal expansion of the PP sheet, melting then started and the sheet was completely molten within 2 minutes. The amount of sagging occurring after 6 minutes from the start of the experiment (i.e., in the melting phase of sagging where the polymer sheet flowed under gravity) was taken as the sag distance of the polymer sheet and the experiment became reproducible.
After 12 minutes, the sagged sample was removed from the oven and allowed to cool. The local extension of the polymer sheet was then measured from the deformed grid. The local extension ratio [lambda] was calculated as follow:
[lambda] = Length of the grid distance of the sagged sample/length of the grid distance of the undeformed sample (1)
The local extension ratio of the ABS sample could not be measured accurately because ABS gave very little sagging and there was insignificant deformation of the grids. Therefore the local extension ratio can be expected to be very close to 1.
In this experiment, the sagging of each polymer was measured three times and the average of these values were used for analysis. The percentage error between different experiments was less than 12%.
3. DISCUSSION OF RESULTS
3.1 Dynamic Shear Test Analysis
The storage modulus of PP and ABS as a function of frequency at 190[degrees]C is shown in Fig. 1. ABS has the highest storage modulus over that entire frequency range, which indicates the highly elastic nature of the polymer at this temperature. In the low frequency region, HMS PP shows slightly higher storage modulus than conventional PP and this is believed to be due to the presence of large macromolecules and branching in the polymer [13, 14].
As the frequency is increased, the modulus of the polymers increases as continually shorter chain molecules are involved in the motion . In the high frequency region. the storage modulus is less dependent upon the molecular weight of the polymer . HMS PP has lower storage modulus compared to conventional PP and this implies that it possibly contains fewer short chain molecules than conventional PP.
Loss modulus shows a similar behavior to the storage modulus, as shown in Fig. 2.
The complex viscosity of PP and ABS at 190[degrees]C is shown in Fig. 3. ABS has the highest viscosity over the entire frequency range, which indicates that it is a highly rigid polymer. The HMS PP has a wider molecular weight distribution than conventional PP, and this is shown by its greater shear thinning viscosity [16, 17]. Similar shear characteristic have been observed for HMS PP grades by other researchers .
The complex zero shear viscosity of the polymers can be calculated using the Cross Model as shown in Eq 2, and the calculated results are summarized in Table 1.
[[eta].sup.*] = [[[eta].sup.*].sub.0]/1 + [K.sub.2][([omega]).sup.m] (2)
In Eq 2, [[eta].sup.*] is the complex viscosity at a particular frequency, [[[eta].sup.*].sub.0] is the complex zero shear viscosity, [omega] is the angular frequency, [K.sub.2] and m are constants. The calculated [[[eta].sup.*].sub.0] will be used to correlate with the sagging result in Section 3.3.2.
To further understand the viscoelastic properties of the polymer, tan delta ([delta]) values were calculated. Tan [delta] is a ratio of the loss modulus to storage modulus of a polymer (i.e., G"/G'). If tan [delta] is greater than 1, the viscous component of the deformation dominates and the polymer will behave as a viscous fluid. Polymers will behave as an elastic fluid when tan [delta] is less than 1. For thermoforming, a polymer sheet must have sufficient elastic component to resist sagging, but remain viscous enough to flow into the mold under stress. Hence thermoforming conditions (e.g., temperature range) should be chosen so that tan [delta] of the polymer is close to 1 .
The effect of temperature on the viscoelastic properties of the polymers measured at a frequency of 0.05 rad/s is shown in Fig. 4. Conventional PP has much greater tan [delta] values than HMS PP and ABS in the temperature range covered. Therefore it is expected that conventional PP has a higher tendency to sag when heated above 172[degrees]C.
3.2 Thermal Analysis
Table 2 summarizes the DSC results for PP and ABS. HMS PP has a lower melting peak than the conventional PP. and this implies that it has a lower degree of ordered structure, which is believed to be caused by the branching of the polymer . HMS PP shows a broader melting range (i.e., the difference between the onset temperature of melting and the temperature of melting peak) compared to the conventional PP because of its broad molecular weight distribution characteristics. HMS PP also has a higher crystalization temperature than the conventional PP. The initial slope (Si) of the crystallizations curves for PP has been calculated as outlined by Macauley et al. (1). The higher Si value of HMS PP indicates that the crystallization rate of the polymer is much higher. ABS is an amorphous polymer and has a glass transition temperature of 103.5[degrees]C, which corresponds to the immiscible polybutadiene elastomer component of the polymer .
3.3 Sagging Analysis
The sagging behavior of PP and ABS is shown in Fig. 5. The sagging has been measured for only one sheet thickness. This is based on the work of Cruz ,
who found that the rate of sagging depended mainly upon the area of the sheet and thickness did not play a significant role. Conventional PPs show greater sag distance than HMS PP and ABS. especially for long heating times when the temperature of the polymer sheet is expected to be close to the oven temperature of 190[degrees]C. The greater sagging of the conventional PPs is due to their poor melt strength and high tan [delta] value in the melt phase. Conventional PP with a lower MFI value sags to a smaller extent than PPs with high MFI.
3.3.1 Local Extension of the Sagged Sample
The local extension ratios [[lambda].sub.h] (measured in horizontal direction) and [[lambda].sub.v] (measured in vertical direction) were measured along the center of the four sides of the sagged samples and the average results are shown in Figs. 6a and 6b respectively. Close to the clamping edge of the polymer sheet, [[lambda].sub.h] is much greater than [[lambda].sub.v] for conventional PP. and this suggests that uniaxial stretching has dominated in this region. Near the center of the polymer sheet, the values of [[lambda].sub.h] and [[lambda].sub.v] are similar for all the samples, and this shows that equal-biaxial stretching of the polymer sheet is likely to occur in this region. A similar finding was observed by Cruz  for the sagging of circular samples. The amount of sagging for HMS PP is much less than that for the conventional PPs and as a result both [[lambda].sub.h] and [[lambda].sub.v] values are very close to 1.
No [lambda] results are presented for ABS as very little sagging was measured.
Since there is no change in the volume of the sample, the thickness ratio ([[lambda].sub.t]) along the center of the sagged sample can be calculated by Eq 3, and the results are shown in Fig. 6c.
[[lambda].sub.h][[lambda].sub.v][[lambda].sub.t] = 1 (3)
It can be observed that close to the clamping edge of the conventional PP sheet where stretching dominates, the thickness of the polymer sheet is reduced to the greatest extent. Therefore, it is expected that when thermoforming is performed at a temperature where the polymer exhibits a significant amount of sagging, webbing [a fold of polymer sheet that cannot be stretched flat against a mold surface ] is more likely to be observed on the fanned part because of the excessive prestretching of the polymer sheet. For HMS PP. the amount of sagging is small and the thickness distribution along the sagged sample is more even.
3.3.2 Correlation Between Rate of Sagging and Rheological Parameters
The rate of sagging was determined from the slopes of the sagging curves in Fig. 5, and the values are listed in Table 3. Again, the much better sagging performance of the HMS PP and ABS is highlighted.
Figure 7 shows the relationship between the rate of sagging and the melt strength of the polymers measured at 190[degrees]C. This graph shows that the rate of sagging decreases as the melt strength of the polymers increases. In order to have a rate of sagging that can be matched to that of ABS. PP should possess a certain minimum melt strength [i.e., greater than 20 cN measured at the conditions used in the test].
Figure 8 shows the rate of sagging as a function of zero shear viscosity at 190[degrees]C. Again, there is an inverse relationship with the rate of sagging decreasing as the zero shear viscosity of the polymers increases. Hence, increasing chain length of the large macromolecules will have a beneficial effect on the sagging resistance of the polymer.
The storage modulus, loss modulus, and the complex viscosity results for PP and ABS samples have been used to correlate with thermoforming behavior. ABS had the highest storage modulus and complex viscosity, which indicates that it is a highly elastic polymer. HMS PP showed greater shear thinning behavior than conventional PP, indicative of its broader molecular weight distribution. The higher tan [delta] values of conventional PP shows that these polymers have a greater tendency to sag during melt phase thermoforming than HMS PP and ABS.
The sagging behavior of the polymers was studied in an air circulating oven. The results confirm that conventional PPs give a greater amount of sagging than HMS PP and ABS during melt phase thermoforming (i.e., when the forming temperature is higher than the melting point of the conventional PP, which was measured to be around 160[degrees]C). The rate of sagging was inversely correlated with the melt strength and zero shear viscosity of the polymers. These observations demonstrate that PP should possess a high melt strength (of at least 20 cN measured at the conditions used in the test) and a high zero shear viscosity so that its rate of sag will match the low level of sagging experienced with ABS.
The authors wish to acknowledge Orica Australia and Montell Australia for sponsoring this project and for the helpful technical support and Australian Research Council for providing an Australian Postgraduate Award (Industry) to H. C. Lau.
(*.) Cooperative Research Center for Polymers.
(1.) N. Macauley, E. Harkin-Jones, and W. R. Murphy, Plast. Eng., 52 (7), 33 (1996).
(2.) J. J. M. Cormont, Adv. Polym. Technol., 5 (3), 209 (1985).
(3.) J. L. Throne, Thermoforming, Carl Hanser Verlag, Munich (1987).
(4.) R D. Leaversuch, Mod. Plast. Int., 71, 64 (1994).
(5.) J. F. Lappin and P. J. Martin, Plast. Eng., 52 (7), 21 (1996).
(6.) K. E. McHugh and K. Ogale, ANTEC '90, 452 (1990).
(7.) J. D. Gaspari, Plast. Technol., 35, 51 (1989).
(8.) J. J. Morad, ANTEC '95, 783 (1995).
(9.) K. A. Albert, C. A. Cruz, J. P. Palm, and R. W. Johnson, J. Plast. Film & Sheeting, 9, 293 (1993).
(10.) D. Hylton, ANTEC '91, 580 (1991).
(11.) M. J. Stephenson and M. E. Ryan, ANTEC '97, 734 (1997).
(12.) H. C. Lau, S. N. Bhattacharya, and G. J. Field, Polym. Eng. Sci., 38, 1915 (1998).
(13.) J. T. Bergen, Viscoelasticity: Phenomenological Aspects, Academic Press Inc. Ltd., London (1960).
(14.) G. Eder, H. Janeschitz-Kriegl. S. Liedauer, A. Schausberger, W. Stadlbauer, and G. Schindlauer, J. Rheol. 33 (6), 805 (1989).
(15.) G. Marin and W. W. Graessley, Rheol. Acta, 16 (5), 527 (1977).
(16.) W. Minoshima, J. L. White, and J. E. Spruiell, Polym. Eng. Sci., 20, 1166 (1980).
(17.) M. R. Drickman and K. E. McHugh, ANTEC '92, 496 (1992).
(18.) H. J. Yoo and D. Dane, ANTEC '92, 569 (1992).
(19.) Y. Eckstein and R. L. Jackson. Plast. Eng., 51 (5), 29 (1995).
(20.) J. M. G. Cowie, Polymers: Chemistry and Physics of Modem Materials, International Textbook Company Limited, Bucks (1973).
(21.) Ulf. W. Gedde, Polymer Physics, Chapman & Hall, London (1995).
(22.) C.A. Cruz, Jr., J. Plast. Film & Sheeting, 11, 190 (1995).