Fracture Behavior of Styrene-Ethylene-Propylene Rubber-Toughened Polypropylene.
H.-J. SUE [*]
The morphology and fracture behavior of isotactic polypropylene toughened by styrene-ethylene-propylene (PP/SEP) were investigated. The SEP rubber, having an average particle size of 0.2 [micro]m, is found to be well dispersed in the PP matrix. The fracture toughness of SEP-modified PP is greatly improved. The toughening mechanism investigation shows that a widespread crazing zone is generated in the crack tip damage zone. An intense narrow damage band in the center of crazed zone is formed. Crazing and shear yielding are found to be the dominant toughening mechanisms in PP/SEP. The crazes are initiated only by large SEP particles in the blend. The small SEP particles ([less than] 0.3 [micro]m) can neither cavitate nor trigger crazing. As a result, large scale shear deformation is suppressed in this blend. These findings are consistent with the notion that the crack tip plane strain constraint has to be relieved in magnitude in order for the deviatoric stress to reach a critical value for widespread shear ba nding.
Wide market demands for inexpensive, recyclable engineering plastics have made polypropylene (PP) a very attractive polymer. Unfortunately, PP has poor impact strength, thus limiting its acceptance for automotive and other engineering applications. There are many research programs that have been undertaken to improve the toughness of PP. Rubber toughening is a well-known approach for improving fracture toughness of polymers [1-8]. Several types of rubber, such as ethylene-propylene rubber (EPR) [9-13] and EPR/diene monomer (EPDM) [9, 13-16], have been chosen to toughen PP. The toughening efficiency on PP is found to depend on the type of rubber, its content and the operating toughening mechanisms.
It has been shown that toughening in rubber-modified PP can be achieved via crazing and shear yielding [9, 10, 12]. Rubber particles, such as EPDM [10, 12, 15, 16] and EPR [12, 14], can generate and terminate crazes in PP under certain test conditions. Shear deformation in some rubber-modified PP systems [12, 14, 15] was also observed to be a dominant toughening mechanism. Jang  demonstrated that in EPDM and in styrene-butadiene rubber (SBR) modified PP, the operative fracture mechanisms depend strongly on particle size. Large particles ([greater than] 0.5 [micro]m) favor crazes while small particles induce shear banding.
If the toughness is improved mainly through crazing, the toughener phase must exhibit good interfacial adhesion with the matrix , and acts not only as a stress concentrator but also as a craze stabilizer [2, 18]. On the other hand, if the polymer is toughened via shear banding, Yee and Pearson [19-21] have shown that relief of triaxial tension by rubber cavitation is critical if the matrix is under plane strain constraint. In contrast, if second phase particles can neither cavitate nor trigger massive crazes, the toughening effect due to these particles will be minimal.
In our previous papers [22, 23], we have demonstrated that SEP rubber is an effective compatibilizer for PP/Noryl blends (Noryl is a product of GE Plastics). An addition of a small amount of SEP can dramatically reduce Noryl particle size and improve interfacial adhesion between Noryl particles and PP. This, in turn, significantly increases the toughness of PP/Noryl blends. In this paper, since SEP forms a strong interfacial adhesion with PP and cannot easily undergo internal cavitation because of its unique morphology , it is possible to study whether or not stress concentration alone can toughen PP effectively. Also, it is possible to learn whether or not limited crazing can trigger massive shear banding in the polymer matrix.
The isotactic polypropylene used to conduct this research has an Mn = 100,000 and Mw = 368,000, with a melt flow rate index (MFI) of 2.5. The SEP (Kraton-G1701, Shell Chemical) is a diblock copolymer with glass-transition temperatures ([T.sub.g]) of -59[degrees]C and 95[degrees]C for the EP and PS blocks, respectively. All polymers used in this research are commercial products. A blend of PP and 5% by weight of SEP (PP/SEP) is prepared using a research grade roll milling machine at a roll temperature of 200[degrees]C.
The PP/SEP blend was compression molded into 4 mm thick plaques at 200[degrees]C. Cooling water (23[degrees]C) was used to cool the plaques at approximately 20[degrees]C/min. Single-edge-notch three-point bend (SEN-3PB) specimens were cut from the compression molded plaques. Samples were machined to reach the final dimensions of 63.5 mm X 12.7 mm X 4 mm. The SEN-3PB bars were notched with a 250 [micro]m radius notch cutter to a notch depth of 5.5 mm, followed by a liquid nitrogen-chilled razor blade tapping to open a sharp crack to a total depth of about 6.4 mm (a/W = 0.5-0.55) for J-integral fracture toughness measurements.
The J-integral test was conducted according to ASTM E813-89 using a multiple-specimen technique at room temperature. An Instron (Model 4411) screw-driven mechanical testing machine was used to perform the measurements at a crosshead speed of 2 mm/min. The tensile test was carried out at room temperature using ASTM D 683M-96 Type M-II specimen geometry at a crosshead speed of 2 mm/min.
Differential scanning calorimetry (DSC) measurements were performed using Perkin-Elmer DSC (model Pyris 1) at a heating/cooling rate of 10[degrees]C/min. The dynamic mechanical behaviors of the neat PP and PP/SEP blend were studied using the dynamic mechanical spectroscopy (DMS) machine, Rheometrics RMS-800, in a torsional mode with 2.5[degrees]C per step. A constant strain amplitude of 0.01% and a fixed frequency of 1 Hz were employed.
The detailed procedures for specimen preparation, mechanical testing, thermal analysis, dynamic mechanical analysis, and microscopic analysis of morphology and deformation processes are described in our previous papers (22, 23). They are not repeated here.
Thermal and Dynamic Mechanical Analysis
It is well known that the crystallinity of semicrystalline polymers has a significant effect on mechanical property and fracture behavior in polymers. The melting behaviors of the neat PP and the PP/SEP blend are studied and shown in Fig. 1. The crystallinity, [X.sub.c] melting temperature, [T.sub.m], and crystallization temperature, [T.sub.c] are shown in Table 1. It can be seen that the melting behaviors of neat PP and PP/SEP blend are almost the same except for an additional small melting peak at 148[degrees]C for the PP/SEP blend. This small melting peak indicates that a small amount of [beta] crystal of PP may be induced by the presence of SEP particles (24). The addition of 5% SEP rubber increases the crystallinity by about 3% and [T.sub.c] by about 2[degrees]C, while [T.sub.m] remains the same. The increases in crystallinity and crystallization temperature indicate that SEP rubber acts as an effective nucleating agent. However, our SAXS results show no differences in lamella thickness and the long per iod of crystalline phase between PP and PP/SEP (25).
Dynamic mechanical spectra are shown in Fig. 2. Phase separation of EP portion of SEP is clearly observed as indicated by the Tan [delta] peak at -59[degrees]C. The [T.sub.g] as well as the shear storage modulus of PP above room temperature is not affected by the addition of SEP rubber. This means that the addition of 5% SEP is unlikely to lower the heat deflection temperature of PP.
TOM of polished thin section (about 20 [micro]m) of PP/SEP. not shown here, is compared to that of neat PP specimen (22). The average size of spherulites in PP/SEP is slightly smaller (50 [micro]m vs. 60 [micro]m). The addition of SEP does not have an apparent effect on the PP spherulitical structure.
To further probe the phase morphology in PP/SEP, TEM investigation is performed. As shown in Fig. 3, well-dispersed SEP particles are formed in PP matrix. The average particle size is about 0.2 [micro]m. It is shown that, as expected, the SEP rubber itself shows a two-phase morphology. From the material supplier (Shell Chemical Company), the SEP rubber is a diblock copolymer consisting of a hard polystyrene block and a saturated, soft EP block. It is believed that the light phase is the PS phase and the dark, [R.sub.u][O.sub.4] stained, phase is the EP phase.
The selected mechanical properties and J-R curves of both neat PP and PP/SEP blend are shown in Table 2 and Fig. 4, respectively. The addition of SEP rubber decreases the Young's modulus and yield stress of PP. However, these reductions are not significant. The storage moduli of PP/SEP at room temperature and at 80[degrees]C are almost the same as those of neat PP, suggesting that the addition of SEP at this level does not affect the heat deflection temperature of PP. Nevertheless, the toughness of PP is greatly improved ([J.sub.c] in Table 2 and Fig. 4). The crack propagation resistance Is also improved as indicated by the slope dJ/da values from Fig. 4.
It is noted that the thickness of PP and PP/SEP is thinner than that required for [J.sub.IC] values (ASTM E81389). As a result, we only report [J.sub.c], instead of [J.sub.IC], values for PP and PP/SEP systems.
Toughening Mechanism Investigation
To gain a better understanding of how the failure process proceeds, OM, SEM and TEM investigations were conducted on the DN-4PB tested specimens (23). As shown in Fig. 5, the optical micrographs of the DN-4PB specimen clearly indicate that a crazed zone is formed in front of the crack tip (Fig. 5a). An intense damage band at the center of the crazed zone is also formed. When this section is examined under cross-polarized light, a long, narrow birefringent strip along the crack wake and ahead of the crack tip is clearly resolved (Fig. 5b). This indicates that shear yielding has taken place in these regions. This deformation phenomenon has also been observed by Chou et al. (11) in EPR-toughened PP. The crazing pattern outside the birefringent zone is optically similar to that observed in neat PP (23). A careful inspection of the crazed zone reveals that most crazes run through the centers and between the boundaries of spherulites (not shown) . Since the level of crazing is not as intense as that observed in PP/ Noryl blend (23), massive shear yielding is suppressed. These interesting damage features are further studied using SEM and TEM.
SEM investigations were performed in the crack wake and the crack tip regions. The DN-4PB subfracture surface zone was polished and etched using an oxidizing acid agent as described in previous papers (22, 23). SEP particles and materials inside the plastic strip were etched preferentially and could be easily revealed using SEM. The results are shown in Fig. 6. Two distinct damage features are clearly observed in Fig. 6a, i.e., a featureless shear yielded zone surrounding the crack and encompassed by a large craze-like damage zone. At a distance of 500 [micro]m ahead of the crack tip (Fig. 6b), massive voids are clearly observed inside these bands. These voids are SEP rubber particles that were etched away. When SEM study was performed in the crack wake, it revealed that widespread shear yielding occurred in the region close to the fracture surface (Fig. 6c) . The voids are dramatically elongated and approximately parallel to the fracture surface (see arrows in Fig. 6c). This suggests that a large scale plast ic flow has taken place in this region. Outside this region, the deformation feature is similar to that observed in front of the crack tip (Fig. 6b).
TEM investigations further reveal that a large scale plastic flow occurs at the crack tip as indicated by severely elongated SEP particles (Fig. 7a). Matrix distortion at the crack tip region is also observed. This is caused by unloading of the blunted crack tip. The TEM micrograph taken 200 [micro]m ahead of the crack tip (Fig. 7b), but inside the damage band, shows a certain degree of cavitation (see arrows in Fig. 7b). The SEP particles were also torn and smeared, suggesting a certain degree of matrix shear yielding (see also Fig. 5b). When TEM observation was performed at the crack wake but away from the fracture surface, crazes are also observed (Fig. 7c) . It is noted that the formation of these crazes are only associated with large particles ([greater than] 0.3 [micro]m). This finding is consistent with the results in EPDM-toughened PP (10) and Noryl particles-toughened PP (23).
The deformation and fracture mode of DN-4PB PP/SEP specimen observed in this study is summarized in Fig. 8. Upon loading, crazing occurs first at the crack tip. With increasing strain, a large fan-shape crazed zone is generated and a shear yielded zone at the center of the crazed zone is initiated around the crack tip. At the same time, the crack tip is blunted. It is noted that sparse crazes occur outside the shear yielded zone. This suggests that crazing is the first deformation stage. In the center of the damaged zone, an intense crazed region, the crazing is transformed into shear banding.
As mentioned earlier, crystallinity and crystalline morphology have significant effects on mechanical properties and fracture mechanisms. Generally, the higher the crystallinity, the higher the modulus and yield stress, and the lower the toughness. In the present study, the crystallinity of SEP-modified PP is about 3% higher than that of neat PP while the spherulite size and the lamellae thickness of PP are almost the same as those of PP/SEP. However, the modulus and yield stress of PP/SEP blend are lower than those of neat PP. This inverse relationship is due to the presence of SEP particles that act as stress concentrators to decrease the overall yield stress and modulus. Furthermore, the random dispersion of SEP particles inside the PP spherulites suggests that SEP and PP are compatible to each other. This property allows SEP to become a good compatibilizer between PP and other rigid polymers, such as PS.
To improve toughness, effective toughening mechanisms have to be generated. Massive crazing and shear yielding are the two most frequently encountered energy absorbing mechanisms [2, 18, 26, 27]. Criteria for the occurrence of crazing and shear yielding are widely discussed and understood (2, 19, 20, 26, 28). For massive crazing to occur, a high level of hydrostatic tension needs to exist ahead of the crack tip. Large particles are more effective in triggering crazes (2, 10, 18). However, for a given rubber loading, larger particles lead to fewer number of particles, therefore, fewer craze initiation sites. The occurrence of massive crazing also depends on the nature of the toughener phase and the matrix. In this study, crazing is found both in neat PP (23) and in PP/SEP. The crazing pattern in both materials does not differ much when viewed using optical microscopy. However, there exist an intense damage strip along the crack wake and ahead of the crack tip, and a larger overall crazed zone in PP/SEP specimen. This difference can be explained as follows: The crazes first occur in the weaker regions and the regions with a higher stress concentration. The presence of SEP particles, which induce stress concentration, makes crazing to occur more easily in these regions. Therefore, at the same stress level, a larger crazing zone is seen in PP/SEP.
The formation of the intense plastic strip (Fig. 5b) around the crack tip can be explained by the slip-line field theory (29, 30). In the PP/SEP blend, since the SEP particles are small, a higher stress is needed to trigger crazing. A high level of crazing can be initiated only at sufficiently high stress regions. At the plastic-elastic boundary of the slip-line field, the mean stress reaches the highest value, as shown in Fig. 9 (31-33). As this maximum increases beyond a critical value, further loading causes massive crazing to occur. The presence of SEP rubber particles ahead of the crack tip widens the craze zone. Once the crazes form, they tend to grow on the same plane, which is perpendicular to the maximum principal stress direction, leading to the formation of a highly intense craze band. Consequently, intense crazing only occurs at the crack tip region. The presence of SEP particles makes these crazes grow in a more stable manner [34, 35]. The stable, intense crazes are sufficient to relieve triaxia l tension. As a result, the intense plastic strip is developed, instead of forming a mature crack.
It has been suggested that triaxial tension at the crack tip has to be relieved in order for the deviatoric stress to reach a critical value for shear yielding to occur (19, 20). From TOM and SEM micrographs, only a limited shear deformation zone is observed at the crack tip and on the crack wake. The formation of the limited level of shear banding can be explained as follows.
It is well known that plastic deformation is likely to be developed at the crack tip for ductile polymers. A larger plastic zone will be generated at the crack tip for tougher materials. It is generally recognized that a plane stress condition exists at the free surface of the blunt crack tip. When the crack grows, plastic deformation takes place first, and is responsible for the diffuse shear yielding zone on both sides of the crack and the crack tip. Ahead of the crack tip or at the elastic-plastic boundary, a high level of triaxial stress is present (Fig. 9). To achieve a large scale shear banding, this plane strain constraint has to be relieved through dilatational deformation, such as particle cavitation and crazing. From TEM investigation, it is seen that a small portion of SEP particles inside the large damage strips is cavitated (Fig. 6b and c and Fig. 7b). No particle cavitation is found in the undeformed region (Fig. 6b and c). The crazes found are to be related only to large SEP particles ([greater than] 0.3 [micro]m) (Fig. 7d). Small SEP particles are ineffective in triggering crazing. Unfortunately, most of the SEP particles are small ([less than] 0.3 [micro]m) (Fig. 3). As a result, massive shear yielding is suppressed due to the lack of massive particle cavitation and crazing. The small shear yielding zone at the crack tip and the crack wake observed in PP/SEP indicates that massive shear yielding cannot occur without significant widespread relief of triaxial tension at the crack tip. This finding is consistent with the result obtained in PP/Noryl blend (23). In PP/Noryl, large Noryl particles are effective to bigger crazing so that large crazed zone is developed at the crack tip. However, the level of crazing is still not sufficient to relieve the triaxial stress constraint due to large inter-particle distances. On the other hand, when PP/Noryl blend is compatibilized with SEP, optimal particle size and interfacial adhesion between PP matrix and Noryl particles are formed (23). As a result, massi ve crazes are generated by Noryl particles, which results in the transformation of crazing into shear banding in PP/Noryl/SEP blends. These results suggest that the relief of triaxial stress at the crack tip is the key to widespread shear yielding. The findings are consistent with the toughening principles proposed by Yee and coworkers (19-21).
The morphology, mechanical properties, and toughening mechanisms of SEP-modified PP were studied. It is found that small SEP particles ([approximate] 0.2 [micro]m) are well dispersed in PP matrix. The addition of 5% by weight of SEP into PP can greatly increase the toughness of PP without lowering heat deflection temperature. The toughening mechanism investigations show that a large fan shape crazed zone and small intense plastic strips are developed. Crazing and shear banding are the main toughening mechanisms. It is also found that only the large SEP particles ([greater than] 0.3 [micro]m) are effective in triggering crazes. Since only a small portion of SEP particles are large, massive shear banding is suppressed.
The authors would like to thank Shell Chemical for donating materials for the present work. Special thanks are given to the Defense Logistic Agency (Contract # SPO 103-96-D-0023) for financial support of this work.
(*.) To whom correspondence should be addressed.
(+.) Current address: Avery Dennison. 902 Feehanville Drive. Mt. Prospect. IL 60056.
(1.) D. S. Parker, H.-J. Sue, J. Huang, and A. F. Yee, Polymer, 31, 2267 (1990).
(2.) C. B. Bucknall, Toughened Plastics, Appli. Sci. Pub. Ltd., London (1977).
(3.) S. Wu. Polymer, 26, 1855 (1985).
(4.) R. J. M. Borggreve, R. J. Gaymans, J. Schuijer, and J. F. Ingen Housz, Polymer, 28, 1489 (1987).
(5.) A. S. Argon, R. E. Cohen, O. S. Gebizlioglu, and C. Schwier, in Advances in polymer Science: Crazing in Polymer, vol. 52/53, p. 275, H. H. Kausch, ed., Springer, Berlin (1983).
(6.) R. J. M. Borggreve, and R. J. Gaymans, 7th International Conference on Deformation, Yield and Fracture of Polymers, p. 34, Cambridge, England (1988).
(7.) H. Breuer, F. Haaf, and J. Stabenow, J. Macromol Sci. --Phys., B14 (3), 387 (1977).
(8.) M. E. Boyce, A. S. Argon, and D. M. Parks, Polymer, 28, 1681 (1987).
(9.) B. Z. Jang, D. R. Uhlmann, and J. B. V. Sande, Polym. Eng. Sci., 25, 643 (1985).
(10.) B. Z. Jang, J. Appl. Polym. Sci., 30, 3449 (1985).
(11.) C. J. Chou, K. Vijayan, D. Kirby, A. Hiltner, and E. Baer, J. Mater. Sci., (a) 23, 2521 (1988). (b) 23, 2533 (1988).
(12.) J. U. Starke, G. H. Michler, W. Grellmann, S Seidler, M. Gableitner, J. Fiebig, and E. Nezbedova, Polymer, 39, 75 (1998).
(13.) G.-M. Kim and G. H. Michler, Polymer, 39, 5689 (1998).
(14.) J. Karger-Kocsis and I. Csikai, Polym. Eng. Sci., 27, 241 (1987).
(15.) M. Ishikawa, M. Sugimoto, and T. Inoune, J. Appl. Polym. Sci. 62, 1495 (1996).
(16.) A. van der Wal, J. J. Mulder, and R J. Gaymans, Polymer, 39, 6781 (1998).
(17.) R. R. Durst, R. M. Griffith, A. J. Urbanic and W. J. Van Essen, ACS Div. Org. Coat Plast. Prepr. 34, 320 (1974).
(18.) A. M. Donald and E. J. Kramer, J. Appl. Polym. Sci., 27, 3729 (1982).
(19.) A. F. Yee and R. A. Pearson, J. Mater. Sci., 21, 2462 (1986).
(20.) Ibid. 21, 2475 (1986).
(21.) A. F. Yee, Dongming Li, and Xiaowei Li, J. Mater. Sci., 28, 6392 (1993).
(22.) G.-X. Wei, H.-J. Sue, J. Chu, C. Huang. and K. Gong, Polymer, 41, 2947 (2000).
(23.) G.-X. Wei, H.-J. Sue, J. Chu, C. Huang, and K. Gong, J. Mater. Sci, 35, 555 (2000).
(24.) J. Varga, J. Mater. Sci., 27, 2557 (1992).
(25.) G.-X. Wei and H.-J. Sue, unpublished work.
(26.) A. J. Kinlock and R. J. Young, Fracture Behavior of Polymers, Applied Science, London, 1983.
(27.) H.-J. Sue and A. F. Yee, J. Mater. Sci., 24, 1447 (1989).
(28.) H.-J Sue, PhD thesis, The University of Michigan, Ann Arbor, Mich. (1988).
(29.) R. Hill, in The Mathematical Theory of Plasticity, Oxford University Press, (1950).
(30.) M. Ishikawa, I. Narisawa and H. Ogawa, J. Polym. Sci., 15, 1791 (1977).
(31.) M. Ishikawa and I. Chiba, Polymer, 31, 1232 (1990).
(32.) M. Ishikawa and H. Takahashi, J. Mater. Sci., 26, 1295 (1991).
(33.) M. Ishikawa, M. Sugimoto and I. Chiba, J. Appli. Polym. Sci., 62, 1495 (1996).
(34.) O. S. Gebizlioglu, H. W. Beekham, A. S. Argon, R. E. Cohen, and H. R. Brown, Macromolecules, 23, 3968 (1990).
(35.) Y. Okamoto, H. Miyagi, and S. Mitsui, Macromolecules, 26, 6547 (1993).
|Printer friendly Cite/link Email Feedback|
|Author:||WEI, G.-X.; SUE, H.-J.|
|Publication:||Polymer Engineering and Science|
|Date:||Sep 1, 2000|
|Previous Article:||Barrier Properties of Blends Based on Liquid Crystalline Polymers and Polyethylene.|
|Next Article:||Isothermal and Nonisothermal Crystallization Kinetics of Nylon 10 12.|