摘要:
Paper Shear-induced crystallization of polyethylene studied by small- and wide-angle X-ray scattering (SAXS/WAXS) techniques{ Ellen L. Heeley,*a Ariana C. Morgovan,a Wim Bras,b Igor P. Dolbnya,b Anthony J. Gleesonc and Anthony J. Ryana aDepartment of Chemistry University of Sheffield, Sheffield, UK S3 7HF. E-mail: E.Heeley@sheffield.ac.uk; Fax: 44 (0)114 273 8673; Tel: 44 (0)114 222 9372 bDUBBLE CRG, ESRF, F-38043 Grenoble, France. Fax: 33 (0)476 882412; Tel: 33 (0)476 882375 cDaresbury Laboratory, Daresbury, Warrington, Cheshire, UK WA4 4AD Received 12th July 2002, Accepted 8th October 2002 First published as an Advance Article on the web 17th October 2002 Shear induced crystallization of commercial grade polyethylenes (PE) have been investigated using simultaneous SAXS and WAXS, enabling real time studies of the kinetics and structure development to be performed.The processing of semi-crystalline polymers such as polyethylene (PE) involves the shaping of the molten material in moulds and dies by injection moulding methods or film blowing and fibre spinning techniques.1 Once processed, the polymer shape is stabilised via crystallization, controlling and determining the aesthetic and mechanical properties of the final material. However, procedures like injection moulding and fibre spinning impose shear flow into the polymer, which, in turn affects the crystallization kinetics and thus morphology. The rate of crystallization is profoundly altered by the influence of flow due to the alignment or orientation of segments of the molecular chains.This orientation of the chains is believed to affect the crystallization kinetics by providing sites where nucleation is promoted (in the flow direction). From these nucleation sites oriented lamellae grow radially, perpendicular to the flow direction. The development of oriented lamellar morphologies in bulk polymers has been termed the ‘shish kebab’ model, where the shish are the portions of oriented molecules and the kebabs are the outgrowing lamellae.2 To study the development of shear-induced macrostructure small- and wide-angle X-ray scattering (SAXS/WAXS) are invaluable tools, allowing the long range ordering (lamellae stacking) from SAXS and crystalline atomic ordering from WAXS to be probed. With the use of synchrotron radiation and advances in ‘in-situ’ shearing devices, it is possible to follow the crystallization of semi-crystalline polymers using simultaneous time resolved SAXS and WAXS.In this communication we present some recent data obtained using time resolved SAXS and SAXS/WAXS, to study the morphological development in commercial grade PE using a shear instrument{ adapted for use with X-rays. Here, the PE samples are subjectedtoa pulse of pre-shear at the crystallizationtemperature having been quenched from the melt. The isothermal crystallization temperature (Ti) chosen was such that quiescent crystallization at Ti would take several hours to complete. The PE samples{ investigated have different molecular weights and polydispersities allowing an insight into how these features affect the crystallization kinetics and development of the final macrostructure.Fig. 1, gives the temperature and shear programme followed for the PE samples, where Ti is the chosen crystallization temperature and Tm is the melting point of the PE sample. The two PE samples used in this study were Lupolen PE and PE1220 provided by BASF and Dow respectively.{ Firstly, time resolved SAXS data was recorded for the PE samples at several temperatures after pre-shear (constant shear units). Pre-shearing Lupolen PE for 5 s at 100 s21 (500 shear units) at temperatures close to the melt temperature (112 uC) is seen to give rise to orientation in the SAXS pattern. In Fig. 2, a selection of SAXS frames (at various times and temperatures) are given, showing the development of oriented lamellar structures (to varying degrees) indicated by the classic ‘teardrop’ shape in SAXS.Here, the shear direction is vertical with respect to the 2D-SAXS patterns. The SAXS shows increased meridional scattering, which, indicates stacking of lamellar perpendicular to the shear direction. At lower crystallization temperatures, well-oriented crystalline stacks give sharp SAXS peaks in the meridional direction. However, as the crystallization temperature increases, the crystallization rate decreases and the lamellar structure is less well defined. Some of the orientation is lost as the chains relax back and the SAXS shows a more isotropic signature of lamellae. Fig. 3, gives selected SAXS frames for a PE1220 sample where, the shear rate is 10 s at 100 s21 (1000 shear units).With this sample, the SAXS is less oriented and the chains relax back to give mostly isotropic {This paper was originally presented as a poster at the Faraday Discussion 122 meeting. {A Linkam CSS450 shear device equipped with Kapton windows was used for all shear/SAXS experiments at the DUBBLE CRG beamline BM26B, ESRF, France using an 8 m SAXS camera. SAXS data was recorded on a 2D gas-filled multiwire area detector at a frame rate of typically 5 s. SAXS/WAXS experiments were performed on station 16.1 at the Daresbury SRS, UK. SAXS data was recorded on the Rapid detector systems synchronised with a WAXS Photonics CCD detector. Again, all data frames typically collected at 5 s. The PE samples were provided by BASF (Lupolen, Mw ~ 250 k, Mw/Mn ~ 13.5, Tm ~ 112 ¡2 uC ) and Dow (PE1220Mw ~104 K,Mw/Mn~4.7, Tm~122¡ 5 uC).Fig. 1 Temperature and shear programme for PE. 158 PhysChemComm, 2002, 5(23), 158–160 DOI: 10.1039/b206832j This journal is # The Royal Society of Chemistry 2002crystallization (lamellae having random orientations). The crystallization rate is also seen to be quite slow. However, even with pre-shear the crystallization kinetics of this particular sample are still greatly increased in comparison with the quiescent crystallization at similar temperatures. To show this more clearly, the SAXS development (invariant, Q) has been plotted with respect to time for shear-induced crystallization at different temperatures (Q is estimated by sector integrations of the meridional SAXS intensity with respect to time).Fig. 4, gives, Q, versus time for both the Lupolen and PE1220 samples. The plots include a comparison of crystallization under quiescent conditions indicating the crystallization half-time, t½ (50% of the final crystallinity attained). The Lupolen PE sample shows the normalised invariant at temperatures ranging from 109 to 112 uC; the quiescent crystallization is also represented at 109 uC. Clearly, the lower the crystallization temperature, the faster the kinetics. However, at 109 uC, the quiescent crystallization is very slow, even in comparison with the high crystallization temperatures after pre-shear close to the melting point (y112 uC), where crystallization is expected to take several hours under quiescent conditions. The t½ on the graph in Fig.4 shows that the crystallization rate after preshear at 109 uC, is increased up to 2 orders of magnitude compared with that of the quiescent crystallization. Similarly, for the PE1220 sample, pre-shearing increases the crystallization at all temperatures compared to the quiescent crystallization (at the lowest temperature given). Again, the pre-shear crystallization rate at 117 uC and the quiescent crystallization rate at this temperature are still an order of magnitude apart. The inset in both graphs shows a SAXS pattern representing the quiescent crystallization structure after y60 min at the appro- priate temperature. Contrasting the quiescent SAXS patterns in both PE samples with those shown in Fig. 2 and 3, it is obvious that orientation and lamellar stacking is present after pre-shear and gives rise to the characteristic shape and greater scattering intensity on the meridian.Further experiments were performed on the Lupolen PE sample this time with simultaneous SAXS/WAXS during the crystallization process. In Fig. 5, examples of selected frames of SAXS/WAXS are given (the WAXS is recorded with the detector situated at the equator, perpendicular to the shear direction). Here, the sample has been pre-sheared at two rates and crystallized at different temperatures. The WAXS is seen to develop along with the SAXS at 108 uC (initially, only the amorphous halo is visible), being signified by the appearance of the (110) ring, intensifying with time. Finally, the WAXS (110) displays some orientation on the equator.The other set of Fig. 2 SAXS Lupolen PE after pre-shear at several crystallization temperatures. Fig. 3 SAXS of PE1220 after pre-shear at three crystallization temperatures. Fig. 4 Development of SAXS invariant during pre-shear and quiescent crystallizations of Lupolen PE (blue graph) and PE1220 (red graph). Fig. 5 SAXS/WAXS of Lupolen PE. (A) sheared for 5 s at 100 s21, crystallized at 108 uC. (B) sheared for 5 s at 150 s21, crystallized at 110 uC. PhysChemComm, 2002, 5(23), 158–160 159SAXS/WAXS patterns show the crystallization at a higher temperature after pre-shear. Again the SAXS and WAXS start to develop however, the WAXS appears to lag behind the SAXS development. The WAXS (110) ring is clearly visible with less orientation in the final frame of the crystallization; this is supported by the SAXS, which is less intense and less oriented at the same time.There are systematic changes in the diffuse halo in the WAXS prior to the growth of sharp Bragg reflections. The interpretation of this data with liquid-structure calculations is ongoing. Shear-induced crystallization in PE followed by SAXS and WAXS techniques, has shown that oriented lamellar prevail even at high crystallization temperatures compared with quiescent conditions (where the crystallization rate is slower and isotropic scattering is observed). Overall, the Lupolen PE sample appears to show greater oriented structure from the intense meridional SAXS patterns, whereas the PE1220 sample shows less oriented structure after pre-shear with some relaxation of chains as the crystallization process persists (seen by greater scattering on the meridian initially in the experiment, changing to more isotropic scattering features in the advanced stages).The use of both SAXS and WAXS techniques enables one to get information about the atomic crystalline structure as well as the determination of the lamellar periodicity during the crystallization process. Here, initial experimental evidence has indicated that crystallization at small undercoolings after preshear, suggests that some SAXS or long range ordering occurs prior to any WAXS crystalline structure. This type of precrystalline long range ordering has previously been observed in quiescent crystallization and extrusion experiments for several semi-crystalline polymers and can be related to the formation of pre-crystallization density fluctations.3,4 This interpretation is often disputed and the appearance of SAXS before WAXS is just the initial stage of crystallization, when macromolecules do not have a sufficient degree of order to give rise to well defined diffraction peaks, but enough to locally change the (electronic) density of the polymer.This has provided theoretical and experimental descriptions of a route in which molecular chains unravel from the entangled state to the final regular crystalline structure. However, attempting to understand and determine the influences of molecular characteristics with respect to crystallization kinetics under shear conditions, leads us to discuss the roles that Mw and polydispersity can play. This is an area of intensive research activity by a number of groups,5-9 which has led to questions regarding the influence of ‘long chain molecules’ in the polymer system.5,6 Classically, the long chains or chains above critical Mw 7–9 are thought to be the species which are oriented during shear and persist to provide the nucleation sites or the ‘shish’ on to which the lamellae or ‘kebabs’ grow (Fig.6, shows the mechanism of the ‘shish kebab’ morphology which is though to occur during shear-induced crystallization). The short chains below critical Mw are not oriented but relax back during the shear process. Here, this process is seen to operate if we consider the Mw and polydispersity of the two PE samples. The Lupolen sample has a relatively highMw compared to that of PE1220 and as importantly, a far greater polydispersity.Increased numbers of long chains in the distribution of the Lupolen sample, will give rise to more oriented chains persisting after pre-shear and hence the observations in the SAXS of well oriented structure development. The shorter chains in the distribution (those below the critical Mw) do not orient into nucleation sites but are most likely incorporated into the overgrowing lamellae. In comparison, the PE1220 sample is less polydisperse and one would expect that there would be fewer chains of critical Mw (and above) which will orient and persist without relaxation. This is what we are able to conclude from the SAXS data reported here: reduced overall orientation and noticeable relaxation of some initial oriented structure post shear, even after increased shear units.In summary, the data reported here provides an insight into the crystallization kinetics of commercial grade polyethylenes, which have been subjected to flow. The molecular characteristics of these samples clearly determine the development and extent of shear-induced crystallization and therefore the final macrostructure. However, it is difficult to distinguish the real role of each of these characteristics (e.g. Mw, polydispersity, molecular architecture) in such ill-defined materials. Thus, in the future, to examine these effects it will be necessary to study ‘model’ polymers under similar flow conditions. These model polymer systems will have well-controlled and diverse architectures and narrow polydispersities. Successful techniques to observe the structure development of oriented polymer melts have been illustrated here using simultaneous SAXS and WAXS methods. These techniques allow us to probe the very early and later stages of the overall crystallization process.The very early stages of polymer melt crystallization (quiescent and under flow) is still an area which is poorly understood. Finally, with model polymers and time resolved scattering techniques allowing improved investigations into the melt crystallization process, a deeper understanding of the kinetic mechanisms and pathways will be obtained. This will then eventually, lead to improved and more controlled manufacture of such polymer materials in the future. Acknowledgements We would like to acknowledge the financial support for this research from the EPSRC grant GR/M 60415.Also, support from the Microscale Polymer Processing project for experimental and theoretical backing. Finally, the authors would like to thank all beamline staff and techical support at both the ESRF and Daresbury laboratory synchrotron facilities. References 1 A. N. Wilkinson and A. J. Ryan, Polymer Processing and Structure Development, Kluwer, Dordrecht, 1998. 2 A. Keller and F. M. Willmouth, J. Macromol. Sci., 1972, B6, 493. 3 A. J. Ryan, J. P. A. Fairclough, N. J. Terrill, P. D. Olmsted and W. C. K. Poon, Faraday Discuss., 1999, 112, 13. 4 E. L. Heeley, C. K. Poh, W. Li, A. Maidens, W. Bras, I. P. Dolbnya, A. J. Gleeson, N. J. Terrill, J. P. A. Fairclough, P. D. Olmsted, R. I. Ristic, M. J. Hounslow and A. J. Ryan, Faraday Discuss., 2002, 122, in the press. 5 M. R. Mackley and A. Keller, Polymer, 1973, 14, 16. 6 R. H. Somani, B. S. Hsiao, A. Nogales, S. Srinivas, A. Tsou, I. Sics, F. J. Balta-Calleja and T. A. Ezquerra, Macromol., 2000, 33, 9385. 7 G. Kumaraswamy, A. M. Issaian and J. A. Kornfield, Macromol., 1999, 32, 7537. 8 M. Seki, D. W. Thurman, J. P. Oberhauser and J. A. Kornfield, Macromol., 2002, 35, 2583. 9 U. Goschel, F. H. M. Swartjes, G. W. M. Peters and H. E. H. Meijer, Polymer, 2000, 41, 1541. Fig. 6 Mechanism of ‘shish kebab’ development during shear-induced crystallization. 160 PhysChemComm, 2002, 5(23), 158–160
ISSN:1460-2733
DOI:10.1039/b206832j
出版商:RSC
年代:2002
数据来源: RSC