Elsevier

Polymer

Volume 42, Issue 13, June 2001, Pages 5871-5883
Polymer

A study on solid-state drawn fibers of polyethylene by confocal Raman microspectrometry: evaluation of the orientation profiles of amorphous and crystalline phases across the fiber section

https://doi.org/10.1016/S0032-3861(00)00913-7Get rights and content

Abstract

Polarized Raman macrospectrometry and confocal Raman microspectrometry have been used to evaluate the uniaxial draw of polyethylene (PE) at a draw ratio of ∼6. The draw was conducted by solid-state extrusion. The PE samples tested were synthesized to different molecular weights by slurry and gas phase processes at different temperatures. Raman spectra offer an important evaluation of draw, with bands identifiable for each: the amorphous, crystalline and interphase regions of the morphology. Such an interphase is preferentially located at the crystallite fold surface boundary and it has been associated with the presence of chain loops and entangled chain segments in this region. Orientations of the different phases are estimated by measuring the depolarization factors of the corresponding bands. Further, by confocal Raman microspectrometry, profiles of orientation across the fiber have been identified, by focusing the laser beam at different depths in the drawn PE. It was found that the crystalline component orients along the draw direction, while, across the fiber section, the interphase and the liquid-like amorphous component display preferred orientation normal to the draw direction. Correlations among orientation profiles, synthesis conditions and reactor powder ductility are discussed. Orientation profiles have been compared to the deformation profiles obtained by the displacement of pre-imprinted ink marks on the drawn fibers. A possible interpretation of the orientation profiles is attempted in terms of the relevant modes of deformation and alignment of chain segments in the different phases. The Raman results, in combination with wide angle X-ray analysis reveal steps in the drawn process that have implications for understanding the draw of PE and possibly for the general process of ‘necking’ on draw of semicrystalline thermoplastics.

Introduction

Polymeric fibers make up a class of products of high technological interest. On a weight basis, these fibers can display a set of physical and mechanical properties comparable or even superior to steel products. Among polymeric fibers, those made by ultra-high molecular weight polyethelene (UHMWPE) can be distinguished by their ability to provide, at a low cost, a unique combination of low specific weight with high tensile moduli and strength. These properties can prove particularly useful in a range of applications like in the anchorage of oilrigs, towing ropes, composite reinforcement and in bulletproof vests.

On a molecular basis, the achievement of high tensile moduli has been related to the extension and parallel alignment of polymeric chains so that the macroscopic elongation of the fiber can be transferred to the deformation of stiff covalent bonds and bond angles along the molecular backbone. Such a process can be greatly affected by the presence of entanglements among polymeric chains. Entanglements can act as friction centers and temporary junctions hindering and stopping the extension and the alignment of the chain segments [1], [2]. The maximum achievable draw ratio is expected to depend strongly on the degree of entanglements. However, entanglements can also provide effective stress-transfer connections between different chains and can improve the efficiency of draw [3]. For solid state draw (SSD) of semicrystalline polymers, it has been suggested that, especially in the initial stages of the draw process, higher contents of entanglements can lead to more significant strain of the amorphous phase and to higher crystallite alignment along the applied stress field [4].

Studies on the solid state deformation of semicrystalline PE have shown that the flow profiles corresponding to the changes in the strain rate across the fiber section can take different shapes, from simple flat or concave patterns to more complex W-shaped patterns [5], [6]. Differences in the strain rate along the fiber cross-section should correspond to changes in the orientations of the crystalline and amorphous phases at different depths below the fiber surface. Such data can be relevant for a better understanding of the effects of entanglements on the mechanism of plastic deformation.

Several techniques have been used to measure orientation profiles on draw. Thin sections of the specimens can be investigated by X-ray diffraction [7] or transmission electron microscopy (TEM) to estimate orientation at different depths. Attenuated total reflection infrared spectroscopy (FT-IR/ATR) [8] or multiple internal reflection infrared spectroscopy (MIR) [9] are more suitable and non-destructive techniques, but the range of achievable depths is limited to few μm [9]. Polarized Raman spectroscopy has been employed to measure the orientation function in PE fibers [7], [10], [11], [12], [13]. The technique is easy to implement and its non-destructive character prevents unwanted modifications in the oriented phases because of preliminary sample treatment. However, as such, this technique cannot be used for orientation profiling since the measured scattering is limited to a thin surface layer of the specimen [9]. In contrast, confocal Raman microspectrometry (CRM) has been successfully used for the depth profiling of physical and chemical parameters in several inorganic, polymeric and biological systems [14], [15], [16]. However, a survey of the recent literature does not show that this technique has ever been employed to investigate orientation profiles in polymer fibers. Actually, scattered intensities decrease significantly on moving the focal plane at increasing depths into the specimen. This leads to unreliable values of the depolarization factors required to compute orientation functions. Additional problems arise from optical inhomogeneities in the sample, which can modify the polarization of the scattered beam. Nevertheless, as reported in the following sections, intensity ratios measured at different depths by a particular experimental assembly still show a correlation with depolarization factors. Thus, in this work an attempt is made to use such values of the intensity ratios to estimate the orientation profiles of the crystalline and amorphous phases across the fiber section.

The samples studied in this work were a set of UHMWPE fibers obtained by solid-state coextrusion of nascent reactor powders. It was reported that, under particular conditions, synthesis could provide UHMWPE powders that can be directly drawn [1]. This is a possible indication of a low degree of entanglement in some reactor powders. Such a low degree of entanglement can be preserved by maintaining the process temperature below the melting point through the several stages of conversion of reactor powders into fibers. The SSD of UHMWPE reactor powders has been previously studied with the aim to correlate structural and morphological changes induced by draw with specific sets of synthesis conditions [4], [17], [18], [19]. Such studies have focused on the identification of reliable parameters to estimate powder ductility prior to draw. Possible correlations between the nascent degree of entanglements and synthesis conditions were also investigated [20], [21], [22], [23], [24]. These studies showed that coextrusion of reactor powders at a draw ratio of about six leads to specimens, whose physical and structural properties are still related to the initial morphology and to the persistence of nascent chain arrangements, stemming from specific combinations of synthesis conditions [4], [25]. Thus, correlations between synthetic conditions and orientation profiles of the corresponding SSD fibers can provide information on the local efficiency of draw. Such data might prove relevant to estimate the influence of entanglements on the development of the stress field along the fiber section during the coextrusion stage. They can also be useful to investigate the sequence of molecular processes at ever-higher draw.

Section snippets

Experimental

UHMWPE samples were synthesized at Union Carbide Corporation. Synthesis was performed by Ziegler–Natta heterogeneous catalysis, either by slurry or gas-phase process. Details of the syntheses have been given in Ref. [21]. Fibers have been obtained by pressing reactor powders at 150 kg cm−2 under vacuum at 120°C. Strips cut from the compressed plates were sandwiched between the two halves of a cylindrical split billet of high-density PE (see Fig. 1a) and the billet was coextruded through a bronze

Raman macrospectroscopy

The Raman spectrum of PE has been largely investigated and correlations between internal vibrational modes and observed bands have been published [29], [30], [31], [32]. The assignment of Raman bands to the different phases in semicrystalline PE samples has been the object of extensive studies too [7], [13], [33], [34], [35]. Table 2 lists the Raman wavenumbers of some of the more significant bands, their internal vibrational modes and the most common assignments to the different phases. The

Conclusions

Investigations on the ductility of PE reactor powders have focused on the correlations between synthesis conditions and ductility. The ultimate objective of these investigations is the identification of the most appropriate set of synthesis parameters for lowest degrees of entanglements. Coextruded fibers at EDR ∼6 proved useful for this type of investigations. At low draw ratios nascent structures have not yet been obliterated, while some structural features, that may be related to controlling

Acknowledgements

The authors wish to acknowledge the helpful discussions with Prof. Shaw L. Hsu of the University of Massachusetts on the interpretation of the Raman spectra. Authors are particularly grateful to Prof. T. Kanamoto for providing a fiber coextruded from SGC mats. In addition, the awarding of a NATO-CNR Senior Fellowship to one of the authors (S. Ottani) is gratefully acknowledged.

References (54)

  • C.P Lafrance et al.

    Polymer

    (1993)
  • G Zerbi et al.

    Polymer

    (1989)
  • L.H Wang et al.

    Polymer

    (1991)
  • M.J Gall et al.

    Spectrochim Acta

    (1972)
  • R.G Snyder et al.

    Spectrochim Acta

    (1978)
  • A Gibson et al.

    Polymer

    (1978)
  • J.M Brady et al.

    Polymer

    (1989)
  • A.E Zachariades et al.

    J Polym Sci, Polym Lett Ed

    (1979)
  • P Smith et al.

    Colloid Polym Sci

    (1980)
  • P Smith et al.

    J Polym Sci, Polym Phys Ed

    (1981)
  • S Ottani et al.

    Macromolecules

    (1996)
  • T Kanamoto et al.

    Polym J

    (1979)
  • T Kanamoto et al.

    J Polym Sci, Polym Phys Ed

    (1979)
  • M.R Pereira et al.

    J Polym Sci, Part B: Polym Phys

    (1994)
  • J Maxfield et al.

    J Polym Sci, Polym Phys Ed

    (1978)
  • D.I Bower

    J Polym Sci, Polym Phys Ed

    (1972)
  • D.V Luu et al.

    J Raman Spectrosc

    (1980)
  • M.J Citra et al.

    Macromolecules

    (1995)
  • S Hajatdoost et al.

    Appl Spectrosc

    (1996)
  • W Schrof et al.

    Colloid Polym Sci

    (1998)
  • A.M Radder et al.

    J Mater Sci: Mater Med

    (1995)
  • S Ottani et al.

    Macromolecules

    (1995)
  • S Ottani et al.

    Macromol Rapid Commun

    (1995)
  • S Ottani et al.

    Macromolecules

    (1996)
  • S Ottani et al.

    Polym Commun

    (1990)
  • S Ottani et al.

    J Polym Sci, Part B: Polym Phys

    (1991)
  • L.H Wang et al.

    J Polym Sci, Part B: Polym Phys

    (1991)
  • Cited by (34)

    • Raman spectroscopy of biomedical polyethylenes

      2017, Acta Biomaterialia
      Citation Excerpt :

      More importantly, it is not clear whether the in vitro testing in hip simulator could actually reproduce at the microscopic scale the in vivo frictional interaction between the joint sliding surfaces. In an attempt to clarify the origin of implant degradation and to propose methods for elongating their lifetime, a number of Raman studies have been carried out, which focused on assessing phenomena of chemical and structural degradation (e.g., oxidation and degree of crystallinity) in polyethylene structures [51,54,141–143]. Such Raman studies have provided precious contributions to advance our knowledge of degradation processes in biomedical polyethylene grades.

    • Spectroscopic elucidation of structure-property relations in filaments melt-spun from amorphous polymers

      2017, European Polymer Journal
      Citation Excerpt :

      From detailed studies on polyethylene and ROMP reactions [30–33] as well as on amorphous polyamides [34] and polyamide 12 [35], CC bending (δ) and stretching (ν) vibrational modes were assigned for both COP and CoPA to low wavenumbers (500–1150 cm−1), with rocking (ρ), twisting (τ) and bending CH bands at higher wavenumbers (1150–1500 cm−1). Different authors have ascribed the vibrational bands at 1088, 1167, 1297 and 1449 cm−1 to the amorphous phase regions in polyethylenes and cycloolefins [30–32,36]. In the spectrum of CoPA an additional broad peak is observed at 1646 cm−1 assigned to the keto-amide group, which is particularly sensitive to the state of hydrogen bonding in the polymer [34]; this feature is not present in the COP fiber.

    • Deformation mechanism of high-density polyethylene probed by in situ Raman spectroscopy

      2015, Polymer
      Citation Excerpt :

      Two orientation parameters 〈P2〉 and 〈P4〉 given by the polarized Raman spectroscopy enable us to discuss the microscopic mechanism of deformation in detail. Raman spectroscopy has been mainly applied off line for fibers and drawn specimens [31–34]. It should be emphasized that in situ measurements make it possible to provide transient data of the deformation behavior at the molecular level under tension, excluding the effects by stress relaxation that take place immediately after unloading the specimen.

    • The use of Raman spectroscopy in the analysis of UHMWPE uni-condylar bearing systems after run on a force and displacement control knee simulators

      2013, Wear
      Citation Excerpt :

      All the Raman measurements were made in a fully non-destructive way, without any sample manipulation. Under these experimental conditions, the laser spot size was about 1 μm; in a previous paper [19] the field depth for our confocal setup using the 50 μm-pinhole has been calculated as about 5 μm. Therefore, it can be concluded, according to Zerbi et al. [36] that no deep inside of UHMWPE significantly contributes to the Raman signal.

    • Does knee implant size affect wear variability?

      2013, Tribology International
      Citation Excerpt :

      To gain more insight into the wear behaviour, the morphology of the tested inserts was evaluated by micro-Raman spectroscopy. The Raman data obtained in the present study can be related to the wear behaviour of the specimens since they were demonstrated to be corresponding to the surface of the samples [34,35]. The Raman spectrum of PE has been largely investigated and various authors have tried to identify marker bands of PE morphology and correlate their intensity with the crystallinity of the polymer [33,36].

    View all citing articles on Scopus

    Deceased author.

    View full text