Characterization of strain bursts in high density polyethylene by means of a novel nano creep test
Graphical abstract
Introduction
In several semi-crystalline polymers it was found that the deformation of the crystalline phase is governed by the generation and motion of crystal defects - namely dislocations - especially at deformations within the extended plastic regime (Young, 1974; Shadrake and Guiu, 1976; O'Kane and Young, 1995; Argon et al., 2005; Nikolov and Raabe, 2006; Rozanski and Galeski, 2013). Although the importance of dislocations for the plastic deformation in many semi-crystalline polymers is beyond dispute (Galeski, 2003; Galeski and Regnier, 2009; Bartczak and Galeski, 2010) and the deformation behaviour can be well described by micromechanical models (Ayoub et al., 2010; Uchida and Tada, 2013; Shojaei and Li, 2013; van Dommelen et al., 2017), the kinetic mechanisms of dislocations remain widely unclear.
While diffraction methods like the Multi-Reflection X-Ray Profile Analysis (MXPA) (Ungár and Borbély, 1996; Kerber et al., 2011) and texture measurements (Lee et al., 1993; Galeski, 2003) as well as neutron diffraction (Brown et al., 2007), can be considered as highly phase selective, the situation is more complex for mechanical experiments as they register an overlap of effects originating from the crystalline as well the amorphous phase. Nevertheless, for analysing the nature, mobility and annihilation of defects, the determination of activation volumes and energies from mechanical tests seems indispensable for the understanding and identification of the molecular processes controlling elastic properties, strength and ductility.
The first attempts to investigate the deformation behaviour of the crystalline phase independent of the amorphous phase were made by Rabinowitz and Brown (1967). They observed the onset of plastic deformation in the crystalline phase of high density polyethylene (PE-HD) at a stress of 0.18 MPa by cyclic tension experiments in the micro-strain region with high-resolution strain measurement (strain sensitivity of 10−6). Due to the enormous demands on the measuring accuracy and the lack of commercially available measuring systems, no further investigations on polymers in the microstrain range were carried out over several decades. Only recent developments like the nanointendation (Schuh, 2006; VanLandingham et al., 2001) enable a new approach.
From metals it is known that under certain conditions the flow of deformation can be jerky, which means that the macroscopic deformation curve exhibits significant jumps (strain bursts). Uchic (Uchic et al., 2004) showed that the deformation of small cylindrical nickel single crystals, with a diameter of some μm, occurs almost exclusively through strain bursts. The jumps become smaller with increasing sample size and they disappear for macroscopic specimens. Type of deformation (tension or compression) and crystal orientation have also an influence to the strain burst characteristics (Kim et al., 2012).
The cause of strain bursts in metals are usually dislocation avalanches (Zaiser, 2006). In metallic glasses, shear banding and martensitic transformations can also cause strain bursts (Tong et al., 2016). Since the deformation of the crystalline phase in many semi-crystalline polymers is dislocation controlled (Galeski, 2003; Bartczak and Galeski, 2010), strain bursts should also emerge.
The first evidence for strain bursts in a polymer was found by Li and Ngan (2010) with nanoindentation creep experiments on PE-HD. With some estimations, based on experimental data, they could determine an activation energy of 0.22 eV for the plastic deformation of the crystalline phase. Further comprehensive experiments were reported by Zare Ghomsheh et al. (Zare Ghomsheh et al., 2015; Zare Ghomsheh, 2018) where an influence of the loading rate and the applied load on the number and the height of bursts could be verified.
However, nanoindentation creep experiments have several limitations.
- •
The stress field is strongly inhomogeneous (Kermouche et al., 2008), which requires some assumptions and estimations for the evaluation of the activation energy.
- •
The sample volume tested is very small, and local effects can play a role.
Therefore, an alternative test method would be desirable, which does not suffer from these disadvantages.
In semi-crystalline polymers, the dislocation path length is limited by the lamella thickness and therefore strain bursts can only be of the order of the lamella thickness. In order to verify single strain bursts in polymers, the resolution of the deformation measurement must be in the nm scale. In addition, the deformation rate in the amorphous phase should be not too large. The greater the creep rate in the amorphous phase, the more difficult to detect a single strain burst. A possibility could be a creep experiment above the glass transition temperature with very small loadings in the microstrain region (finite strain near the origin about 10−6). Thereby the soft amorphous phase deforms continuously with very small creep rates (few nm/s), while the deformation in the crystalline phase runs jerky over strain bursts (some nm in typically 0.1 s). This would allow to distinguish the deformation in the amorphous phase from that of the crystalline phase and to study the mechanisms of plastic deformation in the crystalline phase independently.
An important aspect is the application of a homogeneous stress field over a large specimen volume. Thus the occurrence of strain bursts becomes more likely. The separation of deformations in the amorphous and crystalline phase is only possible at very low stresses. If the stress exceeds a critical value, strain bursts do not occur as a single event. So they cannot be separated from the deformation of the amorphous phase. In addition, a too high deformation rate in the amorphous phase can make the detection of individual strain bursts impossible, as the rate may exceed the ones observed during strain bursts. Since strain burst in semi-crystalline polymers cause deformations of several nm (Li and Ngan, 2010), a nm resolution in the deformation measurement is required. The requirement on the temperature stability during a measurement is very high. On the one hand temperature fluctuations can influence the strength of the amorphous phase on the other hand the deformation caused by low thermal fluctuations is, due to the large thermal expansion coefficient of polymers, much higher than the required deformation resolution for strain measurements. For instance in PE-HD (linear expansion coefficient 1/K) a temperature increase of 0.1 °C causes, at a length of 10 mm, a length change of 200 nm. Therefore, a very good temperature stability during the measurement is required. In order to address this problem, a torsional loading with a pure shear deformation is more appropriate. In this load case, volume changes from temperature fluctuations do not play a critical role, since a volume change in a rotationally symmetric specimen does not cause torsional deformations. Furthermore a pure shear stress enables easy determination of the stress in the sliding plane.
Section snippets
Description of the novel nano-creep test
Many of the required criteria are fulfilled in a torsion measurement by means of a rheometer. This was the reason to develop the novel test method for strain bursts on the basis of a torsional rheometer test. For the development and evaluation of the testing method an Anton Paar rheometer MCR 301 with temperature chamber CCD 600 was used. This rheometer has a torque resolution < 1 μNm and an angular resolution < 1 μrad. Data acquisition with 10 Hz was done.
Results
For all evaluations, only measurement data > 1.5 min after the beginning of the experiment were considered. This rule ensures a very small average creep rate. Thus the torsion angle difference from the creep in the amorphous phase is much smaller than the expected jump height of the strain bursts. Fig. 3 shows the result of a nano-creep experiment (red curve) with a relatively high number of strain bursts.
Besides the positive strain bursts (in the direction of the loading), there are also back
Discussion
So far mechanical experiments in the micro-strain region were hardly performed in polymers (Patlazhan and Remond, 2012). The reasons are that commercially available strain gauge systems do not have the required high resolution. Moreover, there are high requirements to the temperature control during the experiment with respect to the large thermal expansion coefficient. One exception are the tensile tests of Brown and Rabinowitz (Brown and Rabinowitz (2002) with cyclic normal stress σ on PE-HD
Conclusions and summary
So far, recent observations of dislocation strain bursts occurring during the deformation of semi-crystalline polymers have been suffering from limitations in the experimental methods (nanoindentation creep) applied. This work reports from the development and application of a new method which allows for a direct control of the experimental parameters especially that of the applied stress, and combines it with an extraordinary high resolution of both the stress and the strain measurement. This
Acknowledgments
This work was supported by the Austrian Science Fund FWF (grant number P22913). In addition, many thanks to Melanie Messner, Rezai Jalal, Niklas Sabizer and Thomas Tröscher for the production of test specimens and Tanja Wagenknecht for the support by the numerous experiments. We would also like to thank Anton Paar for providing the rheometer MCR 301 with the temperature chamber CCD 600.
References (61)
- et al.
Rate mechanisms of plasticity in semi-crystalline polyethylene
Polymer
(2005) - et al.
Modelling large deformation behaviour under loading–unloading of semicrystalline polymers: application to a high density polyethylene
Int. J. Plast.
(2010) Relaxation processes in crystalline polymers: molecular interpretation a review
Polymer
(1985)- et al.
A new strain path to inducing phase transitions in semi-crystalline polymers
Polymer
(2007) Strength and toughness of crystalline polymer systems
Prog. Polym. Sci.
(2003)- et al.
Nano- and micro-mechanics of polymer blends and composites, hanser
- et al.
Morphology of undeformed and deformed polyethylene lamellar crystals
Polymer
(2010) - et al.
Extraction of stress–strain curves of elastic–viscoplastic solids using conical/pyramidal indentation testing with application to polymers
Mech. Mater.
(2008) - et al.
Crystallographic orientation and size dependence of tension–compression asymmetry in molybdenum nano-pillars
Int. J. Plast.
(2012) - et al.
Analysis of the non-linear creep of high-density polyethylene
Polymer
(1995)
Simulation of large strain plastic deformation and texture evolution in high density polyethylene
Polymer
Nanoscale fast relaxation events in polyethylene
Scripta Mater.
The α-factor in the taylor flow-stress law in monotonic, cyclic and quasi-stationary deformations: dependence on slip mode, dislocation arrangement and density
Curr. Opin. Solid State Mater. Sci.
Yielding of polyethylene through propagation of chain twist defects: temperature, stem length and strain-rate dependence
Polymer
The role of dislocations in γ-ipp under plastic deformation investigated by x-ray line profile analysis
Mech. Mater.
Plastic yielding of semicrystalline polymers affected by amorphous phase
Int. J. Plast.
Nanoindentation studies of materials
Mater. Today
Viscoplasticity analysis of semicrystalline polymers: a multiscale approach within micromechanics framework
Int. J. Plast.
Shear avalanches in plastic deformation of a metallic glass composite
Int. J. Plast.
Micro-, meso-to macroscopic modeling of deformation behavior of semi-crystalline polymer
Int. J. Plast.
Micromechanics of semicrystalline polymers: towards quantitative predictions
Mech. Res. Commun.
On the nature of acoustic emission and internal friction during cyclic deformation of metals
Acta Mater.
Evidence of dislocations in melt-crystallised and plastically deformed polypropylene
Mater. Sci. Eng., A
Creep behaviour of high density polyethylene films having well-defined morphologies of stacked lamellae with and without an observable row-nucleated fibril structure
Polymer
Molecular mechanisms of the chain diffusion between crystalline and amorphous fractions in polyethylene
Polymer
Effect of microstructure on crazing onset in polyethylene under tension
Polym. Eng. Sci.
Über die reaktionsgeschwindigkeit bei der inversion von rohrzucker durch säuren
Z. Phys. Chem.
Plasticity of semicrystalline polymers
Macromol. Symp.
Cited by (7)
Recent advances in slow crack growth modeling of polyethylene materials
2023, Materials and DesignIndentation creep dynamics in metallic glasses under different structural states
2023, International Journal of Mechanical SciencesCitation Excerpt :It is noted that the ‘pop-in’ events or serrations are unapparent in the P-h curves due to relatively high loading rates [64]. In Fig. 4, we give a closeup of a serration which plays a dominant role in the understanding of plastic deformation and fracture behavior as well as in controlling and designing the plasticity of metallic glasses [65–68]. As suggested in previous work, in the plastic deformation process of metallic glasses, the soft zones rich with flow defects are preferred to be activated to form STZs.
Understanding creep in TiAl alloys on the nanosecond scale by molecular dynamics simulations
2021, Materials and DesignCitation Excerpt :Nevertheless, it is worth noting that the simulations are spanning several nanoseconds, while only 200 ps were simulated in the previous category I. Recently, [57] reported strain bursts for nano-creep experiments in semi-crystalline polymers and assumed the origin of these bursts in dislocation mobility. Our MD simulations support this interpretation.
The influence of crystallization conditions on the macromolecular structure and strength of γ-polypropylene
2019, Thermochimica ActaCitation Excerpt :A larger proportion of thin sections increases the number of sections lying in a position with high Schmid factor, therefore, increases the probability of plastic deformation; (iii) Variations of original defect (dislocation) density: Different crystallization conditions can result in different initial dislocation densities. For plastic deformation of semi-crystalline polymers, a sufficient dislocation density is necessary [3,8]. If this dislocation density is too low, additional dislocations must be mechanically generated resulting in a higher strength; (iv) Varying number of tie molecules and entanglements: As these are part of the amorphous phase, a variation of their numbers may affect the strength [1,56].
Molecular dynamics simulation of plastic deformation in polyethylene under uniaxial and biaxial tension
2022, Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications