Micro-observation of shear wave attenuation mechanism in nylon-66
Ting Li, Zhiping Tang ⁎, Jian Cai
Key Laboratory for Mechanical Behavior and Design of Materials (LMBD), Department of Modern Mechanics,University of Science and Technology of China, Hefei, 230026, China
Received 30 October 2005; accepted 14 July 2006Available online 11 August 2006
The inclined planar plate impact technique developed byClifton et al. [1] and Gupta et al. [2], respectively, has been usedextensively for investigating the dynamic shear property ofmaterials [3–6]. Recently, Irfan and Prakash [7] applied thistechnique to study the transient friction and sliding behavior ofmetals on the impact surface. Brostow et al. [8] discussed theimportance of tribology in polymer science and engineering,such as friction, scratch resistance and wear. Gupta [3] studiedthe impact response of PMMA subjected to combinedcompression and shear loading and found rapid shear attenua-tion between the impact surface gauge and the interior gauges.Gupta suggested a time-dependent shear failure near the impactsurface. However, what kind of failure is it? What happens nearthe surface, and what is the physical mechanism of this failure?These questions are still unanswered.
To explore the failure mechanism near the impact surface ofpolymers, nylon-66 is chosen for experimental investigation in
the present study, since nylon-66 has a spherical grain structurewhich can be observed under a polarized microscope.
2. Experimental work
The inclined planar plate impact experiments are conductedby using a Φ57 mm keyed gas gun. Fig. 1 shows theexperimental setup: a nylon-66 flyer of 8 mm thick hits anylon-66 target sample of 25 mm thick with velocity uo at aninclined angle α in a magnetic field H. The compression andshear waves are recorded by the electro-magnetic particlevelocity (EMV) gauges embedded in the sample [9,10]. Lengthsof the gauges are 17.0, 15.0, and 13.0 mm, and the locations are2.0, 4.0, and 6.0 mm from the impact surface, respectively.
The impact velocity uo ranges from 100 m/s to 340 m/s andthe inclination angle α is from 10° to 20° (less than the frictionalangle). Fig. 2 is a typical experimental record of transversalparticle velocities in shot 0513 for uo=214.9 m/s and α=10°.The dash line represents the profile of the transmitted shearwave at the impact surface, and its amplitude is calculated fromuo and α theoretically. It can be seen that the shear componentattenuates greatly (about 50%), from the impact surface to thefirst gauge (the distance between them is 2 mm), then
1437T. Li et al. / Materials Letters 61 (2007) 1436–1438
propagates almost constantly from gauge 1 through gauge 3.This phenomenon is the same as that found by Gupta inPMMA [3].
The experimental results show that shear attenuation occurswhen uo exceeds a critical value for a fixed inclination angle α.The critical value of uo is about 214.9 m/s for α=10° and113 m/s for α=20°, respectively. When uo reaches 340.5 m/sfor α=10° or 180 m/s for α=20°, the shear waves recorded bythe gauges have almost disappeared.
A Nikon polarized optical microscope is used to examine themicro-structural change in the samples, after the impact. Therecovered samples are cut into thin films of 5-μm thicknessalong the impact axis. Fig. 3 illustrates the initial microstructureof a sample near the surface, before the experiment. The figureshows a clear spherical grain structure. Since the sample is madeof commercial nylon-66, the spherical grain distribution is nothomogenous at microscopic level.
Fig. 4 is the polarized picture of shot 0402 with impactvelocity uo=134.3 m/s and inclined angle α=20°. The figureshows clearly that there is a dark layer underneath the impactsurface, indicating a non-crystalline layer. This means thatmelting occurs during the shear loading, and re-crystallizationdoes not happen during the unloading process. The thickness of
Fig. 2. Typical experimental record of shot 0513, uo=214.9 m/s, α=10°. G1, G2and G3 are the shear wave profiles of 3 EMV gauges embedded in the sample.Their locations are 2, 4, and 6 mm from the impact surface, respectively.
this layer is about 6–8 μm, and the location is about 8 μm fromits center to the impact surface. Even more interesting is that adiscontinuous un-melted layer about 2–3 μm thick can be foundabove the melted layer, indicating that the melting may not becaused by the sliding and dynamic friction on the impactinterface, and the melting heat is generated inside of the sampleand near the surface.
The polarized photo of shot 0404 with uo=156.3 m/s andα=20° is shown in Fig. 5. It has features similar to thoseshown in Fig. 4. However, the top un-melted layer looksthinner and more discontinuous. This means that there weremore heat produced with the increase in impact velocity andshear stress.
Fig. 6 shows another picture of shot 0402 withuo=134.3 m/s and α=20°, but its location is near the radialperiphery of the sample. Due to the influence of lateralrarefaction waves from the periphery of the cylindrical sample,the shear stress component in this area is smaller than that inthe central part of the sample. The figure shows no meltedlayer in this case, but instead, there appears a bright whiteband located about 10–12 μm from the impact surface to itscenter. The thickness of this band is about 2–3 μm and it is atypical adiabatic shear band. The strong plastic deformation ofthe grains in this band produces more refraction of the
Fig. 3. The original microstructure of a nylon-66 sample near the surface beforeexperiment.
Fig. 4. Microscopic picture near the impact surface of shot 0402 withuo=134.3 m/s and α=20°.
Fig. 5. Polarized photo near the impact surface of shot 0404 with uo=156.3 m/sand α=20°.
Fig. 6. Microscopic picture of shot 0402, located near the periphery of thesample.
1438 T. Li et al. / Materials Letters 61 (2007) 1436–1438
polarized light under the microscope and forms the brighterband in the picture.
3. Discussion
The polarized microscopic observation discloses that whenthe transverse impact velocity increases, plastic deformationnear the impact surface will produce a correspondingly largeamount of heat. The low heat conductivity of polymer and theimpulsive loading condition will lead to the forming of anadiabatic shear band. Since the impact interface is constrainedby longitudinal pressure and transversal friction (the inclinedangle is less than the friction angle), the shear band formsbeneath the impact surface. When the loading amplitude
increases further, heat will cause the shear band to transforminto a melted layer (the melting point of nylon-66 is about250 °C). This shear band can cause the decay of shear waves,and the ensuing melting will severely increase the attenuation ofshear waves.
Since most polymers have low heat conductivity and a lowmelting point, the failure mechanism found in this study fornylon-66 could be applied to other polymers and may be auseful reference for polymer tribology study.
Acknowledgments
The authors are grateful to Prof. Y. Horie and Donna Horie atEglin AFB for the correction of the manuscript. This work issupported by the Chinese National Natural Science Foundation(10202022).
References
[1] A.S. Abou-Sayed, R.J. Clifton, L. Hermann, Exp. Mech. (1976) 127.[2] Y.M. Gupta, D.D. Keough, D.F. Walter, K.C. Dao, D. Henley, A.
Urweider, Rev. Sci. Instrum. 51 (1980) 183.[3] Y.M. Gupta, J. Appl. Phys. 51 (1980) 5352.[4] A. Gilat, R.J. Clifton, J. Mech. Phys. Solids 33 (1985) 263.[5] J.B. Aidun, Y.M. Gupta, J. Geophys. Res. 100 (1995) 1955.[6] K.J. Frutschy, R.J. Clifton, J. Mech. Phys. Solids 46 (1998) 1721.[7] M. Irfan, V. Prakash, Int. J. Solids Struct. 37 (2000) 2859.[8] W. Brostow, J.-L. Deborde, M. Jaklewicz, P. Olszynski, J. Mater. Educ. 24
