Elsevier

Polymer Testing

Volume 25, Issue 8, December 2006, Pages 1095-1100
Polymer Testing

Material Properties
Impact tensile fracture testing of a brittle polymer

https://doi.org/10.1016/j.polymertesting.2006.07.006Get rights and content

Abstract

The fracture behavior of a brittle polymer, methylmethacrylate–butadiene–styrene resin, under impact tensile loading was studied using single-edge-cracked specimens. The dynamic load and displacement were measured with a Piezo sensor and a high-speed extensometer, respectively. The load and displacement diagram, i.e., the external work, Uex, applied to the specimen was used to determine the elastic energy, Ee, and non-elastic energy, En, due to viscoelastic and plastic deformation, and the fracture energy, Ef, for creating new fracture surface, As. The energy-release rates were then estimated using Gt=Uex/As and Gf=Ef/As. The values of Gt and Gf were correlated with the fracture loads and the mean crack velocities determined from the load and time relationships.

Introduction

The fracture behavior of brittle polymers has been investigated under various loading conditions to determine their mechanical reliability. For example, under a static load, the stress intensity factor, K, or energy-release rate, G, is usually determined based on linear elastic fracture mechanics by measuring crack length and the load or external work applied to a tensile or bending specimen. The concept of K or G based on linear fracture mechanics reportedly plays an important role in understanding fracture initiation. K or G is also measured under impact loading to estimate their dynamic values using different experimental techniques [1], [2], [3]. Most polymers generally exhibit some non-elastic effect due to viscoelastic and plastic deformation [4]. Brittle fracture also causes inertia or a dynamic effect, since a crack propagates dynamically within the specimen [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. Hence, these two effects should be determined and considered when evaluating K or G. However, quantitative discussion of these two effects on K or G has been limited.

To study this problem under static loading, Arakawa and Mada [22], [23] measured the brittle fracture of polymethyl methacrylate (PMMA) using a high-speed camera, and a high-speed extensometer consisting of an optical fiber and a position-sensing detector (PSD). The dynamic and non-elastic effects were then estimated from the dynamic response and residual deformation of the specimen after fracture. They suggested that G is overestimated if the dynamic and non-elastic energies are included.

In this study, we examined these two effects under impact tensile loading using single-edge-cracked tensile specimens of methylmethacrylate–butadiene–styrene (MBS) resin. The impact load and displacement of the specimens were measured using a Piezo sensor and high-speed extensometer, respectively, to evaluate the external work, Uex, applied to the specimen. The elastic energy, Ee, and non-elastic energy, En, were estimated from the oscillation of the split specimen just after fracture. The fracture energy, Ef, was then determined and correlated with the fracture load, Pc. For the fracture surface, As, the energy-release rate was evaluated using Gt=Uex/As or Gf=Ef/As. The mean crack velocity, vm, was also estimated as a function of Pc, and their correlation is discussed.

Section snippets

Experimental methods

Experiments were performed on single-edge-cracked tensile specimens of 4 mm-thick MBS resin plates. The specimen geometry is shown in Fig. 1. To change the fracture initiation load of the specimens, sharp pre-cracks of different lengths of 2–4 mm were generated by forcing a razor blade into a pre-machined saw-cut on the specimen edge.

An impact tensile load was introduced using a special loading device, which utilizes the free fall of a weight. As illustrated in Fig. 2, this device consists of an

Definition of fracture energy

Fig. 4 plots the load, P, versus displacement, δ, for the dead weight of the frame (see Fig. 2). In this figure, Po and δo indicate the initial static values due to the dead weight, and Pc and δc denote the critical dynamic values at the onset of fracture. Although Pδ relationships generally differ between static and dynamic loads, they were assumed to be identical to simplify the analysis. The external work, Uex, applied to half of the specimen is then given asUex=Pcδc/2.Eq. (1) is valid when

Results and discussion

The dynamic load, P′, applied to the specimen was measured using the Piezo sensor (see Fig. 2). Fig. 5 plots the value of P′ as a function of time, t. P′ increased with t, and then fell abruptly from Pc as the crack started to propagate. The fall time from Pc to P=0 was about 38 μs, suggesting that the mean crack velocity in the specimen was about 190 m/s.

Fig. 6 shows the dynamic displacement, δ′, measured using the optical fiber and PSD (see Fig. 3). As shown in the figure, δ′ increased with

Conclusions

The brittle fracture of MBS resin was studied using an impact tensile device and an optical high-speed extensometer. The impact load and displacement of single-edge-cracked specimens were measured to determine the external work, Uex, applied to the specimen. The elastic energy, Ee, and non-elastic energy, En, just after fracture were estimated from the oscillation of the split specimen. The fracture energy, Ef, was then determined and correlated with the fracture load, Pc. With the fracture

References (24)

  • A.S. Kobayashi et al.

    Dynamic fracture toughness

    Int. J. Fract.

    (1986)
  • J.W. Dally et al.

    A study of hysteresis in the KID- relation

    Exp. Mech.

    (1990)
  • Cited by (6)

    • Infrared thermographic evaluation of large composite grid parts subjected to axial loading

      2015, Polymer Testing
      Citation Excerpt :

      Hence, about 44% of the total energy dissipates during the formation of micro-defects and, respectively, 56% of the energy is expended in the fracture of composite fibers. It is worth noting that the estimated measurement error was ±11 % in the above-mentioned experiments, and similar results were obtained by Arakawa et al. [21]. The next experiments were performed on 11 standard flat composite specimens.

    • Tensile fracture testing and energy evaluation of a light-cured composite resin

      2010, Polymer Testing
      Citation Excerpt :

      Nonetheless, quantitative appraisals of both effects on fracture toughness have seldom been undertaken due to the difficulties inherent in the experimental techniques currently used. We have explored these issues by measuring brittle fracture in polymers using a tensile loading device we have developed [26–30]. This device consists of a high-speed extensometer that allows quantitative measurements to be made of the static and dynamic displacements near the crack position in single-edge-cracked specimens.

    • Ductile and brittle material failures in low-velocity impact

      2010, Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications
    • Fracture energy evaluation of a dental composite

      2011, ICCM International Conferences on Composite Materials
    View full text