PSI - Issue 82

Koji Uenishi et al. / Procedia Structural Integrity 82 (2026) 72–78 Uenishi et al. / Structural Integrity Procedia 00 (2026) 000–000

75

4

5.610 ms

6.015 ms

6.845 ms

6.860 ms

6.995 ms

7.000 ms

7.035 ms

8.450 ms

8.455 ms

8.595 ms

9.770 ms

10.970 ms

12.165 ms

13.920 ms

Fig. 2. (continued).

under tensile loading (Uenishi and Nagasawa, 2023; Uenishi et al., 2024). Also, if there exists only one single perforation line in a specimen, dynamic fracture can move in an “unzipping” way along the perforation line but the propagation speed fluctuates at subsonic and supershear levels (Uenishi et al., 2025). Based on these observations, as a preliminary study, two-dimensional brittle birefringent polycarbonate rectangular specimens with one single (Fig. 1(a)) or double parallel (Fig. 1(b)) perforation line(s) consisting of straight small-scale cracks (length 4 mm each except for the notches (1 mm)) are prepared by using a digitally controlled laser cutter, and an “axe” (cutter) with a weight (total mass 485 g) is inserted at the top of each specimen. Then, each specimen with the “axe” is dropped from a height of 1,500 mm vertically along the groove of an aluminum frame onto a rigid plate. Dynamic fracture behavior and isochromatic fringe patterns (contours of the maximum in-plane shear stress) inside the specimens are visualized using the technique of dynamic photoelasticity in conjunction with a high-speed video camera (Photron FASTCAM Nova S) at a frame rate of 200,000 frames per second (fps). According to Uenishi et al. (2025), the mass density, shear modulus and Poisson’s ratio of the polycarbonate used here are 1,200 kg/m 3 , 820 MPa and 0.37, respectively, and with these material properties, the longitudinal (P) and shear (S) wave speeds are evaluated to be approximately 1,820 m/s and 830 m/s. Figure 2 shows the dynamic evolution of fracture or “unzipping” (splitting) along the single perforation line with straight small-scale cracks illustrated in Fig. 1(a). At 0 ms (millisecond), the bottom of the falling specimen reaches the rigid plate and it is subjected to upward vertical impact. By the time 0.160 ms, the impact-induced P wave has made approximately two round trips up and down inside the specimen, but there seems to be no influence of the impact-induced waves on fracture extension. Instead, dynamic stress concentration due to the advancement of the “axe”, indicated in the photograph at time 0.230 ms, induces extension of fracture or linking of the small-scale cracks along the perforation line, and the tip of the extending fracture can be clearly identified by the distinctive isochromatic fringe pattern or stress concentration, e.g. at time 0.410 ms. The fracture is then arrested and this tip does not change its position for a while. Finally around time 0.800 ms, stress concentration due to the “axe” advancement becomes sufficiently large, and fracture extension and fracture-induced waves are identifiable at 0.815