PSI - Issue 37

Koji Uenishi et al. / Procedia Structural Integrity 37 (2022) 404–409 Uenishi and Xi / Structural Integrity Procedia 00 (2022) 000 – 000

406

3

Sakurai, 2015). However, these theoretical speculations have not been experimentally confirmed yet, and hence, transfer of stresses and development of waves and fractures in a dry granular slope that is subjected to dynamic impact has been investigated in Uenishi and Goji (2018) by using the experimental technique of dynamic photoelasticity in conjunction with high-speed cinematography. 2. Dynamic fracture of granular slopes In Uenishi and Goji (2018), penny-shaped birefringent particles (diameter 8 mm, thickness 3 mm) have been cut out from polycarbonate plates by a digitally controlled laser cutter and piled up on a rigid horizontal plane to form a two-dimensional granular slope that has some inclination angle (in Fig. 2, 60 degrees). Dynamic impact has been given to the top free surface of the slope by free-falling button-shaped aluminum (top row of Fig. 2) or by an airsoft gun-launched projectile (bottom row of Fig. 2). The time-dependent stress variations and fracture evolution have been recorded by a high-speed video camera at a frame rate of 50,000 frames per second. According to the experimental observations, there exist two distinct modes of stress transfer and ensuing fracture development: (1) unidirectional, force chain-like stress transfer, which may result in opening below the position of impact and granular mass flow or total collapse of the whole slope (top row of Fig. 2); and (2) broadly radiated two-dimensional waves that may lead to opening at a position away from the impact point and then toppling failure-like separation of the slope face only (bottom row of Fig. 2). The emergence of each mode seems to depend on the temporal energy profile associated with the impact. However, these experiments have been pe rformed in the “monolithic” framework and no inhomogeneity in and around slopes has been taken into account. Therefore, here, possible influence of material inhomogeneities on the dynamic stability of a granular slope is examined. Firstly, experimentally, solid plates modeling retaining walls are placed over certain boundaries of the granular slopes, and the mechanical roles of confinement in the generation of dynamic particle motion, stress transfer and structural collapse are qualitatively assessed. In this new series of laboratory experiments, impact load is given to the tope free surface of granular slope by a steel ball (sphere of diameter 15 mm, mass 13.8 grams) free-falling

a

b

10 mm

Aluminum

Aluminum

10 mm

Photoelastic particle

Stress transfer

Stress transfer

Opening beneath the position of impact, leading to mass flow Projectile

Projectile

Slope face separation, similar to toppling failure

Opening away from the position of impact

Wave radiation

Fig. 2. (a) Photographs taken by the photoelastic experiments clearly show the particle motion and isochromatic fringe patterns inside the particle in two-dimensional model slopes. The fringe order is proportional to the maximum in-plane shear stress. (b) Difference of colors between the photographs before and after the dynamic impact illustrates unidirectional stress transfer (top) and two-dimensionally radiating waves (bottom) in the granular slope. In both (a) and (b), top snapshots are related to the dynamic impact by free-falling button-shaped aluminum (diameter 20 mm, thickness 10 mm, mass 8.4 grams, impact velocity 3.5 m/s) while the bottom ones are associated with the impact loading due to a projectile launched by an airsoft gun (sphere of diameter 6 mm, mass 0.2 grams, impact velocity 76 m/s) (modified after Uenishi and Goji (2018)).