THE EXPLOSIVE DISINTEGRATION OF PRINCE-RUPERT-DROPS

A high-speed photographic study has been made of the explosive disintegration of Prince Rupert's drops. The drops were prepared by quenching molten soda-lime glass in water. The disintegration of a drop was initiated by exploding a small (26 mg) lead azide charge or by impacting with a hardened steel chisel on to the tail of the drop. The entire fragmentation process was recorded at framing rates ranging from 6500 frames s-1 to 0.5 x 10(6) s-1. The high-speed photographic sequences revealed that in a disintegrating drop the crack front, having been initiated in the tail, propagated at a high velocity (almost-equal-to 1450-1900 ms-1) within the tensile zone, towards the drop's head. Finger-type bifurcating cracks at the crack front were observed. High-speed photographic observations, combined with an analysis of the fragment sizes, indicated that the fast-moving cracks slowed down dramatically on entering the surface compression zone. Sequences of high-speed shadowgraphs also revealed that the rapidly moving crack front did not produce any strong stress waves in the drops which could have contributed to their explosive disintegration. The main features of crack propagation in a drop were found to be similar to those of the self-propagating cracks in thermally tempered soda-lime glass blocks. Measurements of residual stresses in the drops using an indentation technique showed that the surface compressive stress was almost-equal-to 90-170 MPa and the tension zone extended across almost-equal-to 70% of the drop's diameter at the head. Both of these values are similar to those found for thermally tempered soda-lime glass blocks. A model based on the repeated bifurcation of fast-moving cracks within the tensile zone in the drop has been proposed to explain its explosive disintegration.