A Mechanism for Dislocation Generation in Shock-wave Deformation

Numerous articles describing the changes in structure and mechanical response of metals and alloys induced by shock waves have been published in the past twenty years (I-3). Increases in the concentration of point (4) and line defects (5), formation of twins (6), martensite (7) and precipitates (8) have been observed in the residual structures. Smith (9) proposed the f i rs t model accounting for dislocation generation in shock loading. According to this model, a two-dimensional array of dislocations separated the high-density shockedmaterial from the virgin material; the translation of this two-dimensional interface coincided with the shock-front propagation. Aware of the fact that the passage of such an interface would result in undeformed material, Smith (9) hypothesized the presence of sources and sinks of dislocations, moving with the velocity of the shock front. Hornbogen (lO), based on the fact that shock-loaded iron (between 7 and II GPa) presents a substructure characterized by straight screw dislocations, proposed a modification to Smith's (9) model. Shock loading would induce the formation of loops; the edge components would propagate with the front, basically forming a Smith interface. However, the screw components would stay behind and would therefore be the salient residual feature of the substructure. Cowan ( l l ) suggested that at shock pressures generating shear stresses above the theoretical shear stress of the material a difference in dislocation generation mechanism might be expected. Above this stress, termed by him "supercritical stress," the limitations imposed on dislocation velocity would no longer be valid, and a Smith (9) interface might become possible. No specific mechanism is however proposed by him.