Dislocation Sources in Crystals

Lithium fluoride crystals were subjected to short stress pulses (1–10 μsec in length) in order to study dislocation nucleation in them. It was found that several heterogeneities can lead to dislocation nucleation: (a) cleavage steps, (b) dislocation loops, (c) glide bands, (d) inclusions or ``dirt'' particles, (e) precipitates (as grown), and (f) radiation-induced precipitates. Also, it was found that annealing treatments at moderate temperatures (200–600°C) followed by ``rapid'' cooling (>5°C/hr) made segments of previously pinned dislocations become mobile. These mobile segments multiplied as they moved through the crystals. Several treatments were tried that were ineffective in producing dislocation nucleation. These included: temperature changes, surface wetting, simultaneous irradiation and stressing, reflections from free surfaces, and diagonal bending. Carefully cleaned glass spheres were pressed into contact with dislocation-free regions of a LiF crystal. In this way, shear stresses that were 100 times greater than the normal yield stress of the crystal (∼G/85) were reached without causing homogeneous dislocation nucleation. It is concluded that small foreign heterogeneities cause most of the dislocation nucleation in real crystals. Homogeneous nucleation occurs only at high stresses (∼G/30). Relatively little dislocation multiplication results from classical Frank-Read sources (almost none in LiF). Most dislocation multiplication occurs as a prior dislocation moves through a crystal.