Comment on the brittle-to-ductile transition: A cooperative dislocation generation instability; dislocation dynamics and the strain-rate dependence of the transition temperature

A new theory by Khantha, Pope and Vitek (KPV) and Khantha, which attributes sharp brittle-to-ductile transitions (BDT) of the type observed in silicon to a cooperative Kosterlitz-Thouless instability for dislocation generation, is examined critically. Results of simulations relevant to the KPV model show that contrary to the claim made by Khantha et al., the KPV theory does not predict a strain-rate dependent temperature T c for the sharp transition. Instead, it predicts a strain-rate independent sharp transition, or, in the quasi-brittle regime, a strain-rate dependent gradual transition. The new theory in its present form therefore does not explain the experimentally observed, strain-rate dependent, sharp transitions in silicon. Evidence from experiments and simulations is presented that this transition is essentially due to the non-homogeneous emission of dislocations from the crack tip. Emission starts at certain points along the crack tip, generating a strongly shielding plastic zone, which traverses the whole length of the crack tip at T c before the stress reaches that for brittle fracture. For a given strain-rate T c is therefore controlled by dislocation velocity and a length which depends on the original source distribution. This model, unlike KPV, predicts correctly the strain-rate dependence of the sharp transition, and explains the fact that it is structure sensitive.