Bakerian Lecture: Dislocations, plastic flow and creep

Abstract

An outline is given of the experimental evidence that dislocations normally exist in crystalline solids, that they are formed during crystal growth and that they can move when the crystal is stressed. It is shown also from observations of silver halides that networks of dislocations can be formed, and also from observations of crystal growth that piled-up groups of dislocations of high strength can be stable. A discussion is then given of the theory of Frank & Read of the origin of slip lines. It is suggested that the elements of the networks serve as sources of dislocations. A difficulty in the theory is to explain coarse slip; it is not understood why one source should produce a large number of dislocations before another comes into play. It is not known at present whether the interaction with sound waves is too great to allow moving dislocations to approach the speed of sound; the explanation of coarse slip may depend on whether this is so. The outlines are given of a theory of ductile fracture; this depends on the production of piled-up groups of several hundred dislocations from each of several active sources; in the neighbourhood of each of these there will be a large concentration of stress. Finally, the concept of 'climb’ of dislocations is introduced in connexion with recovery and creep, and a tentative explanation of Andrade’s law for transient creep is put forward.