Abstract
The electron microscope study of high temperature irradiation damage in single crystal and pyrolytic graphites has revealed differences in the defect nucleation behaviour which must be attributed to the differing crystallite perfection (Thrower 1967 a). A study of the formation and annealing of these defects, both vacancy and interstitial loops, has confirmed the earlier suggestion of Turnbull and Stagg (1966) that self-diffusion in graphite is three-dimensional. Furthermore, there are several pointers to this diffusion being by an interstitial mechanism. During both nucleation and annealing there is evidence of interstitial migration perpendicular to the basal planes, Analysis of these results gives a value of 6·55 ± 0·4 ev for the formation energy of an interstitial, and the implications of this value are discussed, the most important being that there must be a significant amount of bonding between the interstitial atom and the lattice. During annealing, loops move parallel to the basal plane by a pipe diffusion mechanism, the activation energy for which is slightly lower in a twist boundary than in the lattice due to the lower strains around the defect. There is also some evidence of loops amalgamating which appear to be non-coplanar, and there is really no satisfactory explanation for this.

This publication has 22 references indexed in Scilit: