It was recalled that, according to the dislocation model for void ordering and swelling saturation, the latter phenomena occurred due to the absorption by voids of perfect glissile dislocation loops that were produced by irradiation. The formation and glide of small interstitial loops was also confirmed by recent molecular dynamics studies of displacement cascades. The cascade mechanism for loop production was shown to explain the absence of visible dislocation loops in some experiments on void lattices. It was concluded that this was an important argument in favor of the present theory. However, according to the molecular dynamics simulations, it seemed that the glide of such loops did not depend upon the stacking fault energy of the host lattice. This was contrary to the predictions of the elastic continuum theory. The latter showed that a high stacking-fault energy, as in most body-centered cubic metals and in face-centered cubic Ni and Al, favored the unfaulting of small loops. This seemed to be in agreement with experimentally observed void lattice formation in these metals, as compared with the resistance of low stacking-fault energy metals (such as Cu, Ag, Au, and most steels) to void-lattice formation. It was suggested that this discrepancy between the continuum theory and the molecular dynamics simulations demonstrated the need for further studies of displacement cascades; especially the modelling of the effects of impurities upon the nature of interstitial clusters in more complex systems. One problem was to find impurities that could facilitate the unfaulting process and, therefore, void ordering and swelling saturation in face-centered cubic metals which were currently supposed to be resistant to the formation of a void lattice.

V.I.Dubinko: Nuclear Instruments and Methods in Physics Research B, 1995, 102[1-4], 125-31