Abstract
For pt.I see ibid., vol.9, no.10, p.1845 (1976). Models for interstitial aggregation in doped fluorides are discussed. Two problems are considered: first the nature of the interstitial-interstitial interaction, where it is shown that earlier treatments based solely on the ionic model must be modified to allow for a large covalent term in the interaction between interstitials. The proposals are supported by a molecular orbital study of F--F- interactions. Problems in the interpretation of diffraction data on heavily doped crystals are also examined. The author is able to rationalize the experimental observations by extensions of earlier cluster models and by further models based on cluster interactions. Lattice energy calculations confirm the proposals.