The potential energy surfaces of atomic O (3P, 1D, 1S), which was trapped in the crystalline specimens, were developed on the basis of known angularly anisotropic pair interactions. The electrostatic limit, with neglect of exchange and spin-orbit interactions, was assumed. By using a classical statistical treatment for the simulation of spectra, the surfaces were shown to reproduce experimental O(1S  1D) emissions from the substitutional and interstitial sites of crystalline Kr. The surfaces were also in accord with the charge transfer emission spectra of O/Xe solids. In the presence of lattice relaxation, the Xe-O(1D)-Xe insertion site became the global minimum, and could therefore act as a stable trap site. This was in accord with experimental observations of a third trapping site in Xe. In order to rationalize a recently reported long-range mobility of O atoms in these solids, the topology of various electronic surfaces was presented. It was shown that the minimum energy paths which connected interstices on the triplet and singlet surfaces were quite different. The triplet path was strongly modulated, and proceeded along the body diagonals of the unit cell. The singlet path was more gently modulated, and proceeded along the face diagonals. These features were consistent with a thermal mobility that was postulated to proceed via triplet-singlet conversion. However, the electrostatic surfaces failed to support the model quantitatively. The site-specific crossing energies, including lattice relaxation, were calculated to range from 1.2 to 1.7eV. These were an order of magnitude larger than the observed experimental activation energies for migration. The inclusion of spin-orbit and charge transfer mixing on these surfaces, which were absent from the present treatment, was expected to reduce the discrepancy.

A.V.Danilychev, V.A.Apkarian: Journal of Chemical Physics, 1994, 100[8], 5556-66