Electronicg-Factor and the Structure ofF-Centers

Abstract

A study is made of the small electronic $g$-factor shifts observed in complex paramagnetic substances. In most cases there is a simple connection between the contribution of a given ion or atom to $\ensuremath{\Delta}g$ and the anisotropic part of the magnetic hyperfine interaction of its nucleus with the electron. This connection is useful now that a double-spin-resonance technique developed recently by Feher can give experimental values for such hyperfine interaction constants.The theory is applied to the case of the $F$-center. It is shown that a central $F$-center wave function need not have a large $g$-state component to account for the order of magnitude of the observed $\ensuremath{\Delta}g$, provided that the function used is orthogonal to the ion-core orbitals. Both the positive and negative ions contribute to $\ensuremath{\Delta}g$. With the aid of Feher's hyperfine-structure measurements in KCl, one gets: $\ensuremath{\Delta}g (\mathrm{theory})=\ensuremath{-}0.0053$ as compared with $\ensuremath{\Delta}g (\mathrm{exp})=\ensuremath{-}0.0070$.Exchange interactions between the $F$-electron and the ion-core electrons are offered as a possible explanation of the positive $\ensuremath{\Delta}g$ found for LiF. Such interactions could polarize the nearest-neighbor halide ions, effectively creating a hole. It is estimated that a 0.5-1% polarization of the ${\mathrm{F}}^{\ensuremath{-}}$ $2p$ orbitals would be required.