Clarification of Local-Moment-Conduction-Electron Resonance

Abstract

Our data on spin resonances in $\mathrm{Ag}$-Er are inconsistent with the $J\stackrel{\ensuremath{\rightarrow}}{\mathrm{S}}\ifmmode\cdot\else\textperiodcentered\fi{}\stackrel{\ensuremath{\rightarrow}}{\mathrm{s}}$ model, and barely consistent with $J(\stackrel{^}{k}, {\stackrel{^}{k}}^{\ensuremath{'}})$. Alternatively, a suggestive temperature dependence of the $J\stackrel{\ensuremath{\rightarrow}}{\mathrm{S}}\ifmmode\cdot\else\textperiodcentered\fi{}\stackrel{\ensuremath{\rightarrow}}{\mathrm{s}}$ conduction-electron $g$ value is compatible with the data. We present general kinetic equations for a local-moment-conduction-electron system. We clarify questions of detailed balance, relaxation to exchange fields, positive energy absorption, and reduction of the kinetic equations for any possible model to either of two standard forms.