MAXIMUM-LIKELIHOOD COMMON-FACTOR ANALYSIS AS A POWERFUL TOOL IN DECOMPOSING MULTICOMPONENT EPR POWDER SPECTRA

An advanced multivariate statistical technique, maximum-likelihood common-factor analysis (MLCFA), is used to decompose a set of multicomponent EPR powder spectra. Once the correct number of constituents is detected by MLCFA, the resulting abstract-factor spectra are transformed into real EPR powder spectra by means of a minimization procedure or a target transformation. The technique is applied to both simulated and experimental EPR powder spectra. The results indicate that the technique MLCFA is potentially of great value in the field of EPR of polycrystalline materials. © 1993 by Academic Press, Inc.