Department of Physics, Loyola University of Chicago, Chicago, Illinois 60626, USA
Abstract:
The g = 4.3 ESR signal of Fe3+ in glass is usually attributed to a “fully rhombic” distortion characterized by E = D/3, where E and D are the splitting constants belonging to the second-order rhombic and axial terms, respectively, of the crystal-field interaction. An alternate statistical approach to the origin of this signal has been given by Peterson et al. on the basis of axial symmetry and broad overlapping distribution of g| and g that are correlated in opposition (negatively) resulting in spiking at g 4.3. The present paper reports detailed computer studies of the distribution of effective g value produced by large variations in both D and E to test whether negative g-value correlations leading to spiking at g 4.3 are generable from site dependent fluctuations in the second-order crystal-field terms. It is found that the ESR line arising from the middle Kramer's doublet in the fully rhombic limit (E → D/3, |D| > hν) uniquely generates the negative g-value correlation required for spiking at g 4.3.