Analysis of ENDOR Spectra

Abstract

The analysis of ENDOR spectra proceeds in the same way no matter whether the spectra were obtained by conventional stationary ENDOR, by optical detection using the MCDA technique or using luminescence or by electrical detection using the NMR and EPR-induced changes of the electrical conductivity. The ENDOR spectra often contain numerous lines. Their analysis is often not very obvious. Unfortunately, there is no direct way from the spectra to a defect model. In principle, one has to “guess” the defect model and then calculate the theoretical ENDOR spectra assuming the correct interaction tensors and the correct symmetry of the defect. This procedure is often called “fitting” of the spectra. This fit will only work if the model and all parameters are correct. Therefore, even if one would be content with interactive parameters less precise than in principle available from the spectra, it can be dangerous not to calculate the spectra with the precision determined by the experimental line resolution, since one may end up with wrong conclusions. For example, a small line splitting may be caused by a tilt of the crystal orientation during the measurement or by the so-called pseudo-dipolar coupling or by a defect symmetry lower than expected from the analysis of the EPR spectra. It will, however, not always be necessary to analyze all the lines of many shells of those distant lattice nuclei if the microscopic model can be determined from those of the nearer neighbors and that is all what was wanted.