It is shown by theoretical argument and by experiment that only in the case of a first order kinetics polarization do the TSD peaks occur invariably at a fixed temperature. Otherwise their position is shifting in a characteristic way with changing initial polarization. In the case of a space charge release, for example, the maximum temperature is increasing with polarization temperature and polarization time. Thus, l~eak position data for varying polarization conditions allow one to decide in particular whether a peak is due to a first order depolarization process, e,g., complex rearientation, or to the release of a space charge 9The thermalIy stimulated depolarization (TSD) of polar crystals by an increasing temperature usually shows several bands or peaks. This indicates that the depolarization is realized by several different processes. Two such processes are well known: the relaxation of aligned impurity-point defect complexes (1) and the relaxation of a space charge caused by mobile point defects accumulated at the electrodes (2. 3). But there are stiI1 other processes which cause TSD peaks which have not yet been identified (1, 4).It is one of the fundamental problems of any investigation of TSD to relate the observed peaks to specific depolarization processes. From the experimental point of view, there are three quantities which characterize a given TSD peak, on which an analysis may be based: the maximum position, the magnitude of the peak, and the slope of the initial rise of the peak. The magnitude of the peak is eventually (1) a measure of the number of defects causing the polarization. The determination of the slope of the initial rise, which should yield the activation energy af the depolarization process, is in general a delicate task and if the peaks overlap too much, possibly no meaningful value can be obtained at all. However, it is comparatively easy to determine the approximate peak positions. We shall show how the peak position depends on the initial polarization of a sample, and that this can provide some information on the depolarization processes causing the peaks in question.
Some General Considerations Concerning TSDThe TSD is usually investigated by measuring the "depolarization current" Im(t,T) ofa polarized sample of cross section A and thickness L which is connected to a gaIvanometer. The measured current density im(t,T) = Im(t,T)/A is equal to the sum of the charge transport rate density e. j(x,t) and the displacement current density OD (x,t)/Ot at any arbitrary cross section of the investigated sample. Under ideal conditions, the measurement is carried out without any potential drop between the electrodes, AV = E(x,t) 9 dx = 0, and without any charge transfer to the electrodes, i.e., j(o,t) = j(L,t) = 0. Under these conditions solely the displacement current densitydue to the redistribution of charges inside the sample is measured. In general, there exist several different causes of polarization. However, we shall restrict ourselves here Key words: thermally stimulated depolarizatio...