Trap-level spectroscopy by thermally stimulated release of trapped carriers

Abstract

Investigations of thermoluminescence (TL), thermally stimulated conductivity (TSC), thermally stimulated capacitance (TSCAP), and edge region TSCAP have been carried out over the last thirty years to study thermal release of trapped carriers. The goal of this work was and remains the measurement of spectroscopic trap level data: energy levels, transition probabilities and trap concentrations. Little, if any, information on the physical nature of traps can be obtained by these methods. Discrimination between electron and hole traps is possible with TSCAP. The analysis of experimental thermal release curves or “glow curves” is traditionally based on phenomenological theories which take the form of a set of coupled nonlinear rate equations. The complete description of thermal emission of trapped carriers and their subsequent recombination or retrapping for an arbitrary distribution of trap levels and recombination centers of arbitrary concentration has proved intractable. In order to remove the mathematical complexity it became necessary to resort to various approximations. As a result, the model descriptions available until recently were only applicable to the simplest trapping and recombination conditions which can rarely be related to a real material. Exact solutions of the commonly employed so-called “single trap level model” have become available only in the last five years. A number of other conceivable models involving several different electron and hole traps pose no principle mathematical problems any longer. After examination of the vast variety of different thermal emission curves obtained from the physically meaningful range of trapping parameters, it became evident that it is extremely difficult to correlate theory and experiment with any degree of confidence. Even the simultaneous investigation of TL and TSC does not yield sufficient information to determine uniquely the kinetical mechanism of the thermally stimulated recombination process. It is at best possible to test the validity of a given model when most of the spectroscopic trap level parameters are known from independent measurement techniques. Several of those techniques have now been developed. The most important of them is deep-level transient spectroscopy (DLTS). By proper choice of the experimental conditions it is possible to measure thermal emission rates, activation energies, trapping rates, concentration profiles of traps in the depletion region of Schottky barriers or in p- n junctions as well as to discriminate between hole traps and electron traps. The analysis of the experimental data makes use of only the most general principles of thermal release of carriers from traps.