Positron lifetime spectroscopy applied to the study of defects in metals and semiconductors

Abstract

Studies have been made of fast timing techniques in various types of positron lifetime spectrometer and of the analysis procedures used for extracting positron lifetime and intensity values. Evaluation is made in terms of their usefulness and reproducibility in high resolution studies of lattice defects and their formation. The optimisation of a Fast-Slow 4-Way Routing Positron Lifetime Spectrometer has been attempted and the results are discussed in terms of changes in important spectrum parameters which have possibly been overlooked on comparative systems. Its viability as a non-destructive testing technique is considered. A Fast-Slow and a Fast-Fast Positron Lifetime Spectrometer have been constructed and studied. The Fast-Fast system has been developed such that it utilises the detector dynode pulses for timing, rather than the conventional anode pulse. By further optimisation of the time-pickoff method, a substantial improvement in the resolution has resulted. The system has been applied to studies of materials in which the defect types, their concentration and their specific trapping rates for positrons, are in question. A study has been made of the uncertainties arising from different approaches to the computer-aided analysis of multi-component decay spectra. Three analysis programs, all in common use, have been investigated and compared. Results are presented that clearly indicate the underlying reason for discrepancies in published positron lifetime data. A temperature study has been made of thermal vacancy creation in polycrystalline indium from 290 to 425 K by observing the variation in the positron trapping rate. Results are discussed in terms of a simple 2-state trapping model although attention has been paid to the question of pre-thermalisation trapping of positrons as the melting point is approached. Values of the monovacancy formation energy, E, are estimated using several types of analysis method for spectra from three sets of similar temperature measurements using the same sample, each made at differing stages of anneal. Differences in trapping rate variation with temperature (and thus in E) is interpreted as a sign of inadequate annealing and this is used to explain the discrepancies between previously published results. The nature of the defects in various types of gallium arsenide single crystal have been investigated by studying the effect of different dopants and their concentrations upon the rate and intensity of positron trapping at room temperature. Positron lifetimes and intensities in e and n°-irradiated samples have been similarly measured. To determine further the characteristics of the defects, an isochronal anneal study was performed using n°-irradiated gallium arsenide over the temperature range 290 to 725 K. This study was combined with infra-red spectroscopy measurements at each anneal stage in an attempt to correlate the variation of specific lifetime component intensities with those of a selection of infra-red spectral peaks whose origin is in question. The high defect concentrations created by the irradiation are shown to give rise to pre-thermalisation trapping of positrons.