AN OPTIMIZED COMPUTATIONAL METHOD FOR DETERMINING BETA SKIN DOSE DISTRIBUTION USING A MULTIPLE-TLD DOSIMETER/SPECTROMETER (DOSIMETRY, TLD, OPTIMIZATION TECHNIQUE)

This thesis describes a new method to determine the beta skin-dose and the beta depth-dose distribution from the responses of a multiple-chip thermoluminescent dosimeter (TLD) system. The system is designed for and has been tested in beta radiation fields. The design can be extended to handle mixed beta/gamma fields. The method combines analytical calculations with experimental measurements to reconstruct the incident beta-particle energy distribution. The analysis is performed with an experimentally verified Monte Carlo computer program based on electron-transport theory. The experimental measurements are made with a packet of three LiF TLD chips with thin shields which are exposed to the beta radiation field. The beta doses measured in the TLDs are used to reconstruct the incident beta-particle energy distribution from data tabulated by the Monte Carlo computer program.;This method is compared with Gesell's algorithm, which is based on purely empirical information, and it shows better agreement with the measured beta skin doses than the results derived from Gesell's algorithm.;The system used to measure beta-particle energy distributions is based on a plastic scintillation detector, which produces acceptable measured results for electron energies above 90 keV. Methods were developed to correct the spectrometer data for energy dispersion and for the backscattering of incident electrons.;The analytical method assumes that the incident beta-particle energy distribution can be constructed using two or more component functions. The calculations produce an extensive table of doses for the 3-chip/absorber TLD system which are derived from these component functions; one function is a linear function, and the other, a sine function. The reconstruction is performed using a four-parameter-based optimization technique; the four parameters are the energy endpoints and the amplitudes of the component functions. The objective functional of the optimization program is the sum of the squared differences between the measured TLD doses and the corresponding calculated doses.