Thermoluminescence characteristics of the 375 degreesC electron trap in quartz.

Abstract

The thermoluminescence (TL) glow peak in quartz, which appears in the uv at a temperature of 375 \ifmmode^\circ\else\textdegree\fi{}C when a ramp heating rate of 20 \ifmmode^\circ\else\textdegree\fi{}C/s is used, was measured for emission in the uv as well as in the green part of the spectrum. Two glow peaks appear separated by about 20 \ifmmode^\circ\else\textdegree\fi{}C. Isothermal decay rates for both spectral components were measured at temperatures near the glow peaks. Heating ramp rates were varied to obtain data for Hoogenstraaten analyses. Additive dose curves up to saturation were obtained. Solar-simulator bleaching studies were performed. Overall, these data appear to suggest a kinetic order between first and second order. It was found possible by computer simulation to reproduce faithfully the observed data by use of a simple interactive model that possesses two separate recombination centers, both simultaneously fed by electrons from a single electron trap and transported through the conduction band. It was found necessary to attribute a distribution of activation energies to the electron trap, to allow for a significant retrapping probability, and to also invoke the presence of a thermally disconnected electron trap. The analysis comparing the theoretical predictions to the data used a single set of program parameters in obtaining computer-generated solutions to 100 coupled differential-integral equations.These parameters successfully generated the observed initial rise values ranging from 0.9 to 1.1 eV (depending on the method employed) and an observed Hoogenstraaten-analysis energy of 1.49 eV both with a distribution central activation energy of 1.45 eV. A glow curve asymmetry shape factor \ensuremath{\mu}=0.46 given by the theory matched the data well, suggesting an apparent intermediate kinetic order. The theory also simultaneously generated the required uv-TL and green-TL glow peak separation in temperature. A model for generating the necessary electron and hole densities during laboratory additive dosing to reproduce the observed dose growth data was developed. This model is a consistent extension of the prior theoretical analyses. The electron activation energy distribution also provides a ready explanation for the midterm fading phenomenon present when radiation doses acquired during long burial times are compared to laboratory administered doses required to produce the same TL signal.