Evidence of Different Types ofFCenters from Thermoluminescence Studies in X-Irradiated Alkali Halides. I. Highly Pure KCl Crystals

Two thermoluminescence peaks are observed at 135 and 190°C in highly pure KCl crystals (divalent impurity concentration <1 ppm) x-irradiated at room temperature and heated at 40°C/min. Utmost care was taken to avoid exposure of the crystals to light and the concentration of $M$ and other aggregate centers remained negligible in these experiments. The growths of the areas ${\Delta }_i$ and ${\Delta }_g$ under the first and the second peaks, with the time of x irradiation, are found to obey the equations given by Mitchell, Wiegand, and Smoluchowski for the growth of the concentrations $f_i$ and $f_g$ of the first-stage and the second-stage $F$ centers, respectively. From this result and from the observation that the ratio $\frac{{\Delta }_i}{f_i}=\frac{{\Delta }_g}{f_g}$ remains constant in a variety of crystals (irradiated for different times or with different x-ray intensities, quenched, slowly cooled, or partially bleached), it is concluded that all the thermoluminescence is due to $F$ centers and that the 135 and 190°C peaks correspond separately to the bleaching of the first-stage and the second-stage $F$ centers, respectively. The values of the parameters in the equations given by Mitchell et al. are derived from the observed growths of ${\Delta }_i$ and ${\Delta }_g$ and are found to be in agreement with the values obtained by Mitchell et al. The reduction in the area under the thermoluminescence curve on partial optical or thermal bleaching is found to be proportional to the reduction in the $F$-center density. However, optical bleaching reduces both thermoluminescence peaks, whereas thermal bleaching up to about 100°C reduces mainly the first peak. It is found that complete thermal bleaching does not destroy all the vacancies generated during the irradiation but disperses some of them more uniformly. Both thermoluminescence peaks are found to correspond to a first-order process; therefore, it is concluded that the retrapping of the electrons released from the $F$ centers is negligible. The activation energies of the two peaks are 1.05±0.05 and 1.15±0.05 eV, respectively. This indicates that thermal bleaching of the second-stage $F$ centers at higher temperatures is not due to the retrapping of the electrons by anion vacancies, as concluded by some earlier workers, but to differences in the energies of thermal ionization of the two types of $F$ centers. Possible mechanisms which can give rise to differences in thermal ionization energies are tentatively suggested. These results are not inconsistent with the view that the first-stage $F$ centers are formed by the vacancies present in the crystal prior to irradiation and are distributed uniformly in the crystal, and that the second-stage $F$ centers are formed by the vacancies generated in the crystal during the irradiation process, perhaps near a defect, and are present in small regions of high local concentrations. Results of preliminary experiments indicate that the calcium impurity introduces a new thermoluminescence peak at 85°C, enhances the 135°C peak, and suppresses considerably the 190°C peak. Deformation introduces an additional peak at 270°C and reduces the rate of growth of the second peak at 190°C. The parameter $a$ definining the observed rate of generation of new vacancies varies approximately as the square of the intensity of x irradiation, in agreement with the observation of Mitchell et al. and the more recent observation of Abramson and Caspari.