A Sensitive TL-Detector For Radon Daughter Monitoring

[Introductory Remarks] Thermoluminescence (TL) detectors are widely used for the dosimetry of X- and[ y]-rays. Also filter devices for radon daughter monitoring have been developed in which the a activity on the filter was measured with TL-detectors (Breslin et al., 1977). For this purpose normal LiF-tablets with a thickness of 0.4 - 1 mm were used. However the lower detection limit of such TLmonitors for radon daughters is relative high, especially in areas with a high [y]-background, due to the low [a /-y]-sensitivity ratio of such thick LiF-detectors. This [a/y]-sensitivity ratio can be increased if TL-detectors are used whose thickness is comparable with the range of [a]-particles. On the basis of this consideration we have developed a simple light weighted and sensitive TL-monitor for the measurement of the cumulative radon daughter exposure in air. [Description of the "Monitor] The monitor consists of two parts: The sampling pump with the accumulator, and the sampling probe which contains the air filter and TL-detector system. Both parts are connected with a plastic tube. For operation in mines a light weight, portable air sampler with a built-in accumulator (DuPont model 125 and 200, 400 g) is used which enables an adjustable, constant flow rate in the range of 7.5 -12 1/h. These air samplers allow a sampling period of 10 hours before recharging is necessary. For radon daughter measurements in houses this pump is replaced by a larger, netoperated pump with a flow rate of 180 1/h. Figure 1 shows a schematic cross section through the sampling probe whose weight is 40 g. The radon daughters are collected on a hydrophob membrane filter (Sartorius) with an effective filter diameter of 11 mm and a pore diameter of 3 µm; its collection efficiency for radon daughters is [>] 95%. Above the filter the detector system is mounted. It consists of two thin Tm-activated CaSO4-TL films of 8 mm diameter on a Al-carrier foil manufactured by Matsushita Inc., Japan. The thickness of the TL-film is 6 mg/cm2, which is comparable with the range of a rays. The first detector is faced in 3.5 mm distance above the filter; in addition to the ambient y background this detector is exposed to the [a] and ß radiation from the radon daughters on the filter. Between the first and the second TL-detector is an Al-absorber of 0.5 mm thickness which shields this detector against the [a] radiation and partly also to the ß radiation from the filter. Both detectors have the same [y] shielding. After air sampling both TL-detectors are read-out with a commercial, hot-air reader (Matsushita Inc. / Japan, Model 505 A). The glow-curves are identical for [a and y] rays (Regulla et al., 1980). The integral TLsignal from both detectors is displayed in mR-units (or mR-equivalent). The difference between the response of the unshielded TL-detector above the filter and the shielded TL-detector is proportional to the time-integral of the filter activity, integrated over the sampling period. Calibration and Intercomparison Measurements] The linearity of the a response was checked with a 241Am-source for an [a] fluence on the detector surface of 5400 [a's/cm2•s] in the range of 102 -105 mR-equivalent. Within the experimental error of 15% the response function is linear. The TLD-monitor was calibrated for radon daughters in air by comparison with a calibrated integrating, monitor (WLM-meter) equipped with a Si-surface barrier detector and a direct electronic read-out; this instrument was developed by us for area monitoring some years ago (Haider et al., 1976). The results of simultaneous measurements with the TL-monitor and the electronic WLM-meter in mine areas and in indoor air of houses are shown in figure 2. On the abscissa the mean potential [a] energy concentration (in WL) during the sampling period is given which was derived from the measurements with the electronic WLM-meter. On the ordinate the corresponding response of the TLDmonitor to radon daughters (difference between both TL-detectors) is displayed, expressed in terms of mRequivalent per liter sampled air. It follows from this comparison a linear relationship over the whole range of 0.005 - 5 WL with a standard deviation of 10%. The resulting calibration factor for the TLD-monitor is 18± 2 mRequiv per WL • 1.