Effect of thermal and dosimetric history on the accuracy of LiF:Mg,Ti and CaSO4:Dy

LiF:Mg,Ti and CaSO4:Dy detectors (50 per group) are re-used about 90 times, including regular control of the individual calibration factors. Nominal dose was 1.5-2 Gy in water either using 60Co gamma or conventional X rays. Three different readers (Harshaw 2000 and 4000 and TLD-04B) were used. High temperature annealing was 400oC for 1 h (LiF:Mg,Ti) and 600oC for 1 h (CaSO4:Dy). The matrix of the calibrating irradiations is analysed by using two-way variance analysis. The reproducibility of clinical measurements is less by a factor of about 2.5 than that observed in well controlled constant 1-2 Gy 60Co irradiations. A standard annealing process may fail to restore the original sensitivity of the dosemeters resulting in short-term fluctuations of the individual calibration factors. For high precision clinical dosimetry the best way is to use a 60Co calibration before each re-use of the group. Control of the TL reader on a regular basis, especially after any service, is mandatory. (author)