Dependence of diode sensitivity on the pulse rate of delivered radiation

Purpose: It has been reported that diode sensitivity decreases by as much as 2% when the average dose rate set at the accelerator console was decreased from 600 to 40 MU/min. No explanation was given for this effect in earlier publications. This work is a detailed investigation of this phenomenon: the change of diode sensitivity versus the rate of delivery of dose pulses in the milliseconds and seconds range. Methods: X-ray beams used in this work had nominal energies of 6 and 15 MV and were generated by linear accelerators. The average dose rate was varied from 25 to 600 MU/min, which corresponded to time between microsecond-long dose pulses of 60-2.7 ms, respectively. The dose-per-pulse, dpp, was changed by positioning the detector at different source-to-detector distance. A variety of diodes fabricated by a number of manufacturers were tested in this work. Also, diodes in three different MapCHECKs (Sun Nuclear, Melbourne, FL) were tested. Results: For all diodes tested, the diode sensitivity decreases as the average dose rate is decreased, which corresponds to an increase in the pulse period, the time between radiation pulses. A sensitivity decrease as large as 5% is observed for a 60-ms pulse period. The diode sensitivity versus the pulse period is modeled by an empirical exponential function. This function has a fitting parameter, t(eff), defined as the effective lifetime. The values of t(eff) were found to be 1.0-14 s, among the various diodes. For all diodes tested, t(eff) decreases as the dpp decreases and is greater for 15 MV than for 6 MV x rays. The decrease in diode sensitivity after 20 s without radiation can be reversed by as few as 60 radiation pulses. Conclusions: A decrease in diode sensitivity occurs with a decrease in the average dose rate, which corresponds to an increase in the pulse period of radiation. The sensitivity decrease is modeled by an empirical exponential function that decreases with an effective lifetime, t(eff), of 1.0-14 s. t(eff) varies widely for different diodes, dpp, and x-ray energy. It is hypothesized that the capture of excess minority carriers by charge traps, cause the observed decrease in diode sensitivity. Also, it is hypothesized that the slow reopening of these traps occurs in the hundreds of milliseconds to seconds range at ambient temperature and this underlies the slow decrease in the diode sensitivity. Calibration of a diode is best done at the average dose rate with which it will be used. This is easily accomplished for radiation deliveries in which the average dose rate is a constant. However, for a VMAT delivery the average dose rate is a variable. For measurements made under these conditions diodes can be calibrated with a median or average dose rate which splits the difference in diode sensitivity that is known to occur with changes in average dose rate. (c) 2013 American Association of Physicists in Medicine. [http://dx.doi.org/10.1118/1.4788763]