Search for a command to run...
The sensitivity S of Si diodes generally increases with an increase in the instantaneous dose rate r of the radiation beam from clinical linear accelerators. A theoretical model is established to understand the physical origin of this dependence. During a radiation exposure, a portion of the excess minority carriers (electrons or holes) generated in the diode is captured by the R-G (recombination-generation) centers and is recombined with the majority carriers. The captured portion depends on the excess minority-carrier concentration delta p (proportional to r), the R-G center concentration N(t), and the minority-carrier capture cross-sections (sigma(n) for electrons and sigma(p) for holes) by the R-G center. When r increases, the R-G center concentration may not be sufficient to keep the recombination portion constant, which leads to an increase in diode sensitivity because a larger fraction of the charge will be collected. Larger majority-carrier concentration increases the recombination probability of the excess minority carriers and thus decreases the r dependence. The ratio of minority-carrier capture cross-sections, sigma(p)/sigma(n), influences the magnitude of the r dependence and also differentiates the r dependence between n-type and p-type diodes. A number of different circumstances can occur in diodes. When sigma(p) > sigma(n), such as for the dominant R-G center generated by electron radiation, the sensitivity is more dependent on r in an n-type diode than in a p-type diode if all the other device parameters are the same. When sigma(p) < sigma(n), the sensitivity is then more dependent on r in a p-type diode than in an n-type diode. The condition of sigma(p) < sigma(n) can occur when R-G centers with this property are generated by the foundry die process. A diode could have very small r dependence due to large R-G center concentration, generated by heavy platinum doping or radiation accumulated dose. Experimental data are compared with theory.