Evaluation of a method for estimating low-dose-rate irradiation response of MOSFETs

Abstract

A simple method for estimating the threshold-voltage shift due to low-dose-rate ionizing irradiation was recently proposed for power MOSFETs. In this work, the physical considerations governing the applicability of the method are examined. In addition to the power MOSFETs discussed in the previous paper, the method is applied to integrated MOSFETs from two different technologies and critically evaluated. For this method to work, the oxide trapped charge due to low-dose-rate irradiation should be the same as that following irradiation at the dose rates specified in MIL-STD-883D Method 1019.4, and the interface-trap density following low-dose-rate irradiation should be the same as that following irradiation at 1019.4 rates and subsequent high-temperature annealing. Of the two integrated technologies evaluated, the method correctly predicts the low-dose-rate threshold-voltage shift for one, but not for the other. In the case where the method yields the correct result, the agreement appears to be coincidental. The results, coupled with the necessity for transistor-level test structures, suggests that the proposed method is applicable primarily to power MOSFETs that exhibit slow annealing of oxide-trapped charge and no rebound during low-dose-rate irradiation. >