Reducing Errors in Dosimetry Caused by Low Energy Components of Co-60 and Flash X-Ray Sources

Abstract

Radiation effects testing for electronic devices is commonly performed using Co-60 sources and flash x-ray (FXR) sources. Dosimetry for such irradiation is usually done by measuring the dose in TLD's and then converting to dose in SiO/sub 2/ using energy deposition coefficients. This dosimetry method is useful because it is simple, it requires little knowledge of the source spectrum, and it requires little knowledge of the details of the source spectrum, and it requires little knowledge of the details of the device geometry. However, this method can lead to large errors for some Co-60 sources and most FXR sources because they have significant low energy components. The goal of this paper is to determine whether dosimetry can be performed using the usual method with an acceptable level of error. The authors show to what extent errors can be reduced using a filter to remove low energy components from the incident spectra. Finally, the implications of these results for dosimetry methods and standards. The approach used in this paper is to perform a series of calculations which lead to an estimate of the maximum error in the dosimetry. Further, it has been assumed that errors of up to about 30-40% aremore » acceptable for an important class of engineering problems where the attainment of better accuracy is difficult or expensive.« less