COINCIDENCE SUMMING CORRECTIONS USING ALTERNATIVE DETECTOR CHARACTERIZATION DATA

Introduction

To reduce the count time to reach a desired precision for an observed activity, or to reach lower MDAs, it is common practice to place a sample as close to the detector as possible. However, a close proximity to the detector causes true coincidence, also called cascade summing, effects. The general quantitative procedure for treatment of true coincidence effect in the case of complex decay schemes has been considered by Andreev et.al ¹ as early as in 1972. Other authors have considered specific cases ², tabulated example correction factors ³, or experimentally checked some of the earlier theoretical work ⁴. There are many other references to various methods of correcting for cascade summing effects. It is not our goal to provide a complete literature survey as part of this paper.

Instead, we report results from a continuation of our previous work ^5,6 in developing a general methodology for correcting for cascade summing effects. Our method requires a single intrinsic peak-to-total efficiency curve and a so-called spatial efficiency characterization for each detector. While characterizing the detector for peak-to-total and spatial efficiency clearly works, it is somewhat tedious to use in practice. An experimental determination of the peak-to-total efficiency requires purchasing a set of single energy gamma sources and using them to measure a set of calibration spectra. Since the peak-to-total efficiency seems to be a characteristic of a given detector, it would seem desirable that the detector manufacturers provide a peak-to-total characterization with a detector when it is purchased. This approach has the drawback for the manufacturer that the single energy gamma sources are rather short lived. Rather than purchasing them over and over again, we have explored the possibility of using Monte-Carlo modeling instead.

The second characterization required for our methodology to work, the detector spatial efficiency calibration ⁷, requires experimental measurements as well as long computer simulations. This type of detector characterization process is quite time consuming, and requires detailed knowledge of the internal construction of the detector. It is particularly difficult to characterize detectors that are already in use. Therefore, it would be quite desirable to be able to use approximate, generic detector characterizations for all detectors with the same nominal efficiency. In this paper, we will show an evaluation of how well one can perform corrections for coincidence summing effects using such approximate characterization data.