Answer to Question #11077 Submitted to "Ask the Experts"
The following question was answered by an expert in the appropriate field:
When I count a sample (liquid dried on aluminum planchette) for gross beta activity (using an end-window GM detector system) and gamma activity (found to contain only 137Cs, using an NaI system), I find that the gamma activity is always more than beta activity. Considering that my sample predominantly contains 137Cs and to an extent, 90Sr-90Y and 99Tc, how is it possible that beta activity is less than the gamma activity? Am I underestimating the beta activity or overestimating gamma? I use 90Sr source for efficiency determination of beta counting system and 137Cs for the NaI gamma spectroscopy system.
Based on what you have said, my first inclination is to judge that you are likely underestimating the beta activity. If the sample contains essentially only 137Cs the beta emissions are dominated by the 511 keV maximum beta energy (156.8 keV average energy and yield of 94.6 percent) with a lesser component with a maximum beta energy of 1173 keV (415.2 keV average energy and yield of 5.4 percent). The average beta energy, considering both transitions, is about 171 keV.
If you used 90Sr/90Y (the 90Y will be in equilibrium with the 90Sr) for determining the beta counting efficiency, the efficiency value would be heavily affected by the high energy 90Y beta particles. Depending on how the 90Sr source was prepared and what type of covering, if any, was over the source, many of the 90Sr beta particles (maximum energy 546 keV, average energy 195.8 keV) might have been attenuated in the source, and the dominance of the high energy 90Y beta particles (maximum energy 2284 keV, average energy 935 keV) may have been enhanced. Even if none of the beta radiation emitted from the 90Sr/90Y in the direction of the detector was attenuated in the source, the average beta energy would have been about 565 keV, markedly higher than that from the 137Cs. The end result is that the determined detection efficiency value would have been too high for the 137Cs beta radiation, thus leading to underestimation of the sample activity.
Naturally, there may also be other influencing factors to consider. For example, the geometries and degree of self attenuation of the sample and that of the gamma standard used should be essentially the same. You should attempt to get a 137Cs standard prepared in a similar fashion to the samples of interest (same materials, same geometry and about the same self absorption) in order to assess the beta counting efficiency (the gamma component will have a very small impact on the beta detection efficiency).
I wish you well in your continuing measurements.
George Chabot, PhD