Total beta and gamma dose rates in trapped charge dating based on beta counting

The uncertainties involved in using total beta counting to derive the dose rate term in trapped charge dating are examined. A new sample preparation technique is described; this gives a stable counting geometry and retains radon. It is confirmed that the count rate to dose rate conversion factor is relatively independent of the relative contributions of the U and Th decay series and K-40, but that there is a significant dependence for the gamma dose rate. High resolution gamma spectrometry radionuclide analyses of > 3700 representative samples are presented; these allow discussion of the typical relative contributions of the three potential sources of dose rate, e.g. it is shown that, on average, the gamma dose rate makes up 33.15 +/- 0.08% of the total (beta and gamma) dose rates. By direct counting of 400 of these samples, it is then shown that beta count rates can be accurately predicted from these radionuclide analyses. However there is an overdispersion in the relative standard deviation in this relationship of similar to 10%. It is argued that correcting for stopping power differences from sample to sample does not reduce this overdispersion significantly. Finally, the total dose rate (beta, gamma and cosmic ray) is derived from the beta count rate in two ways: (i) using a non-linear parameterisation of the observed relationship between total beta count rate and beta and gamma dose rates, and (ii) by simply multiplying the calculated beta dose rate by a constant. It is shown that both approaches give almost equally good predictions of the total dose rates (compared to those derived from gamma spectrometry), with relative standard deviations of 5% and 6%, respectively. It is concluded that for the great majority of samples, the overall dose rate prediction obtained using beta counting alone can have an associated typical uncertainty of similar to 5%, to which must be added the uncertainty in beta counting. (c) 2007 Elsevier Ltd. All rights reserved.