Experimental determinations of beta attenuation in planar dose geometry and application to ESR dating of tooth enamel

Two experiments were performed to determine the distribution of beta dose in a target medium as a function of distance from a planar source (2π geometry). In Experiment 1 planar absorbers were used to simulate depth in the target medium, and the attenuated dose that penetrated the intervening absorbers was detected using thermoluminescence from CaSO₄:Tm. In experiment 2 pellets of powdered tooth enamel were situated in a cavity in solid tooth enamel. Dose response of the electron spin resonance (ESR) signals in the pellets allowed direct detection of dose distribution within the target medium. The dose distributions were found to agree more closely with Monte Carlo calculations and with one-group transport theory (Prestwich et al., 1997 and Brennan et al., 1997) than with approximations used widely in ESR dating applications (Grün, 1986). Although the latter were supported by experiments by Aitken et al. (1985), the results of experiment 1 provided data to suggest that departure from 2π geometry was unacceptably large in the Aitken et al. (1985) configuration. The beta doses obtained by one-group theory in typical target volumes of enamel were 17 to 40% lower than those obtained using the Grün approximation. It was concluded that one-group transport theory provides the best available routine tool for calculations of beta dose attenuation in near-planar target materials (e.g. tooth enamel and shell) irradiated by natural sources, although Monte Carlo methods are better and could eventually be introduced into the ROSY ESR dating program (Brennan et al., 1997). It was also concluded that a revision of previously published ESR ages using one-group transport for beta dosimetry will be needed. For cases where the U concentrations in enamel are generally low (less than a few ppm), previously published ESR ages on tooth enamel are systematically too young by 5–30%.