Analysis probes and statistical parameters affecting the OSL ages of mortar samples; a case study from Italy

Abstract

Mortars stand among the very few materials that theoretically cannot be reused due to irreversible hardening processes. Among the various dating techniques, radiocarbon and Optically Stimulated Luminescence (OSL) struggle for being selected as the most robust and reliable dating technique for the specific material. The principle of mortar OSL dating depends on the bleaching of the quartz grains of the sandy aggregate while mixing and laying the mortar to daylight. Thus, the OSL ages strongly depend on the effective bleaching of the quartz. Here, we deal with OSL ages from the old Roman Theatre of Padua. Preliminary OSL results based on Single Aliquot Regeneration (SAR) protocol using three regenerated doses to multigrain (MG) quartz aliquots, indicated distributions of ED (Equivalent Dose) leading to significant mismatch of these results compared to the expected ages. The present work attempts to discuss the influence of various data analysis approaches such as (a) exclusion criteria like recycling ratio and recuperation values, (b) selection of the suitable region of interest in the signal's integral along with the correct background (late or early) subtraction technique and (c) appropriate statistical analysis by selecting the suitable age model (descriptive statistics, Central Age Model, Minimum Age Model) on the final distribution as well as both value and precision of the ED, enabling thus to obtain ages that are not only more accurate, but also closer to what was expected. Alternative statistical indicators of either the symmetry of the ED distributions (like skewness, kurtosis, etc) or the OSL signal itself (residual distribution analysis, kinetic parameters and percentage contribution of each OSL component to the overall signal) will be assessed in an effort to investigate possible correlations.