Evaluating analytical methods for ancient glass: Validation of micro-XRF quantitative data through comparison with LA-ICP-MS

This study aims to assess the reliability of micro-X-ray fluorescence spectroscopy (μ-XRF) as a quantitative, noninvasive tool for the in-situ compositional analysis of ancient glass. μ-XRF is widely employed in archaeological and materials science contexts due to its ability to rapidly detect major, minor, and trace elements without the need for sampling. However, it is often assumed that its limited sensitivity to light elements (Z < 13) in field applications poses challenges for a comprehensive characterisation of glass-making technologies and raw materials. In this study, we evaluate the analytical performance of μ-XRF for glass analysis, going beyond compositional characterisation by systematically comparing it with Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS). LA-ICP-MS offers superior detection limits and a broader elemental range, but its use is constrained by the need for sample preparation and its non-portable nature. The precision of μ-XRF was assessed through repeated daily measurements, while accuracy was evaluated using the certified reference glasses Corning A and B. Correction trends were developed by comparing μ-XRF results with both the nominal concentrations of these standards and with corresponding LA-ICP-MS data, in order to mitigate inherent limitations and enhance the method's reliability. The validated protocol was then applied to a set of Roman glass samples of unknown composition. The results confirm that, despite its limitations, μ-XRF can deliver accurate and reproducible data, supporting its use as a rapid and versatile method for in-situ compositional analysis of ancient glass.