Evaluation of the performance of modern X-ray fluorescence spectrometry systems for the forensic analysis of glass

Abstract

Micro X-ray Fluorescence Spectrometry (µXRF) is a well-established technique for the elemental analysis of glass in forensic casework. The standard test method for the forensic analysis of glass using µXRF (ASTM E2926) provides recommendations for the number of replicate measurements that should be collected to characterize a known source and the criteria for the comparison between the known and questioned samples. However, these recommendations were based on interlaboratory data collected using µXRF instrumentation equipped with traditional lithium-doped silicon (SiLi) detectors. This interlaboratory study aimed to evaluate the performance of modern µXRF systems equipped with silicon drift detectors (SDDs) for the forensic comparison of glass. While the SDD-µXRF instruments resulted in improved precision and detection limits (1.4 µg·g−1–1386 µg·g−1) and excellent discrimination (>99 %) of different-source samples, the false exclusion rates for same-source samples were relatively high (>20 %). Two methods were evaluated to reduce the high false exclusion rates: increasing the number of fragments collected for the known source and modifying the recommended comparison criteria. To reduce the false exclusion rate to 5 % or less, a minimum of five known fragments were needed. Alternatively, modifying the recommended comparison criterion reduced the false exclusion rate from 23 % to 2 %, while maintaining low false inclusions (<1%). The findings in this study demonstrate the improved sensitivity and precision observed in glass measurements acquired with µXRF-SDD systems. However, these systems may require adjustments to sampling and the comparison criteria to minimize potential error rates in the forensic comparison of glass fragments.