Validation of an analytical method for the refractive index measurement of glass fragments. Application to a hit-and-run incident

Abstract

A method to determine the refractive index (RI) of glass samples was validated in this research article. This validation was performed by studying analytical features such as the working range, precision, robustness, and bias. Locke silicone oil type B, glass standards type B and a sodium D source (589.3 nm) were chosen for a working RI range from 1.50225 to 1.52381, according to the typical RI values of glass samples of forensic interest. Robustness, calculated as Relative Standard Deviation (RSD) of slopes of ten calibration lines within a period of one month was 0.24%. Precision of the method (calculated for the two extreme glass standards B2 and B12 in the working range) was 0.21% and 0.14% (repeatability) and 0.12% and 0.09% (intermediate precision), respectively. A Certified Reference Material (CRM) of glass of known RI was employed to estimate the bias of the method, verifying a relative error of 0.00066%. The method was applied to eight validation samples (six glass fragments from different parts of an automobile, a glass container and an architectural tinted window), which were differentiated through their RI values. Finally, the procedure was applied to interpret the origin of glass evidence taken from a hit-and-run incident. A common origin of recovered and control glass fragments was supposed on the basis of the RI determination of these samples, together with the application of useful match criteria in forensic pairwise comparisons of glass fragments. Results were confirmed by means of a highly discriminatory technique such as Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry (LA-ICP-MS).