Forensic Investigations of Vehicle-Related Crimes: Is it Elemental?

Zachary Andrews*, Oriana Ovide, Ruthmara Corzo, Claudia Martinez-Lopez, Lauryn Alexander, and Tatiana Trejos, Department of Forensic and Investigative Sciences, West Virginia University, Morgantown, WV 26506

Field (Broad Category): Forensics (Physical Sciences & Engineering) 

Student’s Major: Forensic and Investigative Science 

Vehicle glass is an important type of trace evidence that can provide a link between a potential suspect and crime, such as a hit and run, a burglary, or a drive-by shooting. The elemental profile of glass is informative in these investigations and can be analyzed using a variety of methods such as Micro x-ray fluorescence spectroscopy (μ-XRF) and Laser-Induced Breakdown Spectroscopy (LIBS). Although these methods have been previously reported to fit for purpose, recent advances in technology are anticipated to enhance their analytical performance further. The purpose of this study was to evaluate modern μ-XRF and LIBS instrumental configurations, including their discrimination power and error rates. Variability of elemental composition within 100 fragments from a single windshield was evaluated to assess false exclusion rates. Variability between glass originated from 22 different automobiles was used to estimate false inclusion rates. When comparing fragments within the same pane, the variability for the majority of elements was found to be lower than 10% RSD for both methods. Comparison methods simulating casework situations in which one questioned fragment is compared to three known fragments resulted in superior performance than single one-to-one pairwise comparisons, with false exclusions, false inclusions, and discrimination power better than 4.2%, 2.9%, and 97.1% respectively. LIBS and μ-XRF are shown to be complementary methods for the forensic comparisons of vehicle glass. 

Funding: NIST SURF 

Program/mechanism supporting research/creative efforts: NIST SURF