Analysis of Tires and Tire Traces Using FTIR and Py-GC/MS

• NCJ Number: 205482
• Journal: Canadian Society of Forensic Science Volume: 37 Issue: 1 Dated: March 2004 Pages: 19-37
• Author(s): G. Sarkissian; J. Keegan; E. Du Pasquier; J. P. Depriester; P. Rousselot
• Editor(s): B. Yamashita
• Date Published: March 2004
• Length: 19 pages
• Publication Format: Article
• Publication Type: Report (Study/Research)
• Language: English
• Country: Canada

Annotation

This study examined three analytical techniques used to analyze and differentiate tire rubber samples.

Abstract

Vehicles are commonly used in the commission of crimes, and quite often it is the only link between the crime and the criminal. One way this link is formed is through the tires of the vehicle, be it through tread marks left at the scene or markings of rubber on the road. Currently rubber traces are rarely used or collected for analysis. This study examined three analytical techniques to determine their effectiveness at identifying tire rubber samples. A total of 59 tire samples were collected from cars involved in accidents, with 58 of the samples being from summer tires and only 1 sample being from winter tires. The samples were collected in France and included numerous brands, models, sizes, production dates, and countries of manufacture. All 59 samples were analyzed with Pyrolysis-Gas Chromatography/Mass Spectrometry (Py-GC/MS), while only 27 samples were analyzed through Fourier Transform Infrared Spectroscopy (FTIR) using both Attenuated Total Reflectance (ATR) and Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS). The investigation revealed that FTIR, using ATR and DRIFTS for this sample, showed less ability to differentiate samples, and is not recommended for this use due to poor discrimination and poor reproducibility. Py-GC/MS showed promise in this analysis in both reproducibility and discrimination. The results revealed that a large number of samples could be discriminated based on the composition of the tire. The use of linear discriminant analysis (LDA) in tandem with the Py-GC/MS further improved the discrimination of samples, with 98.3 percent of the samples able to be discriminated to a batch level, and 94.9 percent of samples discriminated to a brand level. These findings show that Py-GC/MS used with both principal component analysis (PCA) and LDA provides the analyst with a powerful analytical tool in identifying and classifying trace rubber residues, to the level of which particular production batch a tire came from. 13tables, 7 figures, and 10 references