In the forensic examination of automotive paint, each layer of paint is analyzed individually by FTIR. The more unique the paint layers are, the more information is contained in the sample, and the stronger are the forensic conclusions that can be drawn.  Laboratories in the United States and Canada will hand-section each layer of the paint and present each separated layer to an infrared spectrometer for analysis, which is time consuming.  Sampling too close to the boundary between adjacent layers can also occur which produces an IR spectrum that is a mixture of two layers.  Not having a “pure” spectrum of each layer prevents a meaningful comparison between each paint layer or, in the situation of searching an automotive paint database, will prevent the scientist from developing an accurate hit list of potential automotive suspects. These problems have been addressed by collecting concatenated IR data from all paint layers in a single analysis by scanning across the cross-sectioned layers of a microtomed paint sample using an FTIR imaging microscope.  Decatenation of the concatenated IR data is achieved using a variation of polytopic vector analysis to select the starting points for a modified alternating least squares algorithm to obtain a “pure” IR spectrum of each automotive paint layer.  This approach saves time, eliminates the need to analyze each layer separately and ensures that the final spectrum of each layer is “pure” and not a mixture.  By integrating this imaging experiment with a prototype pattern recognition IR library searching system to identify the make and model of an automotive vehicle within a limited production year range, we will demonstrate that forensic examination of automotive paints can be facilitated in terms of both the speed and accuracy of the analysis.  The proposed decatenation and library searching methodology is shown to work well in both transmission and attenuated total reflection mode, across different instrument platforms, and with small and large data sets.<br/>

Infrared imaging and multivariate curve resolution for the forensic examination of automotive paint chips

251st National Meeting (2016)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.