Centres Científics i Tecnològics UB

News

25.04.2017

Publication of "Confocal Raman imaging and chemometrics applied to solve forensic document examination involving crossed lines and obliteration cases"

Dr. Tariq Jawhari, responsible of Raman Spectroscopy Technology, has participated as co-author in the publication of the paper: “Confocal Raman imaging and chemometrics applied to solve forensic document examination involving crossed lines and obliteration cases” published in “Analyst” Journal nº 142.

The abstract of the paper is:

"This article describes a non-destructive analytical method developed to solve forensic document examination problems involving crossed lines and obliteration. Different strategies combining confocal Raman imaging and multivariate curve resolution-alternating least squares (MCR-ALS) are presented. Multilayer images were acquired at subsequent depth layers into the samples. It is the first time that MCR-ALS is applied to multilayer images for forensic purposes. In this context, this method provides a single set of pure spectral ink signatures and related distribution maps for all layers examined from the sole information in the raw measurement. Four cases were investigated, namely, two concerning crossed lines with different degrees of ink similarity and two related to obliteration, where previous or no knowledge about the identity of the obliterated ink was available. In the crossing line scenario, MCR-ALS analysis revealed the ink nature and the chronological order in which strokes were drawn. For obliteration cases, results making active use of information about the identity of the obliterated ink in the chemometric analysis were of similar quality as those where the identity of the obliterated ink was unknown. In all obliteration scenarios, the identity of inks and the obliterated text were satisfactorily recovered. The analytical methodology proposed is of general use for analytical forensic document examination problems, and considers different degrees of complexity and prior available information. Besides, the strategies of data analysis proposed can be applicable to any other kind of problem in which multilayer Raman images from multicomponent systems have to be interpreted."