Propellant’s differentiation using FTIR-photoacoustic detection for forensic studies of improvised explosive devices

• Published in: Forensic science international 2017-11, Vol.280, p.169-175
• Main Authors: Álvarez, Ángela, Yáñez, Jorge, Contreras, David, Saavedra, Renato, Sáez, Pedro, Amarasiriwardena, Dulasiri
• Format: Article
• Language: English
• Subjects: Algorithms; Ammonium; Ammonium nitrate; Ammunition; Analytical chemistry; Cellulose esters; Cellulose nitrate; Chemometric analysis; Chromatography; Classification; Crime; Differentiation; Discriminant analysis; Explosive devices; Explosive forming; Explosives; Explosives detection; Forensic chemistry; Forensic engineering; Forensic science; Forensic sciences; Formulations; Fourier transforms; Guanidine; Identification; Infrared analysis; Infrared spectroscopy; Mass spectrometry; Nitrates; Nitroglycerin; Nondestructive testing; Pattern recognition; Photoacoustic; Photoacoustic spectroscopy; Principal components analysis; Propellant; Scientific imaging; Statistical analysis; Sulfur
• ISSN: 0379-0738; 1872-6283
• DOI: 10.1016/j.forsciint.2017.09.018

•PAS combined with advanced statistics has been used to differentiate the propellant brands.•SIMCA and PLS-DA algorithms succeed for brand prediction and classification of propellant.•FTIR-PAS is a nondestructive and micro sampling technique suitable for forensic studies. The use of propellant for making improvised explosive devices (IED) is an incipient criminal practice. Propellant can be used as initiator in explosive mixtures along with other components such as coal, ammonium nitrate, sulfur, etc. The identification of the propellant’s brand used in homemade explosives can provide additional forensic information of this evidence. In this work, four of the most common propellant brands were characterized by Fourier-transform infrared photoacoustic spectroscopy (FTIR-PAS) which is a non-destructive micro-analytical technique. Spectra shows characteristic signals of typical compounds in the propellants, such as nitrocellulose, nitroglycerin, guanidine, diphenylamine, etc. The differentiation of propellant components was achieved by using FTIR-PAS combined with chemometric methods of classification. Principal component analysis (PCA) and soft independent modelling of class analogy (SIMCA) were used to achieve an effective differentiation and classification (100%) of propellant brands. Furthermore, propellant brand differentiation was also assessed using partial least squares discriminant analyses (PLS-DA) by leave one out cross (∼97%) and external (∼100%) validation method. Our results show the ability of FTIR-PAS combined with chemometric analysis to identify and differentiate propellant brands in different explosive formulations of IED.