Classification of ANFO samples based on their fuel composition by GC-MS and FTIR combined with chemometrics

Praew Suppajariyawat; Mathieu Elie; Mark Baron; Jose Gonzalez-Rodriguez

doi:10.1016/j.forsciint.2019.06.001

Classification of ANFO samples based on their fuel composition by GC-MS and FTIR combined with chemometrics

Forensic Sci Int. 2019 Aug:301:415-425. doi: 10.1016/j.forsciint.2019.06.001. Epub 2019 Jun 11.

Authors

Praew Suppajariyawat¹, Mathieu Elie², Mark Baron², Jose Gonzalez-Rodriguez³

Affiliations

¹ School of Chemistry, University of Lincoln, Joseph Banks Laboratories, Green Lane, Lincoln, Lincolnshire LN6 7DL, United Kingdom; Central Institute of Forensic Science, Ministry of Justice, 499 Sukpraprute Building 16th Floor, Prachachuen Rd., Bang-Sue, Bangkok 10800, Thailand.
² School of Chemistry, University of Lincoln, Joseph Banks Laboratories, Green Lane, Lincoln, Lincolnshire LN6 7DL, United Kingdom.
³ School of Chemistry, University of Lincoln, Joseph Banks Laboratories, Green Lane, Lincoln, Lincolnshire LN6 7DL, United Kingdom. Electronic address: jgonzalezrodriguez@lincoln.ac.uk.

PMID: 31228699
DOI: 10.1016/j.forsciint.2019.06.001

Abstract

Ammonium nitrate fuel oil (ANFO) is one of the most favorite explosives used in terrorist attacks. This explosive is a complex mixture of 95-96% ammonium nitrate (AN) and 4-5% liquid hydrocarbons (fuel oil). In this study, we analyze a variety of ANFO explosive mixtures in order to classify their different sources of origin by observing the difference in fuel components. The study was performed by mixing ammonium nitrate with eight different diesel brands collected in Lincoln, UK in two seasons (winter and summer). The samples were extracted using appropriate solvent and extracts were subsequently analyzed in sextuplicate by gas chromatography-mass spectrometry (GC-MS) and Fourier transform infrared spectroscopy (FTIR). A classification model was performed using principal component analysis (PCA) and Lineal Discriminant Analysis (LDA). In this study, four fatty acid methyl ester (FAME) contents were observed by GC-MS in all summer samples but found lack in some winter sample resulting in seasonal variation effect. The classification of pre-blast ANFO samples was achieved using GC-MS and FTIR in a combination with PCA/LDA. The results significantly showed the variation of specific diesel components and providing different classification performance among ANFO samples with high classification performance. Therefore, this study can be beneficial in forensic investigation that the use of diesel components are able to classify among different ANFO samples.

Keywords: ANFO explosives; Chemometrics; Diesel; FAMEs; FTIR; GCMS.