Classification of the Residues after High and Low Order Explosions Using Machine Learning Techniques on Fourier Transform Infrared (FTIR) Spectra

Infrared to terahertz identification of chemical substances used for the production of IEDs

In our modern times, improvised explosive devices (IEDs) have become more sophisticated than ever, capable of causing destruction and loss of life. The creative use of homemade substances for IEDs manufactures has led to efforts in developing sensitive detection methods that can anticipate, identify and protect against improvised attacks. Laser-based spectroscopic techniques provide rapid and accurate detection of chemicals in improvised explosives, but no single method can detect all components of all explosives. In this study, two spectroscopic methods are used for the sensitive identification of 8 explosive chemical substances in the form of powders and vapors. Absorption spectra of benzene, toluene, acetone and ethylene glycol were examined with CO2 laser photoacoustic spectroscopy. The photoacoustic signals of the samples were recorded in the CO2 laser emission range from 9.2 to 10.8 µm and a different spectral behavior was observed for each analyzed substance. Time-domain spectroscopy with THz radiation was used to analyze ammonium nitrate, potassium chlorate, dinitrobenzene, hexamethylenetetramine transmission spectra in the 0.1-3 THz range, and it was observed that they have characteristic THz fingerprint spectra. CO2 laser photoacoustic spectroscopy and THz time domain spectroscopy have met the criterion of proven effectiveness in identifying explosive components. The combination of these spectroscopic methods is innovative, giving a promising new approach for detection of a large number of IED components.