Classical Least Squares Discriminant 1 Analysis of High Explosives Detected on Cotton Fabrics by Quantum Cascade Laser Spectroscopy

Quantum cascade laser spectroscopy was used to detect the presence of residues of highly energetic materials (HEMs) on cotton fibers. The discrimination of the vibrational signals of HEMs from a highly mid-infrared (MIR) absorbing medium was achieved by a simple and fast spectral evaluation using th...

Full description

Autores:
Pacheco-Londoño, Leonardo C.
Aparicio-Bolaños, Joaquín A.
Galán-Freyle, Nataly J.
Román-Ospino, Andrés
Hernandez, Samuel
Tipo de recurso:
Fecha de publicación:
2017
Institución:
Universidad Simón Bolívar
Repositorio:
Repositorio Digital USB
Idioma:
eng
OAI Identifier:
oai:bonga.unisimon.edu.co:20.500.12442/1766
Acceso en línea:
http://hdl.handle.net/20.500.12442/1766
Palabra clave:
Quantum cascade laser (QCL) spectroscopy
Explosives detection
Classical least squares (CLS)
Cotton fabrics
Discriminant analysis (DA)
Highly energetic materials (HEMs)
Rights
License
Licencia de Creative Commons Reconocimiento-NoComercial-CompartirIgual 4.0 Internacional
Description
Summary:Quantum cascade laser spectroscopy was used to detect the presence of residues of highly energetic materials (HEMs) on cotton fibers. The discrimination of the vibrational signals of HEMs from a highly mid-infrared (MIR) absorbing medium was achieved by a simple and fast spectral evaluation using the classical least squares (CLS) algorithm without preparation of standards. CLS focuses on minimizing the differences between spectral features of real spectra acquired by direct MIR spectroscopy and the spectral features of calculated spectra modeled from linear combinations of the spectra of the neat components: HEMs and the cotton fibers, and the bias. HEMs samples in several combinations with cotton fibers were used to validate the methodology. Three (3) independent sets of experiments considering binary, ternary, and quaternary combinations of components, including cotton, TNT, RDX, and PETN, were performed. The models parameters obtained from linear combinations of the calculated spectra were used to perform discrimination analyses and to determine the sensitivity and selectivity of the studied HEM with respect to the substrates and to each other. However, the discrimination analysis was not necessary to achieve successful detection of HEMs samples on cotton substrates. The only requirement to achieve HEM detection (determine the presence or absence of HEM on a substrate) is that the library contains the spectra of all the HEMs and substrates or that the later be added in the field, on the fly. In addition, the extracted spectral signals of several amounts of RDX on cotton (> 0.02 mg) were used to calculate the limit of detection (LOD) based on the spectral signalto- noise ratio (S/N). The calculated S/N values were obtained from the spectra for cotton dosed with several amounts of RDX deposited in decreasing mass order until the calculated S/N reached a value of 3. The LOD determined for RDX on cotton was 22 ± 6 μg.