3.1 QCL MIR spectra
Reflectance spectra for TNT and RDX were acquired. Fig. 2(a)shows the characteristic spectroscopic signals for each HE. For TNT, the prominent signal located at 1567 cm-1(-NO2 asymmetric stretch), 1472 cm-1(CH3 deformation), 1445 cm-1 (C-C ring stretch), 1359 cm-1 (ring stretch), 1199 cm-1 (CH in-plane ring bend), 1173 cm-1 (CH in-plane ring bend scissoring), 1089 cm-1 (CH3 bend twisting), and 1025 cm-1 (CH in-plane ring bend rocking; CH3 deformation) were observed. [52] For RDX, the important band at 1321 cm-1 for -N-NO2 symmetric stretch and the signal at 1593 cm-1 for the -N-NO2 asymmetric stretch were observed. Other signals that stand out include the band at 1234 cm-1 and the band at 1034 cm-1 for -N-C-N stretch; the signals at 1016 cm-1 for the combination -N-NO2stretch and in-plane C–H bending. [53-55]
Fig. 2(b) shows the reflectance spectra of the seven soil samples, identified by Soil-1, Soil-2, Soil-3, Soil-4, Soil 5, Soil-6, and Soil-7. A Principal Components Analysis (PCA) was carried out with the spectra of standard samples of clays and sand, and it was found that PC2 could differentiate between the classes clay and sand. The PC model enabled to conclude that Soils 1-4 have a higher clay proportion (60-65%) in comparison with Soils 5-7, with a clay proportion of 22-29%, considered mostly sandy soils (data not shown). These percentages were calculated from the distances of the PC2 scores with respect to the average of the classes (sand and clay).
Figs. 2(c) and 2(d) show a comparison of the simulated spectra with the experimental spectra considering the mixes at high concentrations and with the same type of soil (Soil 1). The characteristic fingerprint signatures were observed for both TNT and RDX with Soil 1, respectively, Figs. 2(c) and 2(d) . However, some differences in the spectral patterns between the simulated and experimental spectra are observed.