RESUMO
We examined the feasibility of using handheld mid-infrared (MIR) Fourier-Transform infrared (FT-IR) instrumentation for detecting and analysing cyanide (CN) contamination in field contaminated soils. Cyanide spiking experiments were first carried out, in the laboratory, to test the sensitivity of infrared Fourier transform (DRIFT) spectrometry to ferro- and ferricyanide compounds across a range of reference soils and minerals. Both benchtop and handheld diffuse reflectance infrared spectrometers were tested. Excellent results were obtained for the reference soils and minerals, with the MIR outperforming the near-infrared (NIR) range. Spectral peaks characteristic of the -C≡N group were observed near 2062 and 2118cm-1 in the MIR region for the ferro- and ferricyanide compounds spiked into soils/minerals, respectively. In the NIR region such peaks were observed near 4134 and 4220cm-1. Cyanide-contaminated samples were then collected in the field and analyzed with the two spectrometers to further test the applicability of the DRIFT technique for soils containing aged CN residues. The prediction of total CN in dry and ground contaminated soils using the handheld MIR instrument resulted in a coefficient of determination (R2) of 0.88-0.98 and root mean square error of the cross-validation (RMSE) of 21-49mgkg-1 for a CN range of 0-611mgkg-1. A major peak was observed in the MIR at about 2092cm-1 which was attributed to "Prussian Blue" (Fe4[Fe(CN)6]3·xH2O). These results demonstrate the potential of handheld DRIFT instrumentation as a promising alternative to the standard laboratory method to predict CN concentrations in contaminated field soils.