Partial least square regression method for the detection of polycyclic aromatic hydrocarbons in the soil environment using laser-induced fluorescence spectroscopy

Abstract

Polycyclic aromatic hydrocarbons (PAHs) are currently of great concern due to their potential carcinogenicity. The standard methods of identification and quantification of PAHs are costly and time-consuming. Laser-induced fluorescence (LIF) spectroscopy can overcome these limitations and make it possible to simultaneously identify and quantify several PAHs without the aid of chemical pre-processing steps. LIF system and multivariate chemometric analysis, especially the partial least square (PLS) regression method, have been employed to investigate the potential correlations between LIF data and soil properties, especially their physical composition and moisture content, and confirm the capability for the detection of PAHs in soils, using the soil samples artificially comprised of sand and silt/clay. Phenanthrene and pyrene were selected as representative PAHs. The diffuse reflectance has also been recorded to investigate the capability to normalize the LIF data. Data analysis, through a multivariate chemometric analysis, showed significant statistical correlations between the LIF intensity and the PAH concentrations in soil samples. The LIF intensity has been shown to depend on the sample properties, with the diffuse reflectance at 532 nm also depending on these factors. It has been demonstrated that the errors induced by the physical composition could be reduced by dividing the LIF intensity by the diffuse reflectance at 532 nm. The LIF concentration data were compared with those measured by high performance liquid chromatography (HPLC), with values of correlation coefficient (R) of 0.96 and 0.90 for phenanthrene and pyrene, respectively. The results obtained through a data analysis algorithm, based on the linear mixture model and the normalization by diffuse reflectance, demonstrated that the PAHs in artificial soil samples could be discriminated and quantified by LIF spectroscopy.