Multivariate curve resolution analysis excitation-emission matrices of fluorescence of humic substances.

Excitation-emission matrices (EEM) of fluorescence of aqueous solutions of humic substances (HS), and sets of EEM acquired as function of the HS concentration, were analysed by multivariate curve resolution alternating least squares (MCR-ALS). Three types of HS samples were studied: one commercial humic acid; two samples of fulvic acid (FA) extracted from a pinewood soil; two samples of FA extracted from recycled wastes. The fluorescence measurements were carried out at HS concentration between 5 and 100mg/L and at pH 6. The application of MCR-ALS algorithm on each individual EEM, as well as on column-wise augmented matrices, allows the identification of three major fluorophores in all HS samples analysed. The emission and excitation spectra of these fluorophores were recovered and are characteristic of each sample. Moreover, the variation of the fluorescence intensities of each fluorophore with HS concentration shows deviations from linearity at HS concentration higher than 30mg/L, depending on the fluorophore and/or sample. This behaviour reveals the existence of inner filter effects that affect the proportionally between the fluorescent signal and concentration but do not provoke measurable distortions on the fluorescence spectra of the detected fluorophores.