Synchronous fluorescence spectroscopy (SFS) is a rapid, sensitive and non-destructive method suitable for the analysis of multifluorophoric mixtures. In this study non linear variable angle synchronous spectrofluorimetry was applied to the determination of three fluoroquinololes in urine. Although this technique provides very good results, total resolution of multicomponent mixtures is not always achieved when the spectral profiles strongly overlap. Partial least-squares regression (PLS-1) was utilized to a develop calibration model that related synchronous fluorescence spectra to the analytical concentration of fluoroquinolones in the presence of urine. The same multicomponent mixture was determined using excitation emission matrix fluorescence (EEMF) along with N-way partial least squares regression (N-PLS and U-PLS). The determination was carried out in micellar medium 0.01 M with a pH of 4.8 provided by 0.2 M sodium acetate/acetic acid buffer. A central composite design was selected to obtain a calibration matrix of 25 standards plus a blank sample. The proposed methods were validated by application to a test set of synthetic samples. The results show that SFS with PLS-1 is a better method compared to EEMF with N-PLS or U-PLS because of the low RMSEP values of the former.
Copyright © 2012 Elsevier B.V. All rights reserved.