IQUIR   05412
INSTITUTO DE QUIMICA ROSARIO
Unidad Ejecutora - UE
artículos
Título:
Three-way partial least-squares/residual bilinearization study of second-order lanthanide-sensitized luminescence excitation-time decay data. Analysis of benzoic acid in beverage samples.
Autor/es:
LOZANO, V. A.; IBAÑEZ, G. A.; OLIVIERI, A. C.
Revista:
ANALYTICA CHIMICA ACTA
Editorial:
Elsevier
Referencias:
Año: 2008 vol. 610 p. 186 - 195
ISSN:
0003-2670
Resumen:
Lanthanide-sensitized luminescence excitation-time decay matrices were employed for achieving the second-order advantage using as chemometric algorithms parallel factor analysis (PARAFAC) and multidimensional partial least-squares with residual bilinearization (N-PLS/RBL). The second-order data were measured for a calibration set of samples containing the analyte benzoic acid in the concentration range from 0.00 to 5.00mgL−1, for a validation set containing the analyte and the potential interferent saccharin (in the range 0.00–6.00mgL−1), and for real samples of beverages containing benzoic acid as preservant,  saccharin, and other potentially interfering compounds. All samples were treated with terbium(III), trioctylphosphine oxide as a synergistic ligand, and contained a suitable imidazol buffer, in order to ensure maximum intensity of the luminescence signals. The results indicate a slightly better predictive ability of the newly introduced N-PLS/RBL procedure over standard PARAFAC, both in what concerns the comparison with nominal analyte concentrations in the validation sample set and with results provided by the reference high-performance liquid chromatographic technique for the real sample set. saccharin, and other potentially interfering compounds. All samples were treated with terbium(III), trioctylphosphine oxide as a synergistic ligand, and contained a suitable imidazol buffer, in order to ensure maximum intensity of the luminescence signals. The results indicate a slightly better predictive ability of the newly introduced N-PLS/RBL procedure over standard PARAFAC, both in what concerns the comparison with nominal analyte concentrations in the validation sample set and with results provided by the reference high-performance liquid chromatographic technique for the real sample set.