Advantages of data fusion. First multivariate curve resolution analysis of fused liquid chromatographic second-order data with dual diode array-fluorescent detection

GABRIELA IBAÑEZ; ROCÍO B. PELLEGRINO VIDAL; GRACIELA ESCANDAR

ANALYTICAL CHEMISTRY

AMER CHEMICAL SOC

Lugar: Washington; Año: 2017 vol. 89 p. 3029 - 3035

0003-2700

Forthe first time, liquid chromatography-diode array detection (LC-DAD) andliquid-chromatography fluorescence detection (LC-FLD) second-order data,collected in a single chromatographic run, were fused and chemometricallyprocessed for the quantitation of coeluting analytes. Two different experimentalmixtures composed of fluorescent and non-fluorescent endocrine disruptors wereanalyzed. Adequate pretreatment of thematrices before their fusion was crucial to attain reliable results.Multivariate curve resolution-alternating least-squares (MCR-ALS) was appliedto LC-DAD, LC-FLD and fused LC-DAD-FLD data. Althoughdifferent degrees of improvement are observed when comparing the fused matrixresults in relation to those obtained using a single detector, clear benefitsof data fusion are demonstrated through: (1) the obtained limits of detectionin the ranges 2.1?24 ng mL?1 and 0.9?6.3 ng mL?1 for the two evaluated systems and (2) the low relative predictionerrors, below 7% in all cases, indicating goodrecoveries and precision. The feasibility of fusing data andits advantages in the analysis of real samples was successfully assessed through the study of spikedtap, underground and river water samples.