Liquid chromatography with diode array detection and multivariate curve resolution for the selective and sensitive quantification of estrogens in natural waters

• Published in: Analytica chimica acta Vol. 835; pp. 19 - 28
• Main Authors: Pérez, Rocío L., Escandar, Graciela M.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 04.07.2014
• Subjects: Algorithms; Analytical chemistry; Arrays; Chromatography, Liquid - methods; Chromatography, Liquid - standards; detection limit; Diode array detection; Diodes; Drinking Water - chemistry; estradiol; estriol; Estrogens; Estrogens - analysis; estrone; Groundwater - chemistry; least squares; Liquid chromatography; Mathematical models; Mineral Waters - analysis; Multivariate curve resolution; Natural waters; prediction; river water; Rivers - chemistry; Underground; Water Pollutants, Chemical - analysis; DERIV; TW; Estrogens; SPE; LW; Diode array detection; SPW; WWTPE; US; WWTPI; LOD; UW; MS; DSW; MW; Liquid chromatography; LOQ; CORT; BPA; MIP; MES; ENNFM; DLLME; FD; RW; SBSE; CP-CPE; MS/MS; DHS; FW; MM-SPE-MPS; Multivariate curve resolution; W; GC; WW; OP; HF-MMLLE; CF; SW; ANDR; DAD; ECF; DIS; PPs; SWW; DES; UPLC; LC; Natural waters; LVI; PROG
• ISSN: 0003-2670; 1873-4324; 1873-4324
• DOI: 10.1016/j.aca.2014.05.015

[Display omitted] •Potent endocrine disruptors are easily analyzed using non-sophisticated instrumental.•Selectivity is successfully achieved by applying multivariate curve resolution.•Quantification in real samples is accomplished using green-chemistry principles. Following the green analytical chemistry principles, an efficient strategy involving second-order data provided by liquid chromatography (LC) with diode array detection (DAD) was applied for the simultaneous determination of estriol, 17β-estradiol, 17α-ethinylestradiol and estrone in natural water samples. After a simple pre-concentration step, LC–DAD matrix data were rapidly obtained (in less than 5min) with a chromatographic system operating isocratically. Applying a second-order calibration algorithm based on multivariate curve resolution with alternating least-squares (MCR-ALS), successful resolution was achieved in the presence of sample constituents that strongly coelute with the analytes. The flexibility of this multivariate model allowed the quantification of the four estrogens in tap, mineral, underground and river water samples. Limits of detection in the range between 3 and 13ngL−1, and relative prediction errors from 2 to 11% were achieved.