LIQUID CHROMATOGRAPHY WITH MULTIVARIATE DETECTION AND CHEMOMETRIC DATA PROCESSING: A BASIS FOR GREEN-ANALYTICAL METHODS

M.D. CARABAJAL; J.A. ARANCIBIA; R.B. PELLEGRINO VIDAL; G.M. ESCANDAR

Barcelona

Congreso; XVI Chemometrics in Analytical Chemistry; 2016

The general objective of the present work was the development of methods based on green-analytical chemistry principles for the quantification of organic pollutants in complex samples [1,2], by coupling liquid chromatography, multivariate detection and chemometrics data processing [3].Liquid chromatography with dual UV/diode array (DAD) and fluorescence (FLD) detections was carried out in a single run, and the second-order DAD-elution time and FLD-elution time data obtained were treated with MCR-ALS (multivariate curve resolution/alternating least-squares) algorithm [4]. In this way, while analytes are measured through their more appropriate (absorbance and/or fluorescence) signals, chemometric treatment of the corresponding matrices allows the resolution of total or partial overlapped bands, and to overcome the presence of interferences in real samples.The two investigated systems involved emerging contaminants belonging to plastics derived endocrine disruptors (bisphenol A, nonylphenol, octylphenol, diethyl phtalate, dibutyl phthalate and diethylhexyl phthalate), and ten agrochemicals of frequent use, namely fungicides, herbicides, insecticides and a plant growth regulator. The former were investigated in both soft and alcoholic beverages (mineral water, soda, juice, wine and beer) while agrochemicals were quantified in land cultivated vegetables, including mushroom, lettuce, alfalfa sprout, cucumber, and celery.The particularities of each investigated system and the differences in the corresponding data treatment are presented and discussed.The coupling of chromatographic approaches with second-order chemometric calibration allows us to considerably simplify the sample pretreatment and to significantly reduce the analysis time, resulting in analytical green methods that favorably compare with those usually employed in complex systems such as the ones here investigated. Low detection limits were achieved, even in the presence of partially overlapped chromatographic and spectral bands among analytes and potential interferents. The detection capabilities allow one to monitor the presence of the analytes in natural samples complying with international regulations.Acknowledgement: The authors are grateful to the Universidad Nacional de Rosario (Project BIO 237), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET, Project PIP 0163), and Agencia Nacional de Promoción Científica y Tecnológica (ANPCyT, Project PICT 2013-0136) for financial support.References:[1]S. Armenta, S. Garrigues, M. de la Guardia, Trends Anal. Chem. 27, 497 (2008).[2] M. Farré, S. Pérez, C. Gonçalves, M.F. Alpendurada, D. Barceló, Trends Anal. Chem. 29, 1347 (2010).[3] A.C. Olivieri, G.M. Escandar, Practical three-way calibration, Elsevier, Waltham, USA (2014).[4]R. Tauler, M. Maeder, A. de Juan, Multiset data analysis: extended multivariate curve resolution, in: S. Brown, R. Tauler, B. Walczak (Eds.), Comprehensive Chemometrics, 2, Elsevier, Oxford, 473 (2009).