On-line preconcentration system using a short-liquid chromatography column coupled with Liquid chromatography diode array detection for the solid-phase extraction and clean-up of basics drugs in enviromental water samples

M. MARTÍNEZ GALERA; M.D. GIL GARCÍA; M.J. CULZONI; H.C. GOICOECHEA

Congreso; XV Reunión Nacional de la Sociedad Española de la Química Analítica; 2009

Human and veterinary drugs are used in large quantities and significant fractions of the original substances are excreted (as un-metabolized form or active metabolites) via urine or faeces. Moreover, some of them are not degraded in waste treatment plants and enter into hydrologic systems in their original form, with hazardous effects on human health and aquatic ecosystems. In the last decades, different studies carried out in some countries showed that most of these pharmaceuticals can be found in wastewater, river water and even in drinking water at levels of up to a few cg L−1. Therefore, several methods have been developed for the determination of pharmaceuticals in water in the lower ng L-1 range using normally off-line solid phase extraction (SPE) techniques combined with LC or GC. These extraction techniques involve the manipulation of samples, and therefore, they may lead to the partial loss of some of the compounds, as well as an increase of labour, time and cost. The main goal of the present work is the optimization of a simple and automated method using a first LC short column for the preconcentration of several basic drugs in river water samples coupled with a second conventional LC column for separation and diode array detection. Eleven pharmaceuticals compounds, among the more commonly used in human, have been selected in this study. The use of a short LC column (C-1) coupled to LC analytical column (C-2) allows the preconcentration, clean up and analysis of large volumes of water samples and the determination of drugs at ng L-1 range. In order to obtain the lower quantitation limits, the optimized parameters for peconcentration step were percentage of organic modifier in the water sample, flow and time of preconcentration in the first column. The best results were found with a flow rate of 1.5 mL min-1 during 20 min (corresponding to a load of water sample of 30 mL) and 0.4 % of methanol as organic modifier in the aqueous sample. Transference and separation of analytes in C-2 column were performed using a ternary gradient mobile phase composed of 0.25 mol L-1 KH2PO4 adjusted at pH 3.0, methanol and acetonitrile. Several mobile phases were evaluated in order to allow good peak resolutions for the eleven analytes although non resolution of chromatographic peaks was found for some analytes even for long time of analysis. In addition, when real river water samples were analyzed interferent peaks and matrix effect were found for some analytes. In order to solve these problems, standard addition methodology combined with Multivariate Curve Resolution (MCR) has been proposed as a complementary tool to traditional chromatographic approaches. The proposed methodology was validated using spiked blank river water samples. Quantitation limits were ranged between 0.1 and 0.5 cg L-1 with recovery percentages of basic drugs in the river water near to 100 %.