Analysis of pharmaceuticals in soil and organic waste products: an experimental design approach to study the factors influencing the extraction step

Background and objective :Pharmaceutical products excreted by treated animals or humans are transferred to manure or wastewater, respectively. In wastewater treatment plants, they can be degraded and/or sorbed to sludge. Therefore pharmaceuticals may be indirectly disseminated into the environment from the recycling of organic waste products in agriculture. In order to evaluate the potential ecotoxicological effects of the pharmaceutical residues in the environment, it is necessary to develop analytical methods for their quantification. Several different extraction conditions are proposed in the literature and most of them have been developed for restricted groups of pharmaceuticals. It is therefore difficult to select the best method to extract pharmaceuticals in the context of multi-residue analysis. Our objective was to determine the influencing factors on the extraction recovery of pharmaceuticals by the experimental design approach, with emphasis on the extraction medium. We selected pharmaceuticals with various physical-chemical properties and representative of different families, to make possible the extension of the conclusions to other compounds belonging to the same families. Two contrasted solid samples were chosen: soil and sewage sludge. Methods and results : Fourteen pharmaceuticals including eight antibiotics were extracted from solid samples by Ultrasonic-Assisted Extraction (UAE), purified according to an adapted QuEChERS method and analysed by online solid-phase extraction coupled to ultra-high performance liquid chromatography and tandem mass spectrometry (online SPE-UHPLC-MS-MS) [1]. Three successive experimental designs were performed for each matrix. Preliminary experimental designs allowed appreciating the influence of four parameters presumed to be crucial for UAE efficiency: the nature of the organic solvent and its proportion, the nature of the aqueous phase (buffers, modifying agents) and its pH. A response surface design was then performed to better understand the influence on the extraction recovery of the organic solvent proportion in the extractant, the amount of ethylenediaminetetracetic acid (EDTA) added to the buffer and the final pH of the buffer, as well as the interactions between these factors. Our results showed a great influence of interactions between factors for all the fourteen studied pharmaceuticals, responsible for the shapes of response surfaces. Experimental designs were also used to compromise and propose two methods for multi-residue analysis of pharmaceuticals in soil and sewage sludge. The methods were validated with the accuracy profile methodology [2]. Discussion and conclusion : The experimental design approach was a powerful tool to demonstrate that not only parameters of the extraction medium but also interactions between these parameters influence the extraction recovery. The results were explained considering the nature of the solid matrices and the physical-chemical properties of the pharmaceuticals, mainly considering the ionisation state of the pharmaceuticals, their solubility in the extraction medium and their interactions with the solid matrix constituents. Experimental designs also allowed optimising extraction methods that were relevant to analyse environmental samples (see the presentation of Houot et al.). [1] M. Bourdat-Deschamps et al. J. Chromatogr. A 1349 (2014) 11–23 [2] S. Ferhi et al. Anal. Bioanal. Chem. (2016) submitted.