ORGANIC MICROPOLLUTANTS ANALYSIS AND THEIR METROLOGICAL TRACEABILITY TO SI UNITS

The PhD activity described in the present work concerned the determination of organic micropollutants in food matrices at trace level, with the aim of correctly evaluating the measurement uncertainty and establishing metrological traceability for the results of these measurements. The metrological traceability of organic micropollutants measurements is a relevant issue due to the potential adverse effects that these substances can act on human health and on natural ecosystems. Many organic micropollutants have been classified as Persistent Organic Pollutants (POPs) by the United Nations Environment Programme (UNEP), in the framework of the Stockholm Convention (2001) [1]. The research activity was addressed to the study of some food matrices and the attention was focused on two different matrices, green tea powder and milk, and on two organic molecules: - endosulfan, an organochlorine pesticide banned in 2011 due to its toxicity, analysed in green tea; - melamine, a molecule commonly used in the plastic industry which became famous for its fraudulent use as adulterant of the protein content of milk in 2008. Ad hoc methods were developed for the extraction of these molecules from food samples and for their quantification by means of two different analytical techniques: - gaschromatography coupled with mass spectrometry (GC-MS), for the analysis of endosulfan; - Surface Enhanced Raman Scattering (SERS) spectroscopy for the analysis of melamine. For the development of these methods a metrological approach was adopted and correct metrological traceability chains to the SI units were established, considering all the steps which constitute an analytical method. The activity concerning the Organochlorine Pesticides (OCPs) started for the participation in 2012 in an international comparison carried out in the framework of the Comité Consultatif pour la Quantité de Matière (CCQM), namely the “Pilot Study CCQM-P136 Mid-Polarity Analytes in Food Matrix: Mid-Polarity Pesticides in Tea”. Metrological traceability was established using suitable Certified Reference Materials (CRMs), both for the calibration of the analytical instrumentation and for the evaluation of the recovery efficiency of the analytes from the matrix. For the uncertainty evaluation, two approaches were followed: the classical GUM approach based on the law of propagation of uncertainty [2] and the Monte Carlo method [3]. The activity for the determination of melamine in milk consisted in the set up of an extraction procedure followed by the analysis by means of Raman spectroscopy, performed after mixing the samples with gold nanoparticles. Their synthesis is carried out in order to obtain particles having a diameter of 40 nm and a negatively charged surface, which allows their interaction with the melamine molecules (positively charged at acidic pH). These agglomerates are responsible for the amplification of the Raman signal by means of the SERS effect, which generates a considerable intensity enhancement of the Raman signal when the sample is put in contact with a metallic irregular surface or constituted of metallic nanoparticles. This procedure was validated by studying different parameters (linearity, repeatability, limit of detection, limit of quantification and recovery). The developed method allows the quantification of melamine in milk samples in accordance with the European law limits, fixed by the Codex Alimentarius Commission, due to its potential toxicity, to 1 mg/l for powder infant formula and 2.5 mg/l for other foods and animal feed [4]. For the determination of the calibration curve of the Raman spectrophotometer and of the uncertainties of the related coefficients, an algorithm based on Total Weighted Least Squares [5] was used, which takes into account the uncertainty contributions deriving from the instrumental repeatability and the calibration solutions used to calibrate the measuring instrument. References: [1] www.unep.org [2] JCGM 100:2008 “GUM 1995 with minor corrections – Evaluation of measurement data – Guide to the expression of uncertainty in measurements” [3] JCGM 101:2008 “Evaluation of measurement data – Supplement 1 to the “Guide to the expression of uncertainty in measurement” – Propagation of distributions using a Monte Carlo method” [4] Report on the Thirty-Third Session of the Joint FAO/WHO Food Standards Programme (2008) [5] Malengo A., Pennecchi F,. “A weighted total least-squares algorithm for any fitting model with correlated variables, Metrologia 50 (2013) 654-662