Search

Optimization and validation of a multi-mycotoxin method by LC-MS/MS is presented. The method covers the EU-regulated mycotoxins (aflatoxins, fumonisins, ochratoxin A, deoxynivalenol, zearalenone, T-2 and HT-2), as well as nivalenol and 3- and 15-acetyldeoxynivalenol for analysis of cereals, cocoa, oil, spices, infant formula, coffee and nuts. The proposed procedure combines two clean-up strategies: First, a generic preparation suitable for all mycotoxins based on the QuEChERS (for quick, easy, cheap, effective, rugged and safe) protocol. Second, a specific clean-up devoted to aflatoxins and ochratoxin A using immunoaffinity column (IAC) clean-up. Positive identification of mycotoxins in matrix was conducted according to the confirmation criteria defined in EU Commission Decision 2002/657/EC while quantification was performed by isotopic dilution using (13)C-labeled mycotoxins as internal standards. Limits of quantification were at or below the maximum levels set in the EC/1886/2006 document for all mycotoxin/matrix combinations under regulation. In particular, the inclusion of an IAC step allowed achieving LOQs as low as 0.05 and 0.25μg/kg in cereals for aflatoxins and ochratoxin A, respectively. Other performance parameters like linearity [(r)(2)>0.99], recovery [71-118%], precision [(RSDr and RSDiR)<33%], and trueness [78-117%] were all compliant with the analytical requirements stipulated in the CEN/TR/16059 document. Method ruggedness was proved by a verification process conducted by another laboratory., (Copyright © 2014 Elsevier B.V. All rights reserved.)

Stir bar sorptive extraction (SBSE)-thermal desorption (TD) procedure combined with gas chromatography mass spectrometry (GC-MS) and the statistical variance component model (VCM) is applied to the determination of semi-volatile compounds including organochlorine and organophosphorus pesticides in various synthetic and real fruit-based soft drink matrices. When the matrix effects are corrected using isotopically labelled or non labelled internal standard, but matrix/calibration run-induced deviations are still present in the measurements, the adoption of a variance component model (VCM) in the quantitative analysis of various matrices via an overall calibration curve is successful. The method produces an overall calibration straight line for any analyte accounting for the uncertainty due to all the sources of uncertainty, namely matrix-induced deviations, calibration runs performed at different times, measurement errors. Small increases in the detection limits and in uncertainty in the concentration values obtained in the inverse regression face favourably the decrease in times and costs for routine analyses., (Copyright © 2010 Elsevier B.V. All rights reserved.)