D-optimal experimental design coupled with parallel factor analysis 2 decomposition a useful tool in the determination of triazines in oranges by programmed temperature vaporization–gas chromatography–mass spectrometry when using dispersive-solid phase ex

• The analysis of triazines in oranges by PTV-GC/MS according to SANCO/12495 is done.
• The QuEChERS procedure and the injection step are optimized with D-optimal designs.
• The PARAFAC2 decomposition solves coeluents and shifts in the retention time problems.
• The unequivocal identification avoids false negatives even with unknown co-eluents.
• Decision limits found are far below the applicable MRLs for the triazines in oranges.

The determination of triazines in oranges using a GC–MS system coupled to a programmed temperature vaporizer (PTV) inlet in the context of legislation is performed. Both pretreatment (using a Quick Easy Cheap Effective Rugged and Safe (QuEChERS) procedure) and injection steps are optimized using D-optimal experimental designs for reducing the experimental effort. The relative dirty extracts obtained and the elution time shifts make it necessary to use a PARAFAC2 decomposition to solve these two usual problems in the chromatographic determinations. The “second-order advantage” of the PARAFAC2 decomposition allows unequivocal identification according to document SANCO/12495/2011 (taking into account the tolerances for relative retention time and the relative abundance for the diagnostic ions), avoiding false negatives even in the presence of unknown co-eluents. The detection limits (CCα) found, from 0.51 to 1.05 μg kg−1, are far below the maximum residue levels (MRLs) established by the European Union for simazine, atrazine, terbuthylazine, ametryn, simetryn, prometryn and terbutryn in oranges. No MRL violations were found in the commercial oranges analyzed.