Comparison of three multi-way models for resolving and quantifying bifenthrin and tetramethrin using gas chromatography–mass spectrometry

Gas chromatography–mass spectrometry (GC–MS) with second order calibration methods was used for the analysis of bifenthrin and tetramethrin. Under experimental conditions, bifenthrin and tetramethrin showed one and three peaks, respectively, that of bifenthrin completely overlapping with the last one of tetramethrin. A three-way data structure was constructed (sample × elution time × selected ion monitoring (SIM)), which successfully decomposed using three-way models. The data were then used to evaluate the multi-way methods of multivariate curve resolution alternating least squares (MCR-ALS), parallel factor analysis (PARAFAC) and parallel profiles with linear dependencies (PARALIND) and their performance was assessed in the quantitative analysis of the compounds. Using the PARAFAC model, certain initial conditions and restrictions were applied to the data and a new application of PARALIND model was developed for these types of data, where several peaks are observed for one of the compounds and the peaks from different compounds overlap. Finally, the results of the constructed models were compared.