High-resolution 1H Nuclear Magnetic Resonance spectrometry combined with chemometric treatment to identify adulteration of culinary spices with Sudan dyes

An efficient method for detecting the adulteration of commercial culinary spices with Sudan I–IV dyes is proposed using a combination of High-resolution 1H Nuclear Magnetic Resonance and chemometric treatment. The variables were reduced and selected on the basis of the difference between the NMR spectra from the non-contaminated commercial spices and the spices spiked with one of the four Sudan dyes. Partial Least Squares–Discriminant Analysis (PLS–DA) was applied to the most important NMR variables selected. The commercial spices studied were curry, turmeric, and mild and hot paprika, distributed in five classes: non-contaminated spices and spices spiked independently with one Sudan dye. The prediction probabilities provided by PLS–DA were satisfactory for all the classes. Only one spiked sample was misclassified in another contaminated class and it should be stressed that no spices from any of the contaminated classes is assigned to the non-contaminated one. This is very important from the point of view of consumer health, since a suspicious sample which might contain Sudan dyes will be correctly recognised as adulterated.