An automated screening method for the fast, simple discrimination between natural and artificial colorants in commercial saffron products

A fast, simple screening method for discriminating between natural and adulterated saffron was developed that reduces the need for time-consuming techniques. The method uses a flow manifold where saffron samples containing adulterants in a phosphate buffer are passed through a column packed with cotton that retains the adulterants for subsequent elution with dilute ammonia. Samples testing positive for adulterants are subjected to a multivariate calibration technique such as partial least-squares regression (PLS) in order to quantify the adulterant analytes. This allows Methyl Orange, tartrazine and Ponceau-4R to be accurately quantified. The proposed method provides a yes/no overall response and allows the presence of adulterant species in positive samples to be confirmed. The calibration model was optimized by using an appropriate number of factors and the most suitable wavelength ranges to construct the Box-Behnken data matrix. The method was validated by application to saffron samples spiked with both individual adulterants and binary and ternary mixtures thereof.