H NMR-based metabolomics combined with chemometrics to detect edible oil adulteration in huajiao (Zanthoxylum bungeanum Maxim.).

Huajiao is a highly valued spice that is susceptible to fraudulent adulteration, particularly the addition of edible oils to increase weight and improve color. Nuclear magnetic resonance (H NMR) and chemometrics were used to analyze 120 huajiao samples adulterated with different types and levels of edible oils. Using untargeted data and partial least squares-discriminant analysis (PLS-DA), the discrimination rate between types of adulteration reached 100% accuracy, and the R value of the prediction set for the level of adulteration using the targeted analysis dataset combined with PLS-regression methods reached 0.99. Triacylglycerols, major components of edible oils, were identified as a marker of adulteration through the variable importance in projection of the PLS-regression. A quantitative method based on the sn-3 triacylglycerol signal was developed that can achieve a detection limit of 0.11%. Testing of 28 market samples showed adulteration with various edible oils, with adulteration rates ranging from 0.96% to 4.41%.