Rapid prediction of deoxynivalenol contamination in wheat bran by MOS‐based electronic nose and characterization of the relevant pattern of volatile compounds

BACKGROUND Deoxynivalenol (DON) is a mycotoxin, mainly produced by Fusarium sp., most frequently occurring in cereals and cereal‐based products. Wheat bran refers to the outer layers of the kernel, which has a high risk of damage due to chemical hazards, including mycotoxins. Rapid methods for DON detection in wheat bran are required. RESULTS A rapid screening method using an electronic nose (e‐nose), based on metal oxide semiconductor sensors, has been developed to distinguish wheat bran samples with different levels of DON contamination. A total of 470 naturally contaminated wheat bran samples were analyzed by e‐nose analysis. Wheat bran samples were divided in two contamination classes: class A ([DON] ≤ 400 µg kg −1 , 225 samples) and class B ([DON] > 400 µg kg −1 , 245 samples). Discriminant function analysis (DFA) classified wheat bran samples with good mean recognizability in terms of both calibration (92%) and validation (89%). A pattern of 17 volatile compounds of wheat bran samples that were associated (positively or negatively) with DON content was also characterized by HS‐SPME/GC–MS. CONCLUSIONS These results indicate that the e‐nose method could be a useful tool for high‐throughput screening of DON‐contaminated wheat bran samples for their classification as acceptable / rejectable at contamination levels close to the EU maximum limit for DON, reducing the number of samples to be analyzed with a confirmatory method. © 2018 Society of Chemical Industry