Detection of einkorn flour adulteration in flour and bread samples using Computer-Based Image Analysis and Near-Infrared Spectroscopy
Künye
Ayvaz, H., Korkmaz, F., Polat, H., Ayvaz, Z., & Barış Tuncel, N. (2021). Detection of einkorn flour adulteration in flour and bread samples using Computer-Based Image Analysis and Near-Infrared Spectroscopy. Food Control, 127, 108162. https://doi.org/10.1016/j.foodcont.2021.108162Özet
Einkorn (Triticum monocum L.) is an old variety of wheat and is considered an ancient grain. Currently, limited amounts of einkorn are produced within a few regions of Europe and the US, and therefore it is sold at higher prices than common wheat. Either for unfair economic gain (adulteration) or to compensate its weaker gluten structure, einkorn flour tends to be adulterated with bread wheat flour, which is frequently encountered in commercial einkorn flour or einkorn baked products. In this study, einkorn and bread wheat flours were initially analyzed for their common quality properties following the traditional analytical methods. Then, two rapid methods, Computer-Based Image (CBI) Analysis and Near-Infrared Spectroscopy (NIRS), were evaluated to rapidly estimate the % level of bread wheat flour in both einkorn-wheat flour mixtures and the bread made of those mixtures. For this purpose, binary mixtures of einkorn flour and the adulterating bread wheat flour were prepared for calibration (46 samples) and external validation (18 samples) sets, with wheat flour content in the mixtures ranging between 5 and 95% (w/w). Then, for each binary mixture, a loaf of conventional bread was produced. Flour mixtures and loaves of bread samples produced were analyzed by both CBI and NIRS. Our results suggested that CBI could only yield high correlation levels between the wheat flour content and some color properties in bread samples (>0.96), while no sufficient correlations were observed in flour mixtures. Regarding the NIRS, highly accurate models were developed for both flour mixtures (correlation coefficients > 0.99, standard errors < 1.39% and RPD level of 19.3) and bread samples (correlation coefficients > 0.94, standard errors < 2.64% and RPD level of 10.1). Our results indicate that both NIRS and CBI may be implemented in the rapid and easy screening of wheat flour adulteration in einkorn bread, while only NIRS is suggested to be used for the same purpose in flour mixtures.