Prediction of wheat flours composition using fourier transform infrared spectrometry (FT-IR)

Abstract

Physicochemical characteristics of seventy wheat flours of different species namely Einkorn (Triticum monococcum), Spelt (Triticum spelta) and common wheat (Triticum aestivum) were analyzed using different standard methods. The wheat grains were analyzed for moisture, ash, protein, wet gluten, sedimentation index, pH, acidity, fat, starch, falling number, damage starch and Glutograph parameters stretching and relaxation. The relationship between physicochemical characteristics and wheat samples were analyzed using the principal component analysis. For almost all the physicochemical data except moisture and damage starch were obtained significant differences between species. These differences exist in especially between common wheat samples and ancient wheat species. The highest protein, wet gluten, fat, ash, acidity were obtained for ancient species whereas the common wheat were richer in starch and presented the highest stretching and relaxation Glutograph values. Also, the Fourier transforms infrared (FTIR) spectroscopy was used as a nondestructive method to analyze wheat flour composition. On basis of peaks the chemical values of grain were determined. The FTIR spectra was subjected to different spectral pre-treatment in order to improve the prediction of some physicochemical parameters (moisture content, protein content, starch content, gluten content, FN and UCDc) of wheat using partial least squares regression (PLS-R). 1st and 2nd derivates pre-treatments of spectra provided the most suitable model for the prediction of the studied parameters.