Detection and Identification of Starch and Flour Adulteration by Digital Colorimetry and Fourier-Transform Near-IR Spectroscopy

Abstract

A colorimetric device is proposed for identifying and detecting the adulteration of various types of starch and flour by diffuse reflection of UV and IR radiation from LEDs. The color characteristics of the samples (RGB channel values) were determined using cameras on OnePlus 10 Pro and iPhone 14 smartphones with installed applications PhotoMetrix PRO, ColorGrab, and RGBer. Near-infrared spectra (4000–10 000 cm^–1) were recorded on a Fourier-transform infrared spectrometer. Specialized software packages, including TQ Analyst 9, The Unscrambler X, and XLSTAT, processed the dataset of colorimetric and spectral characteristics. The identification features included clustering patterns for different types of starch and flour in principal component analysis and hierarchical cluster analysis. Optimal wavelengths for determining the quality of adulteration of the tested samples were identified: for starch, the simultaneous use of all LEDs (365, 390, 850, and 880 nm); for flour, LEDs with wavelengths of 365 and 390 nm. The qualitative adulteration was assessed using graphs of the dependence of the F1 component on the mass fraction of the added foreign substance in the starch or flour. The effectiveness of the colorimetric method was confirmed by Fourier-transform infrared spectroscopy in the near-infrared region.