BACKGROUND
Invert syrup is a common adulterant in honey falsification, thus generating risk for consumers. The most of developed methods are tedious and time‐consuming for manufactures and consumers. However, terahertz spectroscopy provides analytical information in a simple, rapid and environmentally friendly manner. Subsequently, three terahertz spectroscopic characteristics data, the absorption coefficient, the slope of the absorption coefficient spectra and the area of the absorption coefficient spectra, were employed for determination of Acacia honey adulterated with invert syrup.
RESULTS
Single linear regression (SLR) models with different terahertz spectroscopic features were adopted to predict the syrup adulterant proportion in Acacia honey, respectively. The best SLR model was used the area of the absorption coefficient, displaying an adjusted correlation coefficient of 0.985 and a root mean square error of 3.201. Meanwhile, multiple linear regression (MLR) models using successive projections algorithm for variables selection were implemented. MLR model considered the integral area of the absorption coefficient spectra as the inputs yielded the best result with less selected variables, higher Rc2 and Rp2, lower root mean square error of calibration and prediction as well as higher residual predictive deviation.
CONCLUSIONS
The results indicate terahertz spectroscopy combined with the integral area of the absorption coefficient spectra is reliable enough for invert syrup proportion quantification in Acacia honey and is also a rapid and non‐destructive determination method for other honey adulterants.
This article is protected by copyright. All rights reserved.
