ABSTRACT
Zinc citrate (ZC) has been widely used in food as an important nutritional supplement. Accurate detection of ZC in food is important for health and safety. In this study, THz time-domain spectroscopy (THz-TDS) is used to quantitatively detect ZC in flour and milk powder mixtures. In our research, 15 different contents of ZC in flour and milk powder mixtures were prepared and measured by THz-TDS. A partial least squares (PLS) model was established based on the quantitative analysis of the absorption coefficient data of these two mixtures at 0.5-3.0 THz. The R2 and rms error (RMSE) given by the PLS model prediction were, respectively, 0.999 and 0.14% ZC in flour and 0.999 and 0.20% ZC in milk mixtures, indicating the predictions of the PLS model are in excellent agreement with the experimental measurements. The results show that combining THz-TDS with the PLS model can be used for accurate, quantitative analyses of ZC in food mixtures.