Detection and Quantification of Adulteration in Turmeric by Spectroscopy Coupled With Chemometrics Publisher



Tamiji Z^{1, 2, 3} ; Habibi Z³ ; Pourjabbar Z³ ; Khoshayand MR^{3, 4} ; Sadeghi N^{3, 5} ; Hajimahmoodi M^{1, 2, 3} Show Affiliations
Source: Journal fur Verbraucherschutz und Lebensmittelsicherheit Published:2022

Abstract

One of the main food safety concerns is the extensive fraud in food production that necessitates the development of rapid and accurate diagnostic detection methods. Recently, the near-infrared spectroscopy has been used as a non-destructive method in combination with multivariate analysis methods to evaluate the quality of foods. Spices are very useful in our daily lives. They improve the flavor, color and nutritional value of the foods. Turmeric is a member of the ginger family and its main ingredient is a polyphenol, which is called curcumin and is widely used in the pharmaceutical industry. For commercial profit, this spice can be mixed with a variety of cheap and harmful additives. In this research, the ability of near-infrared spectroscopy as a cost effective, fast and non-destructive method in combination with chemometrics was investigated to monitor and control the adulteration of turmeric with wheat flour, pistachio hull waste and dry bread. For this purpose, the near-infrared spectrum from 54 adulterated mixtures and 102 samples purchased from Tehran supermarkets was measured in the range of 12,000–4000 cm−1. The spectra were pre-processed and principal component analysis (PCA) and partial least squares regression (PLSR) were used for qualitative and quantitative analysis, respectively. PCA in the first dataset, including all three sets of turmeric spectrum, explains 99% of the variance of spectral data. These results showed a very good discrimination in all three types of fraud based on different concentrations. According to the PCA model, most of the samples in the second dataset were either turmeric samples adulterated with dry bread powder or pure turmeric samples. Therefore, the PLSR model was used to quantify the turmeric samples adulterated with dry bread powder. In PLSR, the root mean square error and determination coefficient (R2) of calibration for validation and prediction were 0.265–0.306 and 97.9–98.6%, respectively. Therefore, this method is highly useful to detect adulteration in turmeric. © 2022, Bundesamt fur Verbraucherschutz und Lebensmittelsicherheit (BVL).