Near Infrared Spectroscopy as an efficient tool for the Qualitative and Quantitative Determination of Sugar Adulteration in Milk

Abstract

The present work focuses on detecting the presence of sugar as an adulterant in milk using Near Infrared Spectroscopy. A chemometric model was formulated to evaluate the presence of sugar content in milk, qualitatively as well as quantitatively using multivariate analysis. Total 24 samples were prepared using three different varieties of milk, out of which three samples were pure and the rest were having sugar present in them. Those 21 milk samples were adulterated with sugar at seven different levels: 0.2%, 0.4%, 0.6%, 0.8%, 1.6%, 3.2%, 0.8% and 6.4% of sugar respectively for each kind of milk. The data collected from NIRS instrument was analyzed using chemometric software (CAMO Unscrambler version X 10.3). The Principal Component Analysis was run on the sample set to know the relation between the different samples on the basis of the Near Infrared spectral data. It was observed that the PCA score plot could classify the samples in three different groups on the basis of their adulteration: low, medium and high adulteration. Partial least square (PLS) regression analysis was used to develop a statistical model to predict the percentage of sugar in the adulterated milk samples by selecting vital wavelengths. It was noticed that the regression model revealed quite good results for the prediction of sugar adulterated milk samples with the coefficient of correlation higher than 0.9 and the root means square error of validation (RMSEV) was 0.04. Thus, it was concluded that NIR spectroscopy could provide dairy industry a simple, efficient, quick, green and non-destructive technique for detection and quantification of milk adulteration.