Spectroscopic method (ftir-atr) and chemometric tools to detect cow's milk addition to buffalo's milk

The aim of this work was to use different chemometrics methods to detect cow's milk addition in the buffalo’s milk, through the spectroscopic technique Fourier's Transform Infrared associated with Attenuated Total Reflectance (FTIR-ATR). Preparations were made with increasing levels of cow's milk added to the buffalo's milk, ranging from 0%-100%, totalizing 495 samples. 15 peaks were obtained in the spectral region between 4000 cm-1-600 cm-1 through the FTIR, which were submitted to Principal Component Analysis (PCA) and Artificial Neural Network (ANN). The ANN showed up to be more efficient, displaying classification rates of 100% for the reference samples and for most of the groups with the mixture of cow's and buffalo's milk, total classification rate of 95.5% and good prediction with satisfactory correlation and error. The results displayed the potential of the spectroscopic method FTIR-ATR associated with the chemometrics analysis for the detection of the authenticity of buffalo's milk.

Keywords: adulteration, dairy product, multivariate, spectroscopy, chemometrics.