Determinação do teor de proteína em concentrado proteico de soro de leite por FT-MIR-ATR combinada com calibração multivariada

Abstract

The search for better physical conditions and a healthy aging is a trend nowadays and, in many cases, the diet can be enriched with food supplements, especially with products containing whey protein. Whey is the most known supplement in the world, with several brands available in a constant growing market, which favors the lack of control and supervision. The food industry has put effort into the quality control, authentication and characterization of raw materials and final products throughout the years. Thus, it is necessary to develop tools to provide equivalent reliable data reference methods. The objective of this study was to evaluate the Infrared Spectroscopy with Fourier Transform (FTIR), attenuated Total Reflectance (ATR) along with (PCA and PLS) for a rapid and non-destructive characterization of whey Protein Concentrate (WPC) samples adulterated with (SLP), as well as the simultaneous quantitative protein and the amount in grams of WPC or SLP added to the sample. Substitutions in WPC between 0 to 100% by SLP were analyzed in triplicate by FT-MIR-ATR and revealed numerous bands corresponding to the components of the milk. With increased levels of replacement of SLP, there was a decrease of the Amide I and Amide II bands, revealing a strong correlation with the protein content and indication of fraud in WPC. PCA confirmed intense loads in the amide I and II regions, explaining 71% of the total variance. In 40% of substitution of SLP to the WPC, differentiation was found in the spectrum format around the OH carbohydrate (1080 cm-1), with confirmation by analysis of the second derivative spectra. The protein content of each mixture was determined by Kjeldahl methodology and the correlation between concentrations and spectra was done through calibration by PLS modeling, and resulted in low prediction errors and coefficient of determination of the overall adjustment above 0.98. Therefore, the association of FTIR-ATR spectroscopy with PCA and PLS can be used precisely to predict the amount in grams of SLP substitutions in WPC and the resulting protein content in each mixture.