Near-infrared (NIR) spectroscopy as a fast and effective analytical tool to detect adulteration in grated Parmesan cheese

Milk and its derivatives play an essential role in human nutrition, with grated Parmesan cheese standing out for its wide acceptance, versatility, and growing market demand. However, the increase in production and consumption costs has favored the occurrence of fraudulent practices, especially through the adulteration of the product with starch, a practice aimed at reducing costs and maximizing profits. Such adulterations not only compromise the product’s quality but also harm consumers and pose a challenge to food safety. Traditional methods for detecting this type of fraud are often costly, time-consuming, and destructive. In this context, the present study evaluated the performance of near-infrared (NIR) spectroscopy, in the range of 1350 nm to 2550 nm, as a rapid, efficient, and non-destructive analytical technique for detecting adulteration by starchy products in grated Parmesan cheese. For this purpose, samples from different brands of Parmesan cheese manufacturers were acquired from the local market. The samples were grated in the laboratory and initially adulterated with starch in a controlled manner (0%, 5%, and 10%), allowing the determination of the method’s sensitivity and accuracy, as well as the generation of calibration curves for future analyses. The projections of the samples onto the first two principal components (PC1 and PC2) explained 72.72% and 11.47% of the total variance, respectively, indicating that the PCA model represents the variability of the original data very well. The PC1 loadings explain the most influential spectral regions in the separation, related to the following bands: 1450 nm and 1940 nm, attributed to O–H (water) and N–H (proteins) vibrations, predominant in pure cheese; 2100 – 2300 nm, attributed to combined C–H, C–O, and O–H bands, typical of starch and carbohydrates; 2300 – 2500 nm, attributed to C–H (fat) overtones from the cheese. The PC2 loadings explain secondary variation that may be related to differences in moisture or mixture heterogeneity (lumps, cheese texture). These results demonstrated that there is a difference between the spectra of pure cheese samples and those with different starch concentrations, confirming that NIR spectroscopy associated with the PCA method can be used as a rapid methodology for identifying starch adulteration in cheese. Therefore, this methodology is expected to be validated as an effective alternative for regulatory agencies and industries, promoting greater consumer safety and contributing to the identification of fraud in grated Parmesan cheeses available on the market.