Abstract
Milk is an important food component in the human diet and is a target for fraud, including many unsafe practices. For example, the unscrupulous adulteration of soymilk into bovine and goat milk or of bovine milk into goat milk in order to gain profit without declaration is a health risk, as the adulterant source and sanitary history are unknown. A robust and fit-for-purpose technique is required to enforce market surveillance and hence protect consumer health. Nuclear magnetic resonance (NMR) is a powerful technique for characterization of food products based on measuring the profile of metabolites. In this study, 1D NMR in conjunction with multivariate chemometrics as well as 2D NMR was applied to differentiate milk types and to identify milk adulteration.
Ten metabolites were found which differed among milk types, hence providing characteristic markers for identifying the milk. These metabolites were used to establish mathematical models for milk type differentiation. The limit of quantification (LOQ) of adulteration was 2% (v/v) for soymilk in bovine milk, 2% (v/v) for soymilk in goat milk and 5% (v/v) for bovine milk in goat milk, with relative standard deviation (RSD) less than 10%, which can meet the needs of daily inspection.
The NMR method described here is effective for milk authenticity identification, and the study demonstrates that the NMR-based milk metabolite analysis approach provides a means of detecting adulteration at expected levels and can be used for dairy quality monitoring. © 2016 Society of Chemical Industry.