Origin verification of imported infant formula and fresh milk into China using stable isotope and elemental chemometrics

China is the largest importer of infant formula. However, monitoring the authenticity of infant formula has long been a problem in China. The origin of infant formula is more difficult to trace than fresh milk because it contains many foreign additives that cannot be correlated to a specific country. In this study, stable isotope ratios (δ15N, δ13C, δ34S, δ2H and δ18O) of bulk infant formula milk powder and fresh milk from 6 origins (Netherlands, Switzerland, France, Denmark, New Zealand and China) were compared to establish geographical origin isotopic characteristics. N, C and S elemental contents were also investigated to improve origin verification. Artificial Neural Network (ANN) and Linear Discriminant Analysis (LDA) were used to establish origin verification models of imported infant formula and fresh milk. Overall, results showed that infant formula had higher δ18O, δ2H, δ15N and δ34S values, and lower δ13C values than fresh milk. δ18O and δ2H values of infant formula varied more than in fresh milk, as the additives are not necessarily sourced from the same country, while δ15N, δ13C and δ34S values were more negative, most likely due to the presence of numerous additives probably from non-dairy sources such as fats, sugars, amino acids and proteins. This study showed a combination of isotopes and elemental contents could completely discriminate the origin of infant formula imported into China, with a discrimination accuracy of 100% using ANN.