Rapid detection of acid neutralizers adulteration in raw milk using FGC E-nose and chemometrics
Huaixiang Tian,Bin Chen,Xinman Lou,Haiyan Yu,Haibin Yuan,Juan Huang,Chen Chen
DOI: https://doi.org/10.1007/s11694-022-01403-4
2022-04-27
Abstract:A method for detecting raw milk adulteration with acid neutralizers was developed using a flash gas chromatography electronic nose (FGC E-nose) combined with chemometric techniques. The present study was performed using deteriorated raw milk samples neutralized with sodium hydroxide (NaOH), sodium thiocyanate (NaSCN), sodium carbonate (Na2CO3), and sodium bicarbonate (NaHCO3) to simulate adulteration. Preliminary sample analysis showed the principal component analysis (PCA) could not discriminate between adulterated and fresh raw milk samples completely. More powerful chemometric technologies, such as partial least squares discriminant analysis (PLS-DA), random forest (RF), and multi-layer perceptron (MLP), presented better discrimination performance. The accuracy of PLS-DA, RF and MLP were 98.4%, 100%, and 94.7% respectively. Then the quantification of NaHCO3 in adulterated raw milk has also been conducted by FGC E-nose combined with RF and MLP. RF prediction model showed good performance when mtry was 19 and ntree was 500 (default). The error of MLP prediction model remained constant when the number of iterations was 200. The coefficient of the determination (R2) and root mean square error (RMSE) values for predicting the level of adulteration (NaHCO3) were 0.9172 and 0.0913 in the MLP model, respectively, and 0.9587 and 0.0079, respectively, in the RF model. As a result, RF model showed the best performance in detecting raw milk adulteration with acid neutralizers. Altogether, the FGC E-nose is a useful tool that, when combined with chemometric methods, enables a rapid and sensitive technique to determine the authenticity of raw milk.Graphical abstract
food science & technology