Qualitative analysis of caffeine in beverages based on Raman spectroscopy

Caffeine is an alkaloidal stimulant prevalently in food additives. The widespread presence of caffeine in a wide range of beverages has led to increased consumption, commonly exceeding safe levels. In this study, about 30 beverages available in the Chinese market, were analyzed using a 785 nm Raman spectrometer. The raw spectral data were preprocessed by noise reduction and baseline correction. The qualitative model was established using Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA), and the classification discrimination was carried out using K-Nearest Neighbor algorithm (KNN) and Support Vector Machine algorithms (SVM). Especially, the two models could identify “Spirte” (without caffeine) from other beverages. It was demonstrated that Raman spectroscopy combined with machine learning can provide non-destructive, fast and efficient qualitative detection of caffeine in liquid beverages.