Development of simple identification models for four main catechins and caffeine in fresh green tea leaf based on visible and near-infrared spectroscopy

Abstract Catechin polyphenols and caffeine play an important role in tea quality. This study analyzed the visible and near-infrared (Vis-NIR) spectral variation and concentration differences of catechin and caffeine of fresh tea leaves (Camellia sinensis L.) in three varieties and six leaf positions, and a universal spectral model for quickly and accurately measurement of the catechin and caffeine content in the various varieties and leaf positions was established. It was found that all the catechins and caffeine were significantly influenced by variety and leaf position. While, the Vis/NIR spectrum as an indication of internal biochemical substances was successfully used to distinguish different varieties and leaf positions with the discrimination accuracy rates of 98.15% and 100%, respectively. The quantitative relationship between the chemical components and the data obtained by Vis-NIR spectroscopy was established based on multivariate regression analysis such as partial least squares (PLS) and multiple linear regression (MLR). Furthermore, competitive adaptive reweighted sampling (CARS) and successive projections algorithm (SPA) were used to select the characteristic wavelengths for the development of simple models. Results showed that the quantitative determination models obtained good performance with the determination coefficients (R2) of 0.949, 0.893, 0.968, 0.931 and 0.917 for epigallocatechin gallate (EGCG), epicatechin gallate (ECG), epigallocatechin (EGC), epicatechin (EC) and caffeine (CAF), respectively. Such high detection accuracy shows that the spectral detection model has a strong applicability for both varieties and leaf positions. The overall results of the study revealed the potential use of Vis-NIR spectroscopy as a rapid, simple and non-destructive method for the determination of four main catechins and caffeine in fresh tea leaves, and it will play a great role in the real-time detection of tea physiological information in tea garden.