Data fusion of UPLC data, NIR spectra and physicochemical parameters with chemometrics as an alternative to evaluating kombucha fermentation

Abstract

Kombucha consumption has become popular due to health benefits and sensorial properties. Different analytical techniques were associated with unsupervised chemometrics to evaluate kombucha fermentation using a low-level data fusion strategy. Kombucha was prepared from green and black teas (Camellia sinensis). The samples were analyzed by ultra-performance liquid chromatography (UPLC), portable near-infrared spectroscopy (NIR), and physicochemical analyses (ethanol, pH, total reducing sugars, and total titratable acidity). Also, the bioactive phenolic compounds were evaluated in the fermented samples. The merged results were analyzed by principal component analysis (PCA). The PCA was applied to discriminate substrate for kombucha production and the fermentation time. Although the fermentation behavior was similar for both substrates, the kombucha showed quantitative differences in physicochemical parameters. They also showed differences in bioactive compounds for each kombucha. Data fusion strategy was more effective to differentiate the two types of kombucha than single analysis. The portable NIR can be a reliable and robust analysis for fermentation monitoring. Moreover, the results can be used as a base to further studies and mainly to help in which substrate composition is better to be produced on an industrial scale.