Validación de un modelo de calibración para predecir contenido de taninos empleando FTIR

Abstract

Sorghum (Sorghum bicolor) is the fifth most important cereal in the world, after wheat, rice, corn and barley. It is an excellent source of energy for animal and human food. An advantage of sorghum is its high content of secondary metabolites, such as phenolic acids, flavonoids and condensed tannins, which act as antioxidants, colesterol reducers and antidiabetic agents. Tannins, natural antioxidants, inhibit the enzyme amylase, favoring the development offunctional foods for diabetics. Determining their concentration quickly and accurately is important to detect changes the raw material and the process. Among the methods to determine condensed tannins is the colorimetric method of vanillin, which is selective and sensitive to titrate flavanols and proanthocyanidins. However, this method is destructive, expensive, polluting and time-consuming, so it is not suitable for real-time measurement nor is it environmentally friendly. The aim of this study was to develop a calibration model to predict the content of tannins, expressed as catechins and gallic acid, in samples of vegetal and synthetic tannin extracts in a non-destructive and rapid way, using infrared (IR) spectroscopy with spectra preprocessing and the Partial Least Squares (PLS) method. FTIR spectroscopy is an effective, reliable and sensitive method to characterize and discriminate between tannins from different botanical sources, requiring small samples and a short analysis time. The proposed model allows to identify and quantify tannins in solid samples of natural and synthetic tannin extracts.