Predictive analysis of cocoa procyanidins using near-infrared spectroscopy techniques

A near-infrared reflectance (NIR) spectroscopy technique for the prediction of procyanidins in cocon beans (Theobrom cacao) has been developed. A select group of cocon liquors from different origins used for the manufacture of chocolate were analyted to determine quantitative levels of procyanidin oligomers (monomer to decamer) using normal-phase high-performance liquid chromatography. The calibration set included seasonal, environmental, and fermentation variations to cover the concentration range of procyanidins found at naturally occurring levels. The sample set was analyzed on FOSS NIR System II 6500 spectrophotometers and parttial least-squares algorithms developed, Very good calibration statistics were obtained for the prediction of the total procyanidin oligomers (r 2 = 0.983) with standard deviation/standard error of cross validation ratio (SD/SECV) of 5.68. A giobal version of this callbration, involving 20 Foss instruments produced a calibration r2 = 0,98 with a SD/ SECV = 6.20. Procyanidins have attracted increasing attention because of the rapidly growing body of evidence associating these compounds with a wide range of potential health benefits. A rapid method for the analysis of procyanidins in cocon liquors would be beneficial in quality-control environments and would provide cost benefits to manufacturing operations.