Detection of encapsulant addition in butterfly-pea (Clitoria ternatea L.) extract powder using visible-near-infrared spectroscopy and chemometrics analysis

Abstract

Butterfly-pea (Clitoria ternatea L.) extract powder is a functional product with numerous benefits obtained by extraction followed by the drying process. During drying, encapsulations can be added to protect the color and antioxidants of the samples. Using visible-near-infrared (Vis-NIR) spectroscopy, this research aimed to detect maltodextrin and soybean protein isolate (SPI) added as encapsulants to butterfly-pea extract powder. Butterfly-pea extract powder were added with 10, 20, 30, 40, and 50% concentrations of maltodextrin and SPI. Spectral data were acquired using a Vis-NIR fiber optic spectrometer at 350-1,000 nm. The chemometric methods used were principal component analysis (PCA), PCA-discriminant analysis (PCA-DA), partial least square regression (PLSR), and partial least square discriminant analysis (PLS-DA). The results showed that PCA can discriminate pure and maltodextrin-and SPI-Added samples using low principal components. PCA-DA determined the accuracy levels of 88% for maltodextrin and 94.67% for SPIs. The PLSR models predicted the addition of maltodextrin with the following variables: coefficient of determination of calibration (R c2), 0.98; coefficient of determination of prediction (R p2), 0.98; root mean square error of calibration (RMSEC), 2.1%; and root mean square error of prediction (RMSEP), 4.02%. The values for the addition of SPI were R c2 of 0.97, R p2 of 0.97, RMSEC of 2.72%, and RMSEP of 2.83%. The PLS-DA models resulted in an accuracy of 98 and 91% for the identification of maltodextrin and SPI, respectively. In conclusion, this research showed the potency of Vis-NIR spectroscopy combined with a proper chemometric analysis to detect additives in butterfly-pea extract powders.