Quantitative determination of quality control parameters using near infrared spectroscopy and chemometrics in process monitoring of tapioca sweetener production

Tapioca sweetener is an important ingredient in the food industry. Traditional quality control methods for sweeteners are time-consuming and destructive. The objective of this study was to use near infrared (NIR) spectroscopy in combination with chemometrics to create a predictive model for distinguishing between various types of tapioca sweeteners and measuring quality control parameters of the products. NIR spectra were recorded in the range of 12,000-4000 cm(-1) using the transflectance mode. Based on principal component analysis, tapioca sweetener products with different dextrose equivalent (DE) values could be differentiated. In addition, the quality control parameters, such as total soluble solid (TSS), DE, sugar profiles (glucose, maltose, and maltotriose), pH, and SO2, of tapioca sweeteners were evaluated. Partial least squares regression models provided a promising representation, as indicated by residual predictive deviation values higher than 5 for the prediction of TSS content, DE values, and sugar profiles. The predictive models developed were then applied to monitor TSS, DE, and concentrations of sugars in the manufacturing process of tapioca sweeteners to demonstrate their applicability in actual production. The models can be applied for general screening in the laboratory and during industrial process monitoring.

Automated summary

This study successfully developed a predictive model using near infrared spectroscopy and chemometrics to distinguish between different types of tapioca sweeteners and measure quality control parameters. The model demonstrated high applicability in actual production and can be used for general screening in the laboratory and industrial process monitoring.