Prediction of Pasting and Thermal Properties of Mixed Hom-Mali and Glutinous Rice Flours Using Near Infrared Spectroscopy

Abstract

Starch is a major component of rice grain and plays an important role in the grain quality. The pasting properties and thermal properties are important physical qualities of rice flour. Direct measurement is time-consuming and expensive. Therefore, a rapid predictive method based on near-infrared (NIR) spectroscopy is necessary that can be applied to measure these quality parameters. Glutinous flour (RD6, Sakonnakorn and Niew Ubon varieties) and Hom-Mali rice flour (105 variety) were selected for this research. The flour samples were dry milled into a 120-mesh sieve using a turbo mill. In total, 92 mixed flour samples were prepared by blending Hom-Mali rice flour with each of the glutinous rice flour types in various ratios of which 61 mixed flour samples were used to develop calibration models for measuring pasting and thermal properties based on the NIR spectra of the mixed flour samples. The models were validated by a separate prediction set consisting of 31 mixed flour samples. The models for determining the pasting properties using the pasting temperature, set back, final viscosity, breakdown and trough yield all had a similar level of accuracy with R2 values of 0.99, 0.97, 0.95, 0.80 and 0.80, respectively. Furthermore, the thermal properties indicated that the models for onset temperature, peak temperature, and enthalpy produced moderate correlation values with R² values of 0.82, 0.75 and 0.73, respectively, whereas the conclusion temperature could not be predicted by NIR spectroscopy (R² = 0.06).