Comparison of different models for analyzing starch dynamic hydrolysis

• Dynamic hydrolysis degrees of 5 starches were fitted using different models. • The fitting qualities and dynamic hydrolysis parameters were compared. • The single first-order kinetics equation model exhibited the poorest fit. • The log of slope linear equation model presented biphase hydrolysis process. • The combination of parallel and sequential first-order kinetics equation model had the best fit. Dynamic hydrolysis is an important property of starch, and hydrolysis parameters can provide information on starch qualities and applications in food and nonfood industries. The modified Michaelis-Menten equation (MME), single first-order kinetics equation (SKE), log of slope linear equation (LOSLE), or combination of parallel and sequential first-order kinetics equation (CPSKE) models are usually used to fit the dynamic hydrolysis data. In this study, the hydrolysis profiles of five starches were fitted by the MME, SKE, LOSLE and CPSKE models. The fits of the different models were evaluated using the sum of squares of residuals (SUMSQ), the fitting determination coefficient (R 2 ), and the differences between the experimental and fitted data. When tested on the five starches, CPSKE model exhibited the best fit, LOSLE model had a better fit than did MME model, and SKE model had the poorest fit among them. Although these models had significantly different fitting qualities, the maximum extent of hydrolysis predicted by the different models was significantly positively correlated. The hydrolysis rate coefficient k fitted by the SKE model was significantly positively correlated with the k 1 fitted by the LOSLE and CPSKE models, but had no significant correlation with k 2 during phase 2 as fitted by the LOSLE and CPSKE models. The k 1 and k 2 values fitted by the LOSLE model were significantly positively correlated with the k 1 and k 2 parameters estimated by CPSKE model, respectively. This study could provide useful information for choosing fitting models for analyzing starch hydrolysis profiles.