Comparative study on the optimization and characterization of soybean aqueous extract based composite film using response surface methodology (RSM) and artificial neural network (ANN)

Abstract

Soybean aqueous extract (SAE) is a promising raw material for edible film preparation. In this study, SAE-based composite edible film was developed by incorporating beeswax (0.4–1.2%), clove essential oil (0.5–1.5%), and span-20 (0.5–1.5%), and their effect on the physico-mechanical and barrier properties was evaluated. Response surface methodology (RSM) using central composite rotatable design and supervised artificial neural network (ANN) models were used to predict the effect of the independent variables on responses like tensile strength, elongation at break, water vapor permeability, moisture content, water-solubility, and optical parameters. All the independent variables had a significant role (p < 0.05) on the responses, and the experimental data were better predicted by ANN models compared to RSM with higher R2, lower RMSE, MAE and χ2 values. The RSM optimized value of beeswax, clove essential oil, and span-20 was 1.2%, 0.91%, and 0.73%, respectively. Among the independent variables, beeswax and clove essential oil were the most influential in mechanical and water barrier properties, while the level of span 20 affected the color and solubility of the films. A semi-crystalline nature of the composite film was evident from the SEM and XRD analysis. The characterization of the developed film confirmed its utility in food packaging applications.