Improving green thickener on the mechanical and thermal performance of composite-based plant hard capsule film with multi machine learning models

Alginate-based capsules present significant potential in terms of commercialization and market demand as an alternative to gelatin. Nevertheless, their commercial production remains limited due to elevated costs, the presence of chemically toxic thickeners, and relatively inferior thermal and mechanical properties, which are critical for future applications in drug delivery systems. Thus, in this study, we utilized an artificial neural network (ANN) model to predict how to improve the formulation of composite alginate-based capsule films incorporating various types of green thickeners (potassium chloride, lauric acid, and calcium alginate). The experimental results demonstrate that the film with potassium chloride exhibits superior mechanical resistance to deformation and fracture, with a tensile strength (TS) of 3.76±0.11 MPa. In addition, the film displays improved thermal stability and wear resistance, with a thermal mass loss of 30.9% and a glass transition temperature of 125.3 °C. Furthermore, FTIR and SEM measurements are in line with the ANN model predictions based on experimental results, verifying the model's effectiveness in determining the optimal composite alginate-based capsule film formulation. Overall, the laboratory-scale exploration in this study provides preliminary evidence for developing more feasible and environmentally friendly alginate-based capsule film formulations, along with fundamental guidance for subsequent research toward industrial-scale production.