Evaluation of mechanical properties of starch-gelatin-sargassum polymers based on neural networks

In biodegradable polymer research, starch, gelatin, and Sargassum algae waste are gaining attention as a potential solution to mitigate the environmental impact of conventional plastics. This study, conducted at Universidad Tecnológica de Tula-Tepeji, evaluated the mechanical properties of a polymer composed of these components using a CT3 texture analyzer, with neural network models employed to predict such properties. Polymers were prepared with varying Sargassum concentrations, and their mechanical response under tensile loading was analyzed to estimate Young's modulus. The casting process effectively integrated Sargassum into the polymer, offering a potential eco-friendly alternative. Neural network predictions of the material's behavior proved accurate, demonstrating the usefulness of this approach in materials research. These findings support incorporating additional data and Sargassum variations in future work to further enhance predictive capability.