Manufacturing routes for thermoplastic composite materials: Influence of processing parameters on mechanical properties and environmental sustainability

Abstract

The recyclable nature of thermoplastic fiber-reinforced composites and their potential for sustainable manufacturing have drawn a lot of interest. However, their high melt viscosity presents challenges in achieving effective fiber impregnation, often limiting mechanical performance. This study presents an optimized framework for manufacturing thermoplastic composites by systematically analyzing the influence of processing parameters on mechanical properties and environmental sustainability. The proposed approach integrates experimental design with a Multi-Term Physics-Informed Neural Network (MTPINN), which incorporates governing physical laws, including heat transfer, rheology, and void evolution, into the predictive modeling process. Composite systems based on polypropylene (PP), polyamide (PA), and polyether ether ketone (PEEK), reinforced with glass, carbon, and basalt fibers, are investigated across multiple manufacturing routes. The results demonstrate that optimized processing conditions significantly enhance mechanical performance while reducing energy consumption by up to 22%, corresponding to a substantial reduction in the carbon footprint of the manufacturing process. The MTPINN model achieves a prediction accuracy of 99.2% with a mean error of 1.4%, outperforming conventional data-driven approaches. Furthermore, thermoplastic composites retain more than 80% of their mechanical properties after multiple recycling cycles, supporting circular economy objectives. This work provides a comprehensive and scalable framework for optimizing thermoplastic composite manufacturing, balancing high performance with environmental sustainability.