NSF-DFG: Integrated Computational Materials Engineering of Thermoplastic Composites

• Applicant: Professor Dr.-Ing. Jaan-Willem Simon
• Subject Area: Computer-Aided Design of Materials and Simulation of Materials Behaviour from Atomic to Microscopic Scale ; Plastics Engineering
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 549612624

All processes for thermoplastic composites (TPC), such as thermoforming or additive manufacturing, can be considered thermal welding, where bonding is developed across interfaces, either between polymers or between fiber and polymer. The central focus of this proposal revolves around a fundamental question: How do polymer morphology, interdiffusion, crystallization, and residual stresses during processing impact interlaminar strength and fracture toughness of carbon fiber reinforced TPCs? Answering this question requires a multi-physics, multi-scale model that couples the processing with interfacial mechanical characteristics in TPCs. The ambitious scope of this project will bridge several significant gaps in the current state-of-the-art. With MD as the cornerstone, we plan to bring to light the intricacies of matrix-matrix and fiber-polymer interfacial molecular structures and corresponding strengths. Along with imaging and experimental characterization, this will enable the prediction of molecular-level interfacial mechanical properties, setting the stage for micro-scale modeling at the next length scale. Existing micro-mechanics based multi-scale modeling techniques are not sufficient for fully coupled multi-physics computation, particularly as applied to TPCs. This project seeks to address this by emphasizing the importance of coupling mechanical and thermal fields across the length scales and by refining models for inhomogeneous crystallinity evolution that account for nucleation processes in the vicinity of fibers at the micro-scale. Lastly, at the macro-scale, this project will utilize cohesive zone-based interface elements to capture both bonding and debonding processes in laminated structures and the underlying TPC materials. In parallel to the modeling efforts, we will deploy avant-garde experimental tools to characterize TPC structure and mechanical properties at different scales. Validation will be also rigorously pursued through diverse but targeted thermo-mechanical tests. Optimizing TPC performance requires identifying and correlating process parameters to interfacial microstructures, emphasizing that microstructure evolution during processing is critical to achieving peak material performance. To address computational challenges in multiscale optimization, machine learning (ML) surrogates will be utilized, leveraging physics-based recurrent neural network models and genetic algorithms to efficiently map processing conditions to material behavior and predict optimal processing conditions for specific material combinations.

• DFG Programme: Research Grants
• International Connection: USA
• Cooperation Partners: Professor Gregory M. Odegard; Professor Mehran Tehrani, Ph.D.