A spatially varying cohesive zone model for process-induced interfacial adhesion in thermoplastic overmolding

  • Li, Xiangfeng (University of Wuppertal)
  • Liu, Tiansheng (University of Wuppertal)
  • Tabib, Majd (University of Wuppertal)
  • Neumann, Johannes (University of Wuppertal)
  • Giersberg Sola, Martin (RWTH Aachen University)
  • Valsecchi, Michele (Columbia University)
  • Çelik, Hakan (RWTH Aachen University)
  • Hopmann, Christian (RWTH Aachen University)
  • Kumar, Sanat K (Columbia University)
  • Fish, Jacob (Columbia University)
  • Simon, Jaan-Willem (University of Wuppertal)

Please login to view abstract download link

Thermoplastic overmolding enables the integration of continuous fiber-reinforced tapes with short fiber-reinforced substrates, offering immense potential for lightweight applications. However, the mechanical reliability of these components is frequently compromised by inconsistent interfacial adhesion. Process simulations and experimental findings reveal that bond strength is highly sensitive to local state variables. Consequently, spatial gradients in pressure, cooling rates, and temperature drive substantial interfacial inhomogeneity. This complexity underscores the need to advance beyond empirical trial-and-error optimization. To address these limitations, this work presents an advanced integrated modeling framework developed under the NSF-DFG collaborative initiative. Interfacial strength is governed by the competition between polymer chain interdiffusion and crystallization. While elevated temperatures facilitate the diffusion necessary for healing, the progression of crystallization during cooling acts as a physical barrier, effectively arresting chain mobility and locking the interface structure. Since standard homogeneous models cannot account for this complexity, we implemented a spatially varying cohesive zone model (CZM), in which the traction-separation parameters are mapped directly to the local thermomechanical history and crystallinity evolution. The substrate behavior is captured through a coupled thermal-plastic-viscoelastic framework, while the interface constitutive law is calibrated to reflect the sensitivity of bond formation to local field variables. This methodology allows for the precise prediction of spatiotemporally varying mechanical performance, providing quantitative guidance for process optimization. By defining a causal link between processing controls and the local material state, this framework serves as a robust tool to minimize overdesign and ensure reliability. Consequently, it supports the shift toward resource-efficient manufacturing, reducing material waste and costs in alignment with sustainable industrial goals.