ECCOMAS 2024

Predictive Mechanical Modelling for Polybenzoxazine Nanocomposite Blends for Space Applications

  • Kong, Kyungil (University of Bristol)
  • Hamerton, Ian (University of Bristol)

Please login to view abstract download link

This study employs computational simulations to model a novel resin blend composed of polybenzoxazine (PBZ) and a polyhedral oligomeric silsesquioxane (POSS) reagent, tailored for space applications. The investigation focuses on assessing the mechanical responses of these blends by analysing constituent micro-stresses through the implementation of periodic boundary conditions and continuum-based analytical micromechanical models, including the Mori-Tanaka and Halpin-Tsai models. The study explores the effective elastic responses of various catalysed PBZ-POSS blends using periodic boundary conditions, a representative loading history for the representative volume element (RVE), and material constraints within an RVE window. Diverse RVE models illustrating POSS particle distribution, such as periodic, non-periodic, and overlap inclusions, are developed and compared to experimental data. The Monte Carlo model generation algorithm, referencing the Hard-core model, is extended to consider inclusion boundary conditions adjacent to the RVE window and their periodicity level. Adjustments are made to the method of determining the random geometry of POSS in the RVE window to minimize disparities between the actual macroscopic domain and the virtual RVE, accounting for variations in the properties of the polymer matrix and nanoparticle inclusion during manufacturing. A finite element analysis (FEA) methodology, incorporating Monte Carlo geometrical modelling, is applied to three distinct RVE models: periodic, non-periodic, and overlap inclusion. Results from the FEA simulations of the elastic modulus for the RVE models reveal lower values compared to continuum-based analytical micromechanical models. This discrepancy is attributed to the intrinsic processing characteristics of the PBZ resin, known for producing fewer voids and lower cure shrinkage than similar thermosetting polymers. Notably, the overlap inclusion RVE model in the FEA simulation, when applying periodic boundary conditions, demonstrates particularly good agreement with experimental results, especially when considering the presence of aggregated POSS nanoparticles.