Investigation of Heterogeneous Deformation for Discontinuous Fiber Composites Through Combined Experiments and Modeling

Purdue University, West Lafayette IN

Investigators

Abstract

The application of compression / injection molded fiber-reinforced polymer composites in automotive and aerospace applications has the potential to decrease weight while maintaining desirable mechanical performance. These discontinuous fiber reinforced composites offer an alternative to metals for improved efficiency. Even though injection molding has the ease of manufacturing benefit, this method, like all others, produces defects within the material. Uncertainties about how the complex microstructure and defects influence the mechanical response and integrity of discontinuous fiber-reinforced composites have prolonged adoption of these materials for structural applications. Uses of fiber-reinforced composites are subjected to large-scale testing programs, necessary for certification, which lead to excessive time and cost resulting from a trial-and-error approach to manufacturing. Computational models have the ability to cut the time and cost necessary for certification, but first confidence must exist in the model's predictions. In this research, measurements from x-ray tomography during in situ loading provide knowledge of the initial defects, neighboring microstructure, and strain states within the composite materials contributing to material failure. These experiments provide the foundational information and initial states to enable material strength models, in order to predict the strength of these fiber-reinforced composites. Additionally, the project efforts include a multi-objective approach to outreach, in order to educate and mentor third through eighth grade and undergraduate students and broadly disseminate the tools created during this project. In this research, the mechanical behavior of composites will be studied through x-ray tomography scans during in situ loading, followed by state-of-the-art image analysis to reconstruct, identify, and track microstructural features within the composite. Through digital volume correlation, the heterogeneous deformation of individual events such as fiber rotation, pull-out, and breakage will be quantified with respect to the local microstructure during deformation in 3D bulk material. The experiments will provide the cornerstone for strength modeling by quantifying the irreversible deformation with respect to microstructural features. Statistical modeling based on modern probability theory, including rare event modeling using large deviations, will be used to characterize the composites overall strength with respect to the multivariate spatially varying distributions. A forward modeling formulation will be used to design fiber-reinforced thermoplastics with superior strength properties. These composites will be fabricated and characterized via a similar in situ loading, x-ray tomography experiment to validate the predictive nature of the model.