The project focuses on the deformation and failure of lightweight construction materials under crash loads. We contribute our simulation expertise to the modeling and characterization of material behavior. In the picture on the left: real component. Right: fiber orientation distribution in the component from the injection molding simulation.

Characterization and Modelling of Long Fiber Reinforced Thermoplastics

AiF Project (Industrial Research Associations)

Fiber reinforced plastics are lighter, stronger and more resilient than non-reinforced plastics and therefore, they meanwhile have become a preferred material class within the automotive industry. As far as safety-relevant components in the vehicle are concerned, it is particularly important to predict failure and damage. In this project, we support automotive companies with regard to the analysis and prediction of these mechanisms.

Fiber reinforced plastics (FRP) are often used in crash-relevant components due to their stiffness and strength properties. In crash applications, the industry prefers to use very long fibers, as the longer the fiber length, the higher the solidity is. The fact that they are easy to process, e.g. by injection molding, is another major advantage of long fiber reinforced thermoplastics (LFT). However, modelling their material behavior is a challenging task, as, from a macroscopic point of view, the fibers and the matrix should not be seen individually but as a complete system. Their properties are very inhomogeneous, and non-linear effects must also be considered on the microscale. In this research project, we investigate the damage and failure mechanisms of LFT in the automotive industry.


Digital Twins of Composite Materials

Together with our partners from Fraunhofer IWM, we analyse damage and failure mechanisms of LFT on a realistic geometry both experimentally as well as using numerical simulations. With the FLUID solver, we first obtain a prediction of the orientation of the fibers in the whole component. Using CT images of several samples extracted at characteristic locations of the component, we validate that the prediction of the fiber orientation is indeed accurate. After having completely described the geometrical properties of the material, we are able to create digital twins of the composite.

Stress concentration in one of the sample
© Fraunhofer ITWM
Stress concentration in one of the sample.
Cutout from a digital twin of the composite
© Fraunhofer ITWM
Cutout from a digital twin of the composite.

By means of a dynamic mechanical analysis (DMA) on the pure polymer matrix material, we determine the nonlinear material parameters of the composite material, taking into account effects such as plastification, strain rate dependency and damage. Thus, we are able to show that this method allows us to predict the material behavior of the composite material with high accuracy by using our software tool FeelMath. At the same time, the experimental complexity needed for the material characterization remains low.


Virtual Experiments Save Costs and Keep the Experimental Effort at a Minimum

The micromechanical model and the fiber orientation from the injection molding simulation serve several functions. We can perform virtual experiments for the calibration of macroscopic standard models and reduce the experimental costs of material characterization. With the digital twins of the composite material, we investigate the influence of geometric and physical parameters – such as fiber length, fiber orientation or strain rate.