果冻传媒

To stop global warming, reducing the carbon emissions of the automotive industry is very important. One way to do this is to reduce the weight of cars and trucks. However, this should not come at the expense of crash safety. For this reason, there is great interest in improving the properties of so-called advanced high-strength steel grades. Nowadays, the most widely used family of high-strength steel is dual-phase steel. As the name suggests, their microstructure consists of two phases, namely, ferrite, which is soft but ductile, and martensite, which is hard but brittle. By combining both phases, steel is made that has high strength and is easy to form. However, having multiple phases in a microstructure also complicates the deformation mechanisms. Occasionally, the material fails much earlier than expected, which can have catastrophic consequences.

This project was aimed at improving our understanding of these deformation mechanisms in dual-phase steel microstructures. In the project, Job Wijnen worked together with another PhD student, Tijmen Vermeij. Where Tijmen performed experiments on microstructures, I conducted simulations on those same microstructures. The strength of combining experiments with simulations works two ways. First, experimental data can be used to identify and develop accurate models that can predict the observed material behavior. Next, the simulations can be used to gain more insights into experiments and explain the observed behavior.

Insights into material behavior

In conventional mechanical experiments on steel microstructures, the comparison with simulations is severely limited because we can only see the microstructure and its deformation at the surface of a sample; everything underneath the surface remains unknown. To be able to accurately compare simulations and experiments, we developed a coupled experimental-numerical framework. A key aspect of the framework is the samples, which are locally ultra-thin (around a micrometer). As a result, the microstructure and deformations observed from the front and back sides of the samples are very similar. We can use this data to reconstruct the full 3D geometry and microstructure of the samples, which can then be used in simulations. The developed framework allows for very detailed comparisons between simulations and experiments, which can be used to gain insights into the material behavior.

Predict of deformation patterns

Besides the experimental-numerical framework, Wijnen's research focused on developing numerical models that can correctly predict the deformation patterns at the length scales we were considering. When we look with the naked eye at a material that deforms, it seems as if a large area of the material deforms more or less an equal amount, which we call homogeneous. However, if we zoom in on a metal microstructure, the deformation actually consists of many small localized deformation bands, which we call heterogeneous. During hisPhD, he developed a numerical model that takes into account random fluctuations in local properties. This resulted in the same deformation pattern as observed experimentally.

This is a project within the UNFAIL project, in cooperation with Tata Steel and M2i.

Title of PhD thesis: . Supervisors: Marc Geers, Ron Peerlings, and Johan Hoefnagels.