Modelling open cell foam structures based on 3D image data

The macroscopic properties of open cell foams are highly influenced by the foam's microstructure. Recent research has shown that the complex functional behaviour of foams can be studied by an exact characterisation of their microstructure followed by numeric simulations. The key element of such studies is the microstructure model of the foam. Models from stochastic geometry in conjunction with X-ray computed tomography have turned out to be powerful tools in this context. We focus on the realistic reproduction of the foam's microstructure in a model. Using various geometric characteristics of the cell and strut system, we propose a two step model fitting procedure: First, a Laguerre tessellation model is fitted to the cell system of the foam. In the second step, we estimate the local thickness of the foam's edge system and fit a polynomial model to describe the locally varying thickness distribution. The edge system of the tessellation is then adaptively dilated using the polynomial model as size map to form the struts of the foam. The elastic properties of the resulting microstructure model are evaluated using an iterative FFT-based algorithm. By comparing the effective material properties of the real foam and its model, we found the response of the model in very good agreement with the real foam. These findings support the supposition that exact microstructure models play a vital role in the study of the functional behaviour of foams.