Investigations on the yield behavior of metal foam under multiaxial loadings by an imaged-based mesoscopic model

Metal foams are usually subjected to multiaxial loadings in applications. It is significant to investigate the yield behavior and mechanical mechanism of metal foam under multiaxial loadings. In this study, Firstly, the geometry of the metal foam sample was obtained via 3D reconstruction using micro-computer tomography images. Then, an imaged-based mesoscopic model numerical model is established based on Finite Element Method (FEM). From the uniaxial compression, it is noted that the initial crush randomly takes place at the weakest cells firstly, resulting in the plastic Poisson's ratio no longer to be constant in the metal foam. Therefore, the plastic Poisson's ratio at where localized deformation happens should be used in predicting the yield behavior metal foam. In this study, the plastic Poisson's ratio is determined by the imaged-based mesoscopic model. Two kinds of virtual experiments, i.e. triaxial compression based on a cubic model and biaxial compression based on a butterfly-shape model, were carried out to investigate the yield behavior of metal foam. Similarly, the foam cells crush randomly at the weakest region under triaxial compression. Meanwhile, the central section of the butterfly-shape model firstly collapses under biaxial compression. Based on the virtual experiments, the numerical yield points of metal foam under different stress state are obtained, and the yield surface is plotted in the mean-effective stress space. It is known that the Poisson's ratio has significant effect on the Miller's and Deshpande & Fleck's yield criterion. Based on the local plastic Poisson's ratio, more accurate prediction from Miller's yield criterion is obtained when compare with the numerical yield surface.