Determining three dimensional failure surfaces for porous ceramics from FE simulations and mechanical testing
The failure behavior of porous ceramics is dominated by statistical characteristics of the microstructure. In the computational materials modelling group a powerful numerical tool has been developed to study the influence of pore distributions and clustering on the failure behavior of porous ceramics. For the purpose of validation, the investigations were as yet concentrated mostly on quasi two-dimensional perforated disks under uniaxial loading. These simplifications allow for detailed comparison to uniaxial compression tests for which dedicated specimen geometries have been developed. The goal of the current project is to extend the research to three-dimensional microstructures and general loading cases in order to develop a microstructure based failure surface for porous ceramics. At the core of the current research project lays the further development of the Abaqus and python based representative volume element tool to three dimensional simulations and the validation of the simulations through in-situ mechanical testing in X-ray tomography. While mechanical testing will be restricted to a few dedicated loading cases the computational tool shall be used to apply general multiaxial loading. This multiaxial loading allows determining failure locations in stress space which may be fit to established failure criteria for porous materials. The central challenge will be to connect the observations to microstructural characteristics of the porous ceramics beyond the mere porosity. Simulations shall be done both for computer generated microstructures as for those obtained from three-dimensional non-destructive characterization techniques. For the mechanical tests, the know-how on avoiding edge effects needs to be transferred to fully three-dimensional porous structures. For the current project we seek candidates with a solid background in computational methods and materials mechanics, including mechanical testing methods. Furthermore the candidate must be able to drive the multidisciplinary collaborations with the materials development and non-destructive testing performed in partner groups. A recently established computer cluster in the computational materials modeling group provides the means to perform large scale three-dimensional simulations.
Contact: Hochrainer , Bakir