For the production of a ceramic part, the body needs to be created from a powder. The ceramic powder is pressed into shape and subsequently heated. During the heating process the powder particles grow together through diffusion processes and become a single, porous body. This production process is called sintering and technical ceramics as well as traditional ceramics are usually produced in this way. During sintering the dimensions of the body change as the body shrinks. Depending on the geometry and the process parameters, this shrink can be rather in-homogeneous, which makes it difficult to predict the final shape of a part. Today most parts are developed using trial-and-error methods, which goes hand-in-hand with elevated costs.
Over the past, models have been developed that try to predict the shape-change. Most continuum models of sintering have laid the focus on isothermal conditions. But during the heating and cooling phase the body is subject to non-isothermal loading in space as well as in time. The sintering behavior is dependent on the microstructure of the material. As a powder, the material is granular, then grows into a porous structure and then finally into a brittle solid. The microstructure’s influence on the mechanical properties is taken into consideration by a mesoscale model that is solved with the help of a limit analysis technique. These results are used to calibrate the macro-scale continuum model. We want to be able to model the process, beginning from the powder pressing of granular material, over the densification stage up to the final part, as each of the previous steps influences the subsequent ones. To accomplish this we use the Bigoni-Piccolroaz yield function that can undergo large shape changes to account for different types of plastic yielding behavior, and therefore can be used throughout the whole production process. It is employed together with a coupled thermomechanical description of sintering. This way, it is possible to model the complete forming process. We have developed and implemented a thermomechanical model into a Finite Element Method routine that thus extends the classic continuum description to non-isothermal states. The results are compared to computer-tomography tests, to evaluate the predictions for porosity and shape throughout the part.