Finite Element Calculation of Sintering Deformation Using Limited Experimental Data

Abstract

Predicting the sintering deformation of ceramic powder compacts is very important to manufactures of ceramic components. In theory the finite element method can be used to calculate the sintering deformation. In practice the method has not been used very often by the industry for a very simple reason – it is more expensive to obtain the material data required in a finite element analysis than it is to develop a product through trial and error. A finite element analysis of sintering deformation requires the shear and bulk viscosities of the powder compact. The viscosities are strong functions of temperature, density and grain-size, all of which change dramatically in the sintering process. There are two ways to establish the dependence of the viscosities on the microstructure: (a) by using a material model and (b) by fitting the experimental data. The materials models differ from each other widely and it can be difficult to know which one to use. On the other hand, obtaining fitting functions is very time consuming. To overcome this difficulty, Pan and his co-workers developed a reduced finite element method (Kiani et. al. J. Eur. Ceram. Soc., 2007, 27, 2377-2383; Huang and Pan, J. Eur. Ceram. Soc., available on line, 2008) which does not require the viscosities; rather the densification data (density as function of time) is used to predict sintering deformation. This paper provides an overview of the reduced method and a series of case studies.