Three-dimensional grain size distribution and grain structure reconstruction in liquid-phase sintering

The quantification of sintered microstructures is an inherent part of powder metallurgy. Often we need means to explain the effects of process variations with respect to sintered properties. Quantification of grain size in the microstructure is an essential part of that effort. If satisfactory models exist, then the effort might be simple. Along those lines, three dimensional (3-D) grain size distributions can be obtained from two dimensional (2-D) grain section distributions by using a new forward transformation method. The technique is applied to sintered microstructures to verify the procedure, but the method is not limited to a particular material. A computer simulation program is developed. The method can handle grains with different shapes, dihedral angles, and contact numbers. Both the Weibull and normal distribution functions are used to process the data. The results show that dihedral angle and contact number greatly affect the grain size distribution causing the 3-D grain size distribution to become narrower at high solid volume fractions. The gravitational effects on the grain size distribution and the 3-D grain size distribution in the thin film have been studied as special applications of this forward method.; Spatial microstructure reconstruction is an important aspect in the study of liquid phase sintering. Many microstructural properties can be unveiled by this approach. At present, there are no sufficiently well developed techniques to achieve this goal effectively and accurately. In this research, a microstructure is reconstructed by using the data of the 3-D grain size distribution and other features obtained from 2-D sections. Then the program produces serials of 2-D sections which, when compared with the original 2-D section images, can give an improved understanding of the relationship between the 2-D parameters and the 3-D parameters. Furthermore, when biasing the spatial random positions of the grains, gravitational settling, agglomeration, and connection chains can be simulated.