The wide application of advanced structural ceramics is limited by the problems related to structural reliability that are common to all brittle materials. Ceramics do not yield before fracture and therefore fracture strength is often the basis for the prediction of a design limit. Unfortunately, the strength of ceramics is found to be very sensitive to processing variables in ways that are difficult to predict. In recent years much effort has gone into toughening ceramics using mechanisms that result in the need for complex microstructures. Such work is usually isolated from studies in process control resulting in relatively few systematic and quantitative evaluations of the relationships between processing conditions and fracture behavior. This project uses a closely integrated interdisciplinary approach to determine the relationships between ceramics processing variables and fracture behavior of alumina. For this first year effort, correlations will be made between the distribution of microstructural heterogeneities and the structure of both polished and ground surfaces along with a limited number of fracture strength measurements. Ceramics processing, microstructural evolution, hardness distributions and fracture behavior will be linked through quantitative neural network based process models at the microstructural level combined with image analysis. The end result will be an improved quantitative understanding of the effects of processing variables on the fracture of advanced ceramics -- an understanding based on sound statistical analysis of microstructure evolution, and one which can be used to properly manufacture these ceramics.
This project will be a timely application of image analysis and computational intelligence in the processing of advanced materials and will contribute to the emerging paradigm of studying materials as complex systems. Furthermore, the successful demonstration of process modeling of high strength ceramics will motivate a shift of the emphasis in ceramics processing to a quantitative understanding of microstructural control. Finally, the educational outcomes will provide resources to teach materials processing as an interdisciplinary subject to university students, and engage young middle school students in the realms of material science and image processing.
