Predicting porosity content in triaxial porcelain bodies as a function of raw materials contents

This work describes the relationships observed between the porosity of fired ceramic test pieces and the raw materials contents in the unfired mixture. The investigation was carried out using the mixture experiments design approach coupled with response surface methodology, which enables the calculation of statistically significant models for the properties from a limited number of experimental results. Ten formulations of a clay mixture, potash feldspar and quartz sand were processed in the laboratory under fixed conditions, similar to those used on wall and floor tile industrial practice, and characterized. Closed porosity (CP) was estimated from the analysis of back-scattered SEM photomicrographs, open porosity (OP) was calculated as the product of bulk density and water absorption, and total porosity (TP) was calculated from OP and CP values. Characterization results were used to calculate statistically significant and valid regression equations, relating those properties with the raw material contents in the unfired mixture. For the particular raw materials and processing conditions used, the models clearly show how quartz contributes to increasing OP and the crucial role played by feldspar when CP and TP are to be minimized (circa 3 vol.%), and how the clay content can counteract that effect and lead to maximum closed porosity (*14 vol.%).