Quantitative firing transformations of a triaxial ceramic by X-ray diffraction methods

Abstract

The firing transformations of traditional (clay based) ceramics are of technological and archeological interest, and are usually reported qualitatively or semiquantitatively. These kinds of systems present an important complexity, especially for X-ray diffraction techniques, due to the presence of fully crystalline, low crystalline and amorphous phases. In this article we present the results of a qualitative and quantitative X-ray diffraction Rietveld analysis of the fully crystalline (kaolinite, quartz, cristobalite, feldspars and/or mullite), the low crystalline (metakaolinite and/or spinel type pre-mullite) and glassy phases evolution of a triaxial (clay-quartz-feldspar) ceramic fired in a wide temperature range between 900 and 1300 ºC. The employed methodology to determine low crystalline and glassy phase abundances is based in a combination of the internal standard method and the use of a nanocrystalline model where the long-range order is lost, respectively. A preliminary sintering characterization was carried out by contraction, density and porosity evolution with the firing temperature. Simultaneous thermo-gravimetric and differential thermal analysis was carried out to elucidate the actual temperature at which the chemical changes occur. Finally, the quantitative analysis based on the Rietveld refinement of the X-ray diffraction patterns was performed. The kaolinite decomposition into metakaolinite was determined quantitatively; the intermediate (980 ºC) spinel type alumino-silicate formation was also quantified; the incongruent fusion of the potash feldspar was observed and quantified together with the final mullitization and the amorphous (glassy) phase formation.The methodology used to analyze the X-ray diffraction patterns proved to be suitable to evaluate quantitatively the thermal transformations that occur in a complex system like the triaxial ceramics. The evaluated phases can be easily correlated with the processing variables and materials properties. These correlations can be employed for materials characterization, design and processing control.