Abstract Scope |
Over the past 20+ years, we have investigated the mineralogy of porcelain. In 2000, we measured mullite formation, using an internal standard XRD method, and found that it tracked specifically with the solubility of alumina in the glass phase, leading to the determination of the glass formation boundary in aluminosilicate glasses. In short, the amount of mullite formed in porcelains can be precisely predicted from the overall chemistry. In recent years, however, there has been the proliferation of Rietveld analysis data that suggests that mullite levels are random and not predictable, suggesting that our theory of mullite formation was incorrect. To determine if there is a problem, we investigate a broad range of samples, processing conditions, and temperatures, comparing Rietveld to the internal standard method. As we will demonstrate, there is a serious problem with Rietveld, and it is not a reliable technique for the measurement of mullite in porcelain. |