Accuracy in quantitative phase analysis of mixtures with large amorphous contents. The case of stoneware ceramics and bricks

For the first time, this work inspects the accuracy of quantitative phase analysis of both crystalline and amorphous components of stoneware tiles and bricks. A number of variables were considered: the nature of the internal standard, experimental conditions and counting statistics. The so-called G-factor method has also been applied. The results of the X-ray powder diffraction analysis have been compared with the results obtained with optical microscopy and image analysis. Only the mixtures spiked with corundum and silicon yielded accurate weight estimates of the amorphous fraction, whereas the use of highly X-ray absorbing internal standards (such as fluorite, rutile and zincite) resulted in gross underestimations. In fact, microabsorption effects are found to drastically reduce the accuracy of the results when standards with linear X-ray absorption coefficients higher than 100 cm^-1 are employed. It was found that very low counting statistics reduced the calculated amorphous fractions in both bricks and stoneware tiles owing to partial masking of the major peak of the internal standard, namely corundum. The application of the G-factor method to the systems investigated was also evaluated. The results are poorer than those obtained using the internal standard.