Application in Mineralogical Systems

1 A Simple Three-phase Mixture. The first sample of the lUCr CPD round robin on QPA summarized the results obtained from the analysis of a 

The materials used in the study (corundum - a-A Os, zincite - ZnO, and fluorite - CaF2) provided a relatively simple analytical system to determine the levels of accuracy and precision that could be expected under ideal conditions. Rather than being too prescriptive in the detail of the techniques to be used for analysis, the round robin organizers allowed the participants to select and report their own methodology. Thus, the RR also served as a survey of commonly used techniques for QPA. 

A summary of all the results returned shows that while the mean values approximate the weighed values, there is a very wide range of values, indicating that many participants were unable to accurately quantify the amounts of materials present even in this relatively simple phase system. Among the significant causes of deviation from the weighed values were operator errors. 

These were largely due to inappropriate use of the method being applied. For the Rietveld based techniques, these errors included (i) entry of incorrect crystal structure information, including atom coordinates and thermal vibration parameters, (ii) incorrect space group notation, (iii) incorrect atom site occupation parameters, (iv) allowing structure parameters (especially thermal parameters) to refine to physically unrealistic values, and (v) not completing the refinement. 

To test the validity of Rietveld based methods for QPA, the round robin organizers undertook the analysis of all eight mixtures in the Sample 1 suite using the following refinement strategy  

---