D117 Fullpat: A Full-Pattern Quantitative Analysis Program and Method for X-ray Powder Diffraction

Abstract

FULLPAT is a quantitative X-ray diffraction methodology that merges the advantages of existing full-pattern fitting methods with the traditional reference intensity ratio (RIR) method [1]. FULLPAT can explicitly analyze all phases in a sample including partially ordered or amorphous phases such as clay minerals, polymers, or glasses. Use of an internal standard allows unconstrained analyses to be conducted by direct fitting of library standard patterns to each phase in the sample. By fitting entire patterns, including background, all phases (including amorphous or disordered components) are explicitly included in a FULLPAT analysis. Amorphous content need not be determined as the difference from 100% (i.e., %amorphous = 100% Σ crystalline phases) but can be measured directly. If all individual library standards are normalized to be equal on the basis of the intensity of the corundum internal standard, an “external standard” or “adiabatic” method analysis can be conducted without addition of an internal standard to the unknown, but with the constraint that the abundances of all phases sum to 100%. The method is coded in Microsoft EXCEL using standard spreadsheet functions and uses least-squares minimization to optimize the fit between the sum of standard patterns and the observed pattern [2]. The method has been applied to many natural rock samples, including those containing disordered minerals such as clay minerals and amorphous components such as volcanic glass. Results for clay mineral-containing mixtures and for internal-standard analyses of amorphouscontaining mixtures were excellent, with values typically within several percent of the known amounts. Tests with numerous standard mixtures containing well-ordered, disordered, and amorphous components gave an overall uncertainty of 3.4 wt.% at the 95% confidence level. It is important to emphasize that these tests were done with known mixtures, using standards that were different from the components in the mixtures. The precision of analysis would be considerably improved for repetitive analyses using standards that were matched to the unknowns.