Application of the Rietveld refinement method for quantification of mineral concentrations in bauxite residues (alumina refining tailings)

• Number and complexity of minerals in mine wastes and tailings make quantification difficult.
• Techniques for quantification were compared using bauxite residue as a case study.
• Rietveld refinement of X-ray diffraction patterns was identified as most accurate.
• Microabsorption and non-ideal mineral structures pose problems for accuracy.
• Further characterization of desilication product is required to improve accuracy.

Mineral processing tailings are complex mixtures of minerals residual from the parent ore and minerals formed during processing. The number of minerals present in tailings samples, and their unusual composition and structure compared with minerals naturally occurring at the Earth's surface, poses challenges for quantification of mineral concentrations. Determination of mineral concentrations in tailings can be required to meet legislative requirements, to conduct geochemical modeling, and to optimize mineral processing efficiency. This review compares techniques for quantification of mineral concentrations in bauxite residue (alumina refining tailings), as a case study, and highlights major sources of inaccuracy in quantification methods. Although alternative techniques exist for the quantification of mineral concentrations in bauxite residues, Rietveld refinement-based quantitative phase analysis is likely to be more accurate and precise than methods based on chemical extraction, normative calculation, or single X-ray diffraction peaks. The problems of microabsorption, caused by both heavy and light X-ray absorbing minerals being present in a sample, and non-ideal mineral compositions, such as those of minerals formed during ore processing (desilication product minerals, in the case of bauxite residue), are particularly difficult and impractical to eliminate completely; however, they may be minimized by careful sample preparation and analysis, and further characterization of process characteristic minerals.