LAUSR

Abstract In the bauxite industry – exploration, beneficiation and refinery – two main chemical parameters are used for the quality control: available alumina (AvAl 2 O 3 ) and reactive silica (RxSiO 2 ). Both are determined using a procedure that simulates the Bayer process at the laboratory scale. For gibbsitic bauxites, the subject of this study, the available alumina is associated with gibbsite, and the reactive silica is associated with kaolinite. Significant quantities of kaolinite increase the operational costs because under the conventional conditions of the Bayer process, kaolinite reacts with NaOH (leaching agent) to form a sodalite-type desilication product, which results in irreversible NaOH losses. In Northern Brazil (Para State), where the world's largest metallurgical alumina refinery (Hydro Alunorte) is located, the loss of caustic soda due to sodalite formation has been reported to be much higher than the estimated value based on the amount of reactive silica that has been measured by quality control. However, the estimated experimental errors are usually within the deviations expected by the certified values, and thus has prevented obtaining a clear answer to this problem. A great innovation for this industry would be to make this control by mineralogical parameters, i.e., the wt% of gibbsite and wt% of kaolinite via Powder X-ray Diffraction Analysis (PXRD). Even with the fast increase in the last years of studies related to mineralogical phase quantification using PXRD analysis, there are still some challenges related to bauxite quantification. For instance, the lack of crystal structure information related to kaolinite, which has been underestimated when trying to apply the Rietveld method with available Crystallographic Information File (CIF). In this work, a combination of the Rietveld, Le Bail and Internal standard methods was tested, and the results were very promising for a Brazilian gibbsitic bauxite. This combination not only improved the quality of gibbsite and kaolinite quantification, but also decreased computer processing time, making it a more convenient and fast procedure, both requirements for an industrial control method.