The impact of mineralogy on processing for recovery of chromite and PGE in the Bushveld Complex, South Africa

The Lower and Middle Group (LG and MG) chromitites of the Bushveld Complex in South Africa are the source of a very large portion of the global chrome supply. The recovery of platinum group elements and base metals (Ni, Cu) as by-products has the potential to add value to these chrome resources. Yet, the effectiveness of chromite and platinum-group element beneficiation circuits is highly sensitive to variations in feed composition. Mineral assemblages have been noted to be affected by surficial weathering (down to 50 m) and hydrothermal alteration. Of particular relevance is the abundance of alteration silicates, the prevailing base metal sulphides (BMS) and platinum group mineral (PGM) assemblages and mineral association which have a significant impact on recoveries and concentrate grade. The goal of this particular case study was to evaluate the potential recoverability of platinum group elements (PGE) as a by-product during chromite production in the Thaba Mine. As shown by a lot of studies, only a very minor amount of the economically important 3E (Pt,Pd,Rh) is enclosed in chromite and will therefore report to the chromite concentrate. On the other hand, the 6E (Pt, Pd, Rh, Ru, Ir and Au) grades of the mined chromitites are usually below 2 g/t and the feasibility of an additional PGE processing plant will be challenging. Therefore, a versatile and flexible geometallurgical framework is needed to identify potential PGE targets within the mine and to predict the recoverability of PGE in the ore. To extract a maximum of information at a minimal cost and material use, the framework proposed iterates between analytical work and statistical/ mathematical modelling.
Here, we focus on the metallurgical test work results of unweathered LG and MG chromitites. More than 100 different diamond drill core intersections of chromitite seams were used as sample material and analyzed by automated scanning electron-based image analysis. Several properties of each sample were fed into a statistical unsupervised classification scheme to create seven mineralogically distinct clusters for a subsequent metallurgical test work at Mintek. Composited batch samples were milled and fed to a shaking table to separate the chromite as efficiently as possible. Tailings were milled to 80 % < 75 µm, sampled and fed to flotation cells. The tailings of the rougher circuit were discharged and the rougher concentrate was subsequently fed to a cleaner flotation stage. Finally, the cleaner concentrate was sampled and chemically analyzed. Batch sample results display rather homogeneous shaking table feed 3E grades, ranging from 0.43 to 0.69 ppm, while Cr2O3 concentrations display a larger variability, ranging from 35 wt% to 42 wt%. Flotation feed grades range from 0.75 to 1.96 ppm. Cleaner flotation concentrates display grades between ca. 4 ppm and 16 ppm, resulting in upgrading factors between 2 and 11. Overall 3E recovery is between 25 and above 40 %. One reason for these PGE losses could be either large and liberated PGM grains or BMS agglomerates associated with PGM and/or PGE, which would go with the chromite concentrate. To reduce losses at the chromite concentration stage a possible PGM removal upfront with coarse flotation should be considered. Secondly, creating flotation reagent regimes and increasing residence time of the ore in the flotation cells to handle complex PGM assemblages (PGE-alloys and –sulpharsenides) will increase the flotation performance. To further increase the recovery, new flotation technology for better mineral surface cleaning may promote flotation of liberated PGM, as well as increasing liberation (e.g. Mach reactor).