Abstract
Mineralogical and textural variability of ores from different sources commonly leads to processing inefficiencies, particularly when a processing plant is designed to treat ore from a single source (i.e. ore of a relatively uniform composition). The bulk of the Witwatersrand ore in the Klerksdorp goldfield, processed at the AngloGold Ashanti Great Noligwa treatment plant, is derived from the Vaal Reef (>90%), with a comparatively small contribution obtained from the Crystalkop Reef (or C-Reef). Despite the uneven contribution, it is of critical importance to ensure that the processing parameters are optimized for the treatment of both the Vaal and C-Reefs. This paper serves to document the results of a geometallurgical study of the C-Reef at the Great Noligwa gold mine in the Klerksdorp goldfield of South Africa, with the primary aim of assessing the suitability of the processing parameters that are in use at the Great Noligwa plant. The paper also draws comparisons between the C-Reef and the Vaal Reef A-facies (Vaal Reef) and attempts to explain minor differences in the recovery of gold and uranium from these two sources.
Three samples of the C-Reef were collected in-situ from the underground operations at Great Noligwa mine for mineralogical analyses and metallurgical tests. Laboratory-scale leach tests for gold (cyanide) and uranium (sulphuric acid) were carried out using dissolution conditions similar to that in use at the Great Noligwa plant, followed by further diagnostic leaching in the case of gold. The gold in the ore was found to be readily leachable with recoveries ranging from 95% to 97% (as opposed to 89% to 93% for the Vaal Reef). Additional recoveries were achieved in the presence of excess cyanide (96% to 98%). The recovery of uranium varied between 72% and 76% (as opposed to 30% to 64% for the Vaal Reef), which is substantially higher than predicted, given the amount of brannerite in the ore, which is generally regarded as refractory. Thus, the higher uranium recoveries from the C-Reef imply that a proportion of the uranium was recovered by the partial dissolution of brannerite. As the Vaal Reef contain high amounts of chlorite (3% to 8%), which is an important acid consumer, it is considered likely that this could have reduced the effectiveness of the H2SO4 leach in the case of the ore of the Vaal Reef.
Since the gold and uranium recoveries from the C-Reef were higher than the recoveries from the Vaal Reef, the results demonstrate that the processing parameters used for treatment of the Vaal Reef are equally suited to the treatment of the C-Reef. Moreover, small processing modifications, such as increased milling and leach retention times, may well increase the recovery of gold (particularly when e.g. coarse gold, or unexposed gold, is present).
Addressing a workload characterization study to the design of consistency protocols