The Causes for the Poor Flotation Performance of a Double-Refractory Gold Ore

The recovery of gold from certain Carlin-type deposits, with numerous perceived and real recovery problems is quite challenging. These deposits are characterised as “double-refractory” because most of the gold values are locked in sulphide minerals (e.g. pyrite and arsenopyrite) and, therefore, cannot be leached directly with cyanide; pre-concentration of sulphide minerals by flotation followed by either roasting or pressure oxidation is necessary. Furthermore, these ores contain varying amounts of carbonaceous matter which is known to cause “preg-robbing” in the cyanidation stage that must be removed before cyanidation. Despite the numerous studies that have been performed on double-refractory gold ores in many laboratories globally, recovery of sulphide minerals remains low and selective separation between sulphide minerals and carbonaceous matter is quite poor. A review of the published literature on double-refractory gold ores showed that the reasons for their poor separation and recovery by flotation are unknown and that no detailed study investigating the causes of this behaviour has yet been conducted. This thesis investigated the possible causes for the poor flotation response and separation selectivity of a double-refractory gold deposit owned by Barrick Gold Corporation (Goldstrike deposit, BR-22 ore). It was hypothesized that “The poor flotation response and selective separation of gold-bearing minerals in the Goldstrike double-refractory gold ore are mainly caused by certain non-sulphide gangue species, and can therefore be rectified in principle by an appropriate alternative processing strategy for separating the deleterious species from the value species”. To address this hypothesis, a comprehensive suite of analytical techniques including fire-assay, inductively coupled plasma atomic emission spectroscopy (ICP-AES), X-ray Diffraction, optical microscopy and image analysis techniques (Mineral Liberation Analyser) were used for detailed chemical and mineralogical characterisation of the ore. The distribution of gold in different gold bearing minerals was also identified using diagnostic acid leaching methods. Preliminary flotation experiments were performed to improve the flotation recovery and separation selectivity of the ore. To address the hypothesis a novel experimental design strategy was developed and applied. A gravity separation technique (i.e. Mozley separator) in conjunction with a tailored mixture experimental design was used to first isolate/separate mineral components, blend the separated mineral components in controlled proportions and characterize the blends metallurgically using flotation. In the course of the use of this suite of techniques, the effect of certain gangue minerals and mineral associations on the flotation response of the ore was studied. Through the detailed chemical and mineralogical characterisation of the ore, it was found that in order to obtain the adequate degree of liberation, the ore needed to be ground to a very fine P80 of approximately 20 µm. But the preliminary flotation performance (i.e. recovery and separation selectivity) of the fine-grind Goldstrike double-refractory gold ore was quite poor and non-selective despite the minerals of interest being almost fully-liberated. The results of the flotation experiments on the blended components of the ore supported the hypothesis and showed that even a small increase in the mass proportion of non-sulphide gangue minerals had a detrimental effect on the recovery of sulphides and separation selectivity. The very fine size distribution and extensive free surface area of the existing non-sulphide gangue minerals especially clays and carbonates had a noticeable effect on the flotation recovery and separation selectivity of the ore. Therefore, it is postulated that effect of non-sulphide gangue minerals on the surface/pulp chemistry of sulphide minerals caused by existence of fine and fully-liberated non-sulphide gangue minerals may be the reason for the low recovery and poor separation selectivity of the gold bearing minerals. It is recommended that a comprehensive surface chemistry study be conducted for investigating the adsorption of reagents and oxidation on the surface of minerals. Also, spiking flotation experiments as a tool for examining the extent effect of non-sulphide gangue minerals (carbonaceous matter, clays and carbonates) on the flotation performance of sulphides should be considered. Moreover, by combining spiking tests with a surface chemistry study, it may be possible to understand the underlying mechanism of the adverse effect of certain gangue minerals on flotation performance of the double-refractory gold ore type. Finally, as a quick approach for increasing the recovery of sulphides before these additional studies, separation of non-sulphide gangue minerals especially clays and carbonates prior to the sulphide flotation stage is suggested for the Goldstrike double-refractory gold ore