The development of a supercritical extraction method to extract zero valent gold

Abstract

South Africa’s easily accessible, high grade gold ores are on the decline and as a result it is more expensive to mine the deeper, low grade ores. Companies such as DRD Gold have identified secondary gold deposits that now are economically feasible to process. The advantages of mining old tailings around the Witwatersrand basin are that the major processing costs such as crushing and grinding have already been incurred.

These secondary deposits contain free surface gold particles as well as gold which is locked up in refractory silicate and sulphide minerals (i.e. pyrite and quartz). The free gold particles can easily be removed through conventional hydrometallurgical processes such as cyanidation and carbon in leach. DRD Gold recovers around 60% of the gold from these tailings (grade of the tailings is typically around 0.31 g/t). To tap into the remaining 40%, supercritical CO2 extraction has been proposed.

The supercritical fluid (SCF) is created by taking CO2 to conditions above its critical temperature and pressure (>31C and >73.8 Bar). The properties of a SCF are ideal for extracting valuable minerals as the CO2 is able to penetrate into the ore due to its gas like diffusivity and the high pressures within the reactor. However CO2 is non polar and thus a ligand is required to extract the gold. A fluorinated -diketone (hexafluoroacetylacetone (HFA)), non-fluorinated -diketone (acetylacetone (AcAc)) and an ionic liquid (betaine bis(trifluoromethylsulfonyl)imide ([Hbet][TF2N])) were selected as the ligands of interest for this research. To determine what ligand-oxidant combination had the most promising results, the ligands were mixed with either nitric acid (HNO3) or hydrogen peroxide (H2O2). It was determined that the AcAc-HNO3 system achieved the greatest gold extraction.

The following conditions were investigated to determine what effect they had on the amount of gold that could be extracted from a pure gold button; temperature (80-100 C), concentration of HNO3 (0-6.34 M), molar ratio of ligand to oxidant and volumetric ratio of ligand to oxidant. It was determined that the AcAc-HNO3 system exhibited the most promising gold extraction results after an 8-hour extraction period. The gold was in excess so no meaningful percentage extraction could be calculated, however, the process was able to extract 64.65 mg of Au. This occurred at 100C, 100 Bar, using a 1:1 molar ratio of AcAc to HNO3 (6.34 M), the total volume of the oxidant-ligand mixture was 23 mL, in a reactor fitted with a Hastelloy liner. It was found that corrosion in the form of co-leaching of Fe, Cr and Ni occurred for all extraction mixtures. The replacement of the Hastelloy liner with a Teflon liner showed decreased gold extraction pointing to a possible positive kinetic effect of these metals on the extraction reaction.

The newly developed method was used to treat tailings from the DRD Gold ERGO plant, and it was observed that the process could extract an additional 48.38% of the gold present in the ERGO plant tailing.