An overview of the advantages and disadvantages of the determination of gold mineralogy by automated mineralogy

Abstract

The gold mining industry is experiencing boom times with the gold price pushing towards record levels and interest in new projects picking up. It is, however, becoming increasingly important to fully understand the intricacies of each project as more complex ores are encountered and lower grade ores are processed.

This overview aims to highlight the advantages and disadvantages of using automated mineralogical and associated techniques in gaining a greater understanding of gold ore behaviour. Recent developments in techniques, such as QEMSCAN and MLA, have made them more applicable to the analysis of traditionally difficult gold-bearing ores and more accessible to the industry. These techniques are continuously challenged by the need of the industry to gain representative results, but a good understanding of their limitations can allow complementary techniques to be employed, validating the results and providing a comprehensive picture of how the mineralogy of an ore can affect its metallurgy. The nature of the complementary techniques will be examined in relation to their association and use with automated mineralogical methods.

Introduction

The mineralogy of gold-bearing ores is a key factor in predicting their expected behaviour during processing. Although it is widely understood that direct gold mineralogy can have a profound effect on processing, the difficulties and cost associated with comprehensive characterisation of gold-bearing ores means that proper mineralogical analysis is often not undertaken until after a processing problem has been established. It is in this area, among others, that the increased analysis speed and reduced cost that has emerged in automated mineralogical techniques can be of great benefit to the gold mining industry. This becomes especially important with the increased occurrence of refractory and complex ores in new operations.

Metallurgically, gold is defined as ‘free-milling’ or ‘refractory’ depending on its amenability to recovery by conventional cyanidation. Marsden and House (1992) define free-milling gold as occurrences of gold whereby greater than 95% can be recovered by cyanidation, whereas ‘refractory’ gold is defined as that for which recovery is less efficient. Refractory ores can be further defined as mildly refractory, with recovery between 80% and 95% by cyanide, moderately refractory, with recovery between 50% and 80%, or as strongly refractory, with less than 50% recovery (Vaughan, 2004). Complex ores can be defined as gold-bearing ores containing multiple populations of gold that exhibit varying forms of refractoriness.

In the process mineralogy of gold-bearing ores, it is the refractory portion of gold ore that is of most importance to both the mineralogist and metallurgist. Refractoriness in gold ores can be caused by a wide variety of different mineralogical phenomena. Generally, these ores are sulphidic and can contain carbon in some form. The most common causes of refractoriness are physical lock-up, gold coatings, alteration of host minerals, carbonaceous materials or insoluble gold alloys or compounds. A brief overview of each of these probable causes is provided below:

• Physical lock-up of gold can occur when fine particles remain encapsulated within the host phase. This can be overcome by ultra-fine grinding of the ore to liberate the locked gold. However, gold may also be present in solid solution within the mineral matrix in the form of molecular or “invisible’ ” gold and require complete destruction of the mineral to be liberated (Cook and Chryssoulis, 1990).

• Passivation of gold during leaching can be caused by the formation of insoluble coatings that inhibit cyanidation. These coatings may occur as carbonates, sulphides, oxides or metal–cyanide complexes (Lorenzen and Van Deventer, 1992).

• Decomposition of host minerals in the presence of cyanide can lead to increased reagent consumption and consequently reduced leaching efficiency. The minerals of copper, zinc, lead, arsenic and antimony can react with cyanide to form metal–cyanide complexes, which can reduce the cyanide level and lead to environmental issues. Oxygen may also be consumed in the oxidation of minerals such as pyrrhotite (Marsden and House, 1992).

• The presence of carbonaceous material in gold ores can lead to a phenomenon known as preg-robbing, whereby constituents of the ore adsorb the aurodicyanide complex from solution (Goodall et al., 2005a).

• Gold present as an alloy in the form of a telluride, electrum or as aurostibite, may become refractory due to the very slow leaching kinetics in aerated cyanide solutions (Marsden and House, 1992).

All of these causes of refractoriness are controlled by mineralogy of the host rock; hence comprehensive determination of not only the gold distribution and associations but the gangue mineralogy can be vital in identifying the most efficient processing route to be used.

This overview aims to identify the existing technology available for automated determination of mineralogy and which complementary techniques can be used in conjunction as a major benefit to the gold mining industry. To ensure that it is comprehensive, techniques complementary to the strictly automated techniques are considered in relation to how they can enhance the effectiveness and efficiency of a gold deportment study.