Flowsheet: THE GOLDEN GIANT MILL; Part One of Two

The Golden Giant deposit comprises a diverse group of stratabound pyrite-rich, siliceous and locally baritic metasediments of probable chemical sedimentary derivation. Ranging in thickness from 3 to 40 m, the mineralized zone has gold values varying from 4 to 16 g per tonne and contains appreciable molybdenite. Accessory minerals such as stibnite, realgar, and light green mica are characteristic of the ore zone. Some typical contents for Hemlo ore are given in Table 1.1.

Native gold generally occurs as disseminated grains on silicate grain boundaries. The typical gold-to-silver ratio is 6:1, with an average silver content of 1.5 g per tonne. The majority of the antimony occurs in stibnite, while arsenic occurs primarily in realgar and arsenopyrite.

Because of to the presence of antimony, the ore tends to be mildly refractory when leached at a pH level above 10.0. Therefore, the leach and CIP processes were designed to incorporate pre-aeration and ventilation measures, which enable leach circuit operation at low pulp alkalinities. (pH 8.5 to 9.5)

Tailings Facility and Effluent Treatment Plant

Mill tailings, along with mine water, treated sewage and site run-off, is pumped to a tailings impoundment facility which is bounded by an impervious dam on one side and surface relief for the rest of its perimeter. Water from a seepage collection facility downstream of the main dam is returned to the pond.

Tailings reclaim water is pumped from a floating barge back to the mill where maximum recycling is practised. A large portion of the recycled water is used in the grinding circuit. Ground slurry is thickened in a conventional thickener with the thickener overflow pumped through a carbon column circuit for trace gold recovery. The carbon column circuit discharge reports directly to the effluent treatment plant.

The effluent treatment facility was originally designed as a 2-stage plant with the first stage to be used for cyanide removal and the second stage for heavy metal removal. The circuit configuration is shown in Figure 2.1. The two stages are physically identical, consisting of three 195-cu-m woodstave reaction vessels connected in series and a 32-m-diameter Enviroclear clarifier. The plant is designed to treat 300 cu m of water per hour.

Minimizing Dissolution of Contaminants During Leaching

Through extensive laboratory investigations, it has been established that two operating parameters significantly affect the quality of CIP tailings reclaim water. First, antimony dissolution is very sensitive to ph; second, the residual cyanide level in the CIP tailings solution is directly affected by the free cyanide concentration maintained through the leach circuit. Significant cost savings have been realized from reduced cyanide consumption and reduced removal costs from the waste water by operating the leach circuit at a cyanide concentration just above “starvation levels”.

The leaching circuit was designed to operate at a pH of less than 10.0 to ensure maximum gold recovery. In addition to the impact on recovery, it has been established that antimony dissolution is very sensitive to pH as shown in Figure 3.1. A difference of 0.5 pH units through the 9.0-to-10.0 range typically affects the dissolution by some 50%. In practice, it is possible to maintain a cyanidation ph between 9.4 and 8.8 without having to acidify the pulp. Although HCN gas forms at these levels, it does not significantly affect the cyanide consumption.

Unfortunately, it is not possible to operate at the lower alkalinity level through the carbon adsorption circuit without HCN gas evolving to the working environment.

Therefore, the pH is raised to 10.3 during the contact with carbon. While this does affect antimony dissolution, the effect is not as drastic as in the leaching circuit, owing to the much shorter retention time in the CIP circuit.

The leach circuit was initially operated at a free cyanide concen tration of 0.5 g NaCN per L as per industry norm. After a multitude of tests both in the lab and plant, it has been firmly established that the safe threshold level for maximum gold recovery is much lower — typically 0.05 to 0.07 g per L are required.

Plant trials using 0.3 g per l free NaCN were initiated in December, 1985. A further reduction to 0.1 g per l was made in April, 1986. Continued success and monitoring has led to the adoption of this parameter. As shown in Table 3.1, a significant reduction in cyanide consumption has resulted from the changes described above. As expected, residual cyanide levels in the CIP tailings solution have dropped concurrently.

In summary, the adjustment of leaching circuit parameters has reduced antimony dissolution, residual cyanide levels and overall cyanide consumption. Apart from the direct benefits of cost reductions associated with reduced cyanide consumption, there has been an indirect benefit to reducing effluent treatment costs for both cyanide and antimony removal.

The Hemlo Treatment Process — Background, Development and Process

Evaluations of several available effluent treatment processes have been made at the Golden Giant mill on both the lab and plant scale. The objective of these investigations was to ensure that the process chosen would remove both free cyanide and metallocyanide complexes and effectively precipitate heavy metals, including antimony, copper, iron, nickel and molybdenum. Table 4.1 lists the final effluent quality criteria applicable to the Golden Giant mine.

Liquid sulphur dioxide along with copper sulphate was tested initially (as described in Noranda’s patent No. 1183617) for cyanide destruction in the first treatment stage, while ferric sulphate was used for heavy metal co-precipitation in the second stage. A typical ratio of SO2 to the total cyanide in the feed solution during testing was 7:1 on a weight basis. This ratio could be varied according to the needs of a specific process.

Copper additions were made on the basis of cyanide levels in the plant feed and were typically 0.5 parts to one part copper to one part cyanide. Iron additions were similarly made relative to antimony concentrations in the feed solution, with ratios of iron to antimony averaging 12:1 on a weight basis. Table 4.2 shows operating results obtained while using liquid SO2 and ferric sulphate as described above.

The original design called for utilization of liquid SO2 for cyanide destruction. Although sulphur dioxide was effective, there were safety concerns associated with the storage of large quantities of this reagent at the mine site. Both the shaft collar and the fresh air intake for the underground ventilation system were situated within contamination range of the SO2 tank. The potential of an SO2 leak or spill in the vicinity of either of these locations was deemed an unacceptable risk.

Hydrogen peroxide was also tested as a primary oxidizing agent in cyanide destruction with copper sulphate used to catalyze cyanide removal reactions in the first stage of effluent treatment. Extensive plant scale tests indicated that peroxide was effective whenever the total cyanide level in the plant influent was fewer than eight parts per million with correspondingly low levels of copper, iron and nickel. Under such conditions, ratios of addition were three to four parts peroxide to one part cyanide in the feed plus a small amount of copper sulphate which could be in the form of recycled sludge. Ferric sulphate was used again to co-precipitate heavy metals during peroxide trials in the second stage of the process. Some operating results for peroxide
treatment are given in Table 4.3.

The shortcomings of peroxide utilization became apparent at higher cyanide concentrations. This was attributed to the corresponding increases in the amounts of metallocyanide complexes. Acceptable effluent quality was not consistently attained and the cost of treatment was becoming excessive.

In summary, peroxide was found to be better suited as polishing chemical, possibly to be used in conjunction with other cyanide-removal agents.

The concerns identified during the testwork described above led to the search for another effluent treatment alternative. Development work led to the discovery of a process which employs ferrous sulphate to reduce cupric ion in an aqueous solution. The products of this reaction consistently reduced the cyanide level in all process solutions. This discovery formed the basis for the refinement and implementation of a new effluent treatment process (1).

In the Hemlo process, an aqueous solution of copper and ferrous sulphate is added to the waste water at a controlled pH of 6 to 7. Upon addition of the premixed solution to the process, the following reaction is believed to occur:

Cu^2 + Fe^2 + 3OH^- = Cu^+ + Fe(OH)_3

This reaction does not necessarily proceed to completion. The ferrous ion, in the presence of hydroxide ions, is oxidized to ferric hydroxide while cupric ion is simultaneously reduced to cuprous ion. This reaction takes place instantaneously.

Cuprous ion removes free cyanide as an insoluble precipitate of cuprous cyanide, thereby creating a shortage of free cyanide ions in solution. This causes soluble metal cyanide complexes of copper, zinc and nickel to dissociate into simple cyanide and metal ions, leading to further removal of cyanide by the cuprous ions.

The proposed, simplified reaction for describing the removal of free cyanide is given by:

2Cu^+ + 2CN^- = Cu_2(CN)_2 ppte

Examples of metallocyanide decomposition are as follows:

Ni(CN)_4^2- = Ni^2^+ + 4CN

Cu(CN)_4^2- = Cu^2+ + 4CN

Ferrocyanide is believed to precipitate as cupric ferrocyanide immediately upon addition of premix to the reclaim solution. The proposed reaction is:

2Cu^2+ + Fe(CN)_6^4- = Cu_2Fe(CN)_6 ppte

Finally, thiocyanate is partially removed by this process according to the proposed reaction,

Cu + CNS = CuCNS ppte

This may help account for the relatively high demand for copper in the Hemlo process relative to other known processes.

Heavy metals are co-precipitated from solution by the ferric hydroxide formed from the premix solution. Most notably, the antimony and molybdenum are effectively removed. In the case of certain impurities such as nickel, a pH adjustment to between 9.5 and 10.0 is necessary to ensure precipitation after the initial cyanide removal has been completed.

The copper requirements are directly related to the level of cyanide present in the reclaim water. Ratios of copper to total cyanide may vary between one to one and as high as 10 to one on a weight basis. In general, it has been observed that higher copper addition rates result in lower final cyanide levels in the final effluent as shown in Figure 4.1.

Plant experience at Hemlo has indicated that a ratio of three parts copper to every part of cyanide in the reclaim water is sufficient to achieve effective cyanide removal.

Iron requirements are broadly related to copper additions with a typical iron-to-copper ratio being 1.5:1. The iron, in addition to acting as a reducing agent for copper, serves to co-precipitate antimony and molybdenum. To ensure adequate removal, the addition rate may be adjusted according to the concentrations of these metals.

The free cyanide ions liberated by the breakdown of metal cyanide complexes can be removed to acceptable levels by the addition of premix solution alone. It may be more economical however, to complete final cyanide removal by utilization of hydrogen peroxide at high ph. This practice has been adopted at Hemlo. Hydrogen peroxide is added in the second stage of the process at high pH to oxidize residual simple cyanides.

In summary, the Hemlo process uses no hazardous chemicals and allows for the simultaneous removal of cyanide and heavy metals using a premixed solution of copper and ferrous sulphate. Hydrogen peroxide is used as a polishing agent for simple cyanide oxidation. A patent has been applied for the process described above.

]]>