Thiosulfate Leaching As An Alternative To Cyanidation:

Thiosulfate Leaching as an Alternative to Cyanidation: A Review of the Latest Developments W.T. Yen Department of Mining Engineering Queen’s University Kingston, Ontario, Canada, K7L 3N6 Tel: (613)533-2206 FAX: (613)533-6587 E-mail: [email protected] Guy Deschenes MMSL, CANMET Natural Resources Canada Ottawa, Ontario, Canada, K1A 0G1 Tel: (613)992-0415 FAX: (613)996-9041 E-mail: [email protected] Mark Aylmore CSIRO, Division of Minerals Conlon St., Waterford WA Box 90, Bentley, WA 6982, Australia Tel: 61-8 9334 8074 FAX: 61-8 9334 8001 E-mail: [email protected]

33rd Annual Operator’s Conference of the Canadian Mineral Processors 23-25, January 2001 Ottawa, Ontario

Keywords: Thiosulfate, gold, extraction, reagent consumption, leaching.

Thiosulfate Leaching as an Alternative to Cyanidation: A Review of the Latest Development W.T. Yen1, G. Deschenes2 and Mark Aylmore3 1, Department of Mining Engineering, Queen’s University, Kingston, Ontario, Canada 2, MMSL, CANMET, Natural Resources Canada, 555 Booth St., Ottawa, Ontario, Canada 3, CSIRO, Division of Minerals, Bentley, WA, Australia

Abstract Increasing pressure are motivating the gold mining industry to develop a substitute to cyanide for leaching gold. This effort in mainly based on the high toxicity of cyanide and on the lack of efficiency of cyanide in the treatment of mildly refractory gold ores (sulfides and prerobbing). Consequently, in only a few years, consortia have been created to elucidate and implement a new technology. The most promising lixiviant that has been selected to replace cyanide is thiosulfate. This paper will present a critical analysis on the latest developments in thiosulfate leaching and illustrate the potential and limitations of the technology with experimental data. Finally an assessment of the cost and performance of thiosulfate leaching relative cyanidation will be evaluated. Introduction Concerns regarding the toxicity of cyanide and inability of cyanide solution to effectively leach some carbonaceous gold ore or mild refractory gold ores have made the gold mining industry evaluate alternative lixiviants to leach gold. This concern has been heightened recently by the cyanide spill at a mine in Romania, which devastated a 250-mile stretch of the Danube River and its tributaries, and the environmental damage at Ok Tedi mine in Papua New Guinea. Some countries and locations have prohibited the application of cyanidation to recover gold, such as, California and Colorado States, Japan and Turkey etc. There are several potential non-cyanide lixiviants (Sparrow and Woodcock, 1995) for gold extraction, such as thiosulfate, thiourea and halogen. A pilot feasibility test (Marchant and Broughton, 1987) had suggested that thioureation might result in acutely toxic byproducts even after treatment. Scientists of Swedish Academic of Sciences reported that thiourea is a carcinogen. Halogen is also a pollutant to the environment. Thiosulfate leaching appears to have very good potential as an effective and less hazardous procedure for gold and silver extraction from auriferous ores. In addition to being much less toxic and relatively cheap, it forms a stable anionic complex with gold but is less reactive to other metals, for instances, copper and carbonaceous materials, in comparison with cyanide.

However, the chemistry of the thiosulfate system is more complex than that of cyanide and some aspects of the mechanism are not yet well understood. Historically, high reagent consumption, and a lack of a suitable recovery method has made thiosulfate leaching uneconomical. Current research is providing a better understanding of the process, and addressing some of these difficulties, to make thiosulfate leaching a probably viable alternative for gold extraction. Thiosulfate Leach Chemistry Thiosulfate forms strong complexes with Au(I), Cu(I), Cd(II), Bi(II), Hg(II) and Fe(III). In the presence of ammonium thiosulfate with oxygen, gold complexes with thiosulfate as follows (Hiskey and Atluri, 1988): 4Au + 8S2O32- + O2 + 2H2O → 4Au(S2O3)23- + 4OH-

(1)

As a mild reductant, thiosulfate can reduce Cu2+ to Cu+ and complex the Cu+ as follows (Tykodi, 1990; Rabai and Epstein, 1992): 2S2O32- + 2Cu2+ → 2Cu+ + S4O622Cu+ + 2S2O32- → 2Cu(S2O3)222Cu(S2O3)22- → 2Cu(S2O3)- + S4O62-

(2) (3) (4)

These reactions are supported by observations in which the characteristic blue color of Cu2+ (aq.) fades and a colorless solution containing the complex ion Cu(S2O3)22- (aq.) results. In the presence of oxygen, the cupric rather than the cuprous state is favored, and the cupric state is stabilized by the presence of ammonium ion: 2Cu(S2O3)23- + 8NH3 + 0.5O2 + H2O → 2Cu(NH3)42+ + 4S2O32- + 2OH-

(5)

Under alkaline conditions, cupric tetraamine ions Cu(NH3)42+ act as a catalyst for the gold dissolution in ammonium thiosulfate solution, and can be explained as an electrochemical catalytic mechanism (Yen, et al, 1996): Anodic Area: Oxidation: Au → Au+ + eComplexing: Au+ + 2NH3 → Au(NH3)2+ Au(NH3)2+ + 2S2O32- → 2NH3 + Au(S2O3)23-

(6) (7) (8)

Cathodic Area: Reduction: Cu(NH3)42+ + e- → Cu(NH3)2+ + 2NH3 Complexing: 4Cu(NH3)2+ + 8NH3 + O2 + 2H2O → 4Cu(NH3)42+ + 4OHThe overall reaction of anodic and cathodic areas is the equation (1).

(9) (10)

Thermodynamics Current research is detailing the reactions and species involved in thiosulfate leaching of gold and silver so as to enable the optimum molar ratio for ammonia, thiosulfate and copper to be established. Thermodynamic analysis can provide information on which complexes are formed and the Eh-pH relationships that are essential for understanding the redox and solubility behaviour of the species (Aylmore and Muir, 2001). From thermodynamic principles, appropriate conditions for Cu-NH3-S2O32- occur where the copper(II) ammine complex and the copper (I) thiosulfate complex can coexist in solution , as shown in Figure 1 (black rectangle). Eh (Volts) 2.0

Cu(NH3)4+2 (aq)

1.5

Cu+2 (aq)

1.0

CuO CuO

0.5

Cu(S2O3)3-5 (aq)

0.0

CuS

Cu2O Cu2S

-0.5 -1.0

Cu -1.5 -2.0 4

6

8

10

12

14 pH

Figure 1 Eh-pH diagram for the Cu-NH3-S2O32- system (conditions: 0.1M S2O32-; 0.1M NH3/NH4+, 5x10-4M Cu2+). Key Parameters in Thiosulfate Leaching The parameters controlling the thiosulfate leaching of gold are: reagent concentrations (i.e., thiosulfate, ammonia and cupric ion), pH, Eh, dissolved oxygen, pH and retention time. In the period of 1980 to 1992, most researchers conducted the thiosulfate leaching at an elevated temperature between 35Ο and 70ΟC (Tozawa et al., 1981, Kerley, 1981, 1983, Block-Bolten and Torma, 1985, 1986, Zipperian et al. 1988, Hemmati et al, 1989, Caixia and Qiang, 1991, Hu and Gong, 1991, Murthy, 1991, Cao et al., 1992). More recently work, thiosulfate leaching has been carried out at ambient temperature for practical reasons (Langhan et al, 1992, Wan et al, 1994,

Abbruzzese et al, 1995, Groudev et al, 1996, Yen et al, 1996, Wan, 1997, Wan and Briesley, 1997, Yen et al, 1998, 1999). Depending on the ore treated, the optimum thiosulfate leaching conditions were varied. The effect of various leaching parameters will be illustrated in the following example on a Canadian ore containing copper. The composition of the ore was assayed as below: Gold 24.16 g/t Iron 5.14 % Copper 0.31 % Sulfur 2.54 % Total carbon 0.81% The main sulfide minerals present are chalcopyrite and pyrrhotite with trace amounts of sphalerite and marcasite. The main phases are chlorite, quartz and albite. The optimum grind has been determined as 94% minus 200 mesh. The effect of all other variables was conducted under the optimum grind at ambient temperature. Figure 2 shows the comparison of gold extraction for both cyanidation and thiosulfate leaching. The condition for the cyanide leach was 1 g/L NaCN with pH adjusted with lime to 10.5. The condition for the thiosulfate leach was 0.3M (NH4)2S2O3, 0.03M CuSO4 and 3M NH4OH at pH 9.5-10. The results show that the gold extraction by straight cyanidation was slow and required 48 hours to achieve 90% gold extraction. The poor extraction is due to the negative effect of copper sulfide minerals in cyanide leach. Pre-conditioning the slurry with 200 g/t Pb(NO3)2 followed by cyanidation, 92% gold was extracted within 10 hours and 95% gold extraction in 24 hours. The gold extraction by thiosulfate was 95% in 24 hours, which was faster than that observed for the straight cyanidation. The thereafter thiosulfate leaching was conducted at optimum grind, 24 hours retention time at ambient temperature. The effect of ammonia concentration on the gold extraction and the thiosulfate consumption at 0.3M (NH4)2S2O3 and 0.03M CuSO4 is shown in Figure 3 . The result shows that 95% gold extraction was obtained at 1M NH4OH and remained the same up to 5M NH4OH. The thiosulfate consumption was high at high ammonia concentration and decreased with increasing the ammonia concentration. Approximately 38 kg/t (NH4)2S2O3 was consumed at 1M NH4OH whereas 25 kg/t (NH4)2S2O3 was consumed at 4 or 5M NH4OH. Since ammonia was evolved at 4 or 5M NH4OH, the subsequent tests were carried out at 3M NH4OH or lower. The effect of copper sulfate concentration on gold extraction and thiosulfate consumption at 0.3M (NH4)2S2O3 and 3M NH4OH are shown in Figure 4. The result shows that 95% of gold was extracted at 0.01M CuSO4 and higher copper concentration. Thiosulfate consumption increased rapidly with increased copper content in the solution. About 17 kg/t (NH4)2S2O3 was consumed at 0.01M CuSO4, 30 kg/t (NH4)2S2O3 at 0.03M CuSO4 and 42 kg/t (NH4)2S2O3 at 0.04 or 0.05M CuSO4. The results indicated that copper in the solution is a major thiosulfate consumer. Although copper act as an important role in catalysis reaction, the excess amount of copper just waste the thiosulfate addition.

The effect of ammonium thiosulfate concentration on the gold extraction and the thiosulfate consumption with 3M NH4OH and 0.03M CuSO4 are shown in Figure 5. The results indicate that the thiosulfate concentration is very sensitive to the gold extraction. At 0.1M (NH4)2S2O3, only 40% of gold was extracted whereas at 0.02M (NH4)2S2O3 88% of gold was extracted. A concentration of 0.3M (NH4)2S2O3 was required for 95% extraction of the gold. No much improvement in gold extraction was obtained at a concentration greater than 0.4M (NH4)2S2O3. Thus, 0.3M (NH4)2S2O3 was chosen as a standard thiosulfate concentration for all tests. The effect of pH on the gold extraction and the thiosulfate consumption with 3M NH4OH, 0.03M CuSO4 and 0.3M (NH4)2S2O3 is shown in Figure 6. The best gold extraction (95%) was obtained at a pH of 9.7. At pH value greater than 10 or less than 9.5, the gold extraction was reduced slightly. At a pH of 10.5, the gold extraction was 70% whereas pH of 9.2, the gold extraction was 80%. The thiosulfate consumption increased with increasing pH value. 24 kg/t (NH4)2S2O3 consumed with pH value lower than 9.5. Over the pH range of 9.7 – 10, 30 kg/t (NH4)2S2O3 was consumed and pH of 10.5, 43 kg/t (NH4)2S2O3 was consumed. From above test results, it can be concluded the optimum thiosulfate leaching conditions for the tested ore sample was 0.3M (NH4)2S2O3, 3M NH4OH and 0.01M CuSO4 at a pH of 9.7 for 24 hours. Under these condition, 17 kg/t (NH4)2S2O3 was consumed. Thiosulfate Leaching of Ores Since 1980, there has been 22 papers and eight patents published (Aylmore and Muir, 2001) on the thiosulfate leaching of gold ores. Most work has been carried out on complex ores, ore containing high concentration of copper, refractory carbonaceous ores, copper sulfide concentrates and ores containing high concentrations of lead, zinc or manganese. The gold content in these ore have ranged between 1 g/t Au and 62 g/t Au. Most of the laboratory studies which have been published involved the leaching of finely ground material in stirred tanks. The leaching conditions have reported in the literature vary considerably. The thiosulfate and copper concentrations reported range from 0.1 to 2 mole/L and from 0.001 to 0.1 mole/L respectively. The ammonia concentration used varied from 0.1 to 6 mole/L and the pH correspondingly varied from 9 to 10. Relatively short leaching periods of 1 to 4 hours at an elevated temperature of 30 700C or leaching periods of 24-48 hours at ambient temperature with agitation leach has been used. The reagent consumption ranged from 2 kg/t to 40 kg/t (NH4)2S2O3. In addition to laboratory studies, there has been a couple of pilot scale tests and one heap leach test demonstrated.

The pilot plant was build to recover precious metals from the residual tailings of an old cyanidation plant located at LaColorada, Senora, Mexico (Perez and Galaviz, 1987). Mineral associated with the ore contained high concentration of manganese which caused problems for effective cyanide leaching. The tailings were ground with a copper-thiosulfate solution and leached at pH 9.0 to 10.0. The ammonium thiosulfate concentration used was 100 g/L (0.676M (NH4)2S2O3) with a copper concentration of 3 g/L (0.012M CuSO4*5H2O). The overall gold and silver extraction for this process was 85% and 75% respectively. The pilot plant was operated for about four years. Problems observed during the operation of the pilot plant were corrosion of the carbon steel equipment and air pollution. Newmont Gold Co. patented a microbial inoculation/agglomeration process (Wan, Le Vier and Clayton, 1994, Brierley and Hill, 1993, Wan and Brierley, 1997) for rapid initiation of bio-oxidation of refractory ores in heap followed by thiosulfate leaching to avoid the pregrobbing activity of carbonaceous matter. The work was carried out in trials on both laboratory and pilot plant scale experiments. The ore was crushed to 70 – 98% finer than 1.9 cm and contained approximate 1.61 – 3.50 g/t of gold. The ore contained 0.78 – 1.35% sulfide sulfur and 0.67 – 2.42% acid insoluble carbon. Six heap leaches ranging in size from 432 to 25,900 t

(476 to 28,550 st) were used in the test. The process consisted of growing a dense suspension of the naturally occurring iron-oxidizing bacteria Thiobacillus ferrooxidans and Leptospirillum ferrooxidans and then adding this bacteria culture to the ore as the heap was formed. The distribution of bacteria throughout the heap was achieved with noticeable agglomeration of the fine ore particles. The effluents from the heap leach were collected in either a tank or pond and recirculated to the top of the heap. The progress of the biooxidation pretreatment process was monitored by measuring pH, Eh and Fe2+/Fe3 ions in solution. After biooxidation, an agglomeration-neutralization step was carried out to neutralize the acidic ore for leaching with thiosulfate under alkaline conditions. This was achieved by washing the heap in water to displace soluble acid and iron salts. The washed heap is further neutralized by blending in lime and soda ash. A schematic diagram of the heap leach process used by Newmont Gold Co during the thiosulfate leaching stage is shown in Figure 7 (Aylmore and Muir, 2001). After passing through the heap, the pregnant lixiviant solution was recovered at the bottom of the heap and recirculated, either continuously or intermittently. After precious metals values were recovered from the lixiviant solution by precipitation, the solution was recirculated to the static heap. The gold recovery by thiosulfate was 66% while the conventional cyanidation process resulted in only about 17%. The optimum leach solution contained 0.1-0.2M (NH4)2S2O3 and/or Na2S2O3, at least 0.1M NH4OH and up to 60 ppm of Cu(NH3)42+ (0.0005M cupric ion). The pH range was between 9.2 and 10.0. The ammonium thiosulfate consumption was in the range of 5.2 kg/t to 8.9 kg/t.

Heap Ammonium Thiosulfate Cu2+ NH3 Ammonium Thiosulfate

Pregnant Lixiviant Solution Storage

Precious Metals Recovery System Pump

Figure 7, A Schematic Diagram of A Gold and Silver Leaching and Recovery Process for Static Heaps Using Thiosulfate (Wan et al, 1997, Aylmore and Muir, 2001)

The pilot plant test of biooxidation-thiosulfate heap leach was encouraging and process development is continuing. A demonstration plant (Wan, et al, 1997) was designed to process 707,400 t (780,900 st) of low-grade refractory gold-bearing ore per year, which was normally considered as waste material at Newmont. The demonstration plant began producing gold in October 1995. Barrick Gold Corporation have patented a combined pressure oxidation, thiosulfate and resin-in-pulp process (Thomas et al, 1998) for treating refractory gold ores (Figure 8, Aylmore and Muir, 2001). The project was indicated to find a technology to process a refractory gold ore with a high concentration of carbonaceous matter. In this process ore was ground to about 6585% passing 200 mesh and thickened to about 40-50% solids. Sodium carbonate is added to ensure that pressure oxidation is carried out under alkali conditions and NaCl is added to improve the kinetics and to facilitate oxidation. The pressure oxidized ore slurry leaves the autoclave at about 35% solids and is directed to a leaching tank contains 0.03 -0.05M (NH4)2S2O3, 10-100 ppm CuSO4, 500-1000 ppm NH4OH and 0.01-0.05M SO32-. The pregnant slurry contains 1-5 ppm Au is directed to a Resin-in-Pulp (RIP) circuit where gold and copper are loaded onto a strong base resin to about 1-5 kg/t Au and about 10-25 kg/t Cu. Copper is eluted from resin using 200g/L-ammonium thiosulfate and gold is eluted using 200-g/L potassium thiocyanate. The copper-bearing eluate is returned to the leaching circuit while gold eluate is either electrowon or precipitated. However, the results of the work were not successful enough to continue. The main reason may have been due to low gold extraction (about 70% in average) and high thiosulfate consumption. The roaster was installed at Barrick mine to treat the carbonaceous gold ore prior to cyanidation. In addition, Dr. Ellwanger et al (Li, et al, 1995) have applied ammonium thiosulfate process to a small scale production plant for in-situ leaching at an underground mine site in South Africa. An enriched oxide ore (primarily free gold) was leached by ammonium thiosulfate solution in an underground mine site at a depth of about 10,000 ft. All equipment necessary for gold recovery was installed on railcars. A solution flowrate of 50 gal/min was used. Gold was recovered from the thiosulfate solution by cementation using multiple copper sponge columns. The in-situ leaching application has been granted South Africa patents. In China, it was reported (Li and Zhang) that 96% of the gold and 87% of solver were extracted from a complex sulfide concentrate and a silver-bearing pyrite ore by direct thiosulfate leaching the ore. Based on their laboratory work, a pilot plant was successfully operated with a capacity of 20 tons per day, but no details have been disclosed.

Na2CO3/NaCl

O2/H2O

NH4OH (NH4)2S2O3

Ore Grinding

Autoclave

Leach

Water Treatment

CuSO4

RIP Resin

Liquid Optional Liquid/Solid Separation

Cu/(NH4)2S2O3

Copper Elution

(NH4)2S2O3

Solids

Water

Gold Elution

KOCN

Tailings Disposal Electrowinning Or Precipitation

Refining to Dore

Figure 8, Flow Sheet of Pressure Oxidation, Thiosulfate Leaching and Resin-in-Pulp Process for Refractory Gold Ores (Thomas et at, 1998; Aylmore and Muir, 2001)

Summary of Leach Conditions and Thiosulfate Consumption Most investigators have reported the ammonium thiosulfate leaching conditions used but did not mention the reagent consumption. In reality, the reagent consumption is one of major factors affected the feasibility of the process to the actual operation. Table 1 lists some of results reported by several investigators.

Ore Type S2-, Mn Zn-Pb,S2MnO2 Carbon Copper S2-, Cu Copper S2-, C Copper

Table 1, Thiosulfate Leach Conditions and Reagent Consumption S2O3222+ Cu , M pH % Au kg/t Investigators S2O3 , M NH3, M Recovery Consum. 1.21 0.57 0.02 7-9 95 4 Kerley 0.12-0.5 0.75 7-9 90 23 Block etc. 2 4 0.1 10 90 50% Zipperian 0.71 3 0.15 10.5 73 15-19% Hemmati 1-22% 1.4-8.9% 0.05-2% 93.9 40 Caixia 0.2-0.3 2-4 0.047 10-10.5 95 4.8 Cao 0.2 0.09 0.001 11 90 2 Langhans 0.1 0.1 0.005 9 51-66 5.2-8.9 Wan 0.3 3 0.01 9.7 90 17 Yen

The reagent concentration has varied considerably (0.1M - 1.21M (NH4)2S2O3, 0.09M 4M NH4OH and 0.001M - 0.15M CuSO4*5H2O, pH 7 - 11). The consumption of ammonium thiosulfate also varies (2 kg/t to 40 kg/t (NH4)2S2O3). The additive of ammonium sulfite has been used to reduce the thiosulfate consumption. Our investigation found (Yen, 2000) that sulfite may reduce thiosulfate consumption for about 10%. The ore sample used in our study as mentioned in the previous section has thiosulfate consumption of 30 kg/t (NH4)2S2O3. It required some other additives (Yen, 2000) to reduce the thiosulfate consumption to 17 kg/t (NH4)2S2O3. Further work is investigating the possibility of reducing (NH4)2S2O3 consumption to less than 10 kg/t. Gold Recovery from Thiosulfate Leach Solution There are three possible methods to recover the gold from thiosulfate leach solution: cementation, resin adsorption and solvent extraction. The cementation method can use zinc, copper and iron powders. The major difference with cyanidation is that the de-aeration is not required from the thiosulfate leach solution. Figures 9 shows an example of gold recovery by zinc precipitation. The zinc powder amount required depend not only on the gold content in the solution but also on the copper content in the solution since copper do co-precipitate with gold. In a solution containing 12 ppm Au and 0.03M Cu2+ (Figure 9), a Zn/Au ratio of 125 is required to achieve 100% gold recovery with about 40% of copper co-precipitated. As the Zn/Au ratio increases to 200 times, 100% of both gold and copper were precipitated. As the copper content in the solution was reduced to 0.005M Cu2+, only a Zn/Au ratio of 15 was required to recover 100% of the gold from the thiosulfate

solution with copper recovery at 25%. The time to recovery 100% of gold from the thiosulfate solution was only 15 minutes. The reactions of both gold and copper precipitation can be expressed by following equations: 2Au(S2O3)23+ + 2Zn 0 + 4NH3 → 2Au0 + 2S2O32- + Zn(S2O3)22- + 2Zn(NH3)42+

(11)

Cu(S2O3)22- + Zn0 → Zn(S2O3)22- + Cu0

(12)

Gold and silver could also be recovered from the thiosulfate leach solution by iron powder. It was reported (Jiang, 1990) that more than 90% of the gold and silver was reported to have been recovered from leach solutions at pH value ranging from 9.5 to 10.0 with 4 g/L iron powder at 200C. Increasing the solution temperature to 300C resulted in similar recoveries been obtained at pH 9.5 and pH 6.7. Both gold and silver recovery was only 28% at pH value of 8.1. The difference between zinc precipitation and iron precipitation is that low copper recovery (10% - 25%) was obtained in the iron precipitation process. Wan and Brierley (1997) reported that zinc cemented both gold and copper, and resulted in high zinc consumption. Also, there was the possibility of re-dissolution of cemented-gold when the zinc was not longer able to provide cathodic protection. Thus, most of pilot scale test

used copper rather than zinc for the gold recovery from the leach solution. The pilot plant test using copper cementation method was conducted at LaColoarada, Senora, Mexico (Perez and Galaviz, 1997). The heap leach studies at Newmont Gold in U.S.A. (Wan and Brierley, 1997) and the in-situ leach studies in South Africa (Li, et al, 1995) also use copper cementation process to recover the gold from thiosulfate leach solution. The factors affecting copper cementation of gold from ammoniacal thiosulfate solution has recently investigated by Guerra and Dreisinger (1999). They concluded that increased temperature and a high pH/ammonia concentration enhanced cementation performance. In the presence of sulphite and copper ions in solution however has a negative affect on the cementation. Studies carried by us (Yen, 2000) have found that the amount of copper in Cu/Au ratio of 3000 times was required to recover 98% of gold from a solution contains 15 ppm Au, 3M NH4OH, 0.21M S2O3(NH4)2, pH 9.5, in 60 minutes and at ambient temperature. The gold recovery by resin ion exchanger from thiosulfate leach solution can be achieved by using a strong base resin consisting of a quaternary amine attached to a polymer backbone (e.g. polystyrene beads). The resin-in-pulp process has been used to recovery the gold from a pressure oxidation and thiosulfate leach solution (Thomas, 1998). It was reported that higher gold recovery was obtained by using very dilute thiosulfate solutions at 45-550C and adding the resins to the pregnant slurry. Both gold and copper were loaded on resin and could be eluted separately. Firstly, the copper was eluted from the loaded resin with ammonia salt or a mixture containing 100-200 g/L ammonium thiosulfate solution. The eluate containing ammonium thiosulfate and copper (500-1500 ppm) together with about 10% of the gold on the resin was then recycled to the leaching circuit. Secondly, a thiocyanate solution (100-200g/L) was used to elute gold from the resin. The eluate was passed to a gold recovery process such as electrowinning while the resin was recycled to the RIP circuit. Extraction of gold from thiosulfate solution by solvent extraction method was reported by using alkyl phosphorous esters (Zhao, Wu and Chen, 1997) and using amine mixed with neutral donor reagents. The gold extraction by alkyl phosphorus ester increases with the increase in concentration of thiosulfate in the aqueous phase and of alkyl phosphorus esters in the organic phase respectively. The best gold recovery can be obtained at a concentration of 0.8M S2O32-. Compared with aliphatic hydrocarbon diluents, aromatic hydrocarbons test are unfavorable to the gold extraction. The study on the gold extraction by a mixture of neutral reagent and amines indicated that a combination of the primary amine N1923 and the neutral donor reagents, such as alkyl phosphorus esters (tributyl phosphate, TBP), was the most effective solvent for the extraction. The effect of N1923 and TBP in the mixed organic phase on the gold extraction depend was very much depending on the pH range. It was found that N1932 is the main extractant with TBP as a synergistic reagent at pH <9, while N1932 shows a synergistic effect on the extraction of gold with TBP at pH >9. In summary, our results have shown that zinc cementation is a better choice. However, all pilot tests used the copper cementation process. Both ion exchange and solvent extraction are still in the experimental stage and required further work.

Comparison of Cyanidation and Thiosulfate Leaching: From above descriptions, both cyanidation and thiosulfate leaching on the ore sample of Figures 1 ~ 5 can be compared and listed as below: Conditons D80 Pulp Density Reagent Consumption Additive Retention Time, hours Gold Extraction (%) Temperature Gold Recovery Solution Treatment

Cyanidation Thiosulfate Leaching 53 µm 53 µm 45 % solids 45% solids 2.00 kg/t NaCN 17.0 kg/t (NH4)2S2O3 200 g/t PbNO3 0.01M CuSO4 24 hours 24 hours 95 95 Ambient Ambient Zn ppt or Carbon Zn ppt Recycle, Destruction Recycle

When the ore pre-treated with lead nitrate followed by cyanidation, 95% of gold was extracted in 24 hours. The gold recovery from cyanide leach solution can use either zinc cementation or carbon adsorption. The leach solution can be recycled for leaching. For ammonium thiosulfate leaching, the zinc precipitation stage will regenerate the thiosulfate for recycling back to leach circuit but about 40% of copper was co-precipitated with gold and have to make up in the recycling leach solution. If the RIP process is applied to recover the gold from the leach solution, copper can be fully recovered for leaching but the thiosulfate is not regenerated. Ammonia is not consumed but some dilution may require a partial make up. Cost Estimation Based on above data, the reagent cost may be estimated Table 1, assuming both gold percentage extraction and retention time are the same. Table 1, Cost Estimation for Cyanidation and Thiosulfate Leaching Thiosulfate Leaching Cyanidation with Pb(NO3)2 Reagent Consumed Unit cost Total cost Reagent Consumed Unit cost kg/t $/kg $/t kg/t $/kg NaCN 2.0 $1.87 $3.74 Thiosulf 17.0 $0.308 ate CaO 0.056 $2.00 $0.11 NH4OH 0.36 $0.15 Pb 1.10 $0.20 $0.22 CuSO4 2.00 $2.18 Nitrate Total $4.07

Total cost $/t $5.24 $0.06 $4.36 $9.66

The calculated costs of reagent assumed no recycling of the solution for leaching. The results indicate that the cost for thiosulfate leaching is more than two times that of cyanidation.

If the thiosulfate solution was recycled, the reagent cost could be reduced by 60%, ($3.86/t for a comparable with that of cyanidation). On other hand, if an ore sample contains copper higher than 0.8%, the copper sulfate required would be greatly reduced to 0.001M CuSO4 or less (the copper consumption will be $0.22/t). Since the pre-conditioned the copper-bearing ore with ammonia, the copper tetraamine will be produced for catalysis purpose. If the thiosulfate consumption is possible to be reduced to 10 kg/t (NH4)2S2O3 and the cost of thiosulfate is reduced to $3.08, which will be comparable with cyanidation even without recycling thiosulfate solution. Conclusion: The chemistry of thiosulfate leaching is more complicate than cyanidation. The thiosulfate leaching needs three components: ammonium thiosulfate, ammonium hydroxide and cupric ion (copper sulfate). The concentration of these three components controls the gold extraction and the reagent consumption. The thiosulfate dissolves and stabilizes gold in solution, while ammonia and copper accelerate the leaching reaction. Cupric ion was found to have a strong catalytic effect on the rate of oxidation. The presence of ammonia helps to stabilize the cupric ion as a cupric tetraamine complex. Efficient dissolution is achieved by maintaining the appropriate ratio of ammonia to thiosulfate in solution so that copper can easily transfer between the cupric and cuprous state. The presence of oxygen is important to enhance the cathodic reaction. The optimum concentration of thiosulfate, ammonia and cupric ion depend on the characteristic of ore. Thus, a wide range of leaching condition was used by different investigators, i.e., 0.03M – 4.0M (NH4)4S2O3, 0.09M – 4M NH4OH and 0.001M – 0.1M Cu2+, at pH 7 – 11. The thiosulfate leaching has been applied to several mild refractory gold ores, such as, complex ore, refractory carbonaceous ores, copper-bearing gold ores, ores containing lead and zinc, gold ores associated with sulfide minerals and the gold ore associated with manganese. Most of thiosulfate studies were conducted in laboratory scale. There were five pilots and demonstration production scale. The problems encountered are: use of high concentration of reagents, corrosion of mild steel equipment, evolution of ammonia in the air, and high reagent consumption (two or three times higher than cyanide consumption). To make the thiosulfate leaching process economical, more research is required to reduce the thiosulfate consumption. The gold recovery from the leach solution is another problem in the thiosulfate process. Up to this stage, cementation is the more practical method and has been demonstrated in a pilot test. From our laboratory test results, zinc cementation was found to having advantage of faster kinetics and lower Zn/Au ratio required. But the disadvantage of zinc cementation was found to co-precipitated of gold with copper. Ion exchange and solvent extraction were being studied in the laboratory scale. But the efficiency of ion exchange resin has not been demonstrated. It looks likely there are some problem in selective adsorption and desorption. Also, solvent extraction is not a practical approach at this stage.

Acknowledgments The financial support from CANMET is gratefully appreciated. References Abbruzzese, C., Fornari, P., Massidda, R., Veglio, F., and Ubaldini, S. "Thiosulfate Leaching for Gold Hydrometallurgy"; Hydrometallurgy, 39, pp.265-276, 1995. Aylmore, M.G., and Muir, D.M, "Thiosulfate Leaching of Gold - A review"; Mineral Engineering, 2001, in press. Block-Bolten, A., Daita, M.S., Torma, A.E., and Steensma, R., "Gold and Silver Extraction from Complex Sulfide Wastes"; Proceedings for Conference on Recycle and Secondary Recovery of Metals (ed. Taylor, P.R.), Metallurgical Society, AIME, Warrendale, USA, pp. 715-726, 1985. Cao, C., Hu, J. and Gong, Q., "Leaching of Gold by Low Concentration Thiosulfate Solution"; Randol Gold Forum, Vancouver '92, Randol International, Colorado, USA, pp.293-298, 1992. Ciaxia, J. and Qiang, Y., Rare Metals, 10(4), pp.275-280, 1991. Groudev, S.N., Spasova, I.I. and Ivanov, I.M., Mining Engineering, 9(7), pp.707-713, 1996. Guerra, E. and Dreisinger, D.B., “A Study of the Factors Affecting Copper Cementation of Gold from ammoniacal Thiosulfate solution”; Hydrometallurgy, 51, pp.155-172, 1999. Hemmati, M., Hendrix, J.L., Nelson, J.H. and Milosavljevic, "Study of the Thiosulfate Leaching of Gold from Carbonaceous Ore and Quantitative Determination of Thiosulfate in the Leach Solution"; Extractive Metallurgy'89, Symposium Organized by the Institute of Mining and Metallurgy, held in London, England, July, pp.665-678, 1989. Hu Jiexue and Gong Qian, "Substitute Sulfate for Sulfite during Extraction of gold by Thiosulfate Solution"; Randol Symposium, Cains'91, pp.333-336, 1991. Jian Tao, "Gold Extraction Technology on Copper-bearing Ore and Its Theoretical Research", Ph.D. Thesis, Central South Technological University, 1990. Kerley, B. Jr. and Bernard J., "Recovery of Precious Metals from Difficult Ores"; U.S. Patent 4,269,622, 1981. Kerley, B. Jr. and Bernard, J., "Recovery of Precious Metals from Difficult Ores"; U.S.Patent 4,369,061. Hiskey, J.B. and Atluri, V.P., "Dissolution Chemistry of Gold and Silver in Different Lixiviants"; Mineral Processing and Extractive Metallurgy Review, vol.4, pp.95-134, 1988.

Langhans, Jr. J.W., Lei, K.P.V. and Carnahan, T.G., "Copper-Catalyzed Thiosulfate Leaching of Low-Grade Gold Ores"; Hydrometallurgy, 29, pp.191-193, 1992. Li, Y. and Zhang, W., “Process Development for ammoniacal thiosulfate leaching of gold and silver from complex sulphide concentrates containing precious metals”, Nonferrous Metals (in Chinese), Vol. 41, pp. 67-71, 1989. Li, J., Miller, J.D., Wan, R.Y. and Le Vier, M., “The ammoniacal Thiosulfate System for Precious Metal Recovery”; Proceedings of the XIX International Mineral Processing Congress, Vol. 4, Chapter 7, pp.37-42. 1995. Marchant, B. and Broughton, “Comparative Analysis of Cyanidation and Acidothioureaction”; 89th Annual General Meetings of CIM, Toronto, Paper No. 5, 1987. Murthy, D.S.R., Trans. Ind. Inst. Met.; 44(5), PP.349-354, 1991. Murthy, D.S.R. and Prasad, P.M., Hydrometallurgy, 42, pp.27-33, 1996. Perez, A.K. and Garaviz, H.D., "Method for Recovery of Precious Metals from Difficult Ores with Copper-Ammonium thiosulfate"; U.S. Patent 4,654,078, 1987. Rabai,G. and Epstein, I.R., "Equilibria and kinetics of the fast interaction between copper (II) and thiosulfate ions in aqueous solution", Inorganic Chemistry, 31, pp. 3139-3242, 1992. Sparrow, J. and Woodcock, J.T., Mineral Processing and Extractive Metallurgy Reviews, 14, pp.193-247, 1995. Thomas, K.G., Fleming, C., Marchbank, A.R. and Dreisinger, D., "Gold Recovery from Refractory Carbonaceous Ores by Pressure Oxidation, Thiosulfate Leaching and Resin-in-Pulp", U.S. Patent 5,785,736, 1998. Toykodi, R.J., "In praise of thiosulfate", J. Chemical Education, 68, pp. 146-149, 1990. Tozawa, K., Inui, Y. and Unetsu, Y., "Dissolution of Gold in Ammonium Thiosulfate Solution", 11th Annual Meeting of AIME, Chicago, Illinois, TMS paper #A81-25, 1981. Wan, R.Y. and Brierley, J.A., Mining Engineering, 76, pp.76-80, 1997. Wan, R.Y., Le Vier, M. and Clayton, "Hydrometallurgical Process for the Recovery of Precious Metal Values from Precious Metal Ores with Thiosulfate Lixiviant"; U.S. Patent 5,354,359, 1994. Wan, R.Y., "Importance of Solution Chemistry for Thiosulfate Leaching of Gold"; Proceedings of the World Gold'97, Singapore, Society of Mining, Metallurgy and Exploration, Inc., Littleton, U.S.A., pp.159-162, 1997.

Yen, W.T., Progress report on “Gold extraction by ammonium thiosulfate from a gold copper ore”, PWGSC Contract No. 23440-9-1019/001/SQ, April 31, 2000. Yen, W.T., Stogran, K. and Fujita, T., “Gold Extraction from a Copper Bearing Ore by Thiosulfate Leaching”; Resources Treatment Technology, Japan, 43(2), pp.83-87, 1996. Yen, W.T., Aghamirian, M., Deschenes, G. and Theben, S., "Gold Extraction from Mild Refractory Ore Using Ammonium Thiosulfate"; International Symposium on Gold Recovery, Montreal, May 3-6, CIM, 1998. Yen, W.T., Guo, H. and Deschenes, G., "Development in Percolation Leaching with Ammonium Thiosulfate for Gold Extraction of a Mild Refractory Ore"; Proceedings Symposium on Leaching: Theory, Process Development and Industrial Practice; TMS Annual Meetings, San Diego, USA, Feb 28-March 4, pp.441-455, 1999. Zhao, Jin, Wu, Zhichun and Chen, Chia-yung, “Extraction of Gold from Thiosulfate Solutions using Amine Mixed with Neutral Donor reagents”; Hydrometallurgy, 48, pp. 133-144, 1997. Zhao, Jin, Wu, Zhichun and Chen, Chia-yung, “Extraction of Gold from Thiosulfate Solutions with Alkyl Phosphorus Esters”; Hydrometallurgy, 46, pp. 363-372, 1997. Zipperian, D. and Raghavan, S., "Gold and Silver Extraction by Ammoniacal Thiosulfate Leaching from Rhyolite Ore"; Hydrometallurgy, 19, pp.361-375, 1988.