MAHLANGU, T., GUDYANGA, F.P., and SIMBI, D.J.

Leaching of the arsenopyrite/pyrite flotation concentrates using metallic iron in a hydrochloric acid medium. Hydrometallurgy Conference 2009, The Southern African Institute of Mining and Metallurgy, 2009.

Leaching of the arsenopyrite/pyrite flotation concentrates using metallic iron in a hydrochloric acid medium
T. MAHLANGU*, F.P. GUDYANGA, and D.J. SIMBI
*Department of Materials Science and Metallurgical Engineering, University of Pretoria, Pretoria, South Africa Department of Metallurgical Engineering, University of Zimbabwe, Harare, Zimbabwe

This paper describes experimental investigations into the reductive decomposition of arsenopyrite/pyrite (FeAsS/FeS2), a gold and silver bearing iron/arsenic sulphide using metallic iron powder. The main objective was to establish and experimentally confirm the thermodynamic and kinetic feasibility of reductively decomposing the sulphide matrix for precious metals recovery. In this context, cyanidation leach tests were conducted on the reductive leach residues. The arsenopyrite/pyrite flotation concentrate, mainly studied in the hydrochloric acid medium, decomposed through the nonoxidative chemical dissolution reaction and the reductive decomposition reaction for the arsenopyrite and pyrite components, respectively. Desulphurization levels below 65% were achieved at pH values below 0.15 and iron to concentrate ratios above 1, and the system was characterized by overall very slow kinetics. The reaction system had a direct linear relationship with iron to concentrate ratio and an inverse relationship with pH. An analysis of the pyrite/iron galvanic system showed that pyrite forms a partially inert cathodic surface on which the anodic dissolution of iron occurs, supported by the hydrogen evolution reaction. This phenomenon explained the low desulphurization levels and the mineral decomposition seemed to be restricted to the nonoxidative chemical dissolution reaction for the arsenopyrite component. The cyanide leach of the reductive leach residues showed very little improvement in gold recovery. The reductive leach process for arsenopyrite/pyrite has considerable limitations in terms of both desulphurization and precious metals liberation. Keywords: Pyrite, arsenopyrite reductive leaching, hydrogen sulphide, invisible gold
LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES 1

Introduction A significant fraction of gold in the gold ores that are mined in central and southern parts of Zimbabwe occurs in submicroscopic form as finely disseminated particles or in solid solution with the refractory sulphide host minerals, namely, arsenopyrite (FeAsS), pyrite (FeS2) and stibnite (Sb2S3) (Kadenhe and Makande, 1987; Makande, 1988; Swash, 1988; Casparini, 1983; Husein, 1981). Apart from the conventional roasting pretreatment process prior to cyanidation, hydrometallurgical pretreatment processes can be used as alternatives. The hydrogen sulphide route for the decomposition of sulphides has been widely studied in two main areas of non-oxidative dissolution (NOD) (Cservanyak, 1994; Nicol and Scott, 1979; Scott and Nicol, 1977; Awakara et al., 1980; Ingraham et al., 1972) and the reductive decomposition in acidified aqueous medium (House and Kelsall, 1985; Majima et al., 1981; Majima et al., 1985; Gudyanga et al., 1999a , Chifamba, 1996; Gudyanga et al., 1999b; House, 1986). The sulphides decompose to give either the elemental metal or metal sulphide of lower oxidation state with hydrogen sulphide evolution (Kolodzeij and Adamski, 1990; Cservanyak, 1994; Chifamba, 1996) according to the general reactions [1] and [2]: [1] [2] Iron, the proposed reductant, is a first order transitional element with an [Ar]3d64s2 outer shell electronic configuration and is able to exhibit multiple oxidation states of 0, +2 and +3 (Mackay and Mackay, 1986). The chemistry and electrochemistry of iron have been widely investigated and reported in corrosion science and engineering (Wranglen, 1985; Shreir et al., 1995a; Shreir et al., 1995b). In hydrometallurgical leaching systems, iron and its dissolved species (Fe(II) and Fe(III) ionic species) have been discussed by correlating thermodynamic predictions in the form of potentialpH and speciation diagrams, and stability constants for various complexes (Seon-Hyo et al., 1986). Seon-Hyo et al. (1986), reported two stable ferrous chlorocomplexes formed according to reaction [3]: [3] where n is the stability constant of the ferrous chlorocomplexes and n = 1, 2, 3 and 4; the number of chloroligands in each complex. When n is equal to unity and two respectively, the o stability constants for the two complexes FeCl+ and FeCl2 were determined as 2.2908 and 1.0965 respectively (Seon-Hyo et al. , 1986). It was also shown that at chloride ion concentration less than 2M, the Fe2+ species predominates over both the FeCl+ and FeCl2 complexes while the latter (FeCl2) becomes dominant at concentrations in excess of 2.5M. Metallic iron is a strong reducing agent as indicated by the redox potential [4]: [4] [5] The conditions under which reductive and chemical decomposition of sulphide minerals occur are also favourable to the chemical dissolution reaction [5] of iron. The hydrogen evolution side reaction does not take into consideration the effect of the anions present. The
2 HYDROMETALLURGY CONFERENCE 2009

Gibbs free energy change calculations using data from Bard et al. (1985), clearly shows the effect of the hydrochloric acid systems [6]. [6] On the basis of thermodynamic calculations, iron dissolution in chloride systems appears to be more favourable than in non-chloride systems. This has a negative effect on the leach system because the iron dissolution side reaction deprives the reductive leach reaction of any available iron. On the other hand, the Fe(II)/Fe redox system, as a leaching agent in sulphide concentrate leaching, is not expected to form insoluble compounds with the reaction products in the chloride medium. Instead, the formed chlorocomplexes expectedly result in more negative potential conditions that are thermodynamically favourable to the reductive leach process. These thermodynamic predictions indicate a need for critical control of both acid concentration and the iron to concentrate ratio during reductive decomposition of sulphide minerals. The general reactions governing the reductive decomposition of arsenopyrite, pyrite and other iron sulphides have been determined by Majima and Awakura (1979). Although the pyrite decomposition reaction is often classified under the reductive decomposition system, the reaction exhibits non-oxidative dissolution (NOD) of the mineral when considered in terms of iron and sulphur oxidation states. This is primarily because neither iron nor sulphur seems to undergo any change in their original oxidation states. It would be expected that pyrite decomposition follows reduction to either stoichiometric pyrrhotite [7] or elemental iron [8] followed by the chemical dissolution of the two products [9] and [10] respectively. However, strictly speaking, in reaction [8], iron is oxidized by hydrogen ions. [7] [8] [9] [10] Reaction [7] has not been experimentally proven to occur in acidic aqueous systems (Cservanyak, 1994). Also from the potentialpH diagram (Figure 1), pyrite decomposition to elemental iron and hydrogen sulphide is feasible only at potentials below -0.41V vs SHE. The ferrous ions are the main products at any reducing potentials above -0.41 V vs SHE depending on the working pH. Elsewhere (Holdich and Broadbent, 1985; Koch, 1975; Peters and Majima, 1968), ferrous ions were detected as the principal reaction products instead of pyrrhotite and elemental iron. Czevanyak (1994) in discussing the work of Peters and Majima (1968) reported that hydrogen sulphide evolution always occurred after hydrogen evolution has already started. This scenario clearly indicates that the process efficiency of reductive decomposition will be adversely affected by any pH reduction. From the preceding discussion and available thermodynamic data, the application of elemental iron in the reductive decomposition of pyrite can be inferred. Bourgeois et al. (1979) as referenced by Czevanyak (1994), were the first to investigate elemental iron driven dissolution of pyrite. Pyrite and elemental iron, form a galvanic couple with iron being oxidized to ferrous ions and pyrite reduced to ferrous ions with the evolution of hydrogen sulphide [11]. [11]
LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES 3

Figure 1. PotentialpH diagram for the FeSH2O system. Iron activity = 0.001, sulphur = 0.001 (Stabcal software)

The reductive decomposition of arsenopyrite, which occurs together with pyrite in typical refractory gold ores (Swash, 1988; Yannopoulos, 1990; Chifamba, 1996), can be discussed under the same principles as that of pyrite. The half reaction, representing the decomposition of arsenopyrite to elemental arsenic is shown in reactions [12] and [13]. [12] [13] The leaching system will also depend on the solution redox potential. The work of Majima and Peters (1968) gave the following findings: Polarizing a pyrite electrode in 1M HClO4, they determined a hydrogen overpotential of 0.26V; In constant potential experiments, monitoring current and the concentration of H2S evolved, the current efficiency of H2S evolution varied with potential from 34% (at -0.2 V vs SHE) to 20% (at -0.4 V vs SHE); Varying the acid concentration from 0.1 to 4M at -0.3 V vs SHE, hydrogen evolution rate varied linearly with acid concentration and the reaction order for H2S > 1. The preceeding findings limit not only the working pH range but also the potential. Gudyanga et al. (1999b), using Cr(II) ionic species as the reductant, measured the solution redox potential during reductive decomposition of an arsenopyrite/pyrite concentrate and found a rapid increase in potential with time on reducing pH from 0.43 to 0.2. These results also agree with the findings of Klein and Sluvey (1978) who concluded that the hydrogen evolution reaction dominates above pH 0. The present work extends the reductive decomposition and/or dissolution of a gold bearing arsenopyrite/pyrite flotation concentrate using metallic iron in a chloride medium. The work focuses on the factors that enhance desulphurization and the recovery of precious metals, namely gold and silver from the selected sulphide concentrate. The application of iron in the reductive decomposition of arsenopyrite/pyrite flotation concentrate is envisaged to provide an alternative pretreatment method that addresses the technological and environmental limitations that are characteristic of pyrometallurgical routes while enhancing precious metals recovery.
4 HYDROMETALLURGY CONFERENCE 2009

Experimental procedure Materials Flotation concentrate samples The reductive decomposition experiments were conducted on an arsenopyrite/pyrite flotation concentrate collected as a blend sample from the Kwekwe Roasting Plant, Zimbabwe. The mineralogical and chemical composition of the material are tabulated in Tables I and II. A bulk sample of iron shavings was collected, washed, dried, screened to 100% minus 425 m and then stored under dry and oxygen free conditions in order to minimize the oxidation of the iron particles surfaces. All leaching solutions were prepared using analytical grade reagents of hydrochloric acid, ferrous sulphate hepta-hydrate and potassium dichromate diluted with distilled water and where necessary, deoxygenated with high purity nitrogen. The experimental setup is detailed in Mahlangu et al. (2006). Reductive leaching The flotation concentrate was pulped in 500 ml water together with the predetermined mass of iron shavings and purged with a steady stream of nitrogen gas for a period of 4560 minutes. Nitrogen gas purging was used for the removal of oxygen prior to the addition of hydrochloric acid medium. The pH was monitored by periodic withdrawal of 10 ml samples from the reactor, cooling, recording of pH and then reintroducing the sample back into the reactor. Effluent hydrogen sulphide was scrubbed through a 1.8 litre hydrostatic column of hydrochloric (HCl) acidified potassium dichromate solution. The leaching/decomposition reaction was quantitatively followed by a redox titration of residual dichromate ions in the column with ferrous ions (Mahlangu et al. 2006). Sodium diphynlamine was used as an indicator with the colour changing from pale green to purple on reaching the endpoint. The residues and filtrates were then analysed for Au, Ag, Fe, As, Sb, Pb and S. Cyanidation leaching The reductive leach residue was thoroughly washed with distilled water and cyanide leached in a 0.23% NaCN solution for 48 hours at pH between 11 and 12. The solution pH was adjusted using technical grade sodium hydroxide. The leach residues and filtrates were then analysed for Au and Ag.

Table I Mineralogical composition of the FeAsS/FeS2 flotation concentrate FeS2 (%) 58.7 FeAsS (%) 27.2 PbS (%) 0.1 CuFeS2 (%) 0.3 ZnS (%) 1.8 Sb2S3 (%) 0.8 Other (%) 11.1

Table II Chemical composition of the FeAsS/FeS2 flotation concentrate Fe (%) 36.7 As (%) 12.5 Pb (%) 0.1 Cu (%) 0.1 Zn (%) 1.2 Sb (%) 0.6 S (%) 28.7 Au (g/t) 64.7 Ag (g/t) 43.1 Other (%) 20.2

LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES

Results and discussion Effect of pH on the decomposition of FeAsS/FeS2 The data in Figure 2 show the variation of arsenopyrite/pyrite decomposition with pH at an iron to concentrate ratio of 0 and set leaching temperature of 105C. The results exhibit an inverse near linear relationship between arsenopyrite/pyrite decomposition and pH. In the absence of iron, arsenopyrite decomposes according to the chemical nonoxidative dissolution reaction [13]. Under these conditions, pyrite is expected to remain inert as would be expected from the examination of the FeSH2O potentialpH diagram (Figure 1). [14] In this regard, the sulphide mineral nonoxidative dissolution, observed in this work, was mainly due to the dissolution of the arsenopyrite component. The results also exhibit very low levels of desulphurization (< 16%) associated with the chemical nonoxidative or direct acid leaching of the concentrate. This is despite the relatively large negative free energy change value calculated for the system [13]. The introduction of iron, at an iron to concentrate ratio of 0.16, while working in the same pH range, gave results shown in Figure 3. An inverse linear relationship between sulphide minerals decomposition and pH was observed, with very little improvement in the extent of reaction. The results of further arsenopyrite/pyrite reductive decomposition experiments at progressively increasing iron to concentrate ratios are graphically presented in Figures 3(b), (c) and (d). These tests were conducted in the same pH range and temperature. The iron to concentrate ratio of 0.32 corresponds to the calculated stoichiometric requirement of iron in the system. In all the cases, the inverse near linear relationship between arsenopyrite/pyrite decomposition and pH continue to be exhibited. However, a further pH reduction to values below 0.25, marginal increases in arsenopyrite/pyrite decomposition are observed. The results also show a sustained increase in the sulphide minerals decomposition with an increase in the iron to concentrate ratio. There is a very strong influence of iron to concentrate ratio demonstrated at pH values equal to or greater than pH 0.25. In order to understand and explain these phenomena, there is need for the examination of the chemistry and electrochemistry of the two minerals in acidified chloride systems.

Figure 2. Variation of %sulphur leached arsenopyrite/pyrite with pH at different leaching times (iron/concentrate ratio = 0; temperature = 105C)

HYDROMETALLURGY CONFERENCE 2009

Figure 3. Variation of %sulphur leached from arsenopyrite/pyrite with pH at different leaching times (ron/concentrate ratio: (a) = 0.16; (b) = 0.32; (c) = 0.64; (d) = 0.96; temperature = 105C)

In the reductive decomposition of pyrite and arsenopyrite, the reactions governing the leach systems have been calculated and their thermodynamic feasibility demonstrated in reactions [11] and [14]. Pyrite decomposes to ferrous ions with the evolution of hydrogen sulphide, whereas arsenopyrite decomposes both by chemical nonoxidative and reductive dissolution reactions, giving the same reaction products as in pyrite. [15] [16]

[17] The resultant elemental iron from the arsenopyrite decomposition reaction [17] subsequently dissolves either by supporting the sulphide mineral decomposition reactions [15 or 16] or by the hydrogen evolution side reaction [18]. [18] The involvement of the resultant iron in the reductive leach reaction has a complementary effect on the leach process, whereas the latter negatively affects the process efficiency. Thermodynamic calculations, as indicated by the free energy changes for arsenopyrite and pyrite decomposition reactions, predict a much more feasible process than illustrated by the results in Figures 2 to 3. This is in spite of the fact that the FeClSH2O potentialpH diagram (Figure 4) shows that in the presence of the chloride ions, ferrous ions and hydrogen, sulphide become more stable relative to pyrite. The differences are more evident when compared to the FeSH2O system (Figure 1).
LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES 7

Figure 4. The FeClSH2O Eh pH diagram ([Fe] = 0.001M; [S] = 0.01M; [Cl] = 0.01M) (Stabcal, Software)

Table III Redox reactions and Eo values arsenopyrite, pyrite and iron systems in acidic conditions [Standard temperature = 25C and pressure] G kJ/mol +20.91 +16.44 +78.87
o o

Redox reaction FeS2 + 4H+ + 2e- = Fe2+ + H2S E/V vs SHE = -0.108 0.118pH 0.0296 log (Fe2+)(H2S)2 FeAsS + 2H+ + 2e- = Fe + As + H2S E/V vs SHE = -0.085 0.0591pH 0.0296 log (H2S) Fe2+ + 2e- = Fe E/V vs SHE = -0.409 0.0296 log (Fe2+)
o o

E /V vs SHE

-0.108 -0.085 -0.409

Consideration of the galvanic interactions or the galvanic cell formed by arsenopyrite and/or pyrite with iron could possibly explain these low levels of desulphurization. In the galvanic cell, the anodic dissolution of iron can be supported by either the cathodic dissolution of the mineral sulphide or an inert sulphide mineral surface, which promote the iron dissolution reaction with the evolution of gaseous hydrogen. The standard electrode potential values of the pyrite, arsenopyrite and iron are shown in Table III. The relative potential difference between the FeS2/Fe2+, H2S and Fe/Fe2+ (301 mV) systems hugely favours the galvanic reaction since galvanic reactions are known to proceed at potential differences of 200 mV (Jackson, 1986). However, pyrite has been known to remain inert even at potential differences of more than 400 mV in the case of pyrite/sphalerite galvanic interactions. [19] [20] In the above system sphalerite dissolution is cathodically supported by the oxygen reduction reaction [21] on the pyrite mineral particles surface. [21]
8 HYDROMETALLURGY CONFERENCE 2009

Drawing parallels with the pyrite/iron system and the pyrite/sphalerite system, it can be concluded that iron dissolution is supported by the hydrogen evolution reaction on the inert pyrite mineral particle surface. The arsenopyrite/iron redox system has a similar potential difference (324 mV) to that discussed for the pyrite/iron system (301 mV). In this regard, arsenopyrite is likely to remain partly or total inert while iron anodically dissolves on its surface supported by the hydrogen evolution reaction. It has also been demonstrated elsewhere (Peters and Majima, 1968; Klein and Shuey, 1978; Dreisinger and Abed, 2002) that the hydrogen evolution reaction will occur at same potentials at which pyrite reduction takes place. They also found out that hydrogen sulphide evolution commenced only after hydrogen evolution had begun. This possibly explains the low levels of arsenopyrite/pyrite decomposition. While thermodynamics could predict the feasibility of any reaction system, kinetic feasibility can be only experimentally confirmed. Effect of iron to concentrate ratio on the decomposition of arsenopyrite/pyrite. Figure 5 shows the variation of arsenopyrite/pyrite decomposition with iron to concentrate ratio at pH 0.25 and set working temperature of 105C. The data exhibit a direct linear relationship between arsenopyrite/pyrite decomposition and iron to concentrate ratio. These results are typical of a reaction system directly controlled by the concentration of a single reactant (Weis, 1985). As indicated in the earlier results (Figures 2 to 3), iron to concentrate ratio seems to have a very strong influence on the arsenopyrite/pyrite decomposition relative to pH in the pH range studied. Although decomposition increases with an increase in iron to concentrate ratio, the effective desulphurization levels are significantly low (<50%). Further arsenopyrite/pyriteiron to concentrate ratio plots of the experimental results at progressively increasing pH values (Figures 6(a) to 6(d)) show similar trends to those observed in Figure 5. However, the effective dissolution levels progressively decrease in each subsequent figure. The progressive increase in the levels of mineral sulphide decomposition at increasing iron to concentrate ratio and decreasing pH was further investigated at pH values 0.1 and 0.15 and also iron to concentrate ratios of 1.28 and 1.60. The results of these tests are plotted in Figures 7 to 8 and at pH values 0.1 and 0.15, increasing the iron to concentrate ratio from 0.96 to 1.28 improves total sulphur leached. Any further increase in iron to concentrate ratio beyond 1.28 is accompanied by marginal increases in sulphur leached. At longer leaching times of

Figure 5. Variation of %sulphur leached from arsenopyrite/pyrite with iron to concentrate ratio at different leaching times (pH = 0.25; temperature = 105C)

LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES

Figure 6. Variation of %sulphur leached from arsenopyrite/pyrite with iron to concentrate ratio at different leaching times ((a) pH = 0.34; (b) pH = 0.44; (c) pH = 0.54; (d) pH = 0.62; temperature = 105C)

Figure 7. Variation of %sulphur leached from arsenopyrite/pyrite with iron to concentrate ratio at different leaching times (pH = 0.1; temperature = 105C)

300 minutes, a slight decrease in total arsenopyrite/pyrite decomposition is also observed. While the marginal increases in the sulphide mineral decomposition can be appreciable, further pH reductions coupled with the high temperature (105C) provide a highly corrosive environment that will increase both capital and operational costs of the system. Influence of arsenopyrite/pyrite decomposition on the precious metals recovery The arsenopyrite/pyrite ores generally contain significantly large fractions of gold occurring in submicroscopic form as finely disseminated particles or in solid solution (Iglesias, 1994, Swash, 1988). In these ores and concentrates of arsenopyrite/pyrite, gold often deposits along
10 HYDROMETALLURGY CONFERENCE 2009

Figure 8. Variation of %sulphur leached from arsenopyrite/pyrite with iron to concentrate ratio at different leaching times (pH = 0.15; temperature = 105C)

Table IV Sieve analysis of arsenopyrite/pyrite flotation concentrate and gold distribution (where dist = distribution) Size range (m) +212 -212+150 -150+106 -106+75 -75+53 -53 Total Weight (%) 0.4 6.4 11.0 9.6 14.0 58.6 100 Au dist (%) 3.4 10.4 6.1 14.7 65.4 100 Au assay (g/t) 35.2 62.56 41.84 69.92 121.76 66.80 Size (m) 212 150 106 75 53 Cumulative oversize (%) 0.4 6.8 17.8 27.4 41.4 Cumulative Au dist (%) 3.4 13.8 19.9 34.6

the arsenic rich grain boundaries and usually form irregular inclusions. These characteristics define the refractoriness of many arsenopyrite/pyrite ores. The study (Chifamba, 1996; Gudyanga et al., 1998; Mbewe, 1990; Gudyanga et al., 1999b) of flotation concentrates utilized in the present work, proved to be refractory to direct cyanide leach. Chifamba (1996) and Gudyanga et al. (1999b) further demonstrated that reductive decomposition of the same flotation concentrates with desulphurization values in excess of 95% achieved very low gold recovery. The reductive leach process dissolved iron and released sulphur as hydrogen sulphide but elemental or metallic arsenic remained in the residue. The residues were very refractory giving less than 30% gold extractions into the solution. In this regard, the sulphide matrix is not solely responsible for the refractory nature of the arsenopyrite/pyrite flotation concentrates. In fact, subsequent oxidation of the reductive leach residues realized 95% gold extraction values. These findings concurred with the earlier mineralogical results (Jha, 1987; Cook and Chryssoulis, 1990; Cabri et al., 1989). It is therefore clear that most of the gold, in the arsenopyrite/pyrite flotation concentrate exists as invisible or locked gold. In this form, total decomposition of the host mineral is required for the complete liberation and recovery of precious metals. Reductive pretreatment does not achieve the complete decomposition of the host sulphide mineral and therefore presents serious limitations as a pretreatment process. The results presented in this present work focus on gold and silver deportment in the various sieve fractions and also the response of gold extraction to desulphurization.
LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES 11

Sieve analysis and precious metal deportment in the flotation concentrate The results of the sieve analysis of the arsenopyrite/pyrite flotation concentrate and gold distribution in various size fractions are tabulated in Table IV. The results show that over 50% of the material is below 53 m and this size fraction contains more than 60% gold. Silver distribution is tabulated in Table V and over 90% silver occurs in the minus 53 m fraction. In this regard, fine grinding is inevitable to liberate both gold and silver from the flotation concentrate. Fine to ultrafine grinding has been used to treat some refractory ores but presents two problems, namely (i) high operational costs and (ii) does not necessarily liberate invisible or locked gold. In this context, chemical pretreatment methods have often taken precedence over fine to unltrafine grinding (Wills, 1997; Yannopoulos, 1990). The failure by reductive decomposition (Chifamba, 1996; Gudyanga et al. , 1999b) to liberate the invisible or locked gold necessitated a subsequent oxidative pretreatment process on the residue. Also, the work of Dunn and Chamberlain (1997) revealed that the sulphide matrix is not entirely responsible of the refractory nature of the arsenopyrite/pyrite host mineral. In fact, they realized only around 33% gold recovery after the complete removal of sulphur during pyrolysis. It therefore follows that cyanidation of the reductive leach residues gives only a measure of the refractoriness caused by the sulphide matrix.

Table V Sieve analysis of arsenopyrite/pyrite flotation concentrate and silver distribution (where dist = distribution) Size range (m) +212 -212+150 -150+106 -106+75 -75+53 -53 Total Weight (%) 0.4 6.4 11.0 9.6 14.0 58.6 100 Ag dist (%) 1.1 7.5 91.4 100 Ag assay (g/t) 5.12 24.40 71.30 45.67 Size (m) 212 150 106 75 53 Cumulative oversize (%) 0.4 6.8 17.8 27.4 41.4 Cumulative Ag dist (%) 1.1 8.6

Figure 9. Effect of desulphurization on gold extraction during cyanidation of the arsenopyrite/pyrite reductive leach residues

12

HYDROMETALLURGY CONFERENCE 2009

Effect of desulphurization on gold and silver extraction Figure 9 shows the variation of gold recovery from the arsenopyrite/pyrite reductive leach residue during cyanide leaching. The data at zero per cent sulphur removal represents direct cyanidation of the flotation concentrate. The desulphurization levels range between 14% and 63%. Gold extraction is limited to around 15% maximum at around 63% sulphur removal. It is evident that the reductive decomposition does not liberate significant amounts of gold. These results seem to concur with earlier work of Gudyanga et al. (1999b) and Chifamba (1996), working on similar flotation concentrates, achieved in excess of 90% desulphurization but the gold extraction was limited to around 27%. The recovery data for silver is presented in Figure 10. Similar patterns to those obtained in Figure 9 were observed. From the particle size analysis and precious metals deportment data (Tables IV and V), it was shown that over half of both gold and silver occur in the fine fraction, that is, minus 53 m. When this data is considered together with the extraction data (Figures 9 and 10), qualitative conclusions can be made to the effect that gold exists as invisible or locked gold and as such will require the total decomposition of the arsenopyrite/pyrite host mineral. Conclusions The feasibility of the reductive decomposition of the arsenopyrite/pyrite flotation concentrate using elemental iron has been demonstrated in the hydrochloric acid medium with the reactions governing the leach process were: Chemical non-oxidative dissolution reaction for arsenopyrite; Reductive dissolution for pyrite. The reductive leach process followed an inverse relationship with pH and also a direct relationship with iron to concentrate ratio. Desulphurization levels remained relatively low (< 65%) at very low pH (< 0.15) and high iron to concentrate ratios (> 1.5). The analysis of the pyrite/iron galvanic system revealed that pyrite acts as an inert cathode on which iron oxidation is supported by the hydrogen evolution reaction. This phenomenon satisfactorily explained the low levels of the mineral sulphide decomposition. Cyanide leaching of the reductive leach residues showed that there was very little improvement in both gold and silver extraction. Prior to reductive decomposition, gold extraction was around 5% and after 63.5% desulphurization, only 15% gold extraction was realized. In this context, the sulphide matrix was not solely responsible for the refractoriness of the host mineral.

Figure 10. Effect of desulphurization on silver extraction during cyanidation of the arsenopyrite/pyrite reductive leach residues

LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES

13

 Gold and silver exist as invisible or locked gold and as such total dissolution of the arsenopyrite/pyrite flotation concentrate was necessary in order to access the precious metals. The limitations of the reductive leach process both in terms of desulphurization and precious metals recovery makes the whole process less attractive. There is need for coupling it with subsequent processes that will totally dissolve the host mineral and increase the precious metals recovery. References AWAKARA, Y., KAWEI, S., and MAJIMA, H. Kinetic study of nonoxidative dissolution of galena in aqueous acid solution. Met. Trans. B, vol. 11B, 1980. pp. 377381. BARD, A.J., PARSONS, R., and JORDAN, J. Standard potentials in aqueous solutions, Marcel Dekker, New York. 1985. BOURGEOIS, J.P., AUPAIX, N., BLOISE, R., and MILLET, J.L. Proposition DExplication de la Formation DHydrogene Sulfure dans les Stockages Souterrains de Gaz Naturel Par Reduction des Sulfures Mineraux de la Roche Magasin, Revae de lInst. Francais du Petrole, vol. 34, no. 3, 1979, pp. 371386. CABRI, L.J., CHRYSSOULIS, S.L., DE VILLERS, J.P.R., LAFLAMME, J.H.G., and BUSEK, P.R. The nature of invisible gold in arsenopyrite, The Canadian Mineralogist, 27, 1990. 353 pp. CASPARINI, C. The mineralogy of gold and its significance in metal extraction. C.I.M. Bull. March (1983), 1989. 1983. pp. 145153. CHIFAMBA, J. The reductive decomposition of refractory sulphide concentrates for the recovery of precious metals, gold and silver. MPhil Thesis, University of Zimbabwe 1996. COOK, N.J. and CHRYSSOULIS, S.L. Concentrations of invisible gold in common sulphides, The Canadian Mineralogist, vol. 28, 1990. p. 1. CSERVNYAK, I. Electrochemical reduction of pyrite in acidic aqueous electrolytes. PhD Thesis, University of London, 1994. DREISINGER, D. and ABED, N. 2002. A fundamental study of the reductive leaching of chalcopyrite using metallic iron part I: Kinetic analysis. Hydrometallurgy 66, vol. 13, 2002. pp. 3757. DUNN, J.D. and CHAMBERLAIN, A.C. The recovery of gold from refractory concentrates by pyrolysisoxidation, Extended Abstracts, Complex Ores 97, an International Symposium on the Processing of Complex and Refractory Ores, Bulawayo, Zimbabwe March 1997. pp. 810. GUDYANGA, F.P., MAHLANGU, T., CHIFAMBA, J., and SIMBI, D.J. Reductiveoxidative pre-treatment of a stibnite flotation concentrate: Thermodynamic and kinetic considerations. Minerals Engineering, vol. 11, no. 6, 1998. pp. 563580. GUDYANGA, F.P., MAHLANGU, T., CHIFAMBA, J., and SIMBI, D.J. 1999a. Reductive decomposition of galena (PbS) using Cr(II) ionic species in an aqueous chloride medium for silver (Ag) recovery. Minerals Engineering, vol. 12, no. 71999. pp. 787797. GUDYANGA, F.P., MAHLANGU, T., CHIFAMBA, J., and SIMBI, D.J., 1999b. Sequentialreductive pretreatment of an arsenopyrite/pyrite flotation concentrate. Paper presented at the Minerals Engineering 99 Conference, Falmouth, England, September 1999. HOLDICH, R.S. and BROADBENT, C.P. Investigating the dissolution of pyrite in copper(II) chloride solutions, Extractive Metallurgy 85. IMM, London, 1985. pp. 645658.
14 HYDROMETALLURGY CONFERENCE 2009

HOUSE, C.I. 1986. The electrogeneration of Cr(II) and V(II) solutions and the hydrometallurgical reduction of SnO2, PbS and CuFeS2. PhD Thesis, University of London. 1986. HOUSE, C.I. and KELSALL, G.H. Hydrometallurgical reduction of SnO2, CuFeS2 and PbS by electrogenerated Cr(II) and V(II) solutions, Extractive Metallurgy 85, IMM, London, 1985. pp. 685692. HUSEIN, D.M. Process mineralogy of auriferous pyritic ores at Carlin, Nevada. Process Mineralogy, Metal Extraction, Mineral Exploration and Energy Materials. New York: AIME. 1981. pp. 271289. IGLESIAS, N. and CARRAZA, F. Refractory gold ores; A review of treatment methods and recent advances in Biotechnological techniques, Hydrometallurgy 34, 1994. pp. 383395. INGRAHAM, T.R., PARSONS, H.W., and CABRI, L.J. Leaching of pyrrhotite with hydrochloric acid. Can. Met. Quarterly, vol. 11, no. 2, 1972. pp. 425434. JACKSON, E. Hydrometallurgical Extraction and Reclamation. John Wiley and Sons, New York. 1986. JHA, M.C. Refractoriness of certain gold ores to cyanidation: Probable causes and possible solutions. Mineral Processing and Extractive Metallurgy Review, vol. 2, 1987. pp. 331352. KADENHE, R.M. and MAKANDE, E.S. Review of the roasting plant operations and services available to small mines at Kwekwe, Zimbabwe, African Mining, IMM, London. 1987. KLEIN, J.D. and SLUVEY, R.T. Nonlinear Impedence of MineralElectrolyte Interfaces; Part I., Pyrite, Geophysics, vol. 43, no. 6, 1978. pp. 12221234. KOCH, D.F.A. Electrochemistry of sulphide minerals, Modern Aspects of Electrochemistry. vol. 10, J.O.M. Brochris and B.E. Conway (eds.), 1975. pp. 211237. KOLODZEIJ, B. and ADAMSKI, Z. Dissolution of sphalerite in aqueous hydrochloric acid solutions under reducing conditions. Hydrometallurgy, vol. 24, no. 3, 1990. pp. 393406. MACKAY, K.M. and MACKAY, R.A. Introduction to Modern Inorganic Chemistry, 3rd Edition. International Textbook Company. 1986. MAHLANGU, T., GUDYANGA, F.P., and SIMBI, D.J. Reductive leaching of stibnite using metallic iron in a hydrochloric acid medium I: Thermodynamics. Hydrometallurgy, vol. 84, no. 34, 2006. pp. 192203. MAJIMA, H. and AWAKARA, Y. Nonoxidative leaching of base metal sulphide ore, in 13th Int. Min. Proc. Congress, Warsaw. J. Laskowski (ed.), 1979. pp. 936956. MAJIMA, H., AWAKARA, Y., and MISAKI, N. A kinetic study on nonoxidative dissolution of sphalerite in aqueous hydrochloric acid solutions. Met. Trans. B. 12B, 1981. pp 645649. MAJIMA, H., AWAKURA, Y., and MASAKI, N. Leaching of oxides and sulphides in acidic chloride media. Extractive Metallurgy 85, London 912 September, 1985. IMM, pp. 607627. MAKANDE, E.S. Roasting and cyanide treatment of arsenical and antimonical gold concentrates and residues at the Roasting Plant, Kwekwe, Zimbabwe, Perth International Gold Conference, RANDOL. 1988. MBEWE, K. Improving the recovery of silver in the pressure oxidation leach for the refractory gold ores, BSc Eng. (Hons) Project Report, Department of Metallurgy, University of Zimbabwe. 1990. NICOL, M.J. and SCOTT, P.D. The kinetics and mechanism of the nonoxidative dissolution of iron sulphides in aqueous acidic solutions. J. South Afr. Inst. Min. Met. 1979, pp. 298305.
LEACHING OF THE ARSENOPYRITE/PYRITE FLOTATION CONCENTRATES 15

PETERS, E. and MAJIMA, H. Electrochemical reactions of pyrite in acid perchlorite solutions. Can. Met. Quart, vol. 7, no. 3, 1968. pp. 111117. SCOTT, P.D. and NICOL, M.H. The kinetics and mechanisms of nonoxidative dissolution of metal sulphides. Trends in Electrochemistry, J.O.M. Bockris, D.A.J. Rand, and B.J. Welch (eds.), Plenum Press, New York and London, 1977. pp. 303316. SEON-HYO, K., HENEIN, H., and WARREN, G.W. An investigation of the thermodynamics and kinetics of the ferric chloride leaching of galena concentrate. Metallurgical Transactions, 17B, 1986. pp. 2939. SHREIR, L.L., JARMAN, R.A., and BURSTEIN, G.T. Corrosion 1; Metal/Environment Reactions. vol. 1. Butterworth Heinemann. 1995a SHREIR, L.L., JARMAN, R.A., and BURSTEIN, G.T. Corrosion 2; Corrosion control. vol. 2. Butterworth Heinemann. 1995b. SWASH, P.M. A mineralogical investigation of refractory gold ores and their beneficiation, with special reference to arsenical ores. J. S. Afri. Inst. Min. Metall. vol. 88, no. 5. 1988. pp. 173180. WEISS, N.L. SME Mineral Processing Handbook; vol. 2. Society of Mining Engineers of the Americal Institute of Mining, Metallurgical, and Petroleum Engineers, Inc., New York. 1985. pp. 1324. WILLS, B.A. 1997. Mineral Processing Technology: An introduction to the practical aspects of ore treatment and mineral recovery. Sixth Edition. Butterworth Heinemann. 1997. pp. 1113. Wranglen, G. An Introduction to Corrosion and Protection of Metals. Chapman and Hall 1985. 260 pp. YANNOPOULOS, J.C. The Extractive Metallurgy of Ggold, Van Nostrand Reinhold, New York, 1990. pp. 79110.

Thamsanqa Mahlangu Senior Lecturer, University of Pretoria, South Africa Thamsanqa is a Member of the Southern Africa Institute of Mining and Metallurgy (SAIMM), he is also registered as a Professional Engineering (Pr Eng) with the Engineering Council of South Africa (ECSA). August 2004 to dateSenior Lecturer, Department of Materials Science and Metallurgical Engineering, University of Pretoria. June 2003 to May 2004)Post Doctoral Fellow, Mineral Processing Research Unit, Department of Chemical Engineering, University of Cape Town March 1995 to May 2003Research Fellow/Lecturer, Department of Metallurgical Engineering, University of Zimbabwe March 1994 to February 1995Minerals Process Research Engineer, Institute of Mining Research, University of Zimbabwe. My research focus is on the solution purification by both solvent extraction and precipitation techniques. I have also been involved in the leaching of refractory gold ores as well as the general hydrometallurgical processes of leaching and metals recovery.
16 HYDROMETALLURGY CONFERENCE 2009