Review article

Leaching and recovery of gold using ammoniacal

thiosulfate leach liquors (a review)

Andrew C. Grosse, Greg W. Dicinoski, Matthew J. Shaw, Paul R. Haddad*

Australian Centre for Research on Separation Science, School of Chemistry, University of Tasmania, GPO Box 252-75, Tasmania, Australia

Received 30 November 2001; received in revised form 26 July 2002; accepted 9 September 2002

Abstract

A review is presented summarising the leaching of gold with ammoniacal thiosulfate solutions, and evaluating the current use

and development of ion exchange resins for the recovery of gold and silver from such leach liquors. Comparisons are also made

with other recovery processes, including carbon adsorption, solvent extraction, electrowinning and precipitation. Thiosulfate

leaching chemistry is compared with cyanide leaching, and the problems associated with obtaining a high yield of recovered gold

using the former process are discussed. The present limitations of using Resin-in-Pulp (RIP) and Resin-in-Leach (RIL) systems

with thiosulfate liquors are indicated and possible solutions discussed.

D 2002 Elsevier Science B.V. All rights reserved.

Keywords: Gold; Silver; Recovery; Ammoniacal thiosulfate leaching; Ion exchange resins; Polymer adsorbents

1. Introduction

The current use of cyanidation techniques to leach

gold from its various ores is undesirable from an

environmental perspective. In recent years, cyanide

leaching has been banned in many regions due to

environmental concerns. This is due predominantly

to the acute toxicity of cyanide, effectively demon-

strated by the recent Baia Mare tailings dam breach

and subsequent pollution of the Tisza river in Roma-

nia. In addition, problems can occur when leaching

gold from complex ores, such as those containing

copper or carbonaceous material (Marsden and House,

1992), resulting in poor recoveries of gold unless

multi-step extraction and/or elution steps are incorpo-

rated into the process. The use of alternative lixiviants

for gold extraction has been investigated for some

time, and has been discussed in numerous reviews (e.g.

Bhappu, 1990; Fleming, 1998; McNulty, 2001; Spar-

row and Woodcock, 1995; Wan et al., 1993). A

suitable replacement for cyanide may be the use of

thiosulfate in the presence of ammonia and copper(II).

The advantages of using this approach include a far

lower possibility of an adverse environmental impact,

a more cost-effective operation due to the fact that

thiosulfate is substantially less expensive than cyanide,

and thiosulfate facilitates the leaching of complex

materials including manganiferous ores through matrix

degradation. In addition, the thiosulfate anion is a more

effective lixiviant of preg-robbing and high-copper

ores through heap-leaching than cyanide (Ferron et

al., 1998; Langhans et al., 1992; Wan, 1997; Wan and

0304-386X/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.

doi:10.1016/S0304-386X(02)00169-X

* Corresponding author. Tel.: +61-3-6226-2179; fax: +61-3-

6226-2858.

E-mail address: Paul.Haddad@utas.edu.au (P.R. Haddad).

www.elsevier.com/locate/hydromet

Hydrometallurgy 69 (2003) 1–21

Brierley, 1997). However, there are presently certain

problems associated with using thiosulfate to leach

gold and no suitably robust leaching regime exists.

Moreover, the actual leaching mechanism is still not

fully understood, due predominantly to the ease with

which thiosulfate undergoes disproportionation or oxi-

dation in aqueous solutions, forming various other

sulfur species including sulfite, sulfate, di-, tri- and

higher polythionates, as a function of pH and Eh

(Mizoguchi et al., 1976; Valensi, 1973; Webster,

1986; Zipperian et al., 1988). Furthermore, the met-

allo-thiosulfate complexes themselves are readily sus-

ceptible to decomposition to produce metallic sulfides

or other species (Benedetti and Boulegue, 1991).

Nevertheless, the potential benefits of using thiosulfate

as a lixiviant for gold have generated significant

interest worldwide. A recent comprehensive review

of leaching gold ores using the thiosulfate anion has

been prepared by Aylmore and Muir (Aylmore, 2001).

The present review focuses on the use of polymeric

resins to selectively extract gold from complex leach

solutions. It is the authors’ view that these resins will

be the key to high yield extractions of gold from

interfering species present in leach liquors, such as

other metal complexes (in particular those containing

copper) and thiosulfate decomposition products, espe-

cially trithionate. The synthesis and evaluation of a

range of modified polymers for this purpose is cur-

rently a major research focus in the authors’ laboratory.

In this review, cyanide and thiosulfate leaching

processes will be compared, after which the chemistry

involved in using thiosulfates to extract gold from its

ores will be outlined. A detailed evaluation of avail-

able ion exchange resin materials will then be pre-

sented, together with a comparison of these materials

with other techniques for extracting gold from thio-

sulfate leach liquors, including adsorption onto acti-

vated carbon, solvent extraction and electrowinning.

2. Comparison of cyanide and thiosulfate leaching

regimes for refractory ores

It has long been observed that silver and gold will

dissolve in oxidising solutions containing thiosulfate

ions (Aylmore, 2001; Gundiler and Goering, 1993).

This reaction was first utilised in the silver mining

industry as the Patera process for the recovery of silver

from very high-grade ores (Gowland, 1930). The

modern addition of ammonia to these solutions has

improved gold dissolution through the formation of

readily soluble ammine complexes, thereby reducing

the consumption of thiosulfate. It now appears that the

leaching of gold-bearing ores with ammoniacal thio-

sulfate solutions can accomplish the recovery of a

major portion of the refractory gold from a range of

ores (Genik-Sas-Berezowsky et al., 1978; Zipperian et

al., 1988). In comparison with conventional cyanide

processing, the thiosulfate process often has the advan-

tages of greater efficiency and versatility, coupled with

a significantly lower environmental impact (Kerley,

1981; Wan et al., 1993). Thiosulfate liquors are less

prone to fouling by unwanted metal ions, and hence

can be applied to a wide array of refractory ores. The

common thiosulfate salts (Na+, K+, Ca2 + and NH4+) are

biodegradable and are regarded as nonhazardous by

Worksafe Australia (NOHSC, 1999). These thiosulfate

salts are ‘Generally Recognised As Safe’ (GRAS) in

the US, and are not considered dangerous substances

by European standards (Bean, 1997; EEC/FDA, 2001;

Langhans et al., 1992). In addition, conventional treat-

ment of refractory ores often involves the environ-

mentally contentious step of roasting the ore prior to

leaching, with the consequent release of sulfurous

gases. Roasting of some ores could be minimised by

use of the thiosulfate process, due to the partial dis-

solution of the matrix by the lixiviant and a lack of

preg-robbing behaviour. For example, the Goldstrike

Roaster was commissioned by the Barrick Gold in

2000 only after comparative evaluation of an extensive

trial of thiosulfate leaching (Fleming, 1998). In this

context, it is arguable that the thiosulfate process

represents a significant advance in precious metal

recovery techniques.

A disadvantage of thiosulfate leaching is the pres-

ence of ammonia. This volatile and noxious reagent

can readily escape from open leaching vessels and

contaminate the surroundings, necessitating contain-

ment of the leach liquors. Ammonia presents no

significant environmental threat, and may be recovered

and recycled in an efficient system.

Conventional gold ore processing is generally con-

ducted by leaching finely divided ore in a basic

solution of alkali metal cyanide. Zerovalent metals

are oxidised from the metallic state to soluble MX +

ions by dissolved oxygen, and these metal ions are

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–212

then strongly complexed by cyanide ligands (Eq. (1))

(Elsner, 1846).

4Au0 þ O2 þ 8CN� þ 2H2O ! 4½AuðCNÞ2��

þ 4OH� ð1Þ

Lixiviation of undesirable base metals, such as

copper and iron, reduces the efficiency of the process

by consuming additional reagents, and necessitates

further processing to remove these contaminants

(Fagan et al., 1997; Grigorova et al., 1987; Hilton

and Haddad, 1986; Huang et al., 1997). There are

numerous complex ores from which gold cannot be

efficiently recovered by cyanidation technology. These

complex ores are often termed ‘‘refractory’’, implying

a problematic mineral structure resistant to leaching,

which often includes one or more of the following

types (Bhappu, 1990):

� Pyrite Ore—Finely divided gold embedded in a

sulfide matrix.� Cuprous Ore—Copper is abundant in the mineral

matrix.� Base Metal Ores—Lead or zinc sulfides encapsu-

late the gold.� Manganous Ore—High levels of manganese are

present.� Telluride Ore—Gold is associated with tellurium in

the ore.� Carbonaceous Ore—Fine organic ‘preg-robbing’

carbon contaminates the ore.

In most of the above cases, cyanidation is unable to

lixiviate significant quantities of desirable metals with-

out consuming excessive, and typically uneconomical,

quantities of reagents (Cvetkovski et al., 1996; Davi-

son et al., 1961; Torres and Costa, 1994; Wan, 1997).

This may also result in significant contamination of the

leach liquor with undesirable solubilised metals. Alter-

natively, in the case of organic carbon, the gold

cyanide complex is adsorbed onto fine particles of

native carbon. This gold is then lost to the tailings,

hence this organic carbon robs the pregnant leach

liquor of solubilised gold (i.e. ‘preg-robbing’).

Ammoniacal thiosulfate leaching is less sensitive

than cyanidation to contamination by unwanted cati-

ons (Aylmore, 2001). The presence of abundant

ammonia has the effect of hindering the dissolution

of undesirable ions such as silicates and carbonates

(Abbruzzese et al., 1995). Similarly, a number of

metals and minerals, such as CaO, Fe2O3 and MnO2,

are converted into insoluble hydroxides by ammonium

hydroxide at pH>9.5 (Perez and Galaviz, 1987). It has

also been claimed that ammoniacal thiosulfate leach-

ing beneficiates auriferous sulfide ores via corrosion of

chalcopyrite, pyrrhotite, arsenopyrite and, to a lesser

degree, pyrite (Gong et al., 1993; Feng and Van

Deventer, 2002). However, a number of studies have

concluded that dissolution of pyrite ores does not take

place in thiosulfate liquors (Aylmore, 2001; Aylmore

and Muir, 2001; Schmitz et al., 2001). Importantly, it is

also reported that the aurothiosulfate complex is not

significantly adsorbed onto native carbon, thus reduc-

ing losses due to preg-robbing (Gallagher et al., 1989,

1990; Kononova et al., 2001).

3. Species present in thiosulfate leach liquors

3.1. Metal complexes

Thiosulfate is a divalent type ‘A’ soft ligand, which

tends to form more stable complexes with low-spin d8

{PdII, PtII, AuIII} and d10 {CuI, AgI, AuI, HgII} metal

ions (Livingstone, 1965; Wilkinson, 1987). Com-

monly, the thiosulfate ion acts as a unidentate ligand

via the terminal sulfur atom, establishing strong j-bonds with a metal ion which are stabilised by pk–dkback-bonding. Thiosulfate ligands may also act in a

bridging role via the terminal sulfur atom or as a

bidentate ligand through a sulfur and an oxygen atom,

usually resulting in an insoluble complex (see Fig. 1)

(Gmelin, 1973; Livingstone, 1965; Ryabchikov, 1943;

Zhao et al., 1997).

In a thiosulfate leach liquor, the formation of gold

and silver thiosulfate complexes proceeds via the

catalytic oxidation of the zerovalent metal by a suitable

soluble metal complex, which is typically the cop-

per(II) tetra-ammine complex acting as the primary

oxidant (Jiang et al., 1993a; Li et al., 1995; Thomas et

al., 1998; Tozawa et al., 1981). The reduction of the

copper(II) ammine complex is believed to transfer two

ammonia ligands, allowing the kinetically favoured

diaminoaurate(I) complex to form (Chen et al., 1996;

Gowland, 1930; Zhu et al., 1994). This exchanges

ligands with the free thiosulfate ions to form the more

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 3

thermodynamically stable aurothiosulfate complex

(Briones and Lapidus, 1998; Jiang et al., 1992; Pan-

ayotov et al., 1995). At the same time, the reaction

between dissolved oxygen and the copper(I) thiosul-

fate complex regenerates the copper(II) tetrammine

complex. The activation energy required to lixiviate

metallic gold in the presence of ammonia (pH 10),

excess thiosulfate and Cu(II) is 15.54 kJ/mol (Jiang et

al., 1993b). The activation energy rises to 27.99 kJ/mol

in the absence of Cu(II) and ammonia, demonstrating

the catalytic effects of Cu(II). The reactions that

describe the leaching behaviour of gold in the ammo-

niacal thiosulfate system are shown below (Eqs. (2)–

(5)) (Bhaduri, 1987; Ferron et al., 1998; Fleming,

1998; Genik-Sas-Berezowsky et al., 1978; Jiang et

al., 1992; Marchbank et al., 1996; Umetsu and

Tozawa, 1972; Xu and Schoonen, 1995):

Au0 þ ½CuðNH3Þ4�2þ þ 3S2O

2�3

! ½AuðNH3Þ2�þ þ ½CuðS2O3Þ3�

5� þ 2NH3 ð2Þ

½AuðNH3Þ2�þ þ 2S2O

2�3 ! ½AuðS2O3Þ2�

3� þ 2NH3

ð3Þ4½CuðS2O3Þ3�

5� þ 16NH3 þ O2 þ 2H2O

! 4½CuðNH3Þ4�2þ þ 12S2O

2�3 þ 4OH� ð4Þ

Net Reaction: 4Au0 þ 8S2O2�3 þ O2 þ 2H2O

! 4½AuðS2O3Þ2�3� þ 4OH� ð5Þ

In ammoniacal thiosulfate liquors, metal ions can

form a range of complexes with ammonia, thiosulfate

and hydroxide ions. For example, copper(I) is usually

reported only as [Cu(S2O3)3]5�, yet at concentrations

of thiosulfate below 0.05 M, the primary complex is

expected to be [Cu(S2O3)2]3� (Naito et al., 1970;

Zipperian et al., 1988). Copper(II) is almost exclu-

sively found as [Cu(NH3)4]2 +, although it is suggested

that the triammine complex [Cu(NH3)3]2 + may be the

primary oxidising species (Byerley et al., 1973, 1975).

There appears to be a maximum solubility of copper, in

that approximately one gram of copper will be soluble

for every 1% (NH4)2S2O3 (w/w) present in the leach

solution (Johnson and Bhappu, 1969). Under unfav-

ourable conditions, precipitation of Cu2S2O3 or mixed

cuprous-ammonium thiosulfate salts may occur (Chen

et al., 1996; Flett et al., 1983). Soluble thiosulfate

complexes are also known for a number of heavy

metals, with their stepwise stability constants (bn)and coordination numbers illustrated in Fig. 2 (Agadz-

hanyan et al., 1981; Benedetti and Boulegue, 1991;

Fleming, 1998; Gmelin, 1965, 1969, 1972, 1973;

Hubin and Vereecken, 1994; Livingstone, 1965; Nova-

kovskii and Ryazantseva, 1955; Smith and Martell,

1976; Tykodi, 1990; Vasil’ev et al., 1953; Vlassopou-

los andWood, 1990; Wilkinson, 1987; Williamson and

Rimstidt, 1993). Apart from the formation of anionic

thiosulfate complexes, some metal cations are

expected to form ammine complexes, as detailed in

Fig. 3 (Smith and Martell, 1976). Several copper and

palladium complexes bearing both ammine and thio-

Fig. 1. Structure of thiosulfate complexes.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–214

sulfate ligands are also known, although their stability

constants and solubility remain unknown (Wilkinson,

1987).

Comparison of the stability constants in Figs. 2 and

3 reveals that many important metals will form thio-

sulfate complexes in preference to their corresponding

ammine complexes. Thiosulfate complexes are ex-

pected to predominate for gold(I), silver(I), iron(II),

mercury(II) and lead(II), whereas the metal ions cop-

per(I) and cadmium(II) should be found as an equili-

brium mixture of thiosulfate and ammine complexes.

The remaining soluble metal ions should occur pri-

marily as ammine complexes. The ligands of the auro-

thiosulfate complex are believed to be quite labile, as

near-stoichiometric quantities of cyanide added to a

thiosulfate liquor were found to rapidly form the cor-

responding aurocyanide complex (Lulham and Lind-

say, 1991; Marchbank et al., 1996). This is especially

significant when the rapid reaction between cyanide

and thiosulfate ions is taken into account (Eq. (9)).

Fig. 3. Stepwise stability constants for ammine complexes.

Fig. 2. Stepwise stability constants for thiosulfate complexes.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 5

The complexes [Pd(S2O3)4]6� and [Pt(S2O3)4]

6�

have aqueous solubilities above 10 ppb at pH 7 and 25

jC, and quite low oxidation potentials (�E0) of 0.116

and 0.170 mV, respectively (Mountain and Wood,

1988; Plimer and Williams, 1988). However, they are

not thermodynamically stable, and slowly decompose

into insoluble S-bridged oligomers similar to the silver

complex in Fig. 1 (Anthony and Williams, 1994).

Metals in higher oxidation states such as Au(III) and

Fe(III) are readily reduced by thiosulfate ions, and

hence are not significant in leach liquors. Other anions

from the leach liquor or the mineral matrix, such as

chloride, hydroxide or sulfate may also participate in

metal ion solvation. Stable, soluble complexes bearing

a mixture of ligands may be present, similar to the

copper salts [Cu(CN)X (NH3)Y](1� X) found in ammo-

niacal cyanide leach liquors (Muir et al., 1993). How-

ever, relevant mixed ligand thiosulfate complexes have

not been reported in the literature.

3.2. Sulfur-oxygen anions

The sulfoxy anions initially present in an ammo-

niacal thiosulfate leach liquor are thiosulfate, plus

some sulfate and sulfide from mineral sources. How-

ever, thiosulfate is metastable, which means it may be

readily oxidised or reduced according to the initial

solution potential. Depending on the aqueous envi-

ronment, thiosulfate can break down into sulfite,

sulfate, trithionate, tetrathionate, sulfide, polythionates

(SxOy2 �) and/or polysulfides (Sx

2 �). An important

factor in thiosulfate stability is the pH of the solution,

since thiosulfate rapidly decomposes in acidic media

(Li et al., 1995). Certain metal ions and reagents also

cause the breakdown of thiosulfate, as shown below

(Eqs. (6)–(12)) (Abbruzzese et al., 1995; Briones and

Lapidus, 1998; Fleming, 1998; Nickless, 1968; Wil-

liamson and Rimstidt, 1993; Tykodi, 1990; Xu and

Schoonen, 1995). Note that the anions trithionate

(S3O62 �) and tetrathionate (S4O6

2�), which are not

known to have any lixiviating activity (Aylmore,

2001), can interfere with resin-based recovery methods

by displacing metal complexes from ion-exchange

sites (Fleming, 1998; O’Malley, 2001).

4S2O2�3 þ 4Hþ þ O2 ! 2S4O

2�6 þ 2H2O ð6Þ

S2O2�3 þ 2Hþ ! S0ðpptÞ þ SO2ðgasÞ þ H2O ð7Þ

4S2O2�3 þ O2 þ 2H2O ! 2S4O

2�6 þ 4OH� ð8Þ

S2O2�3 þ CN� þ 1

2O2 ! SCN� þ SO2�

4 ð9Þ

2½FeðS2O3Þ�þ ! 2Fe2þ þ S4O

2�6 ð10Þ

S2O2�3 þ Cu2þ þ 2OH� ! SO2�

4 þ H2Oþ CuSðpptÞ

ð11Þ

2Cu2þ þ 2S2O

2�3 ! 2Cuþ þ S4O

2�6 ð12Þ

In addition to the above reactions, thiosulfate is also

consumed by peroxides, phosphines, polysulfides,

permanganates, chromates, the halogens (chlorine,

bromine and iodine), and their oxyanions. In addition,

certain species of fungi, microfauna and microflora can

digest thiosulfate ions, albeit quite slowly (Xu and

Schoonen, 1995).

Thiosulfate leaching is accelerated by light (John-

son and Bhappu, 1969), although this also encourages

the consumption of thiosulfate complexes in reactions

with semiconductor minerals such as TiO2 and Fe2O3

(Benedetti and Boulegue, 1991; Xu and Schoonen,

1995). Although the thiosulfate ligand is also signifi-

cantly less stable than cyanide in similar conditions,

which limits leaching times and increases reagent

consumption, this native instability of thiosulfate

becomes an asset when environmental considerations

are taken into account. The ligand itself is of very low

toxicity, with its common salts being regarded as

nonhazardous, while many of its soluble metal com-

plexes break down to relatively harmless materials

such as sulfates, insoluble sulfides and oxides, which

present a significantly lesser threat to the environment

than cyanides (Bean, 1997; Langhans et al., 1992;

NOHSC, 1999).

The degradation of thiosulfate ions may be caused,

or catalysed, by the presence of certain metal ions.

Iron(III) accelerates the decomposition of thiosulfate

by intramolecular electron transfer. The deep purple

[Fe(S2O3)]+ complex is formed, and decomposition

occurs via reduction of the metal and concomitant

oxidation and dimerisation of the ligand to form the

tetrathionate ion (Eq. (10)) (Perez and Galaviz, 1987;

Williamson and Rimstidt, 1993; Uri, 1947). Similarly,

the salts of arsenic, antimony and tin catalyse the

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–216

formation of pentathionate from thiosulfates, while

metallic copper, zinc, and aluminium result in the

formation of sulfides (Bean, 1997; Xu and Schoonen,

1995). Both mercury and silver tend to form nearly

insoluble sulfides, although complexation in excess

thiosulfate tends to minimise this precipitation (Bean,

1997). Many iron minerals, such as pyrite and haema-

tite, and most TiO2 and SiO2 ores will catalyse the

oxidative degradation of thiosulfate ions into tetrathi-

onate (Benedetti and Boulegue, 1991; Xu and

Schoonen, 1995). However, the side-reactions and

decomposition processes of many metal thiosulfate

complexes are not well characterised. In any case,

when leaching minerals it is essential to allow for the

natural degradation of thiosulfates via ubiquitous O2,

H3O+, trace Fe3 + and other oxidants. One important

example is copper(II) (see Eq. (12)), which also con-

tributes to the essential gold oxidation step (Eq. (2)).

The gradual but inevitable loss of thiosulfate from

leaching solutions necessitates relatively speedy leach-

ing and handling operations to optimise gold leaching

and minimise precipitation.

It has been proposed that the ammoniacal liquor

used in a thiosulfate leaching regime may be substan-

tially recycled in the leach process after precious metal

recovery (Abbruzzese et al., 1995; Kerley, 1983; Wan

and Brierley, 1997). The recyclability of the liquor

strongly depends on the precious metal recovery tech-

nique employed, which may cause significant degra-

dation of the liquor by oxidation, reduction or

contamination (Awadalla and Ritcey, 1990; Benedetti

and Boulegue, 1991; Byerley et al., 1973; Wan, 1997).

Ammonia could be stripped from exhausted liquor

(tailings) by exploiting its significant volatility (March-

bank et al., 1996). By-products and tailings from

thiosulfate processing should consist primarily of low

toxicity metal hydroxides, oxides, sulfates, polythio-

nates, polysulfides and/or insoluble sulfides, although

pilot studies to date have not directly addressed waste

management.

There are many reversible reactions in which thio-

sulfate is either consumed or generated. Some provide

important contributions to thiosulfate leaching by

recycling various breakdown products and regenerat-

ing thiosulfate ions. These equilibria are summarised

below (Eqs. (13)–(29)) where each reaction has been

formulated with thiosulfate as the product (Bean,

1997; Byerley et al., 1975; Fleming, 1998; Hu and

Gong, 1991; Perez and Galaviz, 1987; Roy and Tru-

dinger, 1970; Xu and Schoonen, 1995; Zipperian et al.,

1988):

SO2�3 þ 2OH� þ S4O

2�6 X 2S2O

2�3 þ SO2�

4 þ H2O

ð13Þ

3SO2�3 þ 2S2� þ 3H2O X 2S2O

2�3 þ 6OH� þ S0

ð14Þ

SO2�3 þ S5O

2�6 X S2O

2�3 þ S4O

2�6 ð15Þ

SO2�3 þ S4O

2�6 X S2O

2�3 þ S3O

2�6 ð16Þ

2S5O2�6 þ 6OH� X 5S2O

2�3 þ 3H2O ð17Þ

4S4O2�6 þ 6OH� X 5S2O

2�3 þ 2S3O

2�6 þ 3H2O ð18Þ

2S3O2�6 þ 6OH� X S2O

2�3 þ 4SO2�

3 þ 3H2O ð19Þ

S4O2�6 þ H2S X 2S2O

2�3 þ S0 þ 2Hþ ð20Þ

2S2O2�4 X S2O

2�3 þ S2O

2�5 ð21Þ

2S2�þ4SO2�4 þ8Hþþ 8e� X 3S2O

2�3 þ6OH�þH2O

ð22Þ

2S2� þ 2SO2 þ 2HSO�3 X 3S2O

2�3 þ H2O ð23Þ

2S2� þ 3SO2 þ SO2�3 X 3S2O

2�3 ð24Þ

2HS� þ 4HSO�3 X 3S2O

2�3 þ 3H2O ð25Þ

S6O2�6 þ 3SO2�

3 X 3S2O2�3 þ S3O

2�6 ð26Þ

S0 þ SO2�3 X S2O

2�3 ð27Þ

S2�ðX Þ þ SO2�3 X S2O

2�3 þ S2�ðX�1Þ ð28Þ

S3O2�6 þ S2� X 2S2O

2�3 ð29Þ

Kerley has augmented the leaching system by add-

ing sulfite, with the twofold aims of regenerating some

decomposed thiosulfate (Eq. (25)) and lixiviating re-

fractory manganese dioxide (Eq. (34)) (Kerley, 1981,

1983). The addition of sulfite to the lixiviant has also

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 7

been investigated by other researchers for similar rea-

sons (Flett et al., 1983; Groudev et al., 1996; Guerra

and Dreisinger, 1999; Hemmati et al., 1989; Johnson

and Bhappu, 1969; Langhans et al., 1992; Lulham and

Lindsay, 1991). The benefits of this treatment are

questionable, as sulfite is readily oxidised by the abun-

dant Cu(II), producing Cu(I), sulfate and dithionate

ions (Aylmore, 2001). Augmentation of the system

with excess sulfate has also been examined, with a

view to enhancing thiosulfate stability (Gong et al.,

1993; Hemmati et al., 1989; Hu and Gong, 1991). This

eight-electron redox reaction describing the reduction

of sulfate to thiosulfate may not be feasible (Eq. (22)).

Thiosulfate is ultimately transformed either directly

or via intermediates into sulfide and/or sulfate. Apart

from sulfide and sulfate, the breakdown products of

thiosulfate are not known to form particularly stable

complexes with most metal ions of interest (Aylmore,

2001; Smith and Martell, 1976). Metal sulfide com-

plexes are generally sparingly soluble, while sulfate

has negligible chelating ability, and complexes incor-

porating other polythionate (SxOy2�) ligands are over-

whelmed by the abundant thiosulfate ions. A number

of authors report the in-situ synthesis of thiosulfate

ions from sulfoxy compounds or ions during carefully

controlled oxidative leaching (Chen et al., 1996;

Genik-Sas-Berezowsky et al., 1978; Groves and

Blackman, 1995; Kerley, 1983). As a by-product of

the destruction of a sulfide matrix, the oxidation of

native sulfur or sulfides may be the cheapest source of

lixiviant generation (Eqs. (27)–(28) and (Eqs. (30)–

(33)) (Bean, 1997; Chen et al., 1996; Genik-Sas-

Berezowsky et al., 1978; Groves and Blackman,

1995). The reaction mechanism(s) that permit this

transformation appear to involve the attack on elemen-

tal sulfur by transitory polysulfide species (i.e. NaSxH

in Eq. (33)) (Bean, 1997; Groves and Blackman,

1995). The sulfur dioxide produced in an ore roasting

step may also be used to generate thiosulfate lixiviant

(Aylmore and Muir, 2001). Recovering harmful sul-

furous matter in this fashion also has the advantage of

minimising the environmental impact of the operation.

However, recycling and/or in-situ generation of thio-

sulfate has yet to be implemented at a significant scale.

4S0 þ 6OH� ! S2O2�3 þ 2S2� þ 3H2O ð30Þ

2NH3 þ SO2 þ S0 þ H2O X ðNH4Þ2S2O3 ð31Þ

2ðNH4Þ2Sþ 2SO2 þ O2 ! 2ðNH4Þ2S2O3 ð32Þ

S8 þ 8NaOH ! 2ðNa2S2O3Þ þ 2NaSxHþ H2O

ð33ÞMnO2 þ 2½ðNH4Þ2SO3� þ 2H2O ! MnS2O6

þ 4NH4OH ð34Þ

4. Recovery of gold from thiosulfate leach liquors

The removal and recovery of gold from conven-

tional cyanide leach liquors has been pursued using the

techniques listed below:

� Precipitation� Electrowinning� Solvent Extraction� Carbon Adsorption� Resin Adsorption

Similarly, these same processes have also been

applied to the recovery of gold from ammoniacal thio-

sulfate leach liquors, with varying degrees of success.

4.1. Precipitation

Precipitation of the majority of gold from a preg-

nant leach liquor can be achieved by adding a pulv-

erised metal. Otherwise known as the Merrill–Crowe

process, or cementation, the primary mechanism of

this recovery technique is the redox reaction between

the zerovalent base metal grains and the target noble

metals (Woollacott and Eric, 1994). The precipitant,

carefully chosen for redox potential, stoichiometrically

displaces the precious metals in solution (see Eq. (35)).

The more common precipitants are copper and zinc,

although iron or aluminium are sometimes employed

(Marchbank et al., 1996; Woollacott and Eric, 1994;

Yen et al., 1998).

2Auþ þM0ðsolidÞ ! 2Au0ðpptÞ þM2þ ð35Þ

Metallic precipitants often have the deleterious

effect of reducing thiosulfate ions while producing

unwanted cations, thus complicating the recycling of

the lixiviant. Copper is a reasonable choice, as solu-

tions of copper ions in thiosulfate liquors may be re-

cycled as leach liquor. Some contamination of the solid

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–218

product often occurs, through either undissolved (ex-

cess) precipitant or by co-precipitation with other metal

ions in solution, necessitating further purification.

Precipitation by the addition of sulfide salts or by

chemical reduction with sodium borohydride, hy-

drogen or sulfur dioxide has also been investigated

(Awadalla and Ritcey, 1990; Deschenes and Ritcey,

1990; Johnson and Bhappu, 1969). These techniques

are not highly favoured, as they are less selective and

tend to precipitate most metals from solution as well as

hindering recycling of the leach liquor.

4.2. Electrowinning

The recovery of metal ions from solution by the

application of direct current is known as electrowin-

ning. Aurothiosulfate ions in the solution will migrate

to the cathode and form a metallic deposit (Abbruzzese

et al., 1995; McPartland and Bautista, 1999; Woolla-

cott and Eric, 1994). Electrowinning is especially

problematic in the presence of a great excess of

unwanted cations [i.e. Cu(I) and Cu(II)], as they may

contaminate the metallic product. This results in a

devalued product requiring further purification. Side

reactions involving the oxidation or reduction of

thiosulfate may also interfere (Aylmore, 2001). This

lowers the efficiency of electrowinning by increasing

the energy input required to recover the desired metals

from solution. Due to the abundant copper and thio-

sulfate ions in these liquors, electrowinning does not

appear to be a viable option.

4.3. Solvent extraction

In this technique, the leach liquor is contacted with a

solution of extractant in a water-immiscible organic

solvent. The gold complex is partitioned into the or-

ganic phase, whereas the other metals ideally remain in

the aqueous phase. The organic phase may then be

separated, stripped of gold, and returned to the extrac-

tion circuit. There have been a number of studies in

which gold has been extracted from ammoniacal thio-

sulfate liquors by solvent extraction. Zhao et al. report

the application of a number of potential gold extraction

reagents using varied diluents, such as benzene, kero-

sene, and 1- and 2-octanol. The extractants employed

in these solvents were primary, secondary and tertiary

alkylamines, tertiary amine oxides, phosphines, phos-

phine oxides and phosphate esters (Chen et al., 1996;

Zhao et al., 1997; Zhao et al., 1998a,b,c, 1999). Each of

these neutral reagents bore large alkyl groups to

enhance solubility in the organic phase. The results

of these studies are shown in Table 1.

Artificial thiosulfate liquors (0.8 M) were examined

with and without ammonia. The presence of ammonia

in the liquor was reported to improve extraction in

several cases (Zhao et al., 1997, 1998b). The amines

alone were effective extractants, with efficacy increas-

ing in the order 1j>2j>3j alkylamines. Aromatic

diluents or kerosene performed better than n-octanol

and chloroform, apparently due to inductive electron-

acceptor effects of the latter solvents on the amines

(Chen et al., 1996; Zhao et al., 1998a,b). The organic

phase was stripped of more than 96% of the extracted

gold in 10 min using aqueous sodium hydroxide

(pH>10) (Zhao et al., 1998b). Phosphorus compounds

performed better in the presence of the primary amine

than alone, suggesting synergistic electron-donating

effects (Zhao et al., 1997, 1998a). The performance

of the amine extractants was significantly improved in

the presence of a trialkylamine oxide (TRAO), which

was also accounted for by electron-donating synergism

Table 1

Solvent extraction of gold from thiosulfate liquors

%Gold

extraction

Extraction

reagent(s)

Concentrations

(mol/L)

Effective

diluent(s)

[S2O32�]

(mol/L)

[NH3]

(mol/L)

[Au]

(mmol/L)

pH Reference

f 92 1j–3j amines 0.5 octane 0.2–0.8 0 0.24 < 8.0 Zhao et al., 1998b

95 TRPO+APE 25% v/v octane/none 0.2–0.8 0–0.15 0.24 >10.0 Zhao et al., 1997

f 96 N1923 +APE 0.5 + 0.2� 0.6 kerosene/octane 0.8 0 0.24 < 8.0 Zhao et al., 1998a

f 97 N1923 + TRAO 0.2� 0.3 + 0.3� 0.6 octane 0.8 0 0.24 < 7.5 Zhao et al., 1998c

f 95 N1923 + TRAO 0.5 + 0.15 octane/kerosene 0.8 1.6 0.20 < 9.0 Zhao et al., 1999

Conditions: Ambient temperature (f 20 jC); 1:1 phase ratio; APE: alkylphosphorus esters.

N1923: primary amine with C19 – 23 alkyl group; TRPO: trialkylphosphine oxide; TRAO: trialkylamine oxide.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 9

(Zhao et al., 1998b,c). The addition of ammonia sig-

nificantly enhanced selectivity of the organic phase for

gold from an artificial solution containing copper, zinc,

nickel, gold and silver (Chen et al., 1996; Zhao et al.,

1999). This was supported by the selective extraction

of gold from a real leach liquor (0.8 M thiosulfate + 2

M ammonia) containing Au(I) (10.66 ppm), Ag(I)

(5.59 ppm) and 0.11% soluble copper (Zhao et al.,

1999). The authors have characterised the metal selec-

tivity of the solvent extraction system by the ‘Separa-

tion Factor’ (a), an unquantified term. The results of

these two selectivity tests are illustrated in Table 2.

Another approach, patented by Virnig et al., utilises

a variety of N-alkylated guanidines in aliphatic or

aromatic solvents (Virnig and Sierakoski, 2001). Alter-

natively, the lixiviant can be a mixture of a quaternary

ammonium compound and a weak acid, such as

phenol. These mixtures were contacted with leach

liquors containing thiosulfate (25–200 mM) and

ammonia (>50 mM) at pH 8–10. Although no process

results were reported, it was claimed that the organic

phase could be stripped of gold by aqueous NaOH.

Solid phase sorbents impregnated with the solvent

phase were recommended to facilitate separation of

the gold-loaded solvent.

Solvent extraction has numerous constraints that

limit the scope of possible industrial applications. Eva-

luation requires an analysis of the loss of the organic

phase and extractive reagent by dissolution in the

aqueous phase, through chemical degradation and by

evaporation. In any case, the application of solvent

extraction is limited to clarified liquors, free of partic-

ulate matter (McPartland and Bautista, 1999; Woolla-

cott and Eric, 1994; Zhao et al., 1999). To produce

clarified liquor from a mineral pulp, additional plant

equipment and processing time would be required, and

these steps would result in considerable increases in

capital costs and operating expenses.

4.4. Carbon adsorption

The capture of the dicyanoaurate complex [Au

(CN)2]� onto activated carbon has been the mainstay

of gold hydrometallurgy for several decades. This

technique has superseded precipitative and electro-

chemical techniques due to its high efficiency, rela-

tively low costs and purity of product. Porous granules

of carbon may be contacted with gold-bearing cyanide

liquor, then recovered with the adsorbed gold. The

adsorption step may take place in the presence of the

mineral pulp (CIP: Carbon-In-Pulp), or may be con-

ducted with the clarified liquor. The former technique

is often preferred, as this tends to reduce capital costs.

Clarification of liquors is not only expensive and time

consuming, but higher recoveries of gold can be

achieved using the scavenging behaviour of CIP

techniques.

The gold-bearing carbon is commonly eluted with a

hot caustic cyanide solution, often under pressure. The

stripped carbon is usually regenerated for re-use by

washing with aqueous acid, then heated in an oven at

650 jC for several hours under a nonoxidising atmos-

phere (Bhappu, 1990). Gold in dilute thiosulfate

liquors may be quantitatively converted into the more

stable aurocyanide complex by adding a slight excess

of cyanide ions (Lulham and Lindsay, 1991; March-

bank et al., 1996). Although permitting the recovery of

the gold onto carbon, this introduces cyanide com-

plexes that can contaminate both the product and the

effluent, thus defeating the environmental advantages

of the thiosulfate process.

There appears to be some contention as to the

affinity of the aurothiosulfate complex for carbon.

Activated carbon achieved a 95% gold recovery after

6-h contact at 25 jC with a gold-bearing leach liquor

(15.8 mg/L Au) derived from a Dominican ore via an

ammoniacal thiosulfate lixiviant (2.0 M Na2S2O3 + 4.0

M NH3; pH 10.5) (Abbruzzese et al., 1995; Meggio-

laro et al., 2000). Conversely, a 0.1 M solution of the

aurothiosulfate complex in KOH (10� 4 M; pH 9.7)

was found to have little or no affinity for carbon at

25 jC (Gallagher et al., 1989, 1990). This discrepancy

may be due to a loss of sensitivity to small changes in

the surprisingly high gold concentration in the latter

Table 2

Solvent extraction of metals from thiosulfate liquors using

N1923 + TRAO (Zhao et al., 1999)

Gold Silver Copper Zinc Nickel

Artificial liquor (mmol/L) 0.20 0.24 14.98 0.20 0.21

Approximate recovery (%)a 95 32 < 2 12 < 2

Leach liquor (ppm) 10.66 5.59 1100 – –

Approximate recovery (%)a 95 15 < 1 – –

Au separation factor (a) – 15 1695 – –

a Conditions: Ambient temperature (20 jC); 1:1 phase ratio; 10-

min contact; diluent: n-octane.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–2110

experiment. In other studies, many types of carbon

adsorbents were contacted with ammoniacal thiosul-

fate leach liquor, and all were found to have a poor

capacity for aurothiosulfate (Kononova et al., 2001;

Mohansingh, 2000). Certainly, the [Au(S2O3)2]3 �

complex has significantly less affinity for carbon than

the [Au(CN)2]� anion. It has not been established

whether this effect is due to the larger size or charge

of the thiosulfate complex, or to other factors. Techni-

ques for elution of carbon-adsorbed aurothiosulfate

have not yet been explored.

4.5. Resin adsorption

The use of resin adsorbents for the recovery of

precious metals is a relatively underdeveloped area of

hydrometallurgy. The principal reason behind this is

the abundance and efficacy of cheap activated carbon

adsorbents. Although resins are extensively utilised in

the Commonwealth of Independent States (CIS) for

the recovery of gold from cyanide liquors (Fleming,

1998), resin adsorbents are more expensive than car-

bon, and their application requires the installation of

specialised apparatus (Bolinski and Shirley, 1996). It is

useful to distinguish Resin-in-Leach (RIL) from

Resin-in-Pulp (RIP). In RIL, the adsorbent is added

to the ore at the same time as the leaching reagents,

whereas in RIP, the adsorbent is introduced after an

initial leaching (induction) period. Although similar in

principle to the CIP/CIL techniques, there are funda-

mental differences in the RIP/RIL processes.

Both carbon and ion exchange resins must be

chemically stripped after separation from the leach

pulp. However, in contrast to activated carbon which

must be thermally regenerated (activated) rather than

treated chemically, a careful choice of eluent can allow

the resin elution and regeneration steps to occur

simultaneously. Importantly, activated carbon is under-

stood to adsorb the aurocyanide between optimally

spaced uncharged graphitic layers of the matrix

(Schmitz et al., 2001), whereas ion-exchange resins

have abundant functional groups of like charge which

concentrate solvated ions of opposite charge (counter-

ions). Unlike carbon, the functional groups of a poly-

meric resin can be tailored for selectivity for a partic-

ular ion or complex.

This advantage has been exploited by several

groups in the preparation of resins that are highly

selective for the aurocyanide complex in conventional

and ammoniacal cyanide liquors (Dicinoski, 2000;

MacKenzie et al., 1995). The resins, Aurix and Minix,

have been optimised for gold recovery from cyanide

solutions by altering properties such as functional

group, capacity, active group spacing and geometry,

matrix cross-linking and porosity. Aurix, which is

marketed by Henkel Australia, is reported to bear

guanidine moieties, whereas Minix, which is marketed

by Mintek of South Africa, bears optimally distributed

tributylammonium groups.

An important characterisation of resin performance

is the Selectivity Ratio (S/R):

S=RfAu;Mg ¼ ½Au�R=½M�R,

where M is a metal ion and [M]R is the concentration

of M on the resin phase. This gives an indication of the

preference of a functional group for gold over the

competing metal, M. The Selectivity Ratios for gold

over copper are reported as: S/R(Au,Cu) = 11.1 (Not-

ren), 2.9 (Minix), 6.5 (Guanidine resin) and 3.7 (Car-

bon) (Dicinoski, 2000). These were derived from a

solution containing Cu (200 ppm), Au (1 ppm) and

NH3 (500 ppm). From the data in Table 3, the

calculated S/R(Au, Cu) for Aurix is 6.0 (MacKenzie

et al., 1995). Elution of the adsorbed gold from Minix

resin was achieved using thiourea (1.0 M) in aqueous

H2SO4 (0.1 M) at 60 jC, whereas gold can be eluted

from Aurix with aqueous NaOH (1.0 M).

Leach liquors with a high thiosulfate concentration

are expected to function as an efficient eluent, desorb-

ing silver (and probably gold) from anion-exchange

resins and resulting in a poor recovery (Wan et al.,

1993). An alternative that has been explored is to

Table 3

Performance of gold selective resins in cyanide liquors (MacKenzie

et al., 1995; Dicinoski, 2000)

ppm

(g/t)

Au Ag Cu Zn Ni Co Fe

Initial 6.1 < 1 12 4.5 20 2.1 6.8

Aurixa 10,560 331 571 1942 5284 175 441

Initial 5.0 0.5 10.0 2.0 5.0 1.0 10.0

Minixb 36,500 – 2300 9880 5600 – 292

Notrenb 39,089 – 1000 575 1550 – 106

Carbon 25,200 < 200 < 200 < 200 460 < 200 1500

a 15 min/stage (seven stages); 28 g resin/L.b Selectivity test; 250 ppm CN.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 11

convert the aurothiosulfate complex to the more stable

aurocyanide, and recover that on a resin (Lulham and

Lindsay, 1991; Marchbank et al., 1996). The recovery

of silver from thiosulfate solutions using ion-exchange

resins has been extensively investigated as a cost-

saving measure in the photographic industry (Degen-

kolb and Scobey, 1977; Lasko and Hurst, 1999; Mina,

1980; Tarasova et al., 1994) and analogous treatments

for gold-bearing solutions are currently under develop-

ment using knowledge gained from silver recovery

(Fleming et al., 2001; Kononova et al., 2001; March-

bank et al., 1996; O’Malley, 2001; Thomas et al.,

1998; Yen et al., 1998). For this reason, approaches

used for silver recovery are discussed further below.

4.5.1. Recovery of silver using resins

Silver is used extensively used in the developing

and fixing of photographic films, a process which

generates a thiosulfate effluent solution containing

silver. This effluent often contains up to 50 ppm

Ag(I) in f 0.01 M sodium thiosulfate (Degenkolb

and Scobey, 1977; Lasko and Hurst, 1999) and may

also contain additives such as the Fe(II)–EDTA com-

plex to aid in the fixing process (Degenkolb and

Scobey, 1977; Mina, 1980). Silver needs to be

removed from the effluent to meet environmental

discharge requirements, and to reduce the processing

expenses by recycling. There have been many attempts

to recover silver from thiosulfate solutions by the use

of ion-exchange processes. For example, the Russian

strong-base resin AV-17 (Me3N+ groups on polystyr-

ene gel matrix; 4.7 meq/g) in hydroxide form was used

to adsorb 66% of the silver from a solution of Na2S2O3

(3.4 M) and Ag(I) (4.5 mM) at pH 5–5.5 (Davankov et

al., 1966a). The adsorbed silver was fully eluted from

the resin (8.6 g dry mass) with aqueous Na2S2O3 (600

mL� 3.4 M). However, when strong base resins were

used to recover silver thiosulfate (ca. 25 mg/L Ag)

from simple aqueous solutions, the metal complex

decomposed within the resin to form insoluble Ag2S

(Davankov et al., 1966b; Lurye and Peremyslova,

1953). It was proposed that a thiosulfate ligand was

lost from the adsorbed silver complex, and decompo-

sition of the remaining [Ag(S2O3)]� was then cata-

lysed by the active alkylammonium groups on the resin

(Davankov et al., 1966b). Although this silver could be

recovered using strong acid as eluent, it was not

considered economically viable given the quantity of

reagents consumed by the precipitation and elution

cycle. Ion exchanger-mediated decomposition of thio-

sulfate complexes has not been observed elsewhere.

The proprietary AkwaKlame system claims to have

overcome pre-existing problems relating to the irrever-

sible binding of silver to strong-base ion exchange

resins (Degenkolb and Scobey, 1977). It should be

noted that these researchers provide no details of the

properties of the resin used, nor any characteristics of

the feed liquor or eluent. They believed that [AgS2O3]�

was the main silver-bearing complex adsorbed, a claim

supported by Marcus (1957). However, Mina has

reported that the silver in photographic effluent is

primarily found as the complex ion [Ag(S2O3)2]3�

(Mina, 1980). Recovery of up to 98% of dissolved

silver was achieved at pH 4 using the weak base resin

IRA-68 (Rohm&Haas; 5.6 meq/g; 3j-alkylamine, pKa

9.2); the low pH being used primarily to prevent

fouling by the ferrous EDTA complex found in the

liquor. It was found that the resin loaded more silver

when the thiosulfate concentration in the liquor was

low, indicating that free thiosulfate ligands compete

significantly for ion-exchange sites. Ammonium thio-

sulfate (pH 6; 1–2 M) with some added sulfite was

found to be the most efficacious eluent for removal of

the loaded silver and regeneration of the resin (Cooley,

1981; Mina, 1980).

The Ukranian-made ion-exchanger AM-2B (Atlan-

tis; Me2N and Me3N+ groups, 1.1 meq/mL), com-

monly applied to the recovery of [Au(CN)2]� from

cyanide liquors in the CIS, has also been examined as

an adsorbent for silver from thiosulfate liquors (Agadz-

hanyan et al., 1982). Between 85% and 95% of the

adsorbed silver was eluted with unspecified solutions

of sodium thiosulfate, sodium sulfite, sodium chloride

or ammonium nitrate. Complete recovery of silver

frommodel thiosulfate liquors has been reported, using

Russian strong-base anion exchangers AV-17 (detailed

above) and AM-p (macroporous; 4.2 meq/g) (Tarasova

et al., 1994). Elution was achieved using a thiosulfate

solution (1–3 N), or by a 5% solution of thiourea in

dilute sulfuric acid (see Table 4).

Commercial ion-exchange resins IRA-67 (weak

base, same as IRA-68) and IRA-458 (strong base

quaternary ammonium, 4.4 meq/g), both manufactured

by Rohm&Haas, were used to recover silver (50 ppm)

from weak thiosulfate liquors (5–10 mmol) at various

pH values (Lasko and Hurst, 1999). The weak base

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–2112

resin was only effective between pH 2 and 6, whereas

the strong base resin adsorbed more than 90% Ag(I) at

all pH values studied. The organic polymer Chitosan

(poly(d-glucosamine)) was also examined for its

capacity to adsorb silver from a variety of solutions,

including thiosulfate liquors (Lasko and Hurst, 1999).

Unfortunately, the highest silver retention levels were

obtained in acidic solution, and Chitosan was not

found to be competitive with commercial resins (see

Table 4).

4.5.2. Recovery of gold using resins

The patent of Jay (1999) utilised amine-functional-

ised ion-exchange resins embedded in a porous polyur-

ethane matrix to adsorb gold from a range of gold-

bearing thiosulfate solutions. Quaternary ammonium

moieties were preferred, giving the best capacity and

broadest effective pH range. The presence of a water-

immiscible alcohol such as 1-octanol (i.e. solvent

impregnation) improved the loading characteristics of

the resins. It was claimed that gold could be stripped

from the adsorbent using unspecified solutions of

sodium or ammonium thiocyanate, and more rapidly

by augmenting the eluent with dimethylformamide,

acidic thiourea, zinc cyanide or sodium benzoate.

There is no evidence to support these latter claims,

and insufficient data were provided to calculate the

efficacy of gold recovery or the gold loading on the

sorbents.

Mohansingh has examined the adsorption of gold

from artificial thiosulfate liquors onto activated carbon

and three commercial ion-exchange resins. In sum-

mary, the best results were obtained using the strong

base resin Amberlite IRA-400 (Rohm & Haas, quater-

nary ammonium; 3.8 meq/g) at ambient temperature

and a pH of 9 (Mohansingh, 2000). Elution was more

problematic, with only 76% of the adsorbed gold

eluted with the best performing eluent (5 M NaCl)

over 24 h. The resin selectivity for gold over copper

was not examined.

The use of a weak-base ion exchange resin such as

Amberlite IRA-743 (Rohm & Haas; weak base poly-

amine; 0.6 meq/mL) to recover gold from dilute

thiosulfate leach liquors has been claimed by March-

bank et al. (1996). It was suggested that this might be

followed by a similar recovery of the cationic cupric

tetrammine complex. A subsequent patent by the same

authors described the recovery of aurothiosulfate via a

wide variety of strong base ion-exchange resins in RIP

and RIL processes (Thomas et al., 1998). A single

weak base resin, Amberlite A7 (Rohm & Haas, poly-

amine, 13.9 meq/g), was also reported to be applicable

Table 4

Polymer adsorbents for silver from thiosulfate liquors

Resin data [S2O32�]

(mmol/L)

pH [Ag] (ppm) %R %E Eluent(s) tested (mol/L) Reference

AV-17 (OH) 3400 5–5.5 130 66 94 Na2S2O3 (3.4 M) Davankov et al., 1966a

Wofatite P 2.8–3.5 – 76 – – NaOH, H2SO4 or

HNO3 (5.6 M)

Lurye and Peremyslova, 1953

VDP-1, VDP-2, N-0 – – 216–10,790 100 – NaOH, H2SO4 or Na2S2O3 Davankov et al., 1966b

MVP-2, MVP-10

AV-16-G, AV-17

AM-2B – – – – 95 Na2S2O3, Na2SO3 Agadzhanyan et al., 1982

NaCl or NH4NO3

AV-17, AM-p 100–3000 – 100–1000 – 100 Na2S2O3 (1–3 M) or Tarasova et al., 1994

AN-221 thiourea in H2SO4

Dowex-1 10–4000 – 11 – – – Marcus, 1957

IRA-68 (weak base) 54–517 4–7 200a 98 >90 (NH4)2S2O3 (1 M) Mina, 1980

IRA-68 (weak base) – 4.2 100–150a 98 – (NH4)2S2O3 (0.81 M) +

Na2SO3 (0.16 M)

Cooley, 1981

IRA-67, IRA-458, GT-73 5–10 2–10 50b 96 – H2SO4 Lasko and Hurst, 1999

IRC-718, Chitosan

Conditions: Ambient temperature (20–30 jC); %R= silver adsorbed; %E= silver eluted.a Solution also contains < 500 mg/L Fe.b Solution also contains 10 mM SO3

2�.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 13

to this process. The preferred adsorbents for this pat-

ented process bear quaternary amine functional groups

of Type I (trialkylammonium) or Type II (hydroxyal-

kylammonium), preferably affixed to macroporous

beads. The preferred treatment regime consisted of

12� 1 h pulp contact stages, then multi-stage stripping

as detailed below. The presence of the resin in the pulp

was reported to minimise the degradation of the

thiosulfate leach liquor. Relatively dilute leach liquors

were used, containing 0.03–0.05 M thiosulfate, 0.5–

1.6 mM Cu(II) and 7–100 mM NH3, at pH values

ranging from 7 to 9 (Fleming et al., 2001; Thomas et

al., 1998). The resin(s) were contacted with the leach

pulp in a Pachuca tank for up to 12 h, then isolated by

screening. Copper and gold were then eluted sepa-

rately from the resin. The selective elution of copper

was achieved using ammoniacal ammonium thiosul-

fate (ca. 100–200 g/L, 5–6 bed volumes) or oxy-

genated ammonia buffered by ammonium sulfate. The

latter reagent was preferred, as it facilitated the recy-

cling of the eluent as leach liquor (Fleming et al.,

2001). Gold was then stripped from the resin with a

solution of ammonium, potassium or calcium thiocya-

nate (ca. 100–200 g/L, 5–6 bed volumes).

Thiocyanate counter-ions on the resin are undesir-

able, as they are quite costly and toxic, and their pre-

sence would worsen the environmental impact of an

installation (Marchbank et al., 1996). Subsequent test-

ing has revealed that the thiocyanate eluent may be

substituted with cheaper and less toxic trithionate or

tetrathionate eluents (Fleming et al., 2001). The poly-

thionate eluents cause less deviation in pH levels, thus

minimising osmotic shock and consequent resin attri-

tion. After copper stripping, gold-loaded resins were

eluted with up to 8 bed volumes of a solution of

trithionate and/or tetrathionate (40–200 g/L; see Table

5). This eluent is prepared cheaply by controlled

oxidation of thiosulfate. The resin was regenerated by

flushing with a solution of sodium hydrogen sulfide

(NaSH; f 2 g/L), recycling both tetrathionate and

trithionate into thiosulfate ions (Eqs. (36) and (37)).

The various schemes for elution of the aurothiosulfate

complex from strong base resins are illustrated in

Fig. 4.

S3O2�6 þ S2� ! 2S2O

2�3 ð36Þ

4S4O2�6 þ 2S2� þ 6OH� ! 9S2O

2�3 þ 3H2O ð37Þ

An ammoniacal leach liquor bearing gold and

copper was treated with the macroporous strong base

resin A-500C (Purolite, quaternary ammonium, 1.15

meq/mL) using the above procedure (Fleming et al.,

2001), resulting in a discharge solution retaining very

little gold and copper (see Table 5). Copper was

stripped with 99.9% efficacy over 2 h, using ammo-

nium thiosulfate eluent (150 g/L, 4 bed volumes). The

resin was then flushed for 4 h at a rate of 2 bed

volumes of polythionate liquor per hour (either f 200

g/L trithionate, or f 40 g/L trithionate with f 80 g/L

tetrathionate). Both solutions resulted in >99% elution

of the adsorbed gold. The principal drawback of this

technique is the requirement for two separate elution

stages, and in particular the additional handling time

and reagents this demands. Processing time and related

costs may be minimised by using an adsorbent that is

highly selective for gold, thus eliminating the need for

multi-stage stripping.

A series of novel Russian ion exchange polymers

was examined for their capacity to adsorb gold (9.5–

17.9 ppm) from clarified thiosulfate leach liquors.

These were prepared by applying a solution containing

equimolar amounts of sodium thiosulfate and ammo-

nia (0.5 mol/l each) to an oxidised arsenopyrite flota-

tion concentrate containing high levels of arsenic

(4.8%) and iron (7.1%) (Kononova et al., 2001).

Consistently, more gold was adsorbed at the lower

pH values (in the range 5–11) (see Table 5). The most

efficient sorbent was the trimethylammonium (strong

base) functionalised polymer AV-17-10P, which

achieved a gold recovery of 94% at pH 6 and 85% at

pH 11. Good gold recoveries were also observed at pH

6 using resins with both strong- and weak-base groups

(polyfunctional resins), resins with weak-base groups,

and with amphoteric phosphonic acid –pyridine

copolymers (see Table 6). Elution of resins AV-17-

10P and AP-100 (5.0 g each) was performed at

ambient temperature, using thiourea in aqueous sulfu-

ric acid (0.5 M each). After 1 h, the eluent (120 ml) had

removed 92.7% and 96.8% of the adsorbed gold,

respectively.

The recovery of aurothiosulfate using a set of

commercially available anion exchange resins has

been studied by O’Malley and Nicol (2001). Equili-

bration of each resin with simple solutions of trisodium

aurothiosulfate at pH 9.5 (22 jC) yielded isothermal

ion distributions (isotherms), which clearly demon-

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–2114

Table 5

Polymer adsorbents for gold from ammoniacal thiosulfate liquors

Resin data Type meq/g Matrix [S2O32�] [NH3] pH T (jC) Sorption (h) [Au]a [Cu]a %Au Eluent(s) S/R(Au,Cu) Reference

IRA-743 WB 1.0 – 0.03 M 7–100

mM

7–9 – 12 – – – Cu: SO42�,

S2O32�, NH3

– (1)

Amberlite A7 WB – MP Au: SCN� (2)

Type I/II resins SB – –

A500C (Purolite) R4N+ 1.7 MP 0.05 M 0.1 M 8.0 60 6� 1 1.8 22 99.45 Au: SCN� 0.034 (3)

A500C (Purolite) R4N+ 1.7 MP – – 6–8 20 4 – – – Au: S3O6

2� + S4O62� 0.038 (4)

AV-17-10P etc.

(see Table 6)

R4N+ – – 0.5 M 0.5 M 5–11 – 5 9.5–17.9 – 94.2 Au: thiourea +H2SO4 – (5)

AmberJet 4200

(Rohm & Haas)

R4N+ 3.7 Gel 0.05 Mb 0.8 M 8–9.5 22 6� 2 8.9 8.7 >99.44 Au: SCN�/NO3

� 2.39 (6)

Aurix (Henkel) Guan – – 25–200

mM

0.05–

0.1 M

8.0–10 – – – – – Au: NaOH, CN� and

sodium benzoate

– (7)

AmberJet 4200 R4N+ 3.7 Gel 0.05 M 0.2 M 9.5 – – 10 10 99 Cu: SO4

2�, NH3, O2 – (8)

Vitrokele 911 (R&H) R4N+ 1.7 MR Au: NH4NO3 (2 M)

IRA-400 (R&H) R4N+ 3.8 Gel 1.0 M 0.1 M 9–12 20–60 8 9.27 31.8 94.7 Au: NaCl (5 M) – (9)

Dowex (Dow) R4N+ – – 9.12 75.98

Elchrom SA – – 9.31 9.23

WB: weak base; SA: strong acid; Guan: guanidyl; R4N+: quaternary ammonium (strong base); Matrix: MP=macroporous; MR=macroreticular. Data taken from: (1) Thomas et al.

(1998), (2) Marchbank et al. (1996), (3) Ferron et al. (1998), (4) Fleming et al. (2001), (5) Kononova et al. (2001), (6) O’Malley (2001), (7) Virnig and Sierakoski (2001), (8)

O’Malley and Nicol (2001), (9) Mohansingh (2000).a Reported in ppm.b Contains 1.0 mM sulfite.

A.C.Grosse

etal./Hydrometa

llurgy69(2003)1–21

15

strated that strong-base resins were superior to weak

base resins for this application. Moreover, they loaded

efficiently even at very low solution concentrations of

gold. The larger capacity of strong base resins should

also make them more tolerant to low levels of com-

peting anions. However, Minix, a strong base resin

with excellent selectivity for aurocyanide, performed

as poorly as the weak base/polyamine resins (at pH 8).

The best performing resin, Amberjet 4200 (Rohm

& Haas, quaternary ammonium, 3.7 meq/g), was used

successfully to scavenge gold from a thiosulfate leach

pulp (see Table 5). It was noted that the copper

concentration in the pulp rose as gold displaced it

from the resin. Due to the gradual formation of the

competing ions trithionate and tetrathionate in the

liquor, the resin contact time must be minimised. A

thiosulfate eluent rapidly displaced copper, and elution

of the adsorbed gold could be achieved using an am-

monium thiocyanate solution, or by using the patented

Parker eluent (2 M aqueous NH4NO3) (O’Malley,

Fig. 4. Elution of aurothiosulfate from anion exchange resins.

Table 6

Recovery of gold by polymer sorbents (Kononova et al., 2001)

Adsorbent examined Functional group affixed to polymer TEC

(meq/g)

SBC

(meq/g)

Au %recovery,

pH 5.8–6.1

Au %recovery,

pH 10.8–11.0

AV-17-10P R-N+(CH3)3 4.4 4.1 94.2 85.4

AP-100 1j and 2j-amines +R-N+R3 3.9 0.7 91.3 72.9

AP-2-12P R-(CH3)2N+CH2N(CH3)2 3.7 1.1 90.4 65.8

AP-24-10P R-N(CH2CH2CH2CH3)2 and R-(CH3)2N+Ph 4.1 1.3 86.5 47.9

AN-85-10P R-NHCH2CH2NH2 6.2 – 86.5 51.0

AN-106-7P R-NH(CH2)2NH(CH2)2NH2 8.9 – 88.1 39.0

ANKF-5 (No pores) R-PO(ONa)2+{N(CH3)2}n + poly(vinylpyridine) U – 87.5 54.2

Conditions: Ambient temperature; 5-h contact time; 0.5 M NH3; 0.5 M Na2S2O32�; 9.5–17.9 ppm Au.

TEC: total exchange capacity; SBC: strong base capacity; U: unknown; R: resin backbone.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–2116

2001; O’Malley and Nicol, 2001). The latter was

preferred, as ferric sulfate was required to regenerate

the resin after thiocyanate elution.

In addition to gold, the thiosulfate complexes of

other metals such as lead, copper, zinc and silver are

adsorbed onto strong base resins from ammoniacal

thiosulfate leach liquors (Fagan, 2000; O’Malley,

2001). This reduces the capacity of the resin to adsorb

gold, and unless these can be separately stripped from

the adsorbent, these metals will contaminate the final

gold eluate. Advantageously, aurothiosulfate has been

shown to displace the more rapidly adsorbed copper on

a strong base resin (O’Malley, 2001). The proposed

affinity order for adsorption of thiosulfate complexes

onto strong-base resins, based on mixed metal adsorp-

tion tests, was reported to be:

Au > PbHAg > CuHZn:

The main products of thiosulfate oxidation, namely

trithionate and tetrathionate, can thermodynamically

displace copper and gold thiosulfate complexes from

strong base resins (Fagan, 2000; O’Malley, 2001).

Although kinetically slower to adsorb, the presence

of these anions in leach liquors restricts the maximal

gold recovery that can be achieved. This may become a

major problem for the more aggressive leaching oper-

ations, where considerable thiosulfate is consumed

during leaching and hence polythionates are abundant

in the liquor. Fleming et al. found that trithionate could

only be eliminated from solution by adding sulfide (to

form thiosulfate, Eq. (29)), which has the undesirable

effect of precipitating gold sulfide (Fleming et al.,

2001). Tetrathionate, formed by the oxidation of thio-

sulfate by O2, Fe(III), Cu(II) or H+ (Eqs. (6)–(12)), is

short-lived in leaching conditions and is eventually

transformed into trithionate. Oxidation of thiosulfate

must therefore be minimised in leaching, both to con-

serve the lixiviant and to ensure efficient gold recovery.

Henkel Australia has claimed a patent for the

application of their guanidine functionalised polystyr-

ene resin (Aurix) to the recovery of gold from liquors

containing thiosulfate (25–200 mM) and ammonia

(>50 mM) at pH 8–10 (Virnig and Sierakoski,

2001). The high pKa of guanidine (13.5) permits the

protonated form of the resin to be present in such

liquors, so they should behave similarly to quaternary

ammonium resins. The proponents claim that gold can

be stripped from the sorbent using aqueous NaOH

(pH>11), with the optional additives of NaCN and/or a

carboxylic acid such as sodium benzoate to facilitate

more efficient elution. No process examples or exper-

imental details were provided with this claim.

5. Conclusions

Ammoniacal thiosulfate leaching is a promising

alternative to cyanide based processes, applicable to

refractory ores and to environmentally sensitive sites.

The process cannot be applied commercially without

reliable techniques for recovery of the gold from

ammoniacal thiosulfate liquors. Standard practices of

precipitation or carbon adsorption are not readily

applicable, whereas solvent extraction requires clari-

fied liquors. The only technique appropriate for recov-

ery of gold from commercial thiosulfate leach pulp is

sorption onto an ion-exchange resin.

The resin processes reported thus far are not suffi-

cient to justify commercial application. The principal

difficulties in present applications arise from the com-

petitive adsorption of unwanted anions. Not only are

the gold and copper thiosulfate complexes adsorbed

onto the resins, but also the inevitable by-products

trithionate and tetrathionate. It is felt that both the costs

and time involved in operating a two-stage elution

process for copper and gold, and the additional require-

ment for the elution (and destruction) of adsorbed

polythionates are commercially prohibitive.

Selectivity may be derived more from steric factors

than electrostatic interactions, as the size, shape and

charge of the thiosulfate complexes of gold and copper

are significantly different. The aurothiosulfate com-

plex is linear and bears a 3� charge, whereas the fully

coordinated copper(II) thiosulfate complex is trigonal

and accommodates a 5� charge. This factor may be

exploited when designing a ligand for maximising

gold recovery. Similarly, the interference of the much

smaller divalent anions trithionate and tetrathionate

may be reduced by a custom-made resin. Few resins

specifically for aurothiosulfate recovery have been

made to date, and these have generally been ad-hoc

preparations (Kononova et al., 2001). The competing

ions Cu(I), Pb(II), trithionate and tetrathionate set

upper and lower size exclusion limits for an appropri-

ate ion-exchange functional group. The authors are

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 17

currently pursuing investigations into varying the

substituent ion exchange groups bonded to polymer

supports for the selective extraction of gold from

ammoniacal thiosulfate liquors.

References

Abbruzzese, C., Fornari, P., Massidda, R., Veglio, F., Ubaldini, S.,

1995. Thiosulfate leaching for gold hydrometallurgy. Hydrome-

tallurgy 39, 265–276.

Agadzhanyan, A.E., Ter-Arakelyan, K.A., Arutyunyan, R.V., Ba-

bayan, G.G., 1981. Study of the behaviour of copper and silver

thiosulfate complexes under different conditions. Arm. Khim.

Zh. 34 (1), 18–22.

Agadzhanyan, A.E., Ter-Arakelyan, K.A., Babayan, G.G., 1982.

Some regularities of silver thiosulfate complex sorption on

AM-2B ion-exchanger. Arm. Khim. Zh. 35 (3), 151–155.

Anthony, E.Y., Williams, P.A., 1994. In: Alpers, C.N., Blowes,

D.W. (Eds.), Thiosulfate Complexing of Platinum Group Ele-

ments. In: Comstock, M.J.ACS Symposium Series 550: Envi-

ronmental Geochemistry of Sulfide Oxidation. American

Chemical Society, Washington, pp. 551–560.

Awadalla, F.T., Ritcey, G.M., 1990. Recovery of gold from thio-

urea, thiocyanate or thiosulfate solutions by reduction-precipi-

tation with a stabilized form of sodium borohydride. Randol

Gold Forum, Sept. 1990, Squaw Valley, USA. Randol Interna-

tional, Golden, CO, USA, pp. 295–306.

Aylmore, M.G., 2001. Treatment of a refractory gold–copper sul-

fide concentrate by copper ammoniacal thiosulfate leaching.

Miner. Eng. 14 (6), 615–637.

Aylmore, M.G., Muir, D.M., 2001. Thiosulfate leaching of gold—a

review. Miner. Eng. 14 (2), 135–174.

Bean, S.L., 1997. Thiosulfates. In: Kroschwitz, J.I. (Ed.), Kirk-

Othmer Encyclopedia of Chemical Technology. Wiley, New

York, pp. 51–68.

Benedetti, M., Boulegue, J., 1991. Mechanism of gold transfer and

deposition in a supergene environment. Geochim. Cosmochim.

Acta 55 (6), 1539–1547.

Bhaduri, R.S., 1987. Lixiviation of refractory ores with diethyl-

amine or ammonium thiosulfate. MSc Thesis, University of

Nevada: Reno. 100 pp.

Bhappu, R.B., 1990. Hydrometallurgical processing of precious

metal ores. In: Arbiter, N., Han, K.N. (Eds.), Gold: Advances

in Precious Metal Recovery. Gordon and Breach Science Pub-

lishers, New York, pp. 66–80.

Bolinski, L., Shirley, J., 1996. Russian resin-in-pulp technology,

current status and recent developments. Randol Gold Forum

’96. Randol International, Golden, CO, USA, pp. 419–423.

Briones, R., Lapidus, G.T., 1998. The leaching of silver sulfide with

the thiosulfate–ammonia–cupric ion system. Hydrometallurgy

50, 243–260.

Byerley, J.A., Fouda, S.A., Rempel, G.L., 1973. Kinetics and

mechanism of the oxidation of thiosulfate ions by copper(II)

ions in aqueous ammonia solution. J. Chem. Soc., Dalton Trans.,

889–893.

Byerley, J.A., Fouda, S.A., Rempel, G.L., 1975. Activation of

Copper(II) Ammine complexes by molecular oxygen for the

oxidation of thiosulfate ions. J. Chem. Soc., Dalton Trans.,

1329–1338.

Chen, J., Deng, T., Zhu, G., Zhao, J., 1996. Leaching and recovery

of gold in thiosulfate based system—a research summary at

ICM. Trans. Indian Inst. Met. 49 (6), 841–849.

Cooley, A.C., 1981. Ion-exchange silver recovery for process EP-2

with nonregenerated bleach-fix. J. Appl. Photogr. Eng. 7 (4),

106–110.

Cvetkovski, V., Jovanovic, L., Mitevska, N., Vukovic, M., 1996.

Recovery of gold and silver from sulfide refractory ores. In:

Kemal, A., Akar, C. (Eds.), Changing Scopes in Mineral Pro-

cessing. Balkema, Rotterdam, pp. 549–552.

Davankov, A.B., Laufer, V.M., Zubakova, L.B., 1966a. Elution of

complex silver thiosulfate ions from anion exchangers after ad-

sorption. Russ. J. Appl. Chem. 40 (8), 1656–1659.

Davankov, A.B., Laufer, V.M., Zubakova, L.B., Aptova, T.A., Mir-

onov, A.A., 1966b. Ion exchange extraction of silver from wash

waters of the motion picture film industry. Russ. J. Appl. Chem.

39 (9), 1935–1940.

Davison, J., Read, F.O., Noakes, F.D.L., Arden, T.V., 1961. Ion

exchange for gold recovery. Trans. Inst. Min. Metall. 70,

247–290.

Degenkolb, D.J., Scobey, F.J., 1977. Silver recovery from photo-

graphic wash waters by ion exchange. SMPTE J. 86 (2 (Febru-

ary)), 65–68.

Deschenes, G., Ritcey, G.M., 1990. Recovery of gold from aqueous

solutions. US Patent 4913730 (April 1990). Canadian Patents

and Development, Canada, p. 8.

Dicinoski, G.W., 2000. Novel resins for the selective extraction of

gold from copper rich ores. S. Afr. J. Chem. 53 (1), 33–43.

EEC/FDA, 2001. (i) European MSDS: Directive 67/548/EEC. (ii)

American MSDS: 29 CFP 1910-1200 (US FDA).

Elsner, L., 1846. J. Prakt. Chem. 37, 441–446.

Fagan, P., 2000. Personal communication: Report on the Ballarat

Gold Forum—Gold Processing in the 21st Century: An Interna-

tional Forum, Ballarat, Australia.

Fagan, P.A., Paull, B., Haddad, P.R., Dunne, R., Kamar, H., 1997.

Ion chromatographic analysis of cyanate in gold processing

samples containing large concentrations of copper(I) and other

metallo-cyanide complexes. J. Chromatogr. 770, 175–183.

Feng, D., Van Deventer, J.S.J., 2002. Leaching behaviour of sul-

phides in ammoniacal thiosulphate systems. Hydrometallurgy

63, 189–200.

Ferron, C.J., Turner, D.W., Stogran, K., 1998. Thiosulfate leaching of

gold and silver ores: an old process revisited. CIM 100th Annual

General Meeting, May 1998, Montreal, Quebec., pp. 1–11.

#THAM 61.3.

Fleming, C.A., 1998. The potential role of anion exchange resins in

the gold industry. EPD Congress 1998. The Minerals, Metals

and Materials Society, Warrendale, PA, USA, pp. 95–117.

Fleming, C.A., McMullen, J., Thomas, K.G., Wells, J.A., 2001.

Recent advances in the development of an alternative to the

cyanidation process—based on thiosulfate leaching and resin

in pulp. SME Annual Meeting (2001): Process Mineralogy II:

Precious Metals. Feb. 2001. SME, Denver, CO. 11 pp.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–2118

Flett, D.S., Derry, R., Wilson, J.C., 1983. Chemical study of

thiosulfate leaching of silver sulfide. Trans. Inst. Min. Met-

all. (Sect. C: Miner. Process. Extr. Metall.) 92 (December),

216–223.

Gallagher, N.P., Hendrix, J.L., Milosavljevic, E.B., Nelson, J.H.,

1989. The affinity of carbon for gold complexes: dissolution

of finely disseminated gold using a flow electrochemical cell.

J. Electrochem. Soc. 136 (9), 2546–2551.

Gallagher, N.P., Hendrix, J.L., Milosavljevic, E.B., Nelson, J.H.,

Solujic, L., 1990. Affinity of activated carbon towards some

gold(I) complexes. Hydrometallurgy 25, 305–316.

Genik-Sas-Berezowsky, R.M., Sefton, V.B., Gormely, L.S., 1978.

Recovery of precious metals from metal sulfides. US 4,070,182.

Sherritt Gordon Mines, Toronto, Canada. 12 pp., 3 drawing

figures.

Gmelin, L. (Ed.), 1965. Copper thiosulfate complexes. In: Gmelins

Handbuch der Anorganischen Chemie. Verlag Chemie: Berlin,

pp. (B1) 591; (B3) 998–1003, 1414–1415.

Gmelin, L. (Ed.), 1969. Mercury thiosulfate complexes. In: Gme-

lins Handbuch der Anorganischen Chemie. Verlag Chemie: Ber-

lin, pp. (B1) 52–55; (B3) 1030–1031; (B4) 1406–1409.

Gmelin, L. (Ed.), 1972. Lead thiosulfate complexes. In: Gmelins

Handbuch der Anorganischen Chemie. Verlag Chemie: Wein-

heim, pp. (B1) 364–365; (C) 594–600.

Gmelin, L. (Ed.), 1973. Silver thiosulfate complexes. In: Gmelins

Handbuch der Anorganischen Chemie. Verlag Chemie: Wein-

heim/Bergstrasse, pp. (B3) 110–133.

Gong, Q., Hu, J., Cao, C., 1993. Kinetics of gold leaching from

sulfide gold concentrates with thiosulfate solution. Trans. Non-

ferr. Met. Soc. China 3 (4), 30–36.

Gowland, W., 1930. The metallurgy of the non-ferrous metals, 4th

ed. Bannister, C.O. (Ed.), Vol. 1. Charles Griffin & Co., London,

pp. 378–381, 415–418.

Grigorova, B., Wright, S.A., Josephson, M., 1987. Separation and

determination of stable metallo-cyanide complexes in metallur-

gical plant solutions and effluents by reverse phase ion pair

chromatography. J. Chromatogr. 410, 419–426.

Groudev, S.N., Spasova, I.I., Ivanov, I.M., 1996. Two-stage micro-

bial leaching of a refractory gold-bearing pyrite ore. Miner. Eng.

9 (7), 707–713.

Groves, W.D., Blackman, L., 1995. Recovery of precious metals

from evaporite sediments. US 5,405,430. United States, 17

pages.

Guerra, E., Dreisinger, D.B., 1999. A study of the factors affecting

copper cementation of gold from ammoniacal thiosulfate solu-

tion. Hydrometallurgy 51, 155–172.

Gundiler, I.H., Goering, P.D., 1993. Thiosulfate leaching of gold

from copper-bearing ores. SME Annual Meeting, Feb., SME,

Reno, Nevada. Paper #93-281.

Hemmati, M., Hendrix, J.L., Nelson, J.H., Milosavljevic, E.B.,

1989. Study of the thiosulfate leaching of gold from carbona-

ceous ore and the quantitative determination of thiosulfate in the

leached solution. Extraction Metallurgy ’89. The Institute of

Mining & Metallurgy, London, UK, pp. 665–678.

Hilton, D.F., Haddad, P.R., 1986. Determination of metal- cyano-

complexes by reversed-phase ion-interaction high-performance

liquid chromatography and its application to the analysis of

precious metals in gold processing solutions. J. Chromatogr.

361 (27), 141–150.

Hu, J., Gong, Q., 1991. Substitution of sulfate for sulfite during

extraction of gold by thiosulfate solution. Randol Gold Forum.

Randol International (USA), Cairns, Queensland, pp. 333–336.

Huang, Q., Paull, B., Haddad, P.R., 1997. Optimisation of selectiv-

ity in the separation of metallo-cyanide complexes by ion-inter-

action liquid chromatography. J. Chromatogr. 770, 3–11.

Hubin, A., Vereecken, J., 1994. Electrochemical reduction of silver

thiosulfate complexes: Part 1. Thermodynamic aspects of solu-

tion composition. J. Appl. Electrochem. 24, 239–244.

Jay, W.H., 1999. Process for Recovery of Gold and/or Silver, WO

99/13116: (March 1999), Arton (No. 001) Pty. Ltd. Australia;

Everett and Goodall Electrical Pty. Ltd.: Australia. PCT/AU98/

00722. 39 pp.

Jiang, T., Xu, S., Chen, J., 1992. Gold and silver extraction by

ammoniacal thiosulfate catalytical leaching at ambient temper-

ature. Proceedings of the First International Conference on Mod-

ern Process Mineralogy and Mineral Processing, pp. 648–653.

Jiang, T., Chen, J., Xu, S., 1993a. Electrochemistry and mechanism

of leaching gold with ammoniacal thiosulfate. XVIII Internation-

al Mineral Processing Congress. AusIMM, Sydney, Australia.

Jiang, T., Chen, J., Xu, S., 1993b. A kinetic study of gold leaching

with thiosulfate. Hydrometallurgy: Fundamentals, Technology

and Innovations. Soc. Min., Metall. Explor., Littleton, CO,

USA, pp. 119–126.

Johnson, P.H., Bhappu, R.B., 1969. Chemical mining—a study of

leaching agents. Circular—New Mexico, Bureau of Mines and

Mineral Resources. New Mexico Burea of Mines and Mineral

Resources, Socorro, NM, USA, pp. 1–10.

Kerley, B.J.J., 1981. Recovery of precious metals from difficult ores.

US Patent 4,269,622. Kerley Industries, United States. 4 pp.

Kerley, B.J.J., 1983. Recovery of precious metals from difficult

ores. US Patent 4,369,061. Kerley Industries, United States.

6 pp.

Kononova, O.N., Kholmogorov, A.G., Kononov, Y.S., Pashkov,

G.L., Kachin, S.V., Zotova, S.V., 2001. Sorption recovery of

gold from thiosulfate solutions after leaching of products of

chemical preparation of hard concentrates. Hydrometallurgy

59, 115–123.

Langhans, J.W.J., Lei, K.P.V., Carnahan, T.G., 1992. Copper-cata-

lysed thiosulfate leaching of low-grade gold ores. Hydrometal-

lurgy 29, 191–203.

Lasko, C.L., Hurst, M.P., 1999. An investigation into the use of

chitosan for the removal of soluble silver from industrial waste-

water. Environ. Sci. Technol. 33 (20), 3622–3626.

Li, J., Miller, J.D., Wan, R.Y., Le Vier, M., 1995. The ammoniacal

thiosulfate system for precious metal recovery. Proceedings of

the XIX International Mineral Processing Congress (IMPC),

San Francisco, CA, USA. SME, Littleton, CO, USA.

Livingstone, S.E., 1965. Metal complexes of ligands containing

sulphur, selenium or tellurium as donor atoms. Q. Rev. 19,

386–397.

Lulham, J., Lindsay, D., 1991. Separation process. International

Patent WO 91/11539. Davy McKee (Stockton) Limited, Cleve-

land TS18 3RE (GB), Great Britain. 17 pp.

Lurye, Y.Y., Peremyslova, E.S., 1953. Secondary processes occur-

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 19

ring during ion exchange on ionites. Russ. J. Appl. Chem. 27

(11), 1143–1147.

MacKenzie, J.M.W., Virnig, M.J., Johns, M.W., 1995. Henkel

Aurix(R) resin—an update. Randol Gold Forum ’95. Perth,

West Australia.

Marchbank, A.R., Thomas, K.G., Dreisinger, D., Fleming, C.,

1996. Gold recovery from refractory ores by pressure oxidation

and thiosulfate leaching. US Patent 5,536,297. Barrick Gold

Corporation, Toronto, Canada. 12 pp., 5 drawing sheets.

Marcus, Y., 1957. The anion exchange of metal complexes. The

silver– thiosulfate system. Acta Chem. Scand. 11 (4), 619–627.

Marsden, J., House, I., 1992. The Chemistry of Gold Extraction.

Ellis Horwood, New York.

McNulty, T., 2001. Cyanide substitutes. Min. Mag., 256–261

(May).

McPartland, J.S., Bautista, R.G., 1999. Concentration and reduction

of Au(I) thiosulfate to metallic gold. In: Liddell, K.C., Chaiko,

D.J. (Eds.), Metal Separation Technologies Beyond 2000: Inte-

grating Novel Chemistry With Processing. The Minerals, Metals

and Materials Society, Warrendale, PA, USA, pp. 105–115.

Meggiolaro, V., Niccolini, G., Miniussi, G., Stefanelli, N., Trivellin,

E., Llinas, R., Ramirez, I., Baccelle Scudeler, L., Omenetto, P.,

Primon, S., Visona, D., Abbruzzese, C., Fornari, P., Massidda,

R., Piga, L., Ubaldini, S., Ball, S., Monhemius, A.J., 2000.

Multidisciplinary approach to metallogenic models and types

of primary gold concentration in the Cretaceous arc terranes

of the Dominican Republic. Trans. Inst. Min. Metall. Section

B (Applied Earth Science) 109, B95–B104 (May–Aug).

Mina, R., 1980. Silver recovery from photographic effluents by ion-

exchange methods. J. Appl. Photogr. Eng. 6 (5), 120–125.

Mizoguchi, T., Takei, Y., Okabe, T., 1976. The chemical behaviour

of low valence sulfur compounds: X. Disproportionation of thi-

osulfate, trithionate, tetrathionate and sulfite under acidic con-

ditions. Bull. Chem. Soc. Jpn. 49 (1), 70–75.

Mohansingh, R., 2000. Adsorption of gold from gold copper am-

monium thiosulfate complex onto activated carbon and ion ex-

change resins. MSc thesis, University of Nevada, Reno. 71 pp.

Mountain, B.W., Wood, S.A., 1988. Solubility and transport of plat-

inum-group elements in hydrothermal solutions: thermodynamic

and physical chemical constraints. In: Prichard, H.M., Potts, P.J.,

Bowles, J.F.W., Cribb, S.J. (Eds.), Geo-Platinum ’87, Open Uni-

versity, Milton Keynes. Elsevier, London, pp. 57–82.

Muir, D.M., LaBrooy, S.R., Deng, T., Singh, P., 1993. The mech-

anism of the ammonia–cyanide system for leaching copper–

gold ores. In: Hiskey, J.B., Warren, G.W. (Eds.), Hydrometal-

lurgy: Fundamentals, Technology and Innovation. AIME, War-

rendale, pp. 191–204.

Naito, K., Shieh, M.-C., Okabe, T., 1970. The chemical behaviour

of low valence sulfur compounds: V. Decomposiion and oxida-

tion of tetrathionate in aqueous ammonia solution. Bull. Chem.

Soc. Jpn. 43, 1372–1376.

Nickless, G., 1968. Inorganic Sulphur Chemistry. Elsevier, Am-

sterdam.

NOHSC, 1999. Approved Criteria for Classifying Hazardous Sub-

stances, National Occupational Health and Safety Commission,

Sydney, Australia.

Novakovskii, M.S., Ryazantseva, A.P., 1955. Cadmium complexes

with thiosulfate. Trudy Khim. Fak. 54 (12), 277–281 (In

Russian).

O’Malley, G.P., 2001. The Elution of Gold from Anion Exchange

Resins, International Patent WO 01/23626 A1 (April 2001).

Murdoch University, Perth, WA. 20 pp.

O’Malley, G.P., Nicol, M.J, 2001. Recovery of gold from thiosul-

fate soliutions and pulps with ion exchange resins. TMS Confer-

ence and Annual Meeting (2001): Cyanide: Social, Industrial,

and Economic Aspects: Alternatives. The Minerals, Metals and

Materials Society, Warrendale, PA, USA, pp. 469–483.

Panayotov, V., Stamenov, S., Panayotova, M., Todorova, E., 1995.

Potentiostatic investigation of the Au and Ag dissolution. 6th

Balkan Conference on Mineral Processing, Ohrid, Macedonia.

Perez, A.E., Galaviz, H.D., 1987. Method for recovery of precious

metals from difficult ores with copper–ammonium thiosulfate.

US Patent 4654078 A. 9 pp.

Plimer, I.R., Williams, P.A., 1988. New mechanisms for the mobi-

lisation of the platinum-group elements in the supergene zone.

In: Prichard, H.M., Potts, P.J., Bowles, J.F.W., Cribb, S.J. (Eds.),

Geo-Platinum ’87, Open University, Milton Keynes. Elsevier,

London, pp. 83–92.

Roy, A.B., Trudinger, P.A., 1970. The Biochemistry of Inorganic

Compounds of Sulfur. Cambridge Univ. Press, Cambridge.

Ryabchikov, D.I., 1943. On the structure of dithiosulphatoplatinite.

C. R. (Doklady) Acad. Sci. URSS, XLI (5), 208–209.

Schmitz, P.A., Duyvesteyn, S., Johnson, W.P., Enloe, L., McMullen,

J., 2001. Ammoniacal thiosulfate and sodium cyanide leaching

of preg-robbing Goldstrike ore carbonaceous matter. Hydrome-

tallurgy 60, 25–40.

Smith, R.M., Martell, A.E., 1976. Critical stability constants: Vol. 4.

Inorganic ligands. Critical Stability Constants (Vol. 4). Plenum,

London.

Sparrow, G.J., Woodcock, J.T., 1995. Miner. Process. Extr. Metall.

Rev. 14, 193–247.

Tarasova, A.A., Serova, I.B., Leikin, Y.U.A., 1994. Silver sorption

from solutions formed in production and treatment of motion

picture and photographic materials. Russ. J. Appl. Chem. 67 (6),

854–857.

Thomas, K.G., Fleming, C., Marchbank, A.R., Dreisinger, D.,

1998. Gold recovery from refractory carbonaceous ores by pres-

sure oxidation, thiosulfate leaching and resin-in-pulp adsorption.

US Patent 5,785,736. Barrick Gold Corporation, Toronto, Can-

ada. 24 pp.

Torres, V.M., Costa, R.S., 1994. Characterisation of gold ores and

CIP tailings using a diagnostic leaching technique. Proceedings

of the XIX IMPC, pp. 15–18.

Tozawa, K., Inui, Y., Umetsu, Y., 1981. Dissolution of gold in

ammoniacal thiosulfate solution. AIME 110th Annual Meeting,

vol. A81-25, pp. 1–12.

Tykodi, R.J., 1990. In praise of thiosulfate. J. Chem. Educ. 67 (2),

146–149.

Umetsu, Y., Tozawa, K., 1972. Dissolution of gold in ammoniacal

sodium thiosulfate solution. Bull. Res. Inst. Miner. Dressing

Metall. 28 (1), 97–104.

Uri, N., 1947. Thiosulfate complexes of the tervalent metals iron,

aluminium, and chromium. J. Chem. Soc., 335–337.

Valensi, G., 1973. Electrochemical behaviour of sulfur. Potential-

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–2120

pH equilibrium diagrams of the sulfur–water system at 25j, 1atm. Rapp. Tech.-Cent. Belge Etude Corrosion 121 (207), 22.

Vasil’ev, A., Toropova, V.F., Busygina, A.A., 1953. The use of

ion exchange for the separation of copper, cadmium and zinc

from thiosulfate solutions. Uch. Zap. Kazausk. Un-ta. 113 (8),

91–102.

Virnig, M.J., Sierakoski, J.M., 2001. Ammonium thiosulfate com-

plex of gold or silver and an amine. US Patent 6197214 (March

2001). Henkel. 10 pp.

Vlassopoulos, D., Wood, S.A., 1990. Gold speciation in natural

waters: I. Solubility and hydrolysis reactions of gold in aqueous

solution. Geochim. Cosmochim. Acta 54 (1), 3–12.

Wan, R.Y., 1997. Importance of solution chemistry for thiosulfate

leaching of gold. World Gold ’97 Conference, Sept. 1997, Sin-

gapore, pp. 159–162.

Wan, R.Y., Brierley, J.A., 1997. Thiosulfate leaching following

biooxidation pretreatment for gold recovery from refractory car-

bonaceous–sulfidic ore. Min. Eng. (USA) 49 (8), 76–88.

Wan, R.Y., Le Vier, M., Miller, J.D., 1993. Chapter 27: Research

and development activities for the recovery of gold from non-

cyanide solutions. Proceedings of the Milton E. Wadsworth (IV)

International Symposium on Hydrometallurgy. Hydrometal-

lurgy: Fundamentals, Technology, and Innovations. Society for

Mining, Metallurgy, and Exploration, Salt Lake City, UT, USA.

Webster, J.G., 1986. The solubility of gold and silver in the system

Au-Ag-S-O2-H2O at 25 jC and 1 atm. Geochim. Cosmochim.

Acta 50 (9), 1837–1845.

Wilkinson, G.1987. Comprehensive Coordination Chemistry. Per-

gamon, Oxford.

Williamson, M.A., Rimstidt, J.D., 1993. The rate of decomposition

of the ferric-thiosulfate complex in acidic aqueous solutions.

Geochim. Cosmochim. Acta 57 (15), 3555–3561.

Woollacott, L.C., Eric, R.H., 1994. Hydrometallurgy. Mineral and

Metal Extraction: An Overview. The South African Institute of

Mining and Metallurgy, Johannesburg, pp. 321–370.

Xu, Y., Schoonen, M.A.A., 1995. The stability of thiosulfate in the

presence of pyrite in low-temperature aqueous solutions. Geo-

chim. Cosmochim. Acta 59 (22), 4605–4622.

Yen, W.T., Aghamirian, M.M., Deschenes, G., Theben, S., 1998.

Gold extraction from mild refractory ore using ammonium thi-

osulfate. International Symposium on Gold Recovery. Preprint

Manuscript (#THAM 61.1), 11 pp.

Zhao, J., Wu, Z.C., Chen, J.Y., 1997. Extraction of gold from

thiosulfate solutions with alkyl phosphorus esters. Hydrometal-

lurgy 46 (3), 363–372.

Zhao, J., Wu, Z., Chen, J., 1998a. Extraction of gold from thiosul-

fate solutions using amine mixed with neutral donor reagents.

Hydrometallurgy 48, 133–144.

Zhao, J., Wu, Z., Chen, J., 1998b. Gold extraction from thiosulfate

solutions using mixed amines. Solv. Extr. Ion Exch. 16 (6),

1407–1420.

Zhao, J., Wu, Z., Chen, J., 1998c. Solvent extraction of gold in

thiosulfate solutions with amines. Solv. Extr. Ion Exch. 16 (2),

527–543.

Zhao, J., Wu, Z., Chen, J., 1999. Separation of gold from other

metals in thiosulfate solutions by solvent extraction. Sep. Sci.

Technol. 34 (10), 2061–2068.

Zhu, G., Fang, Z.H., Chen, J.Y., 1994. Electrochemical studies on

the mechanism of gold dissolution in thiosulfate solutions.

Trans. Nonferr. Met. Soc. China 4 (1), 50–53, 58.

Zipperian, D., Raghavan, S., Wilson, J.P., 1988. Gold and silver

extraction by ammoniacal thiousulfate leaching from a rhyolite

ore. Hydrometallurgy 19 (3), 361–375.

A.C. Grosse et al. / Hydrometallurgy 69 (2003) 1–21 21