Hydrometallurgy 134–135 (2013) 124–131

Contents lists available at SciVerse ScienceDirect

Hydrometallurgy
journal homepage: www.elsevier.com/locate/hydromet

A kinetic–thermodynamic study of silver leaching in

thiosulfate–copper–ammonia–EDTA solutions
D.M. Puente-Siller, J.C. Fuentes-Aceituno ⁎, F. Nava-Alonso
Centro de Investigación y de Estudios Avanzados del IPN Unidad Saltillo, Av. Industria Metalúrgica Nº 1062, Parque Industrial Ramos Arizpe, Ramos Arizpe, Coahuila, 25900, Mexico

a r t i c l e i n f o a b s t r a c t

Article history: In this research, an analysis of the effect of ethylenediaminetetraacetic acid (EDTA), thiosulfate and cupric
Received 25 April 2012 ions on the silver leaching kinetics was performed. For that purpose leaching experiments with pure metallic
Received in revised form 31 January 2013 silver were carried out at different concentrations of these reagents at room temperature. The results of this
Accepted 14 February 2013
study showed that small amounts of EDTA accelerate the leaching of silver more efficiently than higher con-
Available online 27 February 2013
centrations due to an increase in the oxidation potential of the leaching solution. These results were in accord
Keywords:
with Pourbaix diagrams and redox potential measurements. A synergistic effect was found in the silver
Leaching leaching kinetics when the thiosulfate and EDTA concentrations were both decreased. On the other hand,
Thiosulfate the silver leaching kinetics was reduced at low Cu(II) concentrations due to the decrease in the oxidizing abil-
EDTA ity of the leaching solution.
Silver Characterization by SEM and EDXS of the silver in the first minutes of leaching revealed that the silver parti-
cles were coated by a layer of copper sulfides and copper oxides. Furthermore, characterization of the solid
residue during the precipitation of silver in the leaching experiment was performed by the same analytical
techniques, showing the presence of silver sulfide on the unreacted silver surface. The mechanism by
which the silver sulfide is precipitated was found to be related to the copper sulfide formation. It was also
observed that an increase in the concentration of EDTA promotes the silver dissolution avoiding the forma-
tion of copper sulfides or oxides; in these cases the process is controlled by the chemical reaction. These ob-
servations were also supported with Pourbaix and species distribution diagrams.
© 2013 Elsevier B.V. All rights reserved.

1. Introduction the gold dissolution kinetics by 18 to 20 times (Aylmore and Muir,

2001).
Cyanide leaching is effective for gold and silver extraction from The chemistry of this alternative leaching system is very compli-
minerals in aqueous solutions. The precious metals are generally re- cated due to the occurrence of several chemical and electrochemical
covered with cyanide as complexing agent and oxygen as oxidant reactions during the precious metal leaching. In addition to the silver
(Zhang, 2004). or gold oxidation reaction, some thiosulfate oxidative degradation oc-
The need of processing difficult-to-treat ores as well as the environ- curs (Heath et al., 2008) involving the formation of tetrathionate and
mental risks associated to cyanide use has encouraged the research and other sulfur compounds (Aylmore and Muir, 2001), as well as several
development of new leaching systems for precious metals. The thiosul- silver and copper complexes with ammonia and thiosulfate.
fate–copper–ammonia system is considered a promising alternative to It has been reported that the leaching kinetics of silver can be fa-
cyanide (Alonso-Gómez and Lapidus, 2009). The method has been stud- vored by an increase in the thiosulfate concentration (Zipperian et
ied formally by different research groups since 1980 (Abbruzzese et al., al., 1988); furthermore, the silver pregnant solution can be continu-
1995; Feng and Van Deventer, 2006). ously electrodeposited avoiding the silver re-adsorption in the ore
In this process cupric ions oxidize the precious metals while thiosul- (Preg-Robbing phenomenon) (Fuentes-Aceituno et al., 2004). These
fate forms complexes with them. Ammonia is also added to stabilize the findings suggest that the leaching of silver in thiosulfate solutions
cupric ions in the alkaline solution, avoiding their precipitation as cop- would be a very promising alternative process; however, the high
per hydroxide (Feng and Van Deventer, 2007). Only few oxidizing consumption of cupric ions and thiosulfate has made the thiosulfate
agents, such as cupric ions, are adequate for working in ammoniacal so- leaching system uneconomical compared with the cyanide leaching,
lutions (Abbruzzese et al., 1995); the presence of cupric ions enhances so the process has not been widely used on a commercial scale
(Aylmore and Muir, 2001).
Consequently, scientists and engineers have focused the research
⁎ Corresponding author. Tel.: +52 844 438 9600x8512. on increasing the oxidizing capability of this leaching system and, at
E-mail address: [email protected] (J.C. Fuentes-Aceituno). the same time, trying to avoid the oxidative loss of thiosulfate. Several

0304-386X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.hydromet.2013.02.010
 D.M. Puente-Siller et al. / Hydrometallurgy 134–135 (2013) 124–131 125

investigations have been carried out to find an additional reagent for the possible redox reactions that could take place. The diagrams
complexing cupric ions with the aim of increasing the silver and gold were constructed considering the equilibrium data contained in the
leaching kinetics but avoiding the thiosulfate oxidation (Alonso-Gómez Hydra database of the Medusa© suite software (Puigdomenech, 2004),
and Lapidus, 2009; Feng and Van Deventer, 2010). complemented with thermodynamic data reported by Aylmore and
Several studies have shown that EDTA (ethylenediaminetetraacetic Muir (2001). Table 1 presents the Gibbs free energy data complemented
acid) is an effective complexing agent for increasing the extraction and used in this work. The software is based in an algorithm developed by
kinetics of heavy metals such as Pb, Cu, Zn and Cd (Manouchehri et Eriksson (1979), which minimizes the Gibbs free energy of reactions in
al., 2006). EDTA is more efficient than other complexing agents like cit- equilibrium that can occur in the aqueous leaching systems, and deter-
rate (Labanowski et al., 2008). mines the predominating species under specific solution conditions.
In the precious metals leaching with the thiosulfate system it has
been demonstrated that the presence of EDTA decreases the thiosul- 3. Results and discussion
fate oxidation rate by the formation of complexes of EDTA with the
Cu(II) ions; the oxidizing potential of the leaching solution decreases The results of this study are discussed in four parts; in the first part
(Alonso-Gómez and Lapidus, 2009; Feng and Van Deventer, 2010). It appears the effect of the thiosulfate and EDTA concentrations on the
has been also established that the addition of small amounts of EDTA silver leaching; the second section analyzes the kinetic data with
improves substantially the thiosulfate leaching of gold (Feng and Van the shrinking core model; the third section presents the thermody-
Deventer, 2011); however, there is very little information as well as a namic analysis of the studied systems, and the fourth, the effect of
lack of understanding about the phenomena that controls the silver the Cu(II) concentration on the silver leaching.
leaching kinetics with EDTA as cupric ion complexing agent.
The objective of this study is to elucidate the phenomena involved in 3.1. Effect of EDTA and thiosulfate concentrations on the silver leaching
the kinetics of silver dissolution with the thiosulfate–copper–ammonia–
EDTA system, in an attempt to increase the silver recovery and to mini- Fig. 1 illustrates the effect of three different EDTA concentrations
mize the reagent consumption. (1.25 × 10−4 M, 0.025 M and 0.05 M) on the silver dissolution kinetics
at room temperature. The thiosulfate and copper concentrations were
2. Materials and methods kept constant at 0.2 M and 0.05 M respectively. From these experi-
ments it was observed that the leaching system with the lowest EDTA
For the leaching tests the composition of the solution was changed concentration (1.25 × 10−4 M) enhanced the silver leaching kinetics.
in order to study the effect of thiosulfate, EDTA and copper on the sil- This behavior was also observed in the leaching experiments for
ver leaching. Pourbaix and species predominance diagrams were 0.1 M and 0.3 M thiosulfate and the same EDTA concentrations
constructed for the system, and the kinetic model of shrinking core (Figs. 2 and 3 respectively). The leaching solution with 0.1 M thiosulfate
was applied to the results obtained. and 1.25 × 10−4 M EDTA reached a silver extraction of 100% in the first
20 min (see Fig. 2); after this time the silver extraction decreased to
2.1. Leaching tests 90%, probably due to the precipitation of silver, which suggests an un-
stable leaching system. All the leaching experiments carried out with
All the silver leaching experiments were performed in an agitated solutions containing 1.25 × 10−4 M and 0.025 M EDTA showed some
batch reactor (250 mL glass beaker), magnetically stirred at 400 rpm. oscillations in the silver extraction curves (Figs. 1, 2 and 3) indicating
The reactor was provided with a pH electrode (SENSOREX) and a reprecipitation and, in some cases, redissolution of the silver. A possible
redox potential electrode with calomel reference (Cole-Parmer). The explanation to this phenomenon is proposed by Abbruzzese et al.
general procedure for the leaching experiments was as follows: (1995): in the initial stage of the leaching process the dissolution of me-
200 mL of the solution containing cupric sulfate, thiosulfate, EDTA tallic silver (Eq. (1)) occurs simultaneously with the reduction of cupric
and ammonium were placed in the reactor at 25 °C. The concentra- ions to cuprous species (Eq. (2)).
tions of cupric ions, thiosulfate and EDTA were varied as follows:
−
7.87 × 10 −4 M and 0.05 M for the cupric sulfate; 0.1 M, 0.2 M and Ag þ 3S2 O32− →AgðS2 O3 Þ35− þ e ð1Þ
0.3 M for thiosulfate; and 1.25 × 10 −4 M, 0.025 M and 0.05 M for −
EDTA. All solutions were prepared with reagent grade chemicals CuðNH3 Þ42þ þ 3S2 O32− þ e →CuðS2 O3 Þ35− þ 4NH3 ð2Þ
and deionized water, and the pH value was adjusted to 10.2 with am-
monium hydroxide. After adjusting pH, 0.125 g of metallic silver (Alfa Besides these reactions, other simultaneous reactions may occur
Aesar, 99.999%) (1.3–3.2 μm) was placed in the reactor while stirring. such as the oxidative decomposition of thiosulfate involving the for-
The pH and redox potential were measured before the silver addition mation of additional sulfur compounds such as tetrathionate, as
and during the leaching experiments. Samples of the leach solution well as the precipitation of copper sulfides species such as CuS or
were withdrawn at different times during the leach period for analy- Cu2S (Eqs. (3) and (4)).
sis of silver by atomic absorption spectrophotometry (VARIAN Spec- 2þ 2− 2− þ
tra AA 240). All samples for silver analysis were analyzed at the end Cu þ S2 O3 þ H2 O→CuS þ SO4 þ 2H ð3Þ
of the experiment, and before the analysis they were kept in dark ves-
sels to prevent the precipitation of silver by light. Once the experiment Table 1
Gibbs free energy for thiosulfate and silver species used in the con-
was concluded, the residue was filtered, rinsed with deionized water struction of thermodynamic diagrams (25 °C, Aylmore and Muir,
and left to air dry. The solid residues were characterized chemically 2001).
and morphologically by SEM (Phillips XL30ESEM). The local chemical
Chemical species ΔG°298 (kJ/mol)
composition of the residues was determined by EDXS (EDAX Genesis).
S2O32−(a) −532.2
Ag2S −40.5
2.2. Elaboration of the thermodynamic diagrams
Ag2+(a) 268.2
Ag+(a) 77.2
In order to elucidate the predominant species in each leaching sys- Ag(S2O3)35−(a) −1598.3
tem, Pourbaix and species distribution diagrams were constructed Ag(S2O3)23−(a) −1058.6
using the Medusa© Software. These diagrams are also useful for de- Ag(S2O3)−(a) −506.3
Ag2O3(c) 87.0
termining the oxidizing ability of the leaching systems, as well as
126 D.M. Puente-Siller et al. / Hydrometallurgy 134–135 (2013) 124–131

Fig. 1. Effect of EDTA concentration on silver leaching for a solution of 0.2 M thiosulfate
Fig. 3. Effect of EDTA concentration on silver leaching for a solution of 0.3 M thiosulfate
(0.125 g silver, 200 mL solution, 0.05 M Cu(II), 0.6 M NH4OH, pH = 10.2, room tem-
(0.125 g silver, 200 mL solution, 0.05 M Cu(II), 0.6 M NH4OH, pH = 10.2, room tem-
perature and 400 rpm).
perature and 400 rpm).

0.3 M thiosulfate because of its lower reagent consumption and

þ þ
2Cu þ S2 O32− þ H2 O→Cu2 S þ SO42− þ 2H ð4Þ higher silver leaching kinetics; nevertheless, from the perspective of
chemical reaction kinetics, it could be expected that an increase in
Once the copper sulfide species are formed, the silver contained in
the thiosulfate concentration would enhance the silver leaching
the thiosulfate complex (Ag(S2O3)35−) could precipitate as silver sul-
kinetics. These results suggest that in systems with 0.2 M and 0.3 M
fide according to Eq. (5) (Zipperian et al., 1988):
thiosulfate, thiosulfate could be decomposed more than in the system
Cu2 S þ 2AgðS2 O3 Þ35− →Ag2 S þ 2CuðS2 O3 Þ35− : ð5Þ with 0.1 M. This behavior does not imply that there is no thiosulfate
decomposition at 0.1 M, but that this decomposition is smaller and
This reaction would explain the decrease on silver concentration can only be noticed further (after 4 h in Fig. 2, for 0.025 M EDTA),
in the leaching curves of Figs. 1, 2 and 3. Once the silver sulfide is when the silver leaching decreases. It can also be observed, in the sys-
formed, it is possible to redissolve it by a substitution of copper by sil- tems with 0.1 M thiosulfate and 1.25 × 10 −4 M or 0.025 M EDTA,
ver in the sulfide matrix, with the consequent formation of a silver that the silver extraction reaches an almost constant maximum
thiosulfate complex in solution (Aylmore and Muir, 2001): value of silver extraction of 92 and 80% respectively. This behavior
is probably related to the precipitation of silver and copper as sulfides
þ þ
2Cu þ Ag2 S→2Ag þ Cu2 S þ 6S2 O32− →2AgðS2 O3 Þ35− : ð6Þ or oxides on the unreacted silver particle surface, generating passive
layers which inhibit a complete silver extraction (see Fig. 2).
In spite of the instability of the leaching systems presented in
The EDTA concentration affects the silver dissolution. It was ob-
Figs. 1–3, it is possible to say that the leaching system with 0.1 M
served in Figs. 1, 2 and 3 that an increase in the EDTA concentration
thiosulfate (Fig. 2) presents fewer oscillations (dissolution–precipita-
decreases the silver extraction. This phenomenon can be explained
tion phenomena) than the system with 0.2 M thiosulfate (Fig. 1).
in terms of thermodynamics and electrochemistry: the Cu (II) com-
From a practical point of view, this leaching system with 0.1 M thio-
plex with EDTA is more stable in 0.05 M EDTA than in 0.025 M
sulfate would be more interesting than the systems with 0.2 and
EDTA solutions. Besides, the oxidizing ability of the leaching solution
decreases in the system with 0.05 M EDTA, causing the diminution in
silver leaching kinetics under this condition. Further evidence of this
phenomenon is presented in Fig. 4a and b, and will be discussed later.
From these results three important characteristics for the thiosul-
fate–EDTA leaching system can be highlighted: a) a decrease in the thio-
sulfate and EDTA concentrations favors the silver leaching; b) the
leaching solution with 0.1 M thiosulfate is more stable than the other
two thiosulfate concentrations; and c) a catalytic effect was found in
the silver leaching kinetics when EDTA concentration was lowered
(see Figs. 1, 2 and 3). If this last statement is true, there should be a dif-
ference in the redox potentials of the leaching solutions, i.e. the redox
potential should become more positive when the EDTA is decreased, re-
vealing that the oxidizing ability of the leaching solution is enhanced,
and therefore the silver dissolution kinetics could be favored.
Redox potential measurements were performed for the leaching so-
lutions in the tests of Figs. 1–3. For example, the leaching solution with
0.025 M EDTA and 0.2 M thiosulfate had a redox potential of 0.4018 V
vs SHE; when the EDTA concentration was increased to 0.05 M for the
same thiosulfate concentration (0.2 M), the redox potential decreased
Fig. 2. Effect of EDTA concentration on silver leaching for a solution of 0.1 M thiosulfate
to 0.323 V vs SHE. This difference in the redox potential indicates that
(0.125 g silver, 200 mL solution, 0.05 M Cu(II), 0.6 M NH4OH, pH = 10.2, room tem- the EDTA concentration has an important effect on the oxidizing ability
perature and 400 rpm). of the leaching solution, and consequently, on the silver dissolution
 D.M. Puente-Siller et al. / Hydrometallurgy 134–135 (2013) 124–131 127

with the shrinking core model. It was found that in some cases the
process was controlled by the chemical reaction.
Eq. (7) is proposed for a process whose kinetics is controlled by a
chemical reaction (Levenspiel, 1972):
 
1=3
t bks C Al =Rp ρB ¼ 1–ð1–XÞ ð7Þ

where t = reaction time, b = reaction stoichiometric coefficient,

ks = rate constant for the surface reaction, CAl = concentration of
the reactive fluid species, Rp = particle radius, ρB = molar density
of silver in the solid sample and X = fractional conversion (extrac-
tion) of the silver.
On the other hand, Eq. (8) can be used for processes whose kinet-
ics is controlled by the diffusion of the fluid species through a porous
layer (Levenspiel, 1972):
 
2 2=3
t 6bDe C Al =Rp ρB ¼ 1–3ð1–XÞ þ 2ð1–XÞ ð8Þ

where De is the effective diffusion coefficient of the reactive fluid spe-

cies through the porous layer.
The silver leaching results shown in Fig. 1 for 0.2 M thiosulfate and
two EDTA concentrations (0.025 M and 0.05 M) were fitted to the
shrinking core model using Eqs. (7) and (8) for the initial leaching
stage (30 min). It was decided to use these EDTA concentrations, and
only the initial leaching data, to avoid the interference of the silver pre-
cipitation as proposed in Eq. (5). It is important to remember that at
0.025 M EDTA the copper can form oxides as shown in Fig. 4a. Further-
more the formation of copper sulfides is also possible as shown in
Eqs. (3) and (4). These copper species could represent a physical barrier
between the fluid reactive species and the unreacted silver sites. There-
fore, Fig. 5 shows the fitted data for the leaching solution with 0.025 M
EDTA (with Eq. (8)) where it can be observed that the regression line
does not represent adequately the porous layer diffusion mechanism.
Fig. 4. Effect of the EDTA concentration on the copper species predominance and oxi- This result suggests that the oxidizing agent (cupric ions) concentration
dizing ability of the leaching solution at 25 °C a) Pourbaix diagram for the Cu–NH3–
is changing during the silver leaching due to the formation of copper ox-
EDTA–H2O system with 0.025 M EDTA, 0.6 M NH4OH and 0.05 M Cu(II). b) Pourbaix
diagram for the same system with 0.05 M EDTA. ides and sulfides. According to Senanayake (2007), one of the require-
ments for using the shrinking core model is that the concentration of
reagents responsible for the surface reaction remains constant; there-
kinetics; i.e. if the redox potential is reduced from 0.4018 to 0.323 V vs fore a different kinetic model is necessary to represent better the silver
SHE, the silver dissolution kinetics will be slower. leaching in this dynamic system (0.025 M EDTA).
A thermodynamic analysis with the aid of Pourbaix diagrams was On the other hand, the data of the silver leaching with 0.2 M thio-
carried out in order to understand better the low leaching efficiency sulfate and 0.05 M EDTA were fitted to Eq. (7) and can be seen in
obtained at high EDTA concentrations (0.05 M). Fig. 4 presents two Fig. 6. In this case, the leaching kinetics is controlled by the silver dis-
predominance or Pourbaix diagrams for the Cu–EDTA–NH3–H2O sys- solution chemical reaction (Eq. (1)). This result is interesting: if the
tem at 0.025 and 0.05 M EDTA (Fig. 4a and b respectively). In these
diagrams the copper species Cu(EDTA) 2− and Cu(NH3)2+ predomi-
nate in the pH range from 1 to 10.4, while the copper oxides predom-
inate at pH values above 10.4. It is important to mention that both
species represent the half cell reduction reaction for the silver
oxidation. It can be observed that the equilibrium potential for the
Cu(EDTA)2−/Cu(NH3)2+ couple is modified considerably with the modifi-
cation of EDTA concentration. An increase in the EDTA concentration from
0.025 to 0.05 M (Fig. 4a–b) causes a more negative equilibrium potential
for the Cu(EDTA)2−/Cu(NH3)2+ couple (from −0.125 to −0.35 V vs SHE
approximately). According to the kinetic theory of electrochemical reac-
tions, it is well known that a decrease in the redox potential of a system
will affect exponentially the oxidation kinetics of the chemical species,
causing, in this case, the slower silver leaching kinetics observed. This is
in accordance with the leaching results shown before as well as with
the redox potential measurements carried out.

3.2. Kinetics of the silver leaching with 0.2 M thiosulfate

Fig. 5. Silver leaching data fitted to Diffusion Control Model, shrinking core in spherical
particles. Silver dissolution during the first 30 min of the leaching experiment (0.125 g
In order to understand the phenomena that control the silver silver, 200 mL solution, 0.2 M thiosulfate, 0.025 M EDTA, 0.05 M Cu(II), 0.6 M NH4OH,
leaching kinetics, some silver leaching experiments were analyzed pH = 10.2, room temperature and 400 rpm).
128 D.M. Puente-Siller et al. / Hydrometallurgy 134–135 (2013) 124–131

process is controlled by the chemical reaction, the silver dissolution is

expected to occur faster for 0.05 M EDTA solutions than for 0.025 M
EDTA due to the absence of copper oxides and sulfides on the silver
particle surface; the presence of copper oxides or sulfides would in-
deed decrease the Cu (II) concentration in the leaching solution.
However, the silver extraction curves of Fig. 1 showed that silver dis-
solution for 0.05 M EDTA reached only a 30% in 6 h. This behavior can
be explained by the decrease in the oxidizing ability of the leaching
solution when the EDTA is increased from 0.025 to 0.05 M, as was
shown in the Pourbaix diagram of Fig. 4b. It is important to note
that silver dissolution is an electrochemical nature reaction, and
hence it is expected that a decrease in the redox potential of the
leaching solution (when EDTA is increased from 0.025 to 0.05 M)
will slow down the silver leaching kinetics.
Fig. 7. Distribution of copper species at different pH values for the Cu–NH3–EDTA–H2O
system. The conditions for the diagram construction were 0.025 M EDTA, 0.05 M
3.3. Thermodynamic study of the Cu–NH3–EDTA–H2O, Cu–S–NH3-EDTA– Cu(II), 0.6 M NH4OH, and EH = 0.40 V at 25 °C.
H2O and Ag–S2O3–NH3–EDTA–H2O systems
solution; all the copper being complexed by the EDTA and ammonia.
It was observed that the silver extraction curves showed some oscil- This diagram indicates that the low efficiency of the leaching sys-
lations (precipitation–redissolution of silver). In addition to this, the tems shown in Figs. 1, 2 and 3 is not related to the copper oxide pre-
precipitation of copper oxides and sulfides affects the silver leaching cipitation but to the decrease in the redox potential of the couple
process with a 0.2 M thiosulfate solution containing 0.025 M EDTA. Cu(EDTA) 2 −/Cu(NH3)2+, as was shown in Fig. 4b. Consequently, the
The formation of this solid species can be explained thermodynamically silver dissolution reaction is the rate determining step for the leaching
by the Pourbaix diagrams of Fig. 4, where the presence of copper oxides system with 0.05 M EDTA, as shown in Fig. 6. The absence of CuO in
occurs at pH values of 10.2 and higher. This pH is very similar to that of the speciation diagram under this leaching condition suggests that the
the leaching conditions, suggesting that the copper oxides form on the silver particle is not coated by any ash layer.
silver particle surface causing a decrease on the silver leaching kinetics. All the silver leaching experiments with 1.25 × 10 −4 and 0.025 M
Fig. 7 presents a species distribution diagram for the Cu–NH3– EDTA presented oscillations. The oscillations were related to the dis-
EDTA–H2O system at 25 °C, considering the next concentrations: solution–precipitation phenomenon: the dissolution of silver with
0.025 M EDTA, 0.05 M Cu(II), 0.6 M NH3 and a fixed potential of thiosulfate (Eq. (1)), and the precipitation of silver sulfide (Eq. (5)).
0.4 V vs SHE (similar to the redox potential of the leaching solution). In this section further evidence is presented for a better understand-
The copper complexation with EDTA and ammonia as well as the ing of the phenomena that cause the so-called oscillations in the
presence of CuO are observed at a pH of 10.2. Furthermore the copper leaching curves.
species Cu(EDTA) 2− and CuO predominate at this pH condition. This Figs. 9, 10 and 11 show the Pourbaix diagrams for the Ag–S–
is indicative that the copper in solution can be stabilized with EDTA, EDTA–H2O systems at 0.2, 0.1 and 0.3 M thiosulfate respectively
but also can precipitate as oxides. From an economic point of view (0.025 M EDTA, 1 × 10 −5 M Ag). The equilibrium potential for the
this system would imply a loss of the oxidizing agent (Cu(II) com- redox couple Ag(S2O3)35 −/Ag at pH 10.2 corresponds to − 0.089,
plexes in solution), increasing the costs of the process. − 0.044 and − 0.119 V vs SHE respectively. Furthermore, the
On the other hand, Fig. 8 shows the species distribution diagram leaching solutions for 0.1 M, 0.2 M and 0.3 M thiosulfate at 0.025 M
under similar conditions as those of Fig. 7, but in this case the EDTA EDTA presented a redox potential of 0.4259, 0.4018 and 0.323 V vs
concentration was increased to 0.05 M and the potential was fixed SHE respectively; if these redox potential operation conditions are visu-
at 0.32 V vs SHE (the redox potential measured for this leaching solu- alized on the Pourbaix diagrams for each system, it can be seen that the
tion). In this diagram, the copper oxide CuO does not appear in the silver predominant species is the complex Ag(S2O3)35− (see Figs. 9, 10
and 11). This result suggests that metallic silver is oxidized to form
the complex Ag(S2O3)35− in accordance with Eq. (1). However the pre-
cipitation of the silver–thiosulfate complex to form silver sulfide is sup-
posed to occur in these systems by the presence of copper sulfide (see
Eqs. (3)–(5)). According to Abbruzzese et al. (1995) the precipitation

Fig. 6. Silver leaching data fitted to Chemical Reaction Control Model, shrinking core in
spherical particles. Silver dissolution during the first 30 min of the leaching experi- Fig. 8. Distribution of copper species at different pH values for the Cu–NH3–EDTA–H2O
ment (0.125 g silver, 200 mL solution, 0.2 M thiosulfate, 0.05 M EDTA, 0.05 M Cu(II), system. The conditions for the diagram construction were 0.05 M EDTA, 0.05 M Cu(II),
0.6 M NH4OH, pH = 10.2, room temperature and 400 rpm). 0.6 M NH4OH and EH = 0.32 V at 25 °C.
 D.M. Puente-Siller et al. / Hydrometallurgy 134–135 (2013) 124–131 129

Fig. 9. Pourbaix diagram for the Ag–S–EDTA–H2O system. The conditions for the dia- Fig. 11. Pourbaix diagram for the Ag–S–EDTA–H2O system. The conditions for the dia-
gram construction were 0.2 M thiosulfate, 0.025 M EDTA and 1 × 10−5 M Ag at 25 °C. gram construction were 0.3 M thiosulfate, 0.025 M EDTA, and 1 × 10−5 M Ag at 25 °C.

of copper sulfides can occur by the oxidative degradation of the thiosul- Contrary to the 0.025 M EDTA system, the leaching system with
fate in the leaching solution. On the other hand, the Pourbaix diagram for 0.05 M EDTA did not show oscillations in any case (Figs. 1, 2 and 3).
the Cu–S–NH3–EDTA–H2O system at 0.2 M thiosulfate and 0.025 M This can be explained thermodynamically: for the leaching solution
EDTA presented in Fig. 12a shows that the equilibrium potential for the containing 0.2 M thiosulfate and 0.025 M EDTA the Pourbaix diagram
redox couple Cu(EDTA)2−/Cu2S is −0.0632 V vs SHE at pH 10.2 approx- of Fig. 9 predicts an equilibrium potential of −0.089 V vs SHE at
imately. These results indicate that it is possible that the reduction of pH 10.2 for the redox couple Ag(S2O3)35−/Ag; on the other hand, the di-
Cu(EDTA)2− to Cu2S and the oxidation of Ag to Ag(S2O3)35− occur simul- agram for the same thiosulfate concentration and 0.05 M EDTA shows
taneously; therefore, the copper sulfide species is also probably formed that the equilibrium potential for the redox couple Cu(EDTA)2−/Cu2S
by this reduction reaction in addition to that mentioned in Eqs. (3) and is −0.112 V vs SHE at pH 10.2 (Fig. 12b). These results indicate that
(4). Once this copper sulfide species is formed during the leaching
process, the silver complex Ag(S2O3)35− can precipitate to form Ag2S
according to Eq. (5).
Two solid samples corresponding to the leaching experiment at
0.2 M thiosulfate and 0.025 M EDTA at 10 and 50 min (Fig. 1) were
characterized by SEM and EDXS (Fig. 13a, b, c and d). Fig. 13a shows a
spherical morphology of the silver particles after 10 min leaching. The
EDXS analysis (see Fig. 13b) revealed the presence of silver, copper, ox-
ygen and sulfur; these last three elements could be related to the forma-
tion of copper sulfides and copper oxide species on the silver surface.
These results are in accordance with the thermodynamic predictions
and support the proposed idea that the silver particle is being coated
by copper sulfides and oxides.
Fig. 13c shows the silver particles coated by a precipitated solid
after 50 min of the leaching process. The EDXS analysis (Fig. 13d) re-
vealed the presence of silver and sulfur, which could be the result of
the precipitation of the silver sulfide (Eq. (5)). Fig. 13d did not
show any copper sulfide on the residue, which was expected as the
copper sulfide had reacted in order to form the silver sulfide.

Fig. 12. Effect of the EDTA concentration on the copper sulfide formation at 25 °C. a)
Pourbaix diagram for the Cu–S–NH3–EDTA–H2O system with 0.2 M thiosulfate,
Fig. 10. Pourbaix diagram for the Ag–S–EDTA–H2O system. The conditions for the dia- 0.025 M EDTA, 0.6 M NH4OH and 0.05 M Cu(II). b) Pourbaix diagram for the same sys-
gram construction were 0.1 M thiosulfate, 0.025 M EDTA, and 1 × 10−5 M Ag at 25 °C. tem with 0.05 M EDTA.
130 D.M. Puente-Siller et al. / Hydrometallurgy 134–135 (2013) 124–131

Fig. 13. Effect of leaching time on the chemical composition of the solid residue. Experimental conditions: 0.2 M thiosulfate, 0.025 M EDTA, 0.6 M NH4OH and 0.05 M Cu(II). a) Morphol-
ogy at 10 min leaching, b) chemical composition of the residue at 10 min, c) morphology at 50 min leaching, d) chemical composition for the residue at 50 min.

the reduction of Cu(EDTA) 2− to Cu2S cannot occur simultaneously with Fig. 6: the silver leaching under these conditions is controlled by the
the oxidation of Ag to Ag(S2O3)35−. In addition, the redox potential mea- chemical reaction in the initial stage.
surements showed that an increase in the EDTA concentration de-
creases the redox potential of the solution and hence the oxidation 3.4. Effect of copper concentration on the silver leaching
ability of the system. Eqs. (3) and (4) show the possibility to form cop-
per sulfide species as a consequence of the thiosulfate oxidative de- In the previous sections it was observed that a decrease of the
struction; however the decrease in the oxidation ability of the system thiosulfate and EDTA concentrations enhances the silver leaching
makes the formation of these copper sulfides less probable. These re- kinetics. This section deals with the effect of decreasing the cupric ion
sults are consistent with the findings mentioned when discussing concentration from 0.05 M to 7.87 × 10−4 M. Fig. 14 presents the silver

Fig. 15. Silver leaching data fitted to Diffusion Control Model, shrinking core in spherical
Fig. 14. Effect of the thiosulfate concentration on the silver extraction. Experimental particles. Silver dissolution during the first 30 min of the leaching experiment (0.125 g sil-
conditions: 0.125 g silver in 200 mL thiosulfate solution 1.25 × 10−4 M EDTA, 0.6 M ver, 200 mL solution, 0.1 M thiosulfate, 1.25 × 10−4 M EDTA, 7.87 × 10−4 M Cu(II),
NH4OH, 7.87 × 10−4 M Cu(II) at room temperature and 400 rpm. 0.6 M NH4OH, pH = 10.2, room temperature and 400 rpm).
 D.M. Puente-Siller et al. / Hydrometallurgy 134–135 (2013) 124–131 131

leaching for the systems containing 0.1 M, 0.2 M and 0.3 M thiosulfate, tion reaction: there is no formation of a copper oxide layer on the
1.25 × 10−4 M EDTA and 7.87 × 10 −4 M Cu(II). All experiments metallic silver particle surface.
showed low silver leaching values: less than 12% in 6 h. If these results 3.- In the leaching systems with EDTA concentrations lower than
are compared with those of Fig. 2 for 1.25 × 10−4 M EDTA, it is clear 0.025 M, the precipitation of copper sulfides species also occurs.
that the decrease in cupric ion concentration affects significantly the These copper sulfide species proceed from the silver–thiosulfate
leaching of silver. This behavior is associated with the decrease in the complex.
oxidizing ability of the leaching solution, and is in accord with the find- 4.- Finally, the slowest silver leaching kinetics was found when the cupric
ings of Abbruzzese et al. (1995), who observed that the presence of a ion concentration was decreased from 0.05 M to 7.87 × 10−4 M. This
high concentration of cupric ions promotes fast leaching kinetics. behavior is related to the precipitation of copper oxides on the silver
Some oscillations also appear in these silver extraction curves, re- particle surface, as well as to the decrease in the oxidizing ability of
lated to the silver precipitation as silver sulfide, as mentioned in the the leaching system.
last section. The experimental data for the silver leaching with solu-
tions 0.1 M, 0.2 M and 0.3 M thiosulfate, 1.25 × 10 −4 EDTA and Acknowledgments
7.87 × 10 −4 Cu(II) were fitted to the shrinking core model in the
same way as in the cases of Figs. 5 and 6. The results for the first The authors are grateful to CONACYT (Mexico) for the postgradu-
30 min are presented in Fig. 15. This range of time was chosen to ate scholarship awarded to Damaris M. Puente-Siller. The authors also
avoid the interference of any silver precipitation (passivation phe- wish to thank Dr. Alejandro Uribe-Salas, Dr. Roberto Pérez-Garibay
nomena). From Fig. 15 it can be said that during the initial stage of and M.Sc. Juan Antonio González-Anaya for their continued support
the leaching the kinetics of the process is controlled by the diffusion regarding the laboratory materials.
of the fluid species through the porous layer. As exposed previously,
this porous layer is composed of copper oxides. The leaching process References
is passivated after 1 h for the tests of curves 0.2 and 0.3 M thiosulfate
Abbruzzese, C., Fornari, P., Massidda, R., Veglió, F., Ubaldini, S., 1995. Thiosulphate
(Fig. 14), probably due to the loss of the oxidizing agent (cupric ions). leaching for gold hydrometallurgy. Hydrometallurgy 39, 265–276.
According to Feng and Van Deventer (2010), the precipitation of cu- Alonso-Gómez, A.R., Lapidus, G.T., 2009. Inhibition of lead solubilization during the
pric ions occurs in the thiosulfate systems. These results suggest leaching of gold and silver in ammoniacal thiosulfate solutions (effect of phosphate
addition). Hydrometallurgy 99, 89–96.
that the leaching of silver would proceed faster with 0.05 M Cu(II) Aylmore, M.G., Muir, D.M., 2001. Thiosulfate leaching of gold. A review. Miner. Eng. 14,
than with 7.87 × 10 −7 M Cu(II) (see Figs. 2 and 14). 135–174.
Eriksson, G., 1979. An algorithm for the computation of aqueous multicomponent,
multiphase equilibria. Anal. Chim. Acta 112, 375–383.
4. Conclusions Feng, D., Van Deventer, J.S.J., 2006. Ammoniacal thiosulphate leaching of gold in the
presence of pyrite. Hydrometallurgy 82 (3–4), 126–132.
Feng, D., Van Deventer, J.S.J., 2007. Effect of hematite on thiosulphate leaching of gold.
On the basis of the results obtained for the metallic silver leaching Int. J. Miner. Process. 82, 138–147.
with the system thiosulfate–EDTA–Cu(II)–ammonia, the following Feng, D., Van Deventer, J.S.J., 2010. Thiosulphate leaching of gold in the presence of
ethylendiaminetetraacetic acid (EDTA). Miner. Eng. 23, 143–150.
could be concluded:
Feng, D., Van Deventer, J.S.J., 2011. The role of amino acids in the thiosulphate leaching
of gold. Miner. Eng. 24, 1022–1024.
1.- In the range of thiosulfate and EDTA concentrations studied (0.1 M, Fuentes-Aceituno, J.C., Frade-Chávez, J.G., Mendoza-Hernández, R.L., Lapidus-Lavine,
0.2 M and 0.3 M for thiosulfate and 1 × 10−4 M, 0.025 M and G.T., 2004. Pregrobbing en los sistemas de tiosulfato y tiourea. XIV Congreso
Internacional de Metalurgia Extractiva (Pachuca, Hidalgo, México).
0.05 M for EDTA), the silver leaching kinetics is enhanced by the Heath, F.A., Jeffrey, M.I., Zhang, H.G., Rumball, J.A., 2008. Anaerobic thiosulfate leaching:
lowest EDTA concentration, reaching near 100% of extraction in development of in situ gold leaching systems. Miner. Eng. 21 (6), 424–433.
the first 20 min of leaching. In the same way the leaching solution Labanowski, J., Monna, F., Bermond, A., Cambier, P., Fernandez, C., Lamy, I., Van Oort, F.,
2008. Kinetic extractions to assess mobilization of Zn, Pb, Cu, and Cd in a metal-
with 0.1 M thiosulfate presented faster leaching kinetics than with contaminated soil: EDTA vs. citrate. Environ. Pollut. 152, 693–701.
0.2 or 0.3 M thiosulfate. A synergistic effect in the silver leaching ki- Levenspiel, O., 1972. Chemical Reaction Engineering, 2nd edition. John Wiley and Sons,
netics was found when the EDTA and thiosulfate concentrations Inc., New York, pp. 361–377.
Manouchehri, N., Besancon, S., Bermond, A., 2006. Major and trance metal extraction
were decreased to 0.1 and 1.25 × 10−4 M respectively. In these from soil by EDTA: equilibrium and kinetic studies. Anal. Chim. Acta 559, 105–112.
leaching conditions the silver precipitation is minimized, suggesting Puigdomenech, I., 2004. Make Equilibrium Diagrams Using Sophisticated algorithms
that this system is more stable than when the thiosulfate concentra- (MEDUSA), Inorganic Chemistry. Royal Institute of Technology, Stockholm, Sweden.
Senanayake, G., 2007. Review of rate constants for thiosulphate leaching of gold from
tion is higher.
ores, concentrates and flat surfaces: effect of host minerals and pH. Miner. Eng.
2.- The precipitation of copper oxides occurs for the leaching systems 20, 1–15.
with 1.25 × 10 −4 and 0.025 M EDTA. According to the SEM results, Zhang, S., 2004. Oxidation of Refractory Gold Concentrates and Simultaneous Dissolu-
this precipitation takes place on the surface of the metallic silver tion of Gold in Aerated Alkaline Solutions. Ph. D. Thesis. Murdoch University
Western Australia, Australia.
particle. On the other hand, when the EDTA concentration is in- Zipperian, D., Raghavan, S., Wilson, J.P., 1988. Gold and silver extraction by ammoniacal
creased up to 0.05 M, the process is controlled by the silver dissolu- thiosulfate leaching from a rhyolite ore. Hydrometallurgy 19 (3), 361–375.