Electrochemical study of binary and ternary copper complexes in ammonia-chloride medium

Article abstract of DOI:10.1016/j.electacta.2007.03.062

Electrochemical and spectroscopic analyses (UV-vis and XRD) have been applied to evaluate the speciation of copper in ammonia-chloride solutions. The conditions for these analyses were established through thermodynamic studies that included predominance existence and E_h-pH diagrams. These studies highlighted the importance of copper ternary complexes and showed how the solubility of Cu(I) and Cu(II) increases over a wide range of pH when copper ternary complexes are formed. Hence, it has been elucidated the copper, ammonia and chloride concentrations that makes existence of these so-called copper ternary complexes, possible.

Full text of DOI:10.1016/j.electacta.2007.03.062

Electrochimica Acta 52 (2007) 6106–6117
Electrochemical study of binary and ternary copper
complexes in ammonia-chloride medium
Jorge Vazquez-Arenas, Isabel Lazaro, Roel Cruz∗
Facultad de Ingenier´ıa-Instituto de Metalurgia, Universidad Auto´noma de San Luis Potos´ı, Av. Sierra Leona No. 550,
Lomas 2a Seccio´n, 78210 San Luis Potos´ı, S.L.P., Mexico
Received 24 December 2006; received in revised form 5 March 2007; accepted 22 March 2007
Available online 25 March 2007
Abstract
Electrochemical and spectroscopic analyses (UV–vis and XRD) have been applied to evaluate the speciation of copper in ammonia-chloride
solutions. The conditions for these analyses were established through thermodynamic studies that included predominance existence and Eh–pH
diagrams. These studies highlighted the importance of copper ternary complexes and showed how the solubility of Cu(I) and Cu(II) increases over
a wide range of pH when copper ternary complexes are formed. Hence, it has been elucidated the copper, ammonia and chloride concentrations
that makes existence of these so-called copper ternary complexes, possible.
© 2007 Elsevier Ltd. All rights reserved.
Keywords: Copper complexes; Ammonia; Chloride; Thermodynamic; Voltammetry
1. Introduction
The characteristics described for the ammonium chloride
system such as improving leaching and also the possibility of
working at a pH value that provokes that any iron and other
minor elements considered as impurities remain in the residue
after leaching [5], have prompted studies that have addressed
thermodynamic and electrochemical aspects. Most of these stud-
ies described Cu(II) and/or Cu(I) complexation and solubility
aspects [7–9]. Early work by Limpo et al. focused mainly on
effects of experimental conditions on the solubility of Cu(II)
[7]. These authors found that the solubility of Cu(II) in this
medium is restricted by either the solubility of Cu(NH3)2Cl2 or
Cu(OH)1.5Cl0.5, the predominance of both precipitates depen-
dent on chloride concentration. They observed that in the range
of 5–9m, the Cu(II) diammine chloride is generally the solid
phase whose solubility regulates that of Cu(II); the hydroxy-
chloride becomes the most insoluble solid compound and hence
the substance which restricts the solubility of the Cu(II), only at
the lowest test levels of chlorides and when the ratio of [NH3]
and [Cu] is less than 2. Hence, in that work they showed the
need of including the formation of ternary complexes of copper
ammonia-chloride with the general formula CuClx(NH3)4−x, in
order to interpret experimental solubility data and get a better
fit with the predicted values obtained with theoretical models.
Likewise, later work by Solis et al., highlighted the impor-
tance of Cu(I) ternary complexes and provided experimental
The electrolytic recovery of copper from sulfate solutions is
the most common process applied in many hydrometallurgical
processes currently in operation, all of this as a result of the gen-
eral advantages that this medium offers, particularly permitting
wire bar grade cathodic deposits [1]. The characteristics of the
sulfate medium have in fact limited the use of other media in
leaching processes because of considerations of having to con-
vert anyway to sulfate, in order to comply with the requirements
of the electrolytic process [2]. Such is the case of the chloride
media that is reported to improve the kinetics of dissolution
but has the drawback of being a corrosive environment [3]. To
overcome the corrosive characteristics of the chloride medium,
some authors have proposed the combination of this medium
with ammonia, which inhibits the formation of chlorine gas and
is also a much stronger complexing agent [4–6]. To our knowl-
edge two hydrometallurgical based ammonia chloride (NH4Cl)
processes have been applied successfully at industrial levels,
the CENIM-LNETI process for base metals and the Ezinex®
process for zinc recovery [5,6].
∗
Corresponding author. Tel.: +52 444 825 43 26; fax: +52 444 825 43 26.
E-mail address: rcuz@uaslp.mx (R. Cruz).
0013-4686/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2007.03.062

J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
6107
thermodynamic data for the Cu(I)–ammonia-chloride system
[8].
(SCE) as reference electrode, with a potential of 241 mV ver-
sus the standard hydrogen electrode (SHE). For convenience,
all potentials reported here have been converted to the SHE
scale. Before making any electrochemical measurements, the
surface of the disk was prepared by being ground on fine sili-
con carbide paper (grade 600 and 2000), polishing to a mirror
finish with 0.05 ␮m alumina paste followed by ultrasonic clean-
ing. The electrochemical measurements were carried out using
an AUTOLABTM model PGSTAT 30 potenstiotat–galvanostat
coupled to a personal computer.
´
Nila and Gonzalez [9] experimentally confirmed that Cu(II)
solubility was limited by formation of the hydroxychloride pre-
cipitate (Cu(OH)1.5Cl0.5). These authors did not report Cu(II)
solubility limitations by formation of Cu(NH3)2Cl2 as did
Limpo et al. [7], but this could be attributed to the fact that
the concentration of chloride they used was never higher than
1 M.
With the exception of some reports [9–11], most of the
research studies have been oriented towards leaching conditions
and in fact recent works are more concerned with the applica-
tion of the Cu(II) tetrammine complex as an oxidizing agent for
thiosulfate leaching of gold [12,13].
UV–vis spectrophotometric analysis of solutions was carried
using a Lambda 35 Perkin-ElmerTM. Solid samples were char-
acterized by X-ray diffraction (XRD) analysis using a Phillips
RigakuTM DMAX 2200 diffractometer.
The dissolution and stabilization of copper in hydrometal-
lurgical operations such as leaching and electrowinning require
very high concentrations of chloride and ammonia [5,6]. In spite
of this the evaluation of the redox system Cu(I)/Cu(II) under
such conditions has been scarcely studied, and even the impor-
tance of the binary and ternary complexes has been overlooked
in electrochemical studies. A better understanding of the charac-
teristics of theses systems could allow an improved assessment
of this medium.
Therefore, the present study aims to further highlight the
existence of binary and ternary complexes for both Cu(I) and
Cu(II) in ammonia-chloride medium by means of electrochem-
ical characterization of Cu(I)/Cu(II) redox couples. Likewise
spectroscopic techniques (UV, XRD) were used to characterize
the predominant soluble and insoluble species under conditions
of high chloride and high ammonia concentrations. The experi-
mental approach was based on thermodynamic predictions.
3. Results and discussion
3.1. Thermodynamic study
In order to establish different thermodynamic aspects of the
Cu–NH4Cl–NH3–H2O system, it is important to have a wide
knowledge of the solution chemistry of the different species
involved. Thus, in this study graphical methods such as species
distribution and predominance area (including Eh–pH) diagrams
were constructed to illustrate the various possible equilibrium
reactions for copper in an ammonia-chloride media. The con-
struction of these thermodynamic diagrams was based on free
energy minimization algorithms reported by Eriksson [14], and
used by the Chemical Equilibrium Software MEDUSA® [15].
The equilibrium constants for complex formation were taken
from the critically selected stability constants published by the
National Institute of Standards and Technology [16] and others
from the literature (Tables 1–3) [7,8,17–20]. Furthermore, these
data were complemented with the HYDRA program database
[15].
2. Experimental
All solutions employed in this work were prepared using
reagents of analytical quality and deionized water (18.2 ␮ꢀ cm)
obtained by inverse osmosis (BarnsteadTM). Unless otherwise
stated all Cu(II) solutions were prepared using CuCl2·2H2O,
NH4Cl and in some instances NH4OH. Variation of solutions
pH was achieved by adjustment and control with either 4 M
NaOH or 10 M H2SO4 solutions. Cu(I) solutions were prepared
by dissolution of cuprous chloride salt (CuCl) in solution 4 M of
NH4Cl and adjusted to the required pH. To warrant the presence
of Cu(I) complexes, hydrazine dihydrochloride was added to the
solution as an stabilizer agent. All Cu(I) solutions were prepared
with deionized water which had been previously deareated with
nitrogen for about 40 min. To prevent oxidation of Cu(I) a nitro-
gen atmosphere was kept during the period of analysis. Fresh
Cu(I) solutions were colorless and it took approximately 2 h for
them to turn to colored solutions indicating presence of Cu(II)
complexes.
For many of the constants there is a wide variety of values
and conditions under which each was obtained, thus in a similar
way as that proposed by Limpo et al. [7], a path was followed
in which the criteria of selection of constants was based on the
value of ionic strength at which they were obtained. Although,
it is acknowledged that aqueous electrolyte solutions with
Table 1
Thermodynamic data for chlorocomplexes, used in the construction of the
Cu(I)/Cu(II)–NH4Cl–NH3–H2O diagrams, at 298 K
Complex
log β
Ionic
strength
Reference
Cu+ + Cl− = CuCl(ac)
2.70
6
5.99 High
10.57
7.38
0.60
5
5
[17]
[17]
[19]
[14]
[14]
[16]
[16]
[16]
[16]
[15]
−
2−
2−
Cu+ + 2Cl− = CuCl2
Cu+ + 3Cl− = CuCl3
2Cu+ + 4Cl− = Cu2Cl4
Cu+ + Cl− = CuCl(s)
Cu2+ + Cl− = CuCl+
–
–
5
5
5
5
0
A conventional three-electrode cell was used for electro-
chemical experiments. A glassy carbon (GC) cylinder was used
to construct a disk electrode. The disk electrode was made by
embedding the GC cylinder in Teflon, allowing the exposure
of only one face with an area of 0.38 cm2. A graphite rod was
employed as counter electrode and a saturated calomel electrode
Cu2+ + 2Cl− = CuCl2(−ac)
0.67
0.30
Cu2+ + 3Cl− = CuCl3
Cu2+ + 4Cl− = CuCl4
−0.64
2−
Cu2+ + 1.5OH− + 0.5Cl− = CuOH1.5Cl0.5(s)
17.3

6108
J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
Table 2
fraction of species present as a function of pH. Given the pre-
dominance of NH4 at pH values below 6, no copper ammine
+
Thermodynamic data for hydroxo-complexes, used in the construction of the
Cu(I)/Cu(II)–NH4Cl–NH3–H2O diagrams, at 298 K [15]
complexes are observed in this pH zone and thus the chloride
complexes account for most of the total Cu(I) concentration. As
shown in Fig. 1, in this pH region the predominant species is
CuCl32−. As pH increases, Cu(NH3)2Cl becomes the predom-
inant species producing the highest solubility of Cu(I) for one
individual species, since its fraction is around 0.95 at pH’s values
within 8.5–12.5. The solid phases evaluated were the CuCl(s) and
Cu2O(s), as it has been reported in the literature [9,21–23]. How-
ever, for the conditions at which the diagram was constructed,
the Cu2O(s), exhibits the highest stability. The solid phase was
determined for pH > 12.5.
Complex
log β
Ionic strength a
Cu+ + OH− = 0.5Cu2O(s) + 0.5H2O
Cu2+ + OH− = CuOH+
14.7
0
3
0.7
1
1
6.8 ( 2)
11.2 ( 4)
14.5
15.6
18.9
Cu2+ + 2OH− = CuOH2(−ac)
Cu2+ + 3OH− = CuOH3
Cu2+ + 4OH− = CuOH4
2−
Cu2+ + 2OH− = CuOH2(s)
1
0.7
Cu2+ + 2OH− = CuO(s) + H2O
19.51
3.1.2. Predominance diagrams for Cu(I)
To determine the formation of soluble and solid phases under
different conditions for the Cu(I)–NH4Cl–NH3–H2O system,
independentpredominancediagramswereconstructedasafunc-
tion of Cu(I), NH3 and Cl− concentrations versus pH.
In all diagrams, the doted lines indicate the conditions to
be evaluated in this study, which are those used for distribu-
tion diagram construction (Fig. 1). The diagrams obtained show
the predominance zone of different complexes and solid phases
(Fig. 2). Hence, it can be observed that for the acidic zone
(pH < 6.57) the trichloride complex (CuCl32−) predominates as
far as the chloride concentration is higher than ∼1 M (pCl− < 0).
For lower chloride concentrations, the formation of the solid
phase CuCl(s) would take place due to the low stability of the
CuCl2− complex [9,21–23].
Fig. 1. Species distribution diagram for copper (I) complexes in a solution
containing a total concentration of 0.2 M Cu(I) and 4 M NH4Cl.
concentration of ions greater than about 1 × 10−6 M do not
behave ideally, it is considered a practical way for treatment
of real hydrometallurgical systems, as they generally involve
solutions highly concentrated (such as the ones in the present
study), and especially those used in the electrolytic stage.
For the neutral region, there is an important variation of pre-
dominant copper species which is strongly affected by changes
in ammonia and chloride concentrations, and that is why we
labeled this zone as ligands interchange zone. Hence, as the
ammonia concentration increases the predominant complexes
change as follows: CuCl32− → CuNH3Cl2− → Cu(NH3)2Cl.
3.1.1. Species distribution diagram for
Cu(I)–NH4Cl–NH3–H2O system
−
For the conditions to be evaluated in this work, CuNH3Cl2
The species distribution diagram for the Cu(I)–NH4Cl–NH3–
H2O system is illustrated in Fig. 1. This diagram shows the
would be the predominant specie at 6.57 < pH < 6.89. In
the neutral-alkaline zone, the Cu(NH3)2Cl exhibits a wide
Table 3
Thermodynamic data for amino-complexes, used in the construction of the Cu(I)/Cu(II)–NH4Cl–NH3–H2O 25 ◦C diagrams, at 298 K
Complex
log β
Ionic strength
Reference
+
NH3 + H+ = NH4
9.4
5.93
10.58
8.92
8.82
11.33
4.2 ( 2)
7.75 ( 5)
10.6 ( 1)
12.9 ( 1)
12.43
14.9
1
2
2
1
1
1
2
2
2
2
2
–
–
–
5
5
[17]
[15]
[15]
[8]
[8]
[8]
[15]
[15]
[15]
[15]
[17]
[17]
[17]
[17]
[7]
Cu+ + NH3 = CuNH3
+
+
Cu+ + 2NH3 = Cu(NH3)2
Cu+ + NH3 + Cl− = CuNH3Cl(ac)
Cu+ + NH3 + 2Cl− = CuNH3Cl2
−
Cu+ + 2NH3 + Cl− = Cu(NH3)2Cl(ac)
Cu2+ + NH3 = CuNH3
2+
2+
Cu2+ + 2NH3 = Cu(NH3)2
Cu2+ + 3NH3 = Cu(NH3)3
2+
Cu2+ + 4NH3 = Cu(NH3)4
2+
2+
Cu2+ + 5NH3 = Cu(NH3)5
Cu2+ + NH3 + OH− = CuNH3OH+
Cu2+ + NH3 + 3OH− = CuNH3(OH)3
−
16.3
15.7
3.62
11.73
Cu2+ + 2NH3 + 2OH− = Cu(NH3)2(OH)2(ac)
Cu2+ + NH3 + 3Cl− = CuNH3Cl3
Cu2+ + 3NH3 + Cl− = Cu(NH3)3Cl+
−
[7]

J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
6109
Fig. 2. Predominance area diagrams for Cu(I) constructed as pX (X = [Cu], [NH3], [Cl−]) functions of pH: (a) pCu, [NH4+ + NH3] = 4 M, [Cl−] = 4 M, (b) pNH3,
[Cu]total = 0.2 M, [Cl−] = 4 M and (c) pCl, [Cu]total = 0.2 M, [NH4+ + NH3] = 4 M. Dotted lines show the changes produced in the predominant species areas when no
ternary complexes are considered.
predominance, which is only limited for the Cu(NH3)2+, when
3.1.3. Species distribution diagram for
Cu(II)–NH4Cl–NH3–H2O
chloride concentration decreases below 0.178 M (Fig. 2a) and
for Cu2O(s) for pH > 13.13 (Fig. 2). The formation of Cu2O(s)
Fig. 3 shows species distribution diagrams for Cu(II) for the
same condition of the Cu(I) system. From this diagram it can
observed that a lower affinity for chloride complexes is obtained
+
is actually promoted by the lack of NH4 that acts as a buffer
and removes OH− ions.
From the diagrams, illustrated in Fig. 2, it is possible to
generate lineal diagrams showing the dominant species as a
function of pH. Thus, for the system under study – 0.2 M
Cu(I) (pCu = 0.698), 4 M NH3 (pNH3 = −0.602) and 4 M Cl−
(pCl− = −0.602) – the corresponding lineal diagram would be
as follows:
(1)
The above lineal diagram shows four pH regions for the pre-
dominant species. This diagram also illustrates in a very simple
way the change of ligands towards species of higher stability.
The ternary complexes cover a wide stability region due to the
combined effect of chloride and ammonia ligands.
Fig. 3. Species distribution diagram for copper(II) complexes in a solution
containing a total concentration of 0.2 M Cu(II) and 4 M NH4Cl.

6110
J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
for Cu(II) in comparison with the Cu(I) system. The total cop-
per concentration is distributed mainly in three different chloride
complexes (CuCl3−, CuCl2, CuCl42−) for pH values below ∼4.
species, but for pH > 4.13 the formation of Cu(OH)1.5Cl0.5(s)
takes place. Earlier publications have reported a coexistence
between Cu(OH)1.5Cl0.5(s) and soluble Cu(II) in the form of
2+
In the region, 4 < pH < 7 a mixed precipitate, Cu(OH)1.5Cl0.5 (s)
,
Cu(NH3)4 at pH values around 7.66 for a total copper con-
would be expected. In agreement with the stability constant of
this solid, the copper is practically insoluble in this region, even
under conditions where ternary complexes such as CuNH3OH+
and Cu(NH3)3Cl+ could exist. For a pH zone around 7–8.2, the
Cu(NH3)3Cl+ is the predominant complex. As the pH increases,
the amount of NH3 increases, and this provokes a displacement
of the Cl− ligand out of the Cu(II) coordination sphere. Hence,
centration of around 0.01 M [9,24]. The incorporation of ternary
complexes data in the present thermodynamic study shows
that the Cu(II) solubility region widens, and complexes such
as Cu(NH3)3Cl+ can be formed at pH ∼ 6.6 for copper con-
centrations of 0.2 M. Contrary to what Limpo reported [7],
Cu(NH3)2Cl2 precipitate was not predicted according to our
calculations, even at a chloride/ammonia concentration ratio
of 10. This fact is attributed to the low Kps of this precipitate
2+
for pH > 8.2, Cu(NH3)4 is the main complex, meanwhile the
(10−1.7). For pH > 8 Cu(NH3)4 and Cu(NH3)5 are formed
due to the high ammonia concentration. It is worth noting, that
even when Fig. 4a and c indicate that for pH > 9.4 (equilib-
rium point of NH3 − NH4+) the predominance of Cu(NH3)52+ is
expected, results of Fig. 4a indicate that this complex would be in
2+
2+
Cu(NH3)52+ is for pH > 9.4. Finally, at pH > 12.3 CuO(s) forma-
tion occurs.
3.1.4. Predominance diagrams for Cu(II)
Similarly to the Cu(I) system, predominance diagrams of
Cu(II) species were independently constructed as a function
of Cu(II), NH3 and Cl− concentrations. Fig. 4 shows that
two solid phases are formed, one in the acidic-neutral zone,
Cu(OH)1.5Cl0.5(s), and another in the alkaline zone, CuO(s). In
the acidic zone pH < 4, the Cu(II) would be soluble under any
system condition. Copper aquocomplexes would be generated
for chloride concentration lower than 2.51 M (pCl > 0.6). For
2+
coexistence with Cu(NH3)4 at the conditions here evaluated.
This fact is better observed when the NH3 + NH4+ concentration
decreases below 4 M. Similarly to the Cu(I) system in the alka-
line zone, as the hydroxide concentration increases the copper
soluble species stability decreases provoking its precipitation,
that for this system is as CuO at pH > 12.3.
From the diagrams shown in Fig. 4, the following lineal dia-
gram was constructed for 0.2 M Cu(II) (pCu = 0.698), 4 M NH3
−
the conditions here evaluated, the CuCl3 is the predominant
Fig. 4. Predominance area diagrams for Cu(II) constructed as pX’s (X = [Cu], [NH3], [Cl−]) functions of pH: (a) pCu, [NH4+ + NH3] = 4 M, [Cl−] = 4 M, (b) pNH3,
[Cu]total = 0.2 M, [Cl−] = 4 M and (c) pCl, [Cu]total = 0.2 M, [NH4+ + NH3] = 4 M. Dotted lines show the changes produced in the predominant species areas when no
ternary complexes are considered.

J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
6111
and Cu(I) systems. In fact for the Cu(I) system, the ammine
complexes are not formed due to the high stability of the ternary
complex Cu(NH3)2Cl.
The precipitate Cu(OH)1.5Cl0.5(s) limits the solubility of
Cu(II) at acid pH and CuO(s) at alkaline pH. While there is no
indication of solubility limitation at acid pH for Cu(I) at alkaline
pH it is limited because of formation of Cu2O(s).
It is clear that the potential required to reduce Cu(II) to Cu(I),
increases as the pH is raised, due to the higher stability of the
ammine complexes. Similarly, the potential required to reduce
Cu(I) complexes to Cu(0), is higher as the pH increase, however,
it was observed that for the neutral pH (ligands’ transfer zone)
the total overpotential to reduce Cu(II) species to Cu(0) is the
lowest, i.e. ∼250 mV for ternary complexes at pH 7, while for
chloride and amine complexes, the overpotentials are 430 and
320 mV, respectively.
Fig. 5. Eh–pH diagram for Cu–NH3–Cl–H2O system at 25 ◦C. [Cu] = 0.2 M,
[Cl−] = 4 M, [NH4+ + NH3] = 4 M.
3.2. Spectroscopic characterization of complexes and solid
phases
(pNH3 = −0.602) and 4 M Cl− (pCl− = −0.602):
3.2.1. UV–vis spectrophotometric analysis of complexes
In order to obtain experimental data to support the results of
the thermodynamic study, solutions of Cu(I) and Cu(II) were
analyzed by UV–vis spectrophotometry. The analyses were
carried out with copper solutions prepared under specific exper-
imental conditions that favor the formation of the predominant
copper soluble species reported in the Eh–pH diagram (Fig. 5;
Table 4).
Forcomparisonpurpose, aUV–visspectrumfora4 MNH4Cl
solution was included in each graph as a blank that allows to
evaluate the existence of the predominant species for each con-
dition. Hence, it was observed than the ammonium and chloride
by themselves only show an absorption band for a wavelength in
the UV region, around 205 nm. For the Cu(I) system, no absorp-
tion was expected for the visible region, given that according to
the electronic configuration of Cu(I), it does not generate col-
ored complexes. However, for wavelengths in the UV region an
important absorption is observed for this system (Fig. 6a), which
is greatly affected by changes in pH. As the pH increases, the
absorption band at wavelengths around 280–300 nm decreases.
Thus, this absorption is attributed to an electronic transition
between Cu(I) and Cl− ions in CuCl32− as it has been reported
by others authors [8]. When the pH increases the Cu(NH3)2Cl
becomes the predominant specie and therefore the absorption
band at wavelengths of ca. 280 nm decreases (Fig. 6b). Note that
(2)
This diagram illustrates the three solubility zones for
Cu(II):chlorocomplexes, ternary complexes and ammine com-
plexes. The chlorocomplexes dominate the acid zone, with the
−
CuCl3 as the most stable of all. The zone of ternary com-
plexes, particularly Cu(NH3)3Cl+, represents a very interesting
option given the advantages of operating processes within the
neutral-basic pH region [7]. Finally, the ammine complexes zone
2+
2+
involves Cu(NH3)4 and Cu(NH3)5 which although highly
stable predominate in a pH region unsuitable because of contin-
uous loss of ammonia.
3.1.5. Eh–pH diagram for Cu(II)–Cu(I)–NH4Cl–NH3
system
From the data above discussed, an Eh–pH diagram was con-
structed for the conditions to be evaluated in this work (Fig. 5).
All the equilibriums involving redox reactions were evaluated
using the Nernst equation. In a similar way to that illustrated by
the lineal diagrams, the different predominant zones are shown
as a function of pH as well as the effect of potential on ligand
coordination changes for both Cu(I) and Cu(II) ions.
As it was shown above, for the experimental conditions here
studied, there are several complexes involved for both Cu(II)
Table 4
Experimental details and results produced by UV–vis spectrophotometric analyses
Solution
Figure
pH
λ (nm)
Species
NH4Cl
4 M NH4Cl + NaOH
Fig. 7
7.5
3
205
240, 280
240
240, ∼300, 873
∼240, 660
∼240, 605
∼240, 620
2−
0.02 M Cu(I) + 4 M NH4Cl
0.02 M Cu(I) + 4 M NH4Cl
0.02 M Cu(II) + 4 M NH4Cl
0.02 M Cu(II) + 4 M NH4Cl
0.02 M Cu(II) + 4 M NH4Cl
0.02 M Cu(II) + 4 M NH4Cl
Fig. 7a
Fig. 7b
Fig. 7c
Fig. 7d
Fig. 7d
Fig. 7d
CuCl3
9.3
3.0
7.4
8.7
9.5
Cu(NH3)2Cl
−
CuCl3
Cu(NH3)3Cl+
2+
Cu(NH3)4
Cu(NH3)5
2+

6112
J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
Fig. 6. UV–vis spectra obtained for solutions prepared according to data presented in Table 4.
an absorption band is observed at wavelengths of ca. 600 nm for
solution of Cu(I) at pH 9.3. This last attributed to Cu(II) species
generated by oxidation of Cu(I) during the experiment (Fig. 6b).
the visible adsorption bands’ position (wavelengths), are in
agreement with the required energy for the complexes reduction
(Eh–pH diagram), which points the stability of the complexes
−
2+
For solutions of Cu(II) at pH 3, where the CuCl3 is the
(CuCl3− < Cu(NH3)3Cl+ < Cu(NH3)4 ≈ Cu(NH3)52+).
predominant specie, the absorption band due to electronic
transition between Cu–Cl is also observed (Fig. 6c). In addition,
an absorption band around 800 nm is observed. This band is
attributed to chloride complexes of copper(II). Fig. 6d shows
the UV–vis spectra for solutions of Cu(II) at a pH of 7.4, 8.7
and 9.5. In this figure, it can be observed that the absorption
band at ∼300 nm disappears, however, a remaining adsorption
it is observed, which decreases as the pH increase, being the
highest adsorption for the spectrum of pH 7.4, due to the
remaining chloride in the complex, Cu(NH3)3Cl+. On the other
hand, an adsorption band is observed between 500 and 700 nm.
The maximum for these absorption bands are located at 660,
605 and 620 nm for pH 7.4, 8.7 and 9.5, respectively. These
adsorption bands have been reported for amino complexes
of Cu(II) [13,25]. A negative displacement in the adsorption
band is expected as the pH increases, due to higher energy
of molecules that present absorption at lower wavelength
(batochromic effect). This effect is observed for pH 7.4 and
The similitude between the complexes energy between
2+
2+
Cu(NH3)4 ≈ Cu(NH3)5 is certainly due the coexistence of
these complexes at the pH evaluated as it is shown in Fig. 3.
3.2.2. X-ray diffraction analysis
X-ray diffraction (XRD) analysis was used to characterize the
solid phases that limit the solubility of copper. Due to the diffi-
culty to generate the necessary amount of Cu(I) precipitates to
perform XRD analyses, only Cu(II) species were characterized.
The solid phases analyzed were produced following experi-
mental conditions established in the thermodynamic diagrams.
Hence, when the pH of the Cu(II) solution was adjusted to a
pH between 4.23 and 6.55 a green precipitate was immediately
formed. The XRD analysis of this solid is shown in Fig. 7, where
it can be observed that the obtained diffraction spectrum fits the
pattern of CuOH1.5Cl0.5 according to the Joint Committee on
Diffraction Standards (JCDS, Fig. 7a) cards [27]. The Kps value
(1 × 10−17.16) for CuOH1.5Cl0.5(s), denotes the importance of
thisspeciesintermsofsolubilitylimitationforCu(II)atthosepH
values. As shown in this study (Fig. 4) and as reported by other
authors the solubility limitation by CuOH1.5Cl0.5(s) is enhanced
when there is high chloride ion concentration [9,18]. However,
there are reports showing that for chloride concentrations higher
2+
8.5 indicating that Cu(NH3)3Cl+ and Cu(NH3)4 are the
dominant species for these conditions. However, for pH 9.5
an inverse displacement is observed because the energy of
the complex is lowered due to the inclusion of the fifth NH3
2+
molecule to form the Cu(NH3)5 [26]. It is worth noting that

J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
6113
was suitable to confirm predictions of existence of all complexes
under study.
Cyclic voltammograms (CV) obtained for each solution
employed are illustrated in Fig. 8. In all CV’s it is observed
that the open circuit potential (OCP) in all solutions is within
the potential region of equilibrium for the Cu(I) complexes, and
it follows the same trend depicted by Eh–pH diagrams, that is, a
negative displacement as the pH increases. Given the difficulty
of stabilizing Cu(I), it was not unexpected to observe the pres-
ence of some Cu(II) in the solutions employed (initial reduction
wave in Fig. 8b and c), fortunately the amount seems to be min-
imal, thus the processes observed account for those of Cu(I)
complexes unless otherwise stated. Therefore the initial Cu(II)
would not significantly contribute to the peak labeled as Ic’, that
corresponds to reduction of Cu(I) to Cu(0). Thus, the reactions
proposed for each pH value would be:
pH
3
CuCl32−(aq) + e− = Cu(s)
pH 7.5 Cu(NH3)2Cl(aq) + H+(aq) + e−
= Cu(s) + NH4+(aq) + NH3(aq) + Cl−(aq)
Fig. 7. XRD diagrams obtained for the precipitates: (a) CuOH1.5Cl0.5 and (b)
CuO. Vertical line groups correspond to the JCPDS card for each solid phase.
pH 9.5 Cu(NH3)2Cl(aq) + e−
= Cu(s) + 2NH3(aq) + Cl−(aq)
than 5.5m the formation of the solid phase Cu(NH3)2Cl2(s) is
predominant [7]. No indication of those observations has been
found at any of the conditions studied in this work. On the other
It can be devised from these equations that for alkaline pH
there would be an excess of ammonia at the interface, which
seems to promote dissolution of Cu(0), this observed as a small
wave (Iaꢀ) before peak IIaꢀ develops. As soon as the excess of
ammonia is consumed the expected process of Cu(0) oxidation
to Cu(I) is observed. Thus, peak IIaꢀ in each voltammogram of
Fig. 8, would correspond to oxidation of Cu(0) to Cu(NH3)2Cl
for pH 7.5 and 9.5 (Fig. 8b and c) and oxidation of Cu(0)
hand, the Kps value for Cu(NH3)2Cl2(s) is higher (1 × 10−1.7
)
than that for CuOH1.5Cl0.5(s)
.
AnothersolidphasedescribedinthediagramsforCu(II)isthe
CuO(s), in order to generate this precipitate the pH of the solution
was risen to a value above 12.5. The XRD spectrum obtained
for this solid phase corresponds to a solid with crystallography
and chemical composition similar to CuO (Fig. 7b).
2−
to CuCl3 at pH 3 (Fig. 8a). In correspondence with ther-
modynamic predictions it is observed that there is a higher
overpotential involved for oxidation of Cu(0) at lower pH, it
however, seems to be a process kinetically favored.
3.3. Voltammetric study
A voltammetric study was carried out in order to character-
ize the electrochemical behavior of Cu(I) and Cu(II) when these
ions are present as complexes in a ammonia-chloride medium.
The experimental set up was based on results shown by ther-
modynamic diagrams (Figs. 1–5). The experimental conditions
used in this work were chosen on the basis of relating these to
hydrometallurgical systems (high chloride and high ammonia
concentrations) and they include predominance regions for most
of the possible binary and ternary complexes for this medium.
3.3.2. Electrochemistry of the oxidation process of Cu(I)
Solutions such as the ones described above were used to
characterize the oxidation behavior of Cu(I) complexes and the
corresponding cyclic voltammograms are illustrated in Fig. 9. In
this case, the redox couples involved include the pH zone around
8.73 and a CV for this pH region is included. In all cases a non-
reversible redox behavior is observed for the Cu(I)/Cu(II) com-
plexes pairs, which present a ꢁEp > > 0.059 V. Thus, it is pro-
posed that for each pH zone studied the reactions involved are:
CuCl32−(aq) → CuCl3−(aq) + e−
3.3.1. Electrochemistry of the reduction process of Cu(I)
On the basis of the thermodynamic results three pH regions
were chosen to characterize the reduction process of Cu(I) com-
plexes in ammonia-chloride medium. Thus, the complexation
forms for Cu(I) at the three pH zones (values in parenthesis)
pH
3
Cu(NH3)2Cl(aq) + (x − 2)NH4+(aq)
→ Cu(NH3)xCly(2−y)+(aq) + (x − 2)H+(aq)
+ (1 − y)Cl−(aq) + e−
where pH 7.5 (x = 3, y = 1), pH 8.73 (x = 4, y = 0) and pH 9.5
(x = 5, y = 0).
2−
were: CuCl3 (3) and Cu(NH3)2Cl (7.5 and 9.5). The solu-
tions employed in these experiments involved a concentration
of 0.02 M Cu(I) which is lower than the used for other experi-
ments as this helped to avoid oxidation to Cu(II). Nevertheless, it

6114
J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
Fig. 8. Voltammetric response of Cu(I) complexes on a glassy carbon electrode at 25 ◦C, scan rate of 100 mV s−1. The scan was initiated from the OCP towards the
negative direction. All solutions employed containing 0.02 M CuCl and 4 M NH4Cl and adjusted at different pH values: (a) 3.0, (b) 7.5 and (c) 9.5.
Fig. 9. Voltammetric response of Cu(I) complexes on a glassy carbon electrode at 25 ◦C, scan rate of 100 mV s−1. The scan was initiated from the OCP towards the
positive direction. All solutions employed containing 0.02 M CuCl and 4 M NH4Cl and adjusted at different pH values: (a) 3.0 and (b) 9.5.

J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
6115
3.3.3. Electrochemistry of the reduction process of Cu(II)
To characterize the electrochemical reduction of Cu(II), four
pH zones were chosen. These zones were chosen on the basis
of the species that are predicted to be predominant at those pH
values (values in parenthesis), that is: CuCl3− (3), Cu(NH3)3Cl+
where pH 7.5 (x = 3, y = 1), pH 8.73 (x = 4, y = 0) and pH 9.5
(x = 5, y = 0).
Thus, in the case of CuCl3− the reduction process would be
CuCl3−(aq) + e− = CuCl32−(aq)
2+
(7.5), Cu(NH3)4 (8.73) and Cu(NH3)52+ (9.3).
As expected in all cases the peak current for process Ic is
practically the same, as in all cases a one electron process is
involved and the displacement in peak potential as predicted it
is a function of the change in sphere coordination.
Cyclic voltammograms (CV) were obtained in solutions of
0.2 M Cu(II) + 4 M NH4Cl, which were adjusted at the pH values
mentioned above and the results are illustrated in Fig. 10. It
is observed that the open circuit potential for each CV shifts
towards negative values as the pH increases and that it follows
the same trend observed in the Eh–pH diagram (Fig. 5). All CV’s
show two cathodic processes labeled as Ic and IIc on the direct
scan and depending on pH several anodic processes labeled as
Ia, IIa and IIIa on the−reverse scan.
Thus, IIc in Fig. 10b–d corresponds to the reduction of
Cu(NH3)2Cl to metallic copper and for Fig. 10a reduction of
2−
CuCl3 to metallic copper, according to the following reac-
tions:
Cu(NH3)2Cl(aq) + 2H+(aq) + e
Except for CuCl3 (Fig. 10a), it is proposed that in all
cases Ic corresponds to the reduction of the Cu(II) complex to
Cu(NH3)2Cl, according to
= Cu(s) + NH4+(aq) + NH3(aq) + Cl−(aq)
CuCl32−(aq) + e = Cu(s) + 3Cl−(aq)
Cu(NH3)xCly(2−y)+(aq) + (x − 2)H+(aq)
+ (1 − y)Cl−(aq) + e−
Different from experiments of direct reduction of Cu(I) com-
plexes to Cu metal, the corresponding potential for peak IIc
(Fig. 10) is more negative than the potential for peak Icꢀ (Fig. 8),
= Cu(NH3)2Cl(aq) + (x − 2)NH4+(aq)
Fig. 10. Voltammetric response of Cu(II) complexes on a glassy carbon electrode at 25 ◦C, scan rate of 100 mV s−1. The scan was initiated from the OCP towards
the negative direction. Thick line corresponds to voltammetric response obtained when potential was switched before reduction of Cu(I) to Cu(0). All solutions
employed containing 0.2 M CuCl2 and 4 M NH4Cl and adjusted at different pH values: (a) 3.0, (b) 7.5, (c) 8.73 and (d) 9.5.

6116
J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
Table 5
Experimental and theoretical potentials calculated for Cu(II)/Cu(I) couples in ammonia-chloride system
ꢀ
ꢀ
ꢀ
(V)
exp.
ꢀ
ꢀ
a
b
pH
Redox couple
CuCl3−/CuCl3
E◦
(V)
E◦
(V)
E◦
|E◦
− E◦exp .| (V)
exp.
Thermo.
Thermo.
2−
3
0.495
0.212
0.104
0.036
0.542
0.104
0.020
−0.029
0.499
0.245
0.117
0.057
0.047
0.108
0.084
0.065
0.004
0.033
0.013
0.021
7.5
8.73
9.3
Cu(NH3)3Cl+/Cu(NH3)2Cl
Cu(NH3)42+/Cu(NH3)2Cl
Cu(NH3)52+/Cu(NH3)2Cl
a
From Cu(II)–NH3–Cl− system, Fig. 10.
From Cu(I)–NH3–Cl− system, Fig. 9.
b
Table 6
this as a result of having a higher concentration of Cu(I) com-
plexes in this case.
Experimental and theoretical potentials calculated for Cu(I)/Cu(0) couples in
ammonia-chloride system
As the sweep is reversed a similar behavior to that observed
for oxidation of Cu(0) formed from Cu(I) solutions at alka-
line pH (Fig. 8b–d) is observed, which as previously suggested
could be triggered by the excess of ammonia at the interface.
Therefore, the same oxidation product is obtained for peak IIa,
Cu(NH3)2Cl. The interface at this point would be the same for
the three alkaline systems (pH 7.5, 8.73, 9.5), however, as indi-
cated by the Eh–pH diagram the subsequent oxidation to Cu(II)
requires a lower overpotential for the change of Cu(NH3)2Cl
ꢀ
ꢀ
pH
Redox couple
Ecross. (V)
E◦
(V)
|E◦
−
Thermo.
Thermo.
ꢀ
E◦exp .| (V)
3
7.5
9.3
CuCl32−/Cu
Cu(NH3)2Cl/Cu
Cu(NH3)2Cl/Cu
−0.092
−0.112
−0.246
0.018
0.110
0.039
0.040
−0.073
−0.286
Table 7
Formation constants for binary and ternary Cu(I) and Cu(II) complexes in
ammonia-chloride medium
to either Cu(NH3)5 or Cu(NH3)42+; hence, the potential dif-
2+
ference between peaks IIa and IIIa is lower as compared to
the oxidation to Cu(NH3)3Cl+ (Fig. 10c and d). For the acidic
zone, a completely different oxidation behavior is observed and
Reaction
log β
log βa
Cu+ + 3Cl− ↔ CuCl3
5.99 [19]
−0.3 [5]
11.3 [8]
6.05
−0.28
11.55
11.34
2−
−
Cu2+ + 3Cl− ↔ CuCl3
2−
sharper anodic peaks depicting oxidation of C−u(0) to CuCl3
Cu+ + 2NH3 + Cl− → Cu(NH3)2Cl
(peak IIa) and subsequent oxidation to CuCl3 (peak IIIa) are
obtained. The oxidation of the predominant species for each
species at each pH zone studied (peak IIIa) was corroborated
by the direct oxidation from Cu(I) to Cu(II) species (CV’s at a
negative switching potential before reduction of Cu(I) to Cu(0)).
Cu2+ + Cl− + 3NH3 → Cu(NH3)3Cl+
11.73 [5]
a
This work.
ꢀ
(Ecross − E◦Thermo), there is good agreement with theoretical
data. The deviation for acid pH is attributed to a lower stabil-
ity of the binary complexes, which require a higher energy for
chargetransferascomparedto theternarycomplexes. Therefore,
this deviation could be considered a deviation from reversibility
for pH 3.
3.3.4. Thermodynamic calculations from experimental data
Formal potentials (E◦ꢀ) for all Cu(I)/Cu(II) couples involved
in voltammograms of Figs. 9 and 10 were determined using
conventional electrochemical methods [28]. These values were
ꢀ
then compared against the thermodynamic values (E◦
)
Finally, supported on these last results and using reported
data (E and β) for well known systems (i.e. Cu(NH)4
Thermo
2+
,
calculated for the Eh–pH diagram. The comparative analysis
(Table 5) ꢀshows thanꢀin both systems there is a higher devi-
ation (E◦Thermo − E◦ ) when the pH approaches the neutral
zone, it, however, is larger for the data obtained with results
of reduction of Cu(II) to Cu(I) (Fig. 10) than the obtained
from results of oxidation of Cu(I) to Cu(II) (Fig. 9). All of this
could be associated to the fact that the Cu(II) system involves
a wider mixture of complexes throughout the whole pH range
and this is notably marked around the neutral zone. Therefore,
such results suggest that the coexistence of multiple complexes
affectsthereversibilityoftheredoxprocessesofthepredominant
species.
Likewise, formal potentials for the Cu(I)/Cu(0) system were
established (Table 6), however, different from the Cu(I)/Cu(II)
system, in this case crossover potentials (Ecross) values were
used instead, since as it has been found Ecross corresponds to the
reversible potential of the metal redox couple deposited with
the ion in solution [20,29]. From data in Table 7, it can be
observed that except for pH 3, which shows a high deviation
Cu(NH3)52+), formation constants (β) for some of the binary
and ternary complexes were calculated (Table 7). Compar-
ison of these results against data reported in the literature
(Tables 1 and 3) shows a good agreement, which further supports
the predictions made for conditions typical of hydrometallurgi-
cal systems.
4. Conclusions
An electrochemical characterization has been performed and
it has been shown the redox characteristics of several binary
and ternary copper complexes. In general ternary complexes are
shown to be more stable and this has been particularly evident
for the Cu(I) system. The existence of the proposed complexes
for each pH region studied has been supported by calculations
of formation constants. The results show a general agreement
with respect to thermodynamic predictions and the deviations
observed in some cases are attributed to the kinetic effect

J. Vazquez-Arenas et al. / Electrochimica Acta 52 (2007) 6106–6117
6117
provoked by ammonia excess at the interface. These results
together with UV–vis and XRD spectra allowed to confirm the
existence of copper species normally not considered.
[6] M. Olper, in: G.W. Warren (Ed.), EPD Congress, The Minerals, Metals and
Materials Society, 1994, p. 513.
[7] J.L. Limpo, A. Luis, M.C. Cristina, Rev. Metal. Madrid 29 (1993) 27.
[8] J.S. Solis, G. Hefter, P.M. May, Aust. J. Chem 48 (1995) 1283.
[9] C. Nila, I. Gonza´lez, Hidrometallurgy 47 (1996) 63.
[10] D. Grujicic, B. Pesic, Electrochim. Acta 50, 22 (2005) 4426.
[11] J. Crousier, I. Bimaghra, Electrochim. Acta 34, 8 (1989) 1205.
[12] J. Black, L. Spiccia, D.C. McPhail, in: C. Young, A. Alfantzi, C. Anderson,
A. James, D. Dreisinger, B. Harris (Eds.), Proceedings of the 5th Interna-
tional symposium of hydrometallurgy, vol. x, Vancouver, Canada, 1990, p.
183.
[13] A. Lam, D.B. Dreisinger, in: C. Young, A. Alfantzi, C. Anderson, A.
James, D. Dreisinger, B. Harris (Eds.), Proceedings of the 5th International
Symposium of Hydrometallurgy, Vancouver, Canada, 2003, p. 195.
[14] G. Eriksson, Anal. Chim. Acta 112 (1979) 375.
[15] I. Puigdomenech, INPUT, SED, and PREDOM: Computer programs,
drawing equilibrium diagrams, TRITA-OOK-3010, Royal Institute of
Technology, September 1983, 12 pp.
[16] National Institute of Standards and Technology, NIST, Critically Selected
Stability Cosntants of Metal Complexes, 2001.
[17] G.W. McDonald, A. Saud, M.S. Barger, J.A. Koutski, S.H. Langer,
Hydrometallurgy 18 (1987) 321.
[18] R.M. Smith, A.E. Martell, CriticalStabilityConstants, vol. 4, PlenumPress,
New York, 1976.
The inclusion of ternary complexes in this study allowed a
better understanding of Cu(I) and Cu(II) solubility in ammonia-
chloride media. Thus, it has been shown the importance of
considering the combined effect of ammonia and chloride con-
centration on the stability of the different copper complexes. In
this way and for the experimental conditions studied, it has been
found that the only solubility limitation for Cu(I) is the forma-
tion of Cu2O at alkaline pH values, however, this limitation is
notably reduced if the copper ternary complex Cu(NH3)2Cl is
formed. Although, Cu(II) presents solubility limitations either
for formation of Cu(OH)1.5Cl0.5 under acidic conditions, and
because of CuO formation at very alkaline pH. It has been high-
lighted that for close to neutral pH the solubility limitation is
reduced by Cu(NH3)3Cl+ formation.
The results here presented, further support the potential appli-
cation of this medium in copper hydrometallurgical processes.
Acknowledgements
[19] M. Wang, Y. Zhang, M. Muhammed, Hydrometallurgy 45 (1997) 53.
[20] D.M. Muir, In: E. Peek, G. Van Weert (Eds.), Chloride Metallurgy, vol. II,
32nd Annual Hydrometallurgy Meeting, 2002, p. 759.
[21] C. Nila, I. Gonza´lez, J. Electroanal. Chem. 401 (1996) 171.
[22] Z.D. Stankovic, Electrochim. Acta 29 (1984) 407.
[23] I.V. Gorbunova, L.I. Lyamina, K.M. Gorbunova, Sov. Electrochem. 23
(1987) 1027.
[24] A. Ramos, M. Miranda, I. Gonzalez, J. Electrochem. Soc. 148 (2001) C315.
[25] M. Valli, S. Matsuo, H. Wakita, T. Yamaguchi, M. Nombra, Inorg. Chem.
35 (1996) 5642.
Financial support for this study has been obtained through
CONACyT grant 39585-Y and UASLP-C03-FRC-06-5.6. J.
´
Vazquez is indebted for the scholarship provided by CONACyT
to carry out M.Eng. studies. An especial acknowledgement to
the reviewers who made comments and suggestions that notably
improved this paper.
[26] F. Cotton, G. Wilkinson, Advanced Inorganic Chemistry, Interscience
Publishers–Wiley, USA, 1990.
References
[27] JCDS-International Centre for Diffraction Data, Ref: Swanson, Tatge, Natl.
Bur Stand (U.S), Circ., vol. 539, no. I, 1953, p. 49.
[1] E. Peters, Hydrometallurgy 29 (1992) 431.
[2] G.M. Swinkels, R.M. Berezowsky, CIM Bull. 71 (1978) 105.
[3] R. Winand, Hydrometallurgy 27 (1981) 285.
[28] A. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and
Applications, Wiley and Sons, USA, 1980.
[4] C. Ek, J. Frenay, J.C. Herman, Hydrometallurgy 8 (1982) 17.
[5] J.L. Limpo, J.M. Figueiredo, S.L. Amer, Hydrometallurgy 28 (1992) 149.
[29] S. Flecther, C.S. Halliday, D. Gates, M. Westcott, T. Lwin, G. Nelson, J.
Electroanal. Chem. 159, 267 (1983).