Minerals Engineering 100 (2017) 9–16

Contents lists available at ScienceDirect

Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng

New insights into the influence of redox potential on chalcopyrite

leaching behaviour
Mohammad Khoshkhoo a,⇑, Mark Dopson b, Fredrik Engström a, Åke Sandström a
a
MiMeR – Minerals and Metallurgical Research Laboratory, Luleå University of Technology, SE-971 87 Luleå, Sweden
b
Centre for Ecology and Evolution in Microbial Model Systems, Linnaeus University, 391 82 Kalmar, Sweden

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

Article history: Chalcopyrite (CuFeS2) is the most economically important and most refractory copper mineral when trea-
Received 2 July 2016 ted in conventional sulphate media leaching systems. In this study, the effect of solution redox potential
Revised 28 September 2016 on leaching of a pure and a pyritic chalcopyrite concentrate was investigated using concentrates with
Accepted 2 October 2016
fresh and aged surfaces. In experiments using concentrates with fresh surfaces, the response to redox
Available online 8 October 2016
potential depended on the presence of pyrite: fresh pyritic concentrate leached more effectively at low
redox potential (in agreement with reductive leaching mechanisms), while the leaching efficiencies from
Keywords:
fresh pure concentrate were similar at high and low redox potentials. The data suggested that the
Chalcopyrite
Hindered dissolution
reductive leaching mechanism does not necessarily result in higher and faster recoveries in the absence
Passivation of the galvanic interaction induced by the presence of pyrite. It was also found that exposure of chalcopy-
Reductive leaching rite to atmospheric oxidation prior to leaching (ageing) had an effect on leaching behaviour in response to
Galvanic interaction redox potential: copper recoveries in leaching of aged concentrates were higher at high redox potentials.
This behaviour was attributed to the presence of iron–oxyhydroxides on the surface of aged concentrates.
Based on the data from this investigation and previous surface studies, it is proposed that iron–
oxyhydroxides play an important role in triggering the hindered dissolution of chalcopyrite.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction compounds on the mineral surface during leaching. The most fre-
quently suggested candidates to form the passivating layers are
As the most abundant copper mineral, chalcopyrite (CuFeS2) is metal-deficient phases, elemental sulphur and jarosite (Klauber,
the most economically important copper resource. Leaching of 2008). Recently iron–oxyhydroxides have also been proposed to
chalcopyrite in sulphate media is mainly preferred for its easy cause passivation (Khoshkhoo et al., 2014a). Similarly, other
handling and compatibility with established chalcocite (Cu2S) aspects of chalcopyrite (bio)leaching, such as the mechanism,
and covellite (CuS) heap leaching/bioleaching processes as well kinetics, role of microorganisms and effect of different leaching
as with conventional solvent extraction and electrowinning parameters are still under debate. A short review of the concepts
technologies (Marsden, 2007). However, the slow dissolution of that are central to the discussions presented in this investigation
chalcopyrite at ambient pressure in sulphate media has hindered is given in the following sections.
its industrial application. Nevertheless, due to the continuous
depletion of high grade ores, future hydrometallurgical alterna- 1.1. Galvanic dissolution of chalcopyrite
tives for chalcopyrite treatment will increase in importance and
applications. As a result, different aspects of chalcopyrite leaching The effect of impurities on the dissolution of chalcopyrite is a
have been extensively researched in recent decades. A summary of relatively old observation (Dutrizac and MacDonald, 1973). For sul-
these studies can be found in Debernardi and Carlesi (2013) and Li phide minerals that are in contact with each other, the mineral
et al. (2013). A review of hydrometallurgical options for chalcopy- with higher rest potential acts as a cathode and is preserved, while
rite treatment is available in Dreisinger (2006) amongst others. the mineral with lower rest potential acts as an anode and is
The slow dissolution of chalcopyrite is termed ‘passivation’ or oxidised (Mehta and Murr, 1983). For chalcopyrite, pyrite gives rise
‘hindered dissolution’ and is believed to be due to formation of to increased copper release as it acts as the cathode and chalcopy-
rite serves as the anode and is oxidised. The main reaction for
⇑ Corresponding author. chalcopyrite dissolution in ferric/ferrous sulphate systems is the
E-mail address: mohammad.khoshkhoo@ltu.se (M. Khoshkhoo). oxidation of chalcopyrite by ferric ions:

http://dx.doi.org/10.1016/j.mineng.2016.10.003
0892-6875/Ó 2016 Elsevier Ltd. All rights reserved.
10 M. Khoshkhoo et al. / Minerals Engineering 100 (2017) 9–16

CuFeS2 þ 4Fe3þ ! Cu2þ þ 5Fe2þ þ 2S0 ð1Þ dissolution rate is optimal. This range is reported to be around
600–680 mV vs SHE (Koleini et al., 2011) which is considered as
Being electrochemical in nature, this reaction can be expressed
a low redox potential level in chalcopyrite leaching systems.
as an anodic half-cell reaction (Eq. (2)) and a cathodic half-cell
The reductive leaching mechanism has also been described by
reaction (Eq. (3)):
another model based on evolution of surface layers measured by
X-ray photoelectron spectroscopy (XPS), time of flight secondary
CuFeS2 ! Cu2þ þ Fe2þ þ 2S0 þ 4e ð2Þ
ion mass spectrometry (ToF-SIMS) and scanning electron micro-
scopy (SEM) (Harmer et al., 2006). In this model, it is suggested
4Fe3þ þ 4e ! 4Fe2þ ð3Þ that chalcopyrite is partially oxidised by ferric ions in the first step,
resulting in a polysulphide phase and release of cupric and ferrous
The limiting step is reported to be the reduction of ferric iron
ions (Eq. (8)). The second step consists of reduction of the polysul-
which is slow on the surface of chalcopyrite due to presence of
phide intermediate by ferrous ions to sulphide (Eq. (9)) and finally,
hindering layers (Dixon et al., 2008). When pyrite and chalcopyrite
the sulphide is oxidised with ferric ions forming elemental sulphur
are in contact, the iron reduction takes place on the pyrite surface,
accompanied by additional release of cupric and ferrous ions into
which is much faster. As a result, the total dissolution rate is
the solution (Eq. (10)). This model also requires a relatively low
increased.
redox potential to satisfy the conditions for oxidation and reduc-
tion reactions.
1.2. Reductive leaching mechanism h
i
4þ
3nCuFeS2ðsolidÞ þ ð12n  12ÞFe3þ
ðaqÞ ! 4ðCuFeÞ  6 S2
n
ðsolidÞ
Reductive leaching of chalcopyrite generally refers to dissolu-
tion of chalcopyrite at low redox potential. The redox potential of þ ð3n  4ÞCu2þ 2þ
ðaqÞ þ ð15n  16ÞFeðaqÞ where n P 2 ð8Þ
a system containing ferric and ferrous ions is mainly determined
h
i
by the ratio of free ferric to ferrous ions, according to the Nernst þ
4ðCuFeÞ4þ  6 S2
n þ ð12n  12ÞFe2þ
ðaqÞ þ ð12n  12ÞHðaqÞ
equation (Eq. (4)): ðsolidÞ
h i
! 4ðCuFeÞ4þ  ð12n  12ÞHþ  6nðS2 Þ þ ð12n  12ÞFe3þ
ðaqÞ
R  T fFe3þ g ðsolidÞ
Eh ¼ E þ ln ð4Þ
n  F fFe2þ g where n P 2 ð9Þ

where E° is the standard electrode potential, R is the gas constant, n

½4ðCuFeÞ4þ  ð12n  12ÞHþ  6nðS2 ÞðsolidÞ þ ð12nÞFe3þ 0
ðaqÞ ! 6nSðsolidÞ
is the charge number, F is Faraday’s constant, and {Fe3+} and {Fe2+}
þ
are activities of ferric and ferrous ions, respectively. From chalcopy- þ 4Cu2þ 2þ
ðaqÞ þ ð12n þ 4ÞFeðaqÞ þ ð12n  12ÞHðaqÞ where n P 2
rite oxidation by ferric ions (Eq. (1)), it seems that higher redox ð10Þ
potentials, i.e. higher ferric to ferrous ions concentrations, should
favour its dissolution. This is not what is usually observed and In the present investigation, results from leaching of a pure
instead, dissolution increases with increasing redox potential and concentrate as well as a pyritic concentrate performed under
then decreases after a critical potential is reached (Kametani and strictly controlled redox potential conditions are presented and
Aoki, 1985). A two-step model based on thermodynamic calcula- the interaction of galvanic dissolution with reductive leaching
tions was proposed to support this observation (Hiroyoshi et al., mechanism is discussed.
2000). The model suggests that in the first step chalcopyrite is
reduced by ferrous ions in the presence of cupric ions to form 2. Materials and methods
chalcocite (Eq. (5)). In the second step, chalcocite which is more
amenable to leaching than chalcopyrite is oxidised by ferric ions 2.1. Aitik chalcopyrite concentrate (high purity concentrate)
or oxygen according to Eqs. (6) and (7).
A high purity chalcopyrite concentrate from the Boliden
CuFeS2 þ 3Cu2þ þ 3Fe2þ ! 2Cu2 S þ 4Fe3þ ð5Þ Mineral AB owned Aitik mine in Sweden was used. The concentrate
was washed with ethanol and water, dried by acetone rinsing at
2Cu2 S þ 8Fe3þ ! 4Cu2þ þ 2S0 þ 8Fe2þ ð6Þ room temperature, divided into 15 g lots, sealed in plastic contain-
ers flushed with nitrogen and kept at 4 °C. The concentrate was
ground in a ring mill for 15 s immediately before addition to the
2Cu2 S þ 8Hþ þ 2O2 ! 4Cu2þ þ 2S0 þ 4H2 O ð7Þ
leaching reactors. This procedure was carried out to assure
This model suggests that the redox potential has to be low presence of fresh chalcopyrite surfaces and comparatively similar
enough for reduction of chalcopyrite in the first step and high initial surface characteristics in all the experiments. The repro-
enough for oxidation of chalcocite in the second. Consequently, ducibility of the grinding procedure was tested in triplicates giving
there is a window of oxidation potential in which the chalcopyrite a d50 of 18 ± 1 lm, a d80 of 42 ± 3 lm and a specific surface area of

Table 1
Chemical and mineralogical composition of the concentrate used in leaching experiments.

Chemical composition (%) Calculated mineralogy (%)

Cu Fe S Zn Pb SiO2 CuFeS2 FeS2 ZnS PbS
Aitik (ground) 32.4 30.9 34.1 0.11 0.05 0.53 94.0 4.00 0.16 0.06
Aitik (38–53 lm) 31.8 30.2 35.2 0.05 0.01 0.45 92.0 6.00 0.07 0.01
Kristineberg (ground) 23.6 34.7 37.5 2.10 0.70 1.10 68.0 26 3.20 0.80
Kristineberg (38–53 lm) 21.6 33.8 39.9 1.7 0.58 0.68 62.4 32.4 2.53 0.67
Aitik Pyrite Conc. 0.04 36.4 39.6 <0.01 <0.01 23.0a 0.12 76.0 0.01 <0.01
a
Total gangue including aluminosilicate minerals (mostly kyanite) for pyrite concentrate.
 M. Khoshkhoo et al. / Minerals Engineering 100 (2017) 9–16 11

mineral in the concentrate (Table 1 and Figs. 1 and 2a). Part of the
Aitik concentrate that had been stored for two years in the original
packing at room temperature was used to investigate the effect of
ageing. The aged concentrate was washed with ethanol and water,
wet sieved, and the particle size fraction between 38 and 53 lm
was dried with acetone rinsing and aliquots sealed in containers
flushed with nitrogen. The sealed containers were kept in a
refrigerator (4 °C) until used. The fraction’s chemical analysis
and calculated mineralogy are given in Table 1 and its surface area
was 0.38 ± 0.01 m2/g.

2.2. Kristineberg chalcopyrite concentrate (pyritic concentrate)

The regular chalcopyrite concentrate from Kristineberg mine,

Boliden Mineral AB contained higher amount of pyrite compared
to the Aitik concentrate (Table 1 and Fig. 1). SEM/EDS analyses
showed that in some grains pyrite and chalcopyrite co-existed
(Fig. 2b). Leaching experiments were performed on the ground
concentrate as well as the aged 38–53 lm fractions (approxi-
mately two years old). The preparation of the concentrate was
similar to the Aitik concentrate. For producing fresh surfaces, the
concentrate was ground for 10 s in the ring mill before each
experiment giving a d50 of 18 ± 1 lm, a d80 of 45 ± 1 lm and a
specific surface area of 0.75 ± 0.02 m2/g. The specific surface area
of the 38–53 lm fractions were 0.35 ± 0.01 m2/g.

Fig. 1. XRD patterns of Aitik and Kristineberg concentrate. Abbreviations: C: 2.3. Aitik pyrite concentrate
chalcopyrite, P: pyrite, G: ganuge.

The pyrite concentrate originating from the Aitik mine was

obtained from tailings of a chalcopyrite flotation at the Boliden
plant and acid washed for removal of the acid soluble gangues.
The amount of pyrite in the concentrate calculated based on the
chemical analyses after the acid washing stage was around 74%
(Table 1). The total amount of the gangue minerals, mostly as
kyanite (Al2SiO5), was around 23%.

2.4. Leaching experiments under controlled redox potential conditions

Batch leaching experiments were performed in 2 l baffled

reactors with a working volume of 1.25 l. Heating was provided
by an electric plate controlled at 80 °C (unless stated otherwise)
by an Ebro GFX 460 Fuzzy Controller. The pH of the solution was
regularly monitored and controlled at 1.5 by manual addition of
5 M sulphuric acid during the experiments. The initial iron concen-
tration (1 g/l (0.02 M) and in some experiments 5 g/l (0.09 M)) was
supplied by addition of ferrous sulphate. When applicable, the ini-
tial concentration of copper was adjusted by addition of copper
sulphate. Stirring was provided by an overhead paddle stirrer at
400 rpm in all experiments. The reactors were equipped with
condensers to minimise evaporation. After the temperature and
redox potential were set at the desired level, 12.5 g of concentrate
(1% (wt/vol) solid content) was added to the reactor. The redox
potential of the solution was continuously measured using a
platinum electrode with a Ag/AgCl reference electrode (Metrohm)
containing 3 M KCl solution as the electrolyte. Unless otherwise
specified, all reported redox potentials are versus the Ag/AgCl/3 M
KCl reference electrode. The electrodes were checked with a refer-
ence solution at frequent intervals and were replaced with a fresh
and clean electrode at least once a day. The redox potential was
adjusted to the set value by automatic addition of a dilute hydro-
Fig. 2. Backscattered electron image of (a) Aitik concentrate and (b) Kristineberg
gen peroxide solution (1.5% v/v). The program for setting the redox
concentrate showing chalcopyrite (C), pyrite (P) and gangue minerals (G).
potential level and controlling the addition of the oxidising agent
was written in LabView software (National Instruments). In exper-
0.90 ± 0.03 m2/g. X-ray diffraction (XRD) and scanning electron iments controlled at a low redox potential level (420–450 mV),
microscopy coupled with energy-dispersive X-ray spectroscopy nitrogen was sparged into the reactors, minimising the oxidation
(SEM/EDS) analyses showed that chalcopyrite was the only copper of ferrous to ferric ions by air, and thus avoiding a loss of redox
12 M. Khoshkhoo et al. / Minerals Engineering 100 (2017) 9–16

potential control. The nitrogen was of chemical quality provided by

AGA with 65 ppm of oxygen and moisture. The described setup
successfully controlled the redox potential ±5 mV of the set values
during the relatively long leaching periods in all the experiments.
One experiment was carried out at a lower redox potential level
(350 mV) where the redox was controlled by automatic addition
of a dilute sodium sulphite solution. The reactors were regularly
sampled and the pulp volume was monitored by weighing before
sampling and adjusted if necessary. The samples were filtered
and the remaining solids were returned to the reactors. The
amount of dissolved copper removed during sampling was
accounted for in the recovery calculations. The pulp was filtered
at the end of the experiments, washed, dried at room temperature
and saved for further analysis. Four selected experiments were
performed in duplicates and the maximum standard deviation in
copper recovery was found to be 1.7.

Fig. 3. Effect of redox potential in leaching of freshly ground Kristineberg

2.5. Calculation of activation energies
concentrate. Leaching conditions: T = 80 °C, pH = 1.4–1.5, initial iron: 1 g/l,
initial copper: 0 g/l. Solid content (S.C.) in the longer experiment at 620 mV was
Initial activation energies at low (420 mV) and high (620 mV) 0.3% (w/v).
redox potentials were calculated by plotting the Arrhenius
equation (Eq. (11)) where k is the leaching rate (mg/h), A is the
pre-exponential factor, Ea is the activation energy, R is the univer- with a lower solid content (4 g instead of 12.5 g) so the redox
sal gas constant and T is temperature (K). potential could be accurately controlled over a longer leaching per-
iod. In leaching of the relatively pure Aitik concentrate, in absence
Ea
ln k ¼ ln A  ð11Þ of pyrite, the effect of reductive leaching was not apparent and the
RT recoveries became more or less similar at low and high redox
Rates from the initial stage of leaching experiments (recoveries potentials (Fig. 4). When pyrite was added to the freshly ground
below 10%) at different temperatures were extracted from the Aitik concentrate and the experiment were repeated at 420 mV,
leaching data by fitting a straight line to at least four data points, the leaching was improved from 50% to around 65% after 24 h of
resulting in R2 of 0.99 or better for all the fits. The natural loga- leaching (Fig. 5; results from leaching of the Kristineberg concen-
rithms of rates (in mg/h) were plotted against the reciprocal of trate at 420 mV is also given for comparison). The amount of pyrite
the temperature (K). The resulting Arrhenius plots were fitted by and chalcopyrite concentrates in this experiment reproduced the
a linear equation and the slopes of the line (Ea/R) were used for pyrite:chalcopyrite weight ratio in the Kristineberg concentrate
activation energy calculations. (0.38) while keeping the total solid content at around 1% (9.04 g
Aitik concentrate mixed with 3.71 g pyrite concentrate). The
2.6. Analyses difference between copper recoveries in Kristineberg leaching
experiment and pyrite added Aitik experiment could be due to
Dissolved iron and copper in the solution samples were the differences in the pyrite source, the presence of pyrite in the
analysed by atomic absorption spectroscopy (AAS; Perkin Elmer, same grains with chalcopyrite in the Kristineberg concentrate
Analyst 100). A CILAS 1064 Liquid apparatus and a Micromeritics (Fig. 2b) as well as a different particle size distribution of chalcopy-
Flowsorb II 2300 were employed for particle size distributions rite in the ground Kristineberg compared to that in the Aitik
and surface area measurements, respectively. The phases and the concentrate. The particle size distribution of the freshly ground
minerals present in the concentrate and residues were examined concentrates were almost similar, but the Kristineberg concentrate
by a PANalytical Empyrean X-ray powder diffractometer. Patterns contained pyrite as well as chalcopyrite and it is possible that their
were collected between 2h angles of 5–90° at Cu Ka radiation of
45 kV, 40 mA. The software HighScore Plus supplemented with
COD 2013 database was utilised to fit and analyse the diffrac-
tograms. A backscattered electron detector on a Ziess (MERLIN
HR) scanning electron microscopy (SEM) was used to produce
SEM photographs and energy dispersive X-ray spectroscopy
(EDS) analyses. The acceleration voltage was set to 20 keV and
the emission current was 1.0 nA.

3. Results and discussion

3.1. Effect of pyrite on reductive leaching

In agreement with the common understanding of the chalcopy-

rite reductive leaching mechanism (Hiroyoshi et al., 2000;
Kametani and Aoki, 1985; Sandström et al., 2005) copper recovery
in leaching of the fresh Kristineberg concentrate performed at low
redox potential (420 mV) was significantly higher than that
obtained at high redox potential (620 mV) after 24 h of leaching
(Fig. 3). Due to high reactivity of the pyrite content in the Fig. 4. Effect of redox potential in leaching of freshly ground Aitik concentrate.
Kristineberg concentrate at 620 mV, the experiment was repeated Leaching conditions: T = 80 °C, pH = 1.4–1.5, initial iron: 1 g/l, initial copper: 0 g/l.
 M. Khoshkhoo et al. / Minerals Engineering 100 (2017) 9–16 13

generally rose with an increase in the redox potential (Fig. 6;

experiments with 1.5 g/l initial copper). A similar trend was
observed in experiments without addition of initial copper that
included one experiment at 350 mV (Fig. 7; see Section 3.3 for
discussion of the effect of the initial copper concentration).
Combining the results in Figs. 6 and 7 suggested that at a redox
potential between 350 and 450 mV, the copper recoveries were
nearly insensitive to redox potential variation but increased when
the redox potential was >500 mV. Copper dissolution from the
aged pyritic Kristineberg concentrate (38–53 lm fractions) was
also greater at higher redox potential (Fig. 8). The leaching results
from fresh and aged concentrates in this investigation cannot be
compared with each other directly, due to their different particle
size distribution. However, it has been previously shown that
when an aged concentrate is leached, the copper dissolution effi-
ciencies are lower compared to leaching of a concentrate with
fresh surfaces under similar leaching conditions (Khoshkhoo
Fig. 5. Effect of pyrite addition in leaching of freshly ground Aitik concentrate.
et al., 2014b).
Leaching conditions: redox potential = 420 mV, T = 80 °C, pH = 1.4–1.5, Initial
iron = 1 g/l, initial copper = 0 g/l. Results from leaching of freshly ground Kristine- Atmospheric oxidation of chalcopyrite (ageing) produces com-
berg concentrate are also given for comparison. pounds such as iron–oxyhydroxides and basic iron–sulphates on
the chalcopyrite surface (Brion, 1980; Buckley and Woods, 1984;
Hackl et al., 1995; Harmer et al., 2006; Klauber et al., 2001;
respective particle size distribution was different from the total Mikhlin et al., 2004). Our previous XPS studies also showed the
particle size distribution. presence of such compounds on the surfaces of the same aged
These results generally indicated that in absence of pyrite the chalcopyrite samples from Aitik (Khoshkhoo et al., 2014a) and
reductive leaching mechanism at low redox potentials would not
improve the leaching efficiencies, compared to high redox poten-
tial experiments. A review of the most cited research articles sug-
gesting higher chalcopyrite dissolution rates within a low redox
potential window with long leaching experiments performed in
ferric/ferrous sulphate media revealed that most studies actually
used a pyritic concentrate (sometimes by addition of extra pyrite)
with a pyrite:chalcopyrite weight ratio of P0.15 (Ahmadi et al.,
2011; Córdoba et al., 2008; Gericke et al., 2010; Kametani and
Aoki, 1985; Koleini et al., 2011; Petersen and Dixon, 2006;
Sandström et al., 2005).
It is difficult to explain the effect of pyrite in accelerating the
reductive leaching of chalcopyrite using the model proposed by
Hiroyoshi et al. (2000) as the role of galvanic interaction is to
provide a surface on which the limiting step of ferric to ferrous
reduction occurs faster compared to on the chalcopyrite surface
(Eq. (3)). In the Hiroyoshi et al. model, ferric to ferrous reduction
occurs during chalcocite oxidation (Eq. (6)) which regardless of
pyrite presence or absence is faster than chalcopyrite oxidation.
Fig. 6. Effect of redox potential in aged Aitik concentrate (38–53 lm) leaching.
Consequently, the dissolution rate of chalcopyrite in the absence
Leaching conditions: T = 80 °C, pH = 1.4–1.5, initial iron: 1 g/l, initial copper: 1.5 g/l.
of pyrite at low redox potentials should still be higher compared
to the experiments at high redox potentials. However, this was
not observed in the long leaching experiments under controlled
redox potential conditions in this investigation (Fig. 4). Instead,
in the absence of pyrite the dissolution rates were comparable at
low and high redox potentials after 72 h of leaching. It is easier
to explain these results by a model similar to the one proposed
by Harmer et al. (2006) whereby the first step is the partial oxida-
tion of chalcopyrite by ferric ions (Eq. (8)). This equation, on one
hand, describes that in the absence of pyrite at low and high redox
potentials similar copper dissolution is obtained. On the other
hand, it supports the significant improvement of copper dissolu-
tion at low redox potential conditions in the presence of pyrite
by providing a surface on which reduction of ferric to ferrous takes
place faster.

3.2. Effect of concentrate ageing in leaching behaviour

The effect of redox potential on leaching behaviour completely

changed when the aged concentrates were used. The copper Fig. 7. Effect of redox potential in aged Aitik concentrate (38–53 lm) leaching.
recoveries from the aged Aitik concentrate (38–53 lm fractions) Leaching conditions: T = 80 °C, pH = 1.4–1.5, initial iron: 1 g/l, initial copper: 0 g/l.
14 M. Khoshkhoo et al. / Minerals Engineering 100 (2017) 9–16

Fig. 8. Effect of redox potential in aged Kristineberg concentrate (38–53 lm)

leaching. Leaching conditions: T = 80 °C, pH = 1.4–1.5, initial iron: 1 g/l, initial
copper: 0 g/l.

Kristineberg (Khoshkhoo et al., 2014b). It is interesting to observe

how the presence of these compounds on the surface, although in
minute amounts, can change the leaching response in regards
to the redox potential conditions. This also implies that the
mechanism of leaching is probably different in presence of ageing
products. The activation energies of both aged (38–53 lm frac-
tions) and freshly ground Aitik concentrate during the initial leach-
ing periods (recoveries below 10%) were >40 kJ/mol and indicated
a chemically controlled reaction (Habashi, 1999) in agreement
with many other reports in the literature (Li et al., 2013). However,
the activation energy of the aged concentrate leached at 420 mV
was much less than that for the fresh concentrate at 420 mV
(76.5 vs. 101.9 kJ/mol) as well as less than that for the aged con-
centrate leached at 620 mV (76.5 vs. 90.9 kJ/mol; Fig. 9, Table 2).
This indicates that the leaching of the aged concentrate at low
redox potential became less sensitive to the temperature variation,
due to the presence of atmospheric oxidation products on the sur-
face of the mineral. At high redox potential, leaching of the aged
concentrate, although considerably slower, was equally tempera- Fig. 9. Arrhenius plots for initial dissolution of (a) aged Aitik concentrate (38–53 lm)
ture sensitive as the fresh concentrate (90.9 vs. 91.8 kJ/mol). It is and (b) freshly ground Aitik concentrate leached at 420 mV and 620 mV.
not completely clear why the ageing products only decreased the
activation energy at low redox potential, but it can be speculated
that the higher activity of ferric ions (main oxidant) at high redox
potential made the reaction equally temperature sensitive in spite Table 2
List of experiments performed for activation energy calculations.
of the aged surfaces. Unfortunately, calculations of activation ener-
gies at later stages of leaching proved to be complicated. The main Experiment Concentrate Redox potential Temp. Ea
problem was that in experiments at low temperature and redox (mV) (°C) (kJ/mol)

potentials where the rate of reactions were too slow (and decreas- A-420-50 Aged Aitik 38–53 lm 420 50 76.5
ing as the leaching progressed) the control of the redox potential A-420-70 420 70
A-420-90 420 90
was lost before significant levels of recoveries were reached.
A-620-30 620 30 90.9
Ageing would not be a concern in industrial scale tank leaching/ A-620-60 620 60
bioleaching as the ore is treated in successive stages of grinding, A-620-80 620 80
flotation and leaching (Khoshkhoo et al., 2014b). However, heap F-420-40 Freshly ground Aitik 420 40 101.9
leaching might take months to years and as a result, atmospheric F-420-60 420 60
oxidation can be a factor in leaching efficiency. It is also important F-420-80 420 80
F-620-40 620 40 91.8
that laboratory investigations take into account the use of aged
F-620-60 620 60
chalcopyrite concentrates. It is normally impossible to identify F-620-80 620 80
ageing by conventional XRD and SEM measurements while the
availability of surface analytical methods such as XPS and auger
electron spectroscopy (AES) is also limited. In addition, atmo-
spheric oxidation is a slow but progressive process that can occur take care to ensure that the concentrate surface characteristics are
in the presence of minute amounts of air and moisture and its similar and of sufficient quality in all experiments. Unfortunately,
extent can be different depending on the age of samples and the effect of ageing is either underestimated or completely ignored
humidity. It has been shown that after three months of ageing in much of the published literature. This might be an important
the concentrate behaved differently under similar leaching condi- factor causing the wide and sometimes contradictory results
tions (Khoshkhoo et al., 2014b). Thus, research laboratories should regarding different aspects of chalcopyrite leaching.
 M. Khoshkhoo et al. / Minerals Engineering 100 (2017) 9–16 15

3.3. Effect of initial copper concentration the results. Nevertheless, it seems that the effect of cupric ion con-
centration in leaching efficiencies is not that profound to be of con-
Two experiments were performed in the absence and presence cern or interest in any possible industrial application.
of 1.5 g/l of initial copper (equivalent to 65% recovery) to examine
if initial cupric ions could possibly have a positive effect on
improving the dissolution rates. The experiments were performed 3.4. Hindered dissolution: a few remarks
on freshly ground Kristineberg concentrate due to its higher
reactivity, the redox potential was controlled at 420 mV and the The formation of iron–oxyhydroxides on chalcopyrite surfaces
temperature was 80 °C. The results were similar in both experi- as a result of atmospheric oxidation is a well–known phenomenon
ments indicating that the presence of cupric ions did not enhance (Brion, 1980; Buckley and Woods, 1984; Hackl et al., 1995; Harmer
the dissolution rates (Fig. 10). et al., 2006; Klauber et al., 2001; Mikhlin et al., 2004). Upon longer
In another series of experiments with 38–53 lm fraction of the exposure and in the presence of moisture, partial or total
aged Aitik concentrate the effect of initial copper at both low hydrolysis of iron–oxyhydroxides results in formation of basic
(420 mV) and high (620 mV) redox potentials was investigated. It ferric–sulphate (Brion, 1980; Parker, 2008). It is reasonable to
was again observed that the leaching rates were higher at high assume that this process would occur in sulphate leaching systems
redox potential experiments (see Section 3.2). At low redox at an accelerated rate. The presence of iron–oxyhydroxides and
potential a consistent trend versus variation of copper concentra- basic iron–sulphates has been reported in leaching investigations
tions was not observed (Fig. 11). At high redox, it was still difficult in sulphate systems and atmospheric pressure (Farquhar et al.,
to find a relation, but generally it seems that the recoveries were 2003; Khoshkhoo et al., 2014a; Parker, 2008). They are also
slightly lower in presence of copper (27% vs. 35% after 72 h). The observed on surface of samples leached in more acidic conditions
reason for this behaviour could have been due to the aged surfaces (Hackl et al., 1995) and in chloride media (Acres et al., 2010; Li
of the concentrate which might have affected the consistency of et al., 2010). Formation of iron-oxyhydroxides on the surface leads
to formation of a phase rich in sulphur and copper below the
oxidised surface, typically regarded as polysulphide or metal-
deficient sulphide. Some investigations considered this phase as
the cause for hindered dissolution (Hackl et al., 1995; Harmer
et al., 2006) while others suggested that this phase can only be a
transient intermediate and has no role in hindering the dissolution
(Holmes and Crundwell, 2013; Klauber, 2008; Parker et al., 2003).
Results from the current investigation are inconclusive for propos-
ing a definitive cause for hindered dissolution. However, the
suggestion that iron–oxyhydroxides could be the precursor for
passivation or hindered dissolution is supported by experiments
showing lower leaching rates in aged concentrates from the outset
even when the conditions that fresh chalcopyrite would react
rapidly were used. The hypothesis that iron-oxyhydroxides trigger
hindered dissolution is also supported by XPS studies in our previ-
ous investigation where the presence of iron-oxyhydroxides
(mixed with elemental sulphur) was related to the slow dissolu-
tion (Khoshkhoo et al., 2014a). It was found that elemental sulphur
Fig. 10. Effect of initial copper concentration in leaching of freshly ground was rigidly bound to the surface such that the XPS measurements
Kristineberg concentrate. Leaching conditions: redox potential = 420 mV, at room temperature, where sublimation of sulphur must occur
T = 80 °C, pH = 1.4–1.5, initial iron = 1 g/l. under ultra-high vacuum in the XPS analysing chamber, did not
result in removal of sulphur from the surface. It can be proposed
that as the dissolution of chalcopyrite in sulphate media is
initiated, elemental sulphur is gradually trapped in already and
newly formed iron–oxyhydroxides (and subsequently basic iron-
sulphates) resulting in a multicomponent layer which hinders
the dissolution.
If this is the case, then it seems that hindered dissolution
cannot be avoided in conventional sulphate leaching systems,
especially in bioleaching systems at high redox potentials and
extended leaching durations where iron–oxyhydroxides would
have enough time to form and become consolidated on the
surface. One possible solution would be to leach under suitable
conditions such that chalcopyrite dissolution reaches an accept-
able level before its surface is extensively covered. Suitable
conditions are fresh surfaces, temperature >80 °C, low redox
potential and presence of pyrite. Such an experiment is shown
in Fig. 3 (freshly ground Kristineberg leached at 420 mV) where
the recovery reached above 70% in eight hours, after which it
started to be hindered. Dixon et al. (2008) and others have shown
Fig. 11. Effect of initial copper concentration in leaching of aged Aitik concentrate
this can be further improved with higher amounts of pyrite and
(38–53 lm). Legend: First number shows the redox potential (mV) and the second
number is the initial copper concentration (g/l). Other leaching conditions:
probably by higher solid contents to increase the collisions
T = 80 °C, pH = 1.4–1.5, initial iron = 1 g/l. between pyrite and chalcopyrite particles.
16 M. Khoshkhoo et al. / Minerals Engineering 100 (2017) 9–16

4. Conclusions Dreisinger, D., 2006. Copper leaching from primary sulfides: options for biological
and chemical extraction of copper. Hydrometallurgy 83, 10–20. http://dx.doi.
org/10.1016/j.hydromet.2006.03.032.
Controlled redox potential leaching experiments showed that Dutrizac, J.E., MacDonald, R.J.C., 1973. The effect of some impurities on the rate of
the reductive leaching mechanism would be beneficial in the chalcopyrite dissolution. Can. Metall. Quart. 12, 409–420.
Farquhar, M.L., Wincott, P.L., Wogelius, R.A., Vaughan, D.J., 2003. Electrochemical
presence of an active galvanic interaction. Experiments using a
oxidation of the chalcopyrite surface: an XPS and AFM study in solution at pH 4.
pyritic concentrate resulted in higher recoveries at low redox Appl. Surf. Sci. 218, 34–43.
potential while the recoveries were similar at low and high redox Gericke, M., Govender, Y., Pinches, A., 2010. Tank bioleaching of low-grade
chalcopyrite concentrates using redox control. Hydrometallurgy 104, 414–
potentials using a relatively pure concentrate in absence of pyrite.
419. http://dx.doi.org/10.1016/j.hydromet.2010.02.024.
This behaviour was shown to be in better agreement with the Habashi, F., 1999. Kinetics of Metallurgical Processes. Metallurgie Extractive
alternative reductive leaching model that suggests chalcopyrite Quebec, Quebec City.
surfaces undergo a first step of partial oxidation before the reduc- Hackl, R.P., Dreisinger, D.B., Peters, E., King, J.A., 1995. Passivation of chalcopyrite
during oxidative leaching in sulfate media. Hydrometallurgy 39 (25–48), 25.
tion reactions occur. This model accounts for the effect of pyrite in http://dx.doi.org/10.1016/0304-386X(95)00023-A.
the reductive leaching mechanism. Harmer, S.L., Thomas, J.E., Fornasiero, D., Gerson, A.R., 2006. The evolution of surface
It was also shown that the reductive leaching was not valid at layers formed during chalcopyrite leaching. Geochim. Cosmochim. Acta 70,
4392–4402. http://dx.doi.org/10.1016/j.gca.2006.06.1555.
all for aged concentrates. Aged concentrates (pyritic and pure) gave Hiroyoshi, N., Miki, H., Hirajima, T., Tsunekawa, M., 2000. A model for ferrous-
steadily increasing recoveries with increased redox potential. The promoted chalcopyrite leaching. Hydrometallurgy 57, 31–38. http://dx.doi.org/
air oxidation products, mainly iron-oxyhydroxide, on the surfaces 10.1016/S0304-386X(00)00089-X.
Holmes, P.R., Crundwell, F.K., 2013. Polysulfides do not cause passivation: results
of the minerals were found to cause such behaviour. Chalcopyrite from the dissolution of pyrite and implications for other sulfide minerals.
ageing can have drastic effects in heap leaching processes where Hydrometallurgy 139, 101–110. http://dx.doi.org/10.1016/j.hydromet.2013.
the ore is exposed for extended periods of time. The wide and 07.006.
Kametani, H., Aoki, A., 1985. Effect of suspension potential on the oxidation rate of
inconsistent chalcopyrite leaching results reported in the literature
copper concentrate in a sulfuric acid solution. Metall. Mater. Trans. B 16, 695–
could also be partly due to using of aged samples. The effect of ini- 705. http://dx.doi.org/10.1007/BF02667506.
tial copper concentration had no influential effect in the leaching Khoshkhoo, M., Dopson, M., Shchukarev, A., Sandström, Å., 2014a. Chalcopyrite
leaching and bioleaching: an X-ray photoelectron spectroscopic (XPS)
rates for possible industrial processes.
investigation on the nature of hindered dissolution. Hydrometallurgy 149,
It was also proposed that iron–oxyhydroxides are the precursor 220–227. http://dx.doi.org/10.1016/j.hydromet.2014.08.012.
of hindered dissolution in sulphate media where their formation Khoshkhoo, M., Dopson, M., Shchukarev, A., Sandström, Å., 2014b. Electrochemical
entraps surface elemental sulphur and other surface species result- simulation of redox potential development in bioleaching of a pyritic
chalcopyrite concentrate. Hydrometallurgy 144–145, 7–14. http://dx.doi.org/
ing to an inevitable formation of a consolidated phase on the 10.1016/j.hydromet.2013.12.003.
surface. It was shown that when suitable conditions are met, high Klauber, C., 2008. A critical review of the surface chemistry of acidic ferric sulphate
copper recoveries can be obtained before the surface is extensively dissolution of chalcopyrite with regards to hindered dissolution. Int. J. Miner.
Process. 86, 1–17. http://dx.doi.org/10.106/j.minpro.2007.09.003.
hindered. Klauber, C., Parker, A., Bronswijk, W.v., Watling, H., 2001. Sulphur speciation of
leached chalcopyrite surfaces as determined by X-ray photoelectron
Acknowledgements spectroscopy. Int. J. Miner. Process. 62 (65–94), 65. http://dx.doi.org/10.1016/
S0301-7516(00)00045-4.
Koleini, S.M.J., Aghazadeh, V., Sandström, A., 2011. Acidic sulphate leaching of
The authors wish to express their gratitude for financial support chalcopyrite concentrates in presence of pyrite. Miner. Eng. 24, 381–386. http://
from the Swedish Research Council Formas (contract number 229- dx.doi.org/10.1016/j.mineng.2010.11.008.
Li, Y., Kawashima, N., Li, J., Chandra, A.P., Gerson, A.R., 2013. A review of the
2009-657). Boliden Mineral AB is highly appreciated for providing
structure, and fundamental mechanisms and kinetics of the leaching of
financial support and mineral samples. Dr. Roland Hostettler is also chalcopyrite. Adv. Colloid Interface Sci. 197–198, 1–32. http://dx.doi.org/
gratefully acknowledged for his invaluable assistance in improving 10.1016/j.cis.2013.03.004.
Li, J., Kawashima, N., Kaplun, K., Absolon, V.J., Gerson, A.R., 2010. Chalcopyrite
the redox potential control program.
leaching: the rate controlling factors. Geochim. Cosmochim. Acta 74, 2881–
2893. http://dx.doi.org/10.1016/j.gca.2010.02.029.
References Marsden, J., 2007. Medium-temperature pressure leaching of copper concentrates -
Part 1: chemistry and initial process development. Miner. Metall. Process. 24,
Acres, R., Harmer, S., Beattie, D., 2010. Synchrotron XPS studies of solution exposed 193–204.
chalcopyrite, bornite, and heterogeneous chalcopyrite with bornite. Int. J. Mehta, A.P., Murr, L.E., 1983. Fundamental studies of the contribution of galvanic
Miner. Process. 94, 43–51. http://dx.doi.org/10.1016/j.minpro.2009.11.006. interaction to acid-bacterial leaching of mixed metal sulfides. Hydrometallurgy
Ahmadi, A., Schaffie, M., Petersen, J., Schippers, A., Ranjbar, M., 2011. Conventional 9, 235–256. http://dx.doi.org/10.1016/0304-386X(83)90025-7.
and electrochemical bioleaching of chalcopyrite concentrates by moderately Mikhlin, Y.L., Tomashevich, Y.V., Asanov, I.P., Okotrub, A.V., Varnek, V.A., Vyalikh, D.
thermophilic bacteria at high pulp density. Hydrometallurgy 106, 84–92. http:// V., 2004. Spectroscopic and electrochemical characterization of the surface
dx.doi.org/10.1016/j.hydromet.2010.12.007. layers of chalcopyrite (CuFeS2) reacted in acidic solutions. Appl. Surf. Sci. 225,
Brion, D., 1980. Etude par spectroscopie de photoelectrons de la degradation 395–409. http://dx.doi.org/10.1016/j.apsusc.2003.10.030.
superficielle de FeS2, CuFeS2, ZnS et PbS a l’air et dans l’eau. Appl. Surf. Sci. 5, Parker, A., Klauber, C., Kougianous, A., Watling, H.R., van Bronswijk, W., 2003. An X-
133–152. http://dx.doi.org/10.1016/0378-5963(80)90148-8. ray photoelectron spectroscopy study of the mechanism of oxidative
Buckley, A., Woods, R., 1984. An X-ray photoelectron spectroscopic study of the dissolution of chalcopyrite. Hydrometallurgy 71, 265–276. http://dx.doi.org/
oxidation of chalcopyrite. Aust. J. Chem. 37, 2403. 10.1016/S0304-386X(03)00165-8.
Córdoba, E.M., Muñoz, J.A., Blázquez, M.L., González, F., Ballester, A., 2008. Leaching Parker, A., 2008. Oxidative Dissolution of Chalcopyrite in Ferric Media: An X-ray
of chalcopyrite with ferric ion. Part II: effect of redox potential. Photoelectron Spectroscopy Study PhD dissertation [PhD Thesis]. Curtin
Hydrometallurgy 93, 88–96. http://dx.doi.org/10.1016/j.hydromet.2008.04.016. University of Technology.
Debernardi, G., Carlesi, C., 2013. Chemical-electrochemical approaches to the study Petersen, J., Dixon, D.G., 2006. Competitive bioleaching of pyrite and chalcopyrite.
passivation of chalcopyrite. Miner. Process. Extr. Metall. Rev. 34, 10–41. http:// Hydrometallurgy 83, 40–49. http://dx.doi.org/10.1016/j.hydromet.2006.03.036.
dx.doi.org/10.1080/08827508.2011.623745. Sandström, Å., Shchukarev, A., Paul, J., 2005. XPS characterisation of chalcopyrite
Dixon, D.G., Mayne, D.D., Baxter, K.G., 2008. Galvanox – a novel galvanically- chemically and bio-leached at high and low redox potential. Miner. Eng. 18,
assisted atmospheric leaching technology for copper concentrate. Can. Metall. 505–515. http://dx.doi.org/10.1016/j.mineng.2004.08.004.
Quart. 47, 327–336. http://dx.doi.org/10.1179/000844308794408317.