Minerals Engineering 58 (2014) 6472

Contents lists available at ScienceDirect

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

The separation of chalcopyrite and chalcocite from pyrite in cleaner

otation after regrinding
Xumeng Chen a, Yongjun Peng a,b,, Dee Bradshaw a
a
b

Julius Kruttschnitt Mineral Research Centre, The University of Queensland, Isles Road, Indooroopilly, Brisbane, QLD 4068, Australia
School of Chemical Engineering, The University of Queensland, St. Lucia, Brisbane, QLD 4072, Australia

a r t i c l e

i n f o

Article history:
Received 7 October 2013
Accepted 9 January 2014
Available online 3 February 2014
Keywords:
Regrinding
Copper minerals
Pyrite activation
Surface oxidation
Galvanic interaction

a b s t r a c t
In this study, the differences between the separation of chalcopyrite and chalcocite from pyrite in cleaner
otation after regrinding were investigated. In the rougher otation prior to regrinding, high chalcopyrite
and chalcocite recovery were obtained in conjunction with high pyrite otation recovery due to the activation of pyrite by copper ions during primary grinding. The rougher otation concentrate was reground
in a rod mill before cleaner otation. It was found that chalcopyrite and chalcocite exhibited different otation behavior and also affected pyrite otation differently in cleaner otation. The mechanism underpinning these phenomena was investigated by a range of techniques including the polarization of mineral
electrodes, X-ray photoelectron spectroscopy (XPS) analyses and ethylene diamine tetraacetic acid
(EDTA) extraction. It was found that the otation behavior of both copper minerals and their effect on
pyrite otation after regrinding were governed by their electrochemical activities and galvanic coupling
with pyrite.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
Regrinding rougher otation concentrates is becoming a
common practice to improve the recovery and grade of valuable
minerals as a result of the need to process low grade and complex
ores. The small grain size of the valuable minerals in low grade ores
requires much ner grinding to provide sufcient liberation. In
some mineral processing plants, the desirable liberation of valuable minerals in low grade ores can only be achieved by regrinding
particles to less than 10 lm (Johnson, 2006). Further, the mineralogy of low grade ores is becoming more complex, therefore, the
separation of valuable minerals from gangue minerals in downstream processes (e.g., otation and leaching) is more challenging
even though minerals are well liberated.
Regrinding not only reduces particle sizes and liberates valuable
minerals from gangue minerals but also provides appropriate surfaces for subsequent otation. Grinding conditions, including the
use of different types of grinding media and pulp chemistry (pH,
Eh and dissolved oxygen), have a signicant effect on mineral otation, and this has been well studied and reviewed (Peng et al.,
2003a, 2003b; Bradshaw et al., 2006; Bruckard et al., 2011). It
has also been found that the otation of ne and ultra-ne particles is more affected by grinding conditions than the otation of
Corresponding author at: School of Chemical Engineering, The University of
Queensland, St. Lucia, Brisbane, QLD 4072, Australia. Tel.: +61 7 3365 7156; fax: +61
7 3365 3888.
E-mail address: yongjun.peng@uq.edu.au (Y. Peng).
0892-6875/$ - see front matter 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.mineng.2014.01.010

intermediate and coarse particles (Frew et al., 1994; Johnson,

2006; Grano, 2009; Peng and Grano, 2010). During regrinding,
most particles are reduced to ne and ultra-ne fractions. It is
therefore anticipated that regrinding conditions have a more signicant effect on the subsequent mineral otation than primary
grinding conditions.
Copper sulde minerals yield most of the copper produced
throughout the world, and are often associated with precious minerals. Due to different formation conditions, their compositions
vary greatly. Among them, chalcopyrite (CuFeS2) and chalcocite
(Cu2S) are the widely oated copper sulde minerals. Chalcopyrite
is usually formed in the primary hydrothermal zone. Chalcocite
can also be formed as part of primary mineralization (e.g. iron
oxide hosted copper-gold deposit), but more frequently as a secondary enrichment mineral (Anthony et al., 1990). With the depletion of chalcopyrite-dominant copper ores, more and more
chalcocite-dominant copper ores with a high pyritic content are
treated. There is a general industrial observation that chalcocitedominant copper ores are more difcult to treat in particular in
cleaner otation after regrinding (Wightman et al., 2011).
The oxidation of copper minerals plays a critical role in copper
otation and also the separation from pyrite. It is well known that
chalcopyrite can be recovered by otation in the absence of collectors under oxidizing conditions, but not under reducing conditions
(Heyes and Trahar, 1977). This is because mild oxidation results in
the formation of a hydrophobic sulfur-rich surface, due to the dissolution of iron ions (Gardner and Woods, 1979; Buckley and

65

X. Chen et al. / Minerals Engineering 58 (2014) 6472

Woods, 1984; Zachwieja et al., 1989). However, extensive oxidation produces a high amount of metal and high valency sulfur oxidation species which are hydrophilic and depress the otation.
Compared to chalcopyrite, chalcocite shows weak collectorless otation, but can be well oated in the presence of xanthate collectors within a certain Eh range (Heyes and Trahar, 1979). The
oxidation of chalcocite has been found to decrease its oatability
due to the produced copper hydroxide layer on the surface and a
decreased number of copper sulde sites for collector adsorption.
However, its otation can be restored by increasing the collector
dosage (Tolley et al., 1996). It is documented that chalcocite is
more reactive than chalcopyrite. Fullston et al. (1999) studied the
oxidation of different copper minerals using zeta potential
measurements, and found that the oxidation of these minerals follows the order: chalcocite > tennantite > enargite > bornite > covellite > chalcopyrite. It is anticipated that chalcocite otation is more
susceptible to an oxidizing grinding condition than chalcopyrite
otation.
Pyrite depression in the cleaner stage plays a critical role in producing a high quality copper otation product. In copper otation
plants, the current strategy is to maximize the recovery of copper
minerals by oating some pyrite at the rougher stage while further
rejecting it at the cleaner stage after regrinding (Wills and
Napier-Munn, 2006). However, despite the minerals being well
liberated by regrinding, the depression of pyrite can be difcult if
pyrite surfaces are activated by copper ions dissolved from copper
minerals. During regrinding, more copper ions can be released
from copper minerals and activate the exposed fresh pyrite
surfaces (Chen et al., 2012, 2013). A number of studies have shown
that more copper ions can be dissolved from chalcocite than from
chalcopyrite (Lascelles and Finch, 2002; Wong et al., 2002). It is
expected that chalcopyrite and chalcocite provide a different
extent of copper activation on pyrite during regrinding, and
therefore pyrite otation.
Moreover, galvanic interactions occur between the copper mineral and pyrite due to a difference in their rest potential resulting
in different electrochemical reactions on both mineral surfaces
and different oatability (Nakazawa and Iwasaki, 1985; Yelloji
Rao and Natarajan, 1989; Ekmeki and Demirel, 1997). It has been
reported by Huang and Grano (2005) that the galvanic current was
increased as the particle size reduced. As a result, the galvanic
interaction is stronger during regrinding than during primary
grinding with a greater effect on mineral otation and separation.
The objective of the current study is to understand the different
separation of chalcopyrite and chalcocite from pyrite in the cleaner
otation after regrinding. Previous studies emphasized primary
grinding and rougher otation, and limited studies have been
undertaken to understand the effect of regrinding on cleaner otation which is the subject of the current study.

2. Experimental details
2.1. Materials and reagents
Chalcopyrite, chalcocite and pyrite single minerals, supplied by
GEO Discoveries, were crushed through a jaw crusher and a roll
crusher, and then screened to collect +0.713.35 mm particle size
fractions. XRD analysis indicated that the pyrite sample was very
pure without detecting any impurity. The chalcocite sample also
had a high purity with a minor amount (<2%) of iron sulde impurity. The purity of the chalcopyrite sample was slightly lower with
about 95% chalcopyrite and 5% pyrite. To be consistent with previous studies, the processed samples were sealed in polyethylene
bags and then stored in a freezer at a temperature of 20 C to
avoid further surface oxidation (Chen et al., 2012, 2013).

Potassium amyl xanthate (PAX) and Interfroth 56 were used as

a collector and frother, respectively. Collector and frother were of
industry grade and used as received. Other chemicals were of AR
grade. The pH was adjusted by the addition of a NaOH solution instead of lime used at plants to eliminate the effect of Ca2+ on otation and surface analyses by XPS (Hu et al., 2000). De-ionized water
was used in all experiments. All chemical solutions were made
fresh daily.
2.2. Grinding, regrinding and otation
This study focused on regrinding of rougher otation concentrates with high sulde concentrations and therefore pure sulde
minerals were used. Crushed chalcopyrite or chalcocite single mineral (100 g), or the mixture of chalcopyrite or chalcocite (50 g) and
pyrite (50 g) was combined with 150 g deionized water, and
ground in a stainless steel rod mill using 4 stainless steel rods
(3750 g) to achieve a P80 = 75 lm determined by traditional sieving. A 2.5% sodium hydroxide solution was added during grinding
to achieve pH 9.0 in a primary grinding discharge.
After grinding, the pulp was transferred to a 1.5 dm3 steel otation cell and conditioned with potassium amyl xanthate (160 g/t)
and Interfroth 56 (200 g/t). 2 min was allowed for the conditioning
of each reagent. During otation, pH was xed at 9.0 by adding a
sodium hydroxide solution. Flotation concentrate was collected
every 10 s, and four concentrates were collected after cumulative
times of 0.5, 2.0, 4.0 and 8.0 min. The air ow rate was 1.5 dm3/
min during the rst 0.5 min otation, and then increased to
3.0 dm3/min from the second concentrate.
The four otation concentrates were combined and reground in
a rod mill with stainless steel grinding media. The total weight of
grinding media was about 10.3 kg. Particle size P80 = 20 lm was
achieved by adjusting the grinding time. The size distribution
was determined by Laser Diffraction with a Malvern MasterSizer
(Malvern Instrument Ltd., U.K.) in line with the laboratory and
plant practice where a laser sizer has been routinely used to control the grind size smaller than 38 lm. A sodium hydroxide solution was added to achieve pH 9.0 in the regrind discharge. The
specic surface area of regrinding feeds and products was measured by the BET method and is shown in Table 1.
After regrinding, the pulp was transferred to a 1.5 dm3 otation
cell. The solid percentage was slightly lower in cleaner otation
than in rougher otation with more than 90% recovery was
achieved in rougher otation. In cleaner otation, different dosages
of collector were added in otation while 200 g/t Interfroth was
added. The operational procedure was the same as used in rougher
otation after primary grinding.
2.3. Surface analysis
Samples for surface analyses (about 10 ml of slurry) were
collected from mill discharges and also otation products. These
samples were frozen in liquid nitrogen to avoid surface oxidation.
2.3.1. EDTA extraction
EDTA extraction was used in this study to determine the
amount of oxidation species from minerals and grinding media. A
Table 1
Specic surface area of regrinding feeds and products (m2/g).
Sample

Regrinding feed

Regrinding product

Chalcopyrite
Chalcocite
Pyrite/chalcopyrite
Pyrite/chalcocite

0.686
0.192
0.760
0.144

1.392
1.317
1.340
1.272

X. Chen et al. / Minerals Engineering 58 (2014) 6472

3 wt% solution of AR grade ethylene diamine-tetraacetic acid disodium salt (EDTA) was made up and the pH was adjusted to 7.5 by
adding sodium hydroxide solution. 95 cm3 of the EDTA solution
was placed in a vigorously-stirred reaction vessel and continuously
purged with nitrogen for 10 min to eliminate oxygen in the solution. Then the sample was added to the EDTA solution, and leached
by EDTA for 10 min while nitrogen was continuously purged
throughout. The slurry was then ltered through a 0.45 lm Millipore lter. Filtrate and dry solids were both weighed and analyzed
by ICP. The sample dry weight was used to calculate the mass of
metal oxidation species per unit mass of solid or unit area of solid
surface.
2.3.2. XPS analysis
XPS measurements were carried out with a KRATOS Axis Ultra
(Kratos Analytical, Manchester, United Kingdom) with a monochromatic Al X-ray source operating at 15 kV and 10 mA
(150 W). The analysis spot size was 300  700 lm. The frozen slurry samples were defrosted just prior to the analysis. The solids
were placed on the stainless steel bar and immediately loaded into
the introduction chamber of the spectrometer. Samples were analyzed at a pressure of 9  1010 Torr at the room temperature. Each
analysis started with a survey scan from 0 to 1200 eV using a pass
energy of 160 eV at steps of 1 eV with 1 sweep. High resolution
spectra of Fe 2p, O 1s, C 1s, S 2p, and Cu 2p were collected at
20 eV pass energy at steps of 100 meV with 2 or 3 sweeps. Binding
energies were charge-corrected by referencing to adventitious
carbon at 284.8 eV.
2.4. Polarization of mineral electrodes
Carefully selected high-purity chalcopyrite and chalcocite were
cut by using an Isomet4000 Precision Saw into 3 mm  3 mm 
10 mm dimension. The electrodes were prepared by attaching
the mineral specimen to copper wires with a conductive silver
based electric conductive adhesive (RS Component Pty Limited),
and then mounting them into non-conductive epoxy resin with
the 3 mm  3 mm side exposed. Before each experiment, the mineral electrode was treated with a grade 600 then a grade 1200 silicon carbide paper, further polished with grit 100,000 diamond
paste (0.25 lm in diameter), rinsed off the diamond using sonic
bath and then thoroughly washed with de-ionized water. The
polarization experiments were conducted by using a standard
three-electrode system composed of the chalcopyrite or chalcocite
mineral electrode, a platinum counter electrode with a surface area
of 31.4 mm2 and a saturated calomel reference electrode (SCE). The
three-electrode system was linked to a Voltalab potentiostat/galvanostat (Model PGZ-100) controlled by a PC with Voltamaster 4
software. The polarization experiments were conducted in 0.1 M
KCl solution at pH 9.0. The potential sweep was positively directed
and at the rate of 10 mV s1. Before each experiment, the open circuit potential (OCP) was recorded for 300 s. The potential recorded
was against SCE.
3. Results and discussion

corrosion potential (or open circuit potential, OCP) of chalcopyrite

and chalcocite is about 20 mV and 90 mV, respectively, in the
same aqueous solution. As a result, chalcocite is more electrochemically active than chalcopyrite. During polarization, the anodic corrosion process is controlled by the dissolution (or oxidation) of the
mineral, while the reduction of dissolved oxygen is the main cathodic process. On the two mineral electrodes, the cathodic reaction,
oxygen reduction is the same:

O2 2H2 O 4e ! 4OH

However, to be the anodic reaction, mineral oxidation is different. The anodic oxidation of chalcopyrite is (Buckley and Woods,
1984):

CuFeS2 3xOH ! CuFe1x S2 xFeOH3 3xe

While the anodic oxidation of chalcocite is (Steger and Desjardins,

1980):

Cu2 S 2xOH ! Cu2x S xCuO xH2 O 2xe

Obviously, reaction (3) is stronger than reaction (2) since the

anodic current density (the right part of the polarization curve at
more positive potentials from the OCP) on chalcocite was much
higher than that on chalcopyrite. The corrosion current density obtained by extrapolating the linear Tafel regions of the polarization
curve to the corrosion potential is an indication of the corrosion
rate (Baboian, 2005). The corrosion current density of chalcopyrite
and chalcocite was about 106.3 and 104.8 A cm2, respectively.
Obviously the corrosion rate of chalcocite was greater than the corrosion rate of chalcopyrite based on the polarization curves. In
other words, the oxidation rate of chalcocite was greater.
During regrinding of the mixed mineral system, a galvanic
interaction occurs between different minerals due to their different
rest potential, which can cause a net current through the surface,
and signicantly inuence the reduction of cathode and oxidation
of anode (Martin et al., 1991). It is generally agreed that pyrite is
more electrochemically noble than chalcopyrite and chalcocite
with higher rest potential as indicated by Chmielewski and Kaleta
(2011). Moreover, the rest potential difference between chalcopyrite and pyrite is smaller than that between chalcocite and pyrite,
with a weaker galvanic interaction. This may affect mineral otation differently, which was further investigated.

Log[i] (A/cm 2 )

66

3.1. Electrochemical properties of chalcopyrite and chalcocite

Chalcocite

The electrochemical property and surface oxidation of copper

sulde minerals play a critical role in their otation, and also inuence the oatability of other minerals (e.g., pyrite) through galvanic interactions and copper activation. The electrochemical
property of chalcopyrite and chalcocite was rstly studied by the
polarization of mineral electrodes. Fig. 1 shows the polarization
curves of chalcopyrite and chalcocite mineral electrodes. The

Chalcopyrite

-0.3

-0.2

-0.1

0.1

0.2

0.3

Applied potential (V, SCE)

Fig. 1. The polarization curves of chalcopyrite and chalcocite electrodes.

X. Chen et al. / Minerals Engineering 58 (2014) 6472

67

3.2. Flotation of chalcopyrite and chalcocite

Before studying the separation of chalcopyrite and chalcocite
from pyrite, the oatability of single copper minerals in the absence of pyrite was investigated. In order to simulate normal otation circuits in plants, in this study, chalcopyrite and chalcocite
recovery were maximized in rougher otation by using stainless
steel media in primary grinding and adding collector in the otation stage. The rougher otation concentrate was reground from
75 lm to 20 lm, and then subjected to cleaner otation.
3.2.1. Flotation of chalcopyrite
As shown in Fig. 2, about 91% chalcopyrite recovery in 8 min of
otation was achieved in the rougher stage. The rougher otation
concentrate was reground from 75 lm to 20 lm. Chalcopyrite otation after regrinding is shown in Fig. 3. Without additional collector, 71% recovery in 8 min of otation was achieved. As the particle
size was reduced, existing surfaces were carried from rougher otation while fresh surfaces were generated as well. From the BET
surface area measurement as shown in Table 1, the specic surface
area was increased from 0.686 m2/g to 1.392 m2/g after regrinding.
About half of the surface existing before regrinding was retained
along with adsorbed collector from the rougher stage, contributing
to a high chalcopyrite recovery in cleaner otation even without
additional collector. As shown in Fig. 3, chalcopyrite recovery
was increased after the addition of extra collector. Maximum
recovery of 95% was achieved after the addition of 160 g/t collector. Further addition of collector decreased chalcopyrite recovery.
During grinding and regrinding in an oxidizing environment
with stainless steel media, the chalcopyrite surface should have
been oxidized to some extent, which affects its otation. To detect chalcopyrite oxidation species, the regrinding feed and product were examined by XPS. The S 2p XPS spectra from mineral
surfaces before and after regrinding are shown in Fig. 4. These
spectra were tted using the 2p1/2 and 2p3/2 doublet with a xed
1:2 intensity ratio and 1.18 eV energy separation. The doublet
with the S 2p3/2 binding energy at 161.1 eV is attributed to S2
from CuFeS2. The doublet at 162.1 eV is due to S2
2 from the metal-decient sulde (CuS2) caused by chalcopyrite oxidation. The
doublet at 162.7 eV is attributed to the formation of polysulde
(Buckley and Woods, 1984; Smart et al., 1999). An energy loss
(EL) feature was tted at approximately 164.1 eV associated with
S 3p to Fe 3d excitation (Harmer et al., 2004; Acres et al., 2010).
The quantication of these sulfur species from regrinding feed
and product is summarized in Table 2. The percentage of monosulde was increased after regrinding due to the generated fresh
surface. Although some polysulde was present in both

Fig. 2. Chalcopyrite recovery as a function of otation time in rougher otation

after primary grinding.

Fig. 3. Chalcopyrite recovery in cleaner otation as a function of collector dosage

after regrinding (P80 = 20 lm).

Fig. 4. S 2p XPS spectra from mineral surfaces of regrinding feed and product of
chalcopyrite mineral.

regrinding feed and product, no sulfate product was observed,

which suggests that chalcopyrite surfaces in both regrinding feed
and product were only mildly oxidized. The metal-decient
sulde and polysulde are hydrophobic, and contribute to the
collectorless otation of chalcopyrite (Gardner and Woods,
1979; Buckley and Woods, 1984; Zachwieja et al., 1989).
EDTA extraction of chalcopyrite after primary grinding and
regrinding was also conducted to study the surface oxidation of
chalcopyrite by measuring copper and iron ion production. As
shown in Table 3, the amount of EDTA extractable iron and copper
ions from chalcopyrite is quite low. The amount of extracted ions is
slightly higher in the regrinding product than in the regrinding
feed due to the further oxidation of generated fresh surfaces.
In summary, mild oxidation occurred on the chalcopyrite surface during regrinding. Metal-decient and polysulde oxidation
products, together with collector carried from rougher otation,
provided surface hydrophobicity, contributing to high chalcopyrite recovery in cleaner otation even without additional
collector. A great amount of fresh surfaces still existed after
regrinding, requiring more collector for further improving the
recovery.

68

X. Chen et al. / Minerals Engineering 58 (2014) 6472

Table 2
S 2p quantication for the regrinding feed and product of chalcopyrite mineral.
Sample

Species

2p3/2 position

Area contribution (%)

Regrinding feed

Monosulde
Disulde
Polysulde
Energy loss

161.1
162.1
162.7
164.1

53.8
18.6
21.1
6.5

Regrinding product

Monosulde
Disulde
Polysulde
Energy loss

161.1
162.1
162.7
164.1

63.0
12.7
18.1
6.2

Table 3
EDTA extractable iron and copper ions of the regrinding feed and product of the
chalcopyrite mineral.
Chalcopyrite

Fe (mole/m2 mineral)
Cu (mole/m2 mineral)

Feed

Product

4.46  106
7.66  106

5.18  106
9.69  106

3.2.2. Flotation of chalcocite

The otation of chalcocite without mixing with pyrite was also
studied. As shown in Fig. 5, in the rougher stage, 92% recovery was
achieved after 8 min otation. The rougher otation concentrate
was reground and oated in the cleaner stage. Chalcocite recovery
as a function of collector dosage is shown in Fig. 6. 59% recovery
was achieved without additional collector in cleaner otation.
After adding extra collector, the recovery was further improved.
At 160 g/t collector, the recovery was about 82% and increased only
slightly with a further addition of collector. Chalcocite recovery
was lower than chalcopyrite recovery in the cleaner otation stage,
which may be associated with the more electrochemically active
nature of chalcocite.
To detect the species on the chalcocite surface, regrinding feed
and product of chalcocite were examined by XPS. The S 2p XPS
spectra from mineral surfaces before and after regrinding are
shown in Fig. 7. The doublet at 161.8 eV is assigned to S2 from
Cu2S (Buckley and Woods, 1984). The doublet at 162.8 eV is attributed to the formation of polysulde (Smart et al., 1999). The broad
peak observed at 167.7 eV is attributed to the sulfate SO2
4 . An energy loss (EL) peak was also tted at 164.0 eV. The quantication of
these suphur species from regrinding feed and product is summarized in Table 4. In the regrinding feed, 75.8 at.% of S is present as
S2, while, in the regrinding product, 79.5 at.% is present as

Fig. 6. Chalcocite recovery in cleaner otation as a function of collector dosage after

regrinding.

Fig. 7. S 2p XPS spectra from mineral surfaces of regrinding feed and product of
chalcocite mineral.

Table 4
S 2p quantication for the regrinding feed and product of chalcocite mineral.

Fig. 5. Chalcocite recovery as a function of otation time in rougher otation after

primary grinding.

Sample

Species

2p3/2 position

Area contribution (%)

Regrinding feed

Monosulde
polysulde
Energy loss
Sulfate

161.8
162.8
164.0
167.6

75.8
6.0
4.9
13.2

Regrinding product

Monosulde
polysulde
Energy loss
Sulfate

161.9
162.9
164.1
167.9

79.5
4.1
5.0
11.4

S2. Therefore, the composition of S does not change signicantly

after regrinding in the chalcocite single mineral system. Sulfate
species are present in both regrinding feed and product. It is hydrophilic and decreases chalcocite oatability.

X. Chen et al. / Minerals Engineering 58 (2014) 6472

69

Table 5
EDTA extractable iron and copper of the grinding and regrinding product of the
chalcocite single mineral.
Chalcocite

Fe (mole/m2 mineral)
Cu (mole/m2 mineral)

Feed

Product

8.93  106
3.19  104

3.57  106
1.52  104

EDTA extraction of chalcocite after primary grinding (P80,

75 lm) and regrinding (P80, 20 lm) was also conducted to quantify
metal oxidation products. As shown in Table 5, chalcocite produced
much more extractable Cu than chalcopyrite. This is in line with
the polarization study, showing that chalcocite is more active than
chalcopyrite. The extracted Cu ion should be in the form of
Cu-hydroxide at pH 9 (Finkelstein, 1997), which is hydrophilic
and decreases the chalcocite otation. The amount of extracted
Cu ions is lower in the regrinding product than in the regrinding
feed, due to a greater surface area generated after regrinding.
It is evident that stronger oxidation occurred during both primary grinding and regrinding of chalcocite compared to primary
grinding and regrinding of chalcopyrite because of the presence
of sulfate species and a higher amount of copper oxidation species.
This explains lower chalcocite recovery in both rougher and
cleaner otation.
3.3. Separation of copper minerals from pyrite in cleaner otation
3.3.1. The separation of chalcopyrite from pyrite
The separation of chalcopyrite from pyrite was studied by using
a mixture of these two minerals. Fig. 8 shows chalcopyrite recovery
as a function of pyrite recovery in rougher otation after primary
grinding. Both chalcopyrite and pyrite displayed high oatability
with more than 90% recovery at the completion of 8 min otation.
Pyrite otation is usually depressed in alkaline solutions without
surface activation. In this system, high pyrite recovery was
achieved due to surface activation by copper ions which emanated
from copper minerals during grinding and conditioning (Leppinen,
1990; Finkelstein, 1997). This test was designed to simulate some
industry otation circuits in which essentially all of the pyrite is
recovered in rougher otation, with the pyrite being rejected in
cleaner otation.
The rougher concentrate was reground from 75 lm to 20 lm
and then subjected to cleaner otation to study the separation of
chalcopyrite from pyrite. Fig. 9 shows chalcopyrite and pyrite
recovery from cleaner otation as a function of collector

Fig. 8. Chalcopyrite recovery as a function of pyrite recovery in rougher otation

after primary grinding.

Fig. 9. Chalcopyrite and pyrite recovery from cleaner otation as a function of

collector dosage.

addition. 89% chalcopyrite recovery was achieved in 8 min otation without additional collector, which is higher than that from
the otation of single chalcopyrite shown in Fig. 3. It suggests
that the presence of pyrite enhances chalcopyrite otation. This
may be caused by the galvanic interaction and surface oxidation,
and will be further discussed. With the addition of extra collector, chalcopyrite recovery was increased slightly and then
decreased, which is similar to that in single chalcopyrite otation shown in Fig. 3. However, pyrite otation was strongly
depressed with 10% recovery in 8 min otation regardless of
the collector addition.
To understand the surface species on both chalcopyrite and pyrite, regrinding feed and product were examined by XPS. The S 2p
XPS spectra from mineral surfaces before and after regrinding are
shown in Fig. 10. The spectra were tted with only two doublets.
The peak at 161.3 is attributable to S2 from chalcopyrite or pyrite.
The other doublet at 162.5 should be due to a combination of S2
2
2
from pyrite, S2
2 from chalcopyrite, and Sn from chalcopyrite. Since
the positions of these peaks are very close, especially S2
2 from pyrite and S2
from chalcopyrite, it is difcult to further decouple
2
them precisely (Smart et al., 1999). Furthermore, a broad peak
was observed at 168.2 eV in the regrinding product, which is
attributed to sulfate SO2
4 (Buckley and Woods, 1984). The quantication of sulfur species from the regrinding feed and product is
summarized in Table 6. Again, a greater monosulde proportion
occurred in the regrinding product due to the generation of new
surfaces.
The Cu 2p XPS spectra from mineral surfaces before and after
regrinding are shown in Fig. 11. The Cu+ Cu 2p3/2 component was
identied at about 932.2 eV (McIntyre and Cook, 1975; Deroubaix
and Marcus, 1992). A second component in the Cu 2p3/2 at about
934.0 eV is attributable to Cu2+-hydroxyl species (Weisener and
Gerson, 2000). The percentages of Cu+ relative to the overall Cu
for the regrinding feed and product are 78.9% and 84.0%, respectively. The slight increase of Cu+ after regrinding is due to the generation of fresh chalcopyrite surfaces. However, for both regrinding
feed and product, Cu2+ ions (oxidation products) are the minor
copper species.
Pyrite otation was strongly depressed after mixing with
chalcopyrite in cleaner otation. As studied previously, pyrite
otation was depressed in alkaline solutions, but the addition of
copper ions signicantly increased pyrite recovery within the
pH 610 range by activating the pyrite surface (Leppinen, 1990;
Finkelstein, 1997; Zhang et al., 1997). Chen et al. (2013) demonstrated that the effect of regrinding with stainless steel media on
pyrite otation at pH 9 was mainly governed by available copper
ions and surface oxidation. As discussed in the previous section,

70

X. Chen et al. / Minerals Engineering 58 (2014) 6472

Fig. 11. Cu 2p XPS spectra from mineral surfaces on regrinding feed and regrinding
product for chalcopyritepyrite mixture.

Fig. 10. S 2p XPS spectra from mineral surfaces of regrinding feed and product for
chalcopyritepyrite mixture.

Table 6
S 2p quantication for the regrinding feed and product for chalcopyritepyrite
mixture.
Sample

Species

2p3/2
position

Area contribution
(%)

Regrinding feed

Monosulde
Disulde/
polysulde
Energy loss

161.3
162.5

24.4
68.1

164.3

7.5

Regrinding
product

Monosulde
Disulde/
polysulde
Sulfate
Energy loss

161.3
162.5

26.4
61.5

168.2
164.4

4.8
7.3

the oxidation of chalcopyrite only produced a small amount of

copper ions which may not sufciently activate pyrite in cleaner
otation. The Cu ion per surface area after primary grinding and
regrinding of chalcopyrite mixed with pyrite was estimated by
assuming that the Cu ion production by chalcopyrite is the same
with and without the addition of pyrite. The Cu ion per surface
area is 9.21  106 mole/m2 after primary grinding and signicantly decreased to 4.06  106 mole/m2 after regrinding. This
decrease in copper ions is associated with the increase in surface
area after regrinding.
Fig. 12 shows S 2p XPS spectra of otation tailing with 96% pyrite and 4% chalcopyrite. Due to the predominant pyrite composition, the XPS spectra illustrate the surface property of pyrite after
regrinding in the presence of chalcopyrite. Fig. 12 indicates a clear
SO2
4 peak at 168.2 eV, suggesting pyrite oxidation. As summarized
in Table 7, 11.1 at.% S occurs in the form of SO2
4 which limits pyrite
activation by copper ions. Pyrite oxidation also indicates the weak
galvanic interaction between chalcopyrite and pyrite with some
oxidation occurring on cathodic pyrite.
In summary, the strong depression of pyrite in the otation of
chalcopyrite mixed with pyrite in the cleaner stage may be caused
by insufcient copper ions originating from chalcopyrite and some
surface oxidation on pyrite.

Fig. 12. S 2p XPS spectra from mineral surfaces on the tailing consisting of 96%
pyrite and 4% chalcopyrite from the cleaner otation of chalcopyrite mixed with
pyrite.

3.3.2. Separation of chalcocite from pyrite

The separation of chalcocite from pyrite was also investigated.
Fig. 13 shows chalcocite recovery as a function of pyrite recovery
in rougher otation after primary grinding. Both chalcocite and
pyrite oated well with more than 90% recovery at the completion
of 8 min otation. Again, the high pyrite recovery was due to the
activation by copper ions which emanated from chalcocite. After
mixing with pyrite, chalcocite otation was slightly lower than
chalcopyrite otation, while pyrite otation was slightly better in
the presence of chalcocite than in the presence of chalcopyrite in
rougher otation with all otation conditions being the same.
After regrinding, chalcocite separation from pyrite without and
with the addition of collector in the cleaner otation stage is
shown in Fig. 14. Without the addition of collector, both chalcocite

X. Chen et al. / Minerals Engineering 58 (2014) 6472

71

Table 7
S 2p quantication for the tailing from cleaner otation of chalcopyrite mixed with
pyrite (160 g/t collector).
Sample

Species

2p3/2 Position

Area contribution (%)

Flotation con1

Monosulde
Disulde
Polysulde
Energy loss
Sulfate

161.5
162.5
163.4
164.5
168.3

10.5
65.6
6.6
6.2
11.1

Fig. 15. Cu 2p XPS spectra from mineral surfaces on regrinding feed and regrinding
product for chalcocitepyrite mixture.

Fig. 13. Chalcocite recovery as a function of pyrite recovery in rougher otation

after primary grinding.

of Cu2+. The percentages of Cu+ to the overall Cu for the regrinding

feed and product are 79.4% and 61.9%, respectively. It suggests that
a greater amount of Cu-hydroxides were formed on the regrinding
product resulting from chalcocite oxidation. Compared to the XPS
results with chalcocite single minerals in the absence of pyrite,
the presence of pyrite signicantly accelerated the oxidation of
chalcocite, which is due to a strong galvanic interaction between
these two minerals as a result of a large difference in their rest
potential.
Based on Table 5, the Cu ion per surface area after primary
grinding is about 4.25  104 mole/m2. It is decreased to
7.9  105 mole/m2 after regrinding, still much higher than that
produced from the chalcopyrite-pyrite mixture after primary
grinding (9.21  106 mole/m2) where pyrite activation was evident. Therefore, chalcocite should be able to provide sufcient copper ions to activate pyrite after regrinding. Furthermore, a strong
galvanic interaction occurs between chalcocite and pyrite, resulting in reduced oxidation on the cathodic pyrite, which can also
benet the copper activation on pyrite.
4. Conclusion

Fig. 14. Chalcocite and pyrite recovery from cleaner otation as a function of
collector dosage.

and pyrite otation were decreased signicantly. Chalcocite recovery was 42.8% in 8 min otation, which is much lower than that in
the single chalcocite otation. Pyrite recovery was 24.8%. With the
collector addition, chalcocite recovery was increased signicantly
to 78% at 120 g/t collector and then increased slightly with further
collector addition. Pyrite recovery was increased in a similar way
with collector amount but was always lower than chalcocite
recovery.
The surface species on regrinding feed and product when chalcocite was mixed with pyrite were examined by XPS. The Cu 2p
XPS spectra from mineral surfaces before and after regrinding are
shown in Fig. 15. The Cu+ Cu 2p3/2 component was identied at
about 932.6 eV (McIntyre and Cook, 1975; Deroubaix and Marcus,
1992). A second component in the Cu 2p3/2 at about 934.2 eV is
attributable to Cu2+-hydroxyl species (Weisener and Gerson,
2000). The broad peak at about 943.3 eV is Cu2+ related shake-up
satellite energy loss structure, which further conrms the presence

The current study demonstrates the different otation behavior

of chalcopyrite and chalcocite and their separation from pyrite obtained in the cleaner stage after regrinding. When chalcopyrite was
mixed with pyrite, a high chalcopyrite recovery was achieved in
the cleaner stage, whilst pyrite otation was strongly depressed
and could not be restored even at a high collector addition. When
chalcocite was mixed with pyrite, both chalcocite and pyrite otation were decreased after regrinding, but the recovery could be
partly restored by additional collector. These phenomena are governed by the electrochemical activity of chalcopyrite and chalcocite and their galvanic coupling with pyrite. Chalcocite is more
electrochemically active than chalcopyrite and therefore a stronger
galvanic interaction occurs between chalcocite and pyrite than
between chalcopyrite and pyrite. As a result, stronger oxidation occurs on chalcocite compared with chalcopyrite, producing a significant amount of hydrophilic oxidation species to depress its
otation, while providing a sufcient amount of copper ions to
activate pyrite.
Chalcopyrite and chalcocite do not affect pyrite otation in the
same way in rougher otation after primary grinding. This is

72

X. Chen et al. / Minerals Engineering 58 (2014) 6472

because the galvanic interaction between minerals in both chalcocitepyrite and chalcopyritepyrite systems is weak at coarse sizes
with small surface areas.
Acknowledgments
The authors gratefully acknowledge nancial support from
Newcrest Mining Limited and the Australia Research Council. The
discussion of this study with Dr. David Seaman, the technical expert from Newcrest Mining Limited, is greatly appreciated. The rst
author also thanks the scholarship provided by the University of
Queensland.
References
Acres, R.G., Harmer, S.L., Beattie, D.A., 2010. Synchrotron XPS studies of solution
exposed chalcopyrite, bornite, and heterogeneous chalcopyrite with bornite.
Int. J. Miner. Process. 94, 4351.
Anthony, J.W., Bideaux, R.A., Bladh, K.W., Nichols, M.C., 1990. Handbook of
Mineralogy (Volume I Elements, Suldes, Sulfosalts). Mineral Data
Publishing, Tucson, Arizona.
Baboian, R., 2005. Corrosion tests and standards: application and interpretation.
ASTM International, West Conshohocken, PA.
Bradshaw, D.J., Buswell, A.M., Harris, P.J., Ekmekci, Z., 2006. Interactive effects of the
type of milling media and copper sulphate addition on the otation
performance of sulphide minerals from Merensky ore Part I: pulp chemistry.
Int. J. Miner. Process. 78, 153163.
Bruckard, W.J., Sparrow, G.J., Woodcock, J.T., 2011. A review of the effects of the
grinding environment on the otation of copper sulphides. Int. J. Miner. Process.
100, 113.
Buckley, A., Woods, R., 1984. An X-ray photoelectron spectroscopic study of the
oxidation of chalcopyrite. Aust. J. Chem. 37, 24032413.
Chen, X., Peng, Y., Bradshaw, D., 2012. Effect of regrinding conditions on the
rejection of pyrite in the cleaner stage. In: 26th International Mineral Processing
Congress, New Delhi, India.
Chen, X., Peng, Y., Bradshaw, D., 2013. Effect of regrinding conditions on pyrite
otation in the presence of copper ions. Int. J. Miner. Process. 125, 129136.
Chmielewski, T., Kaleta, R., 2011. Galvanic interactions of sulde minerals in
leaching of otation concentrate from lubin concentrator. Physicochem. Probl.
Miner. Process., 2134.
Deroubaix, G., Marcus, P., 1992. X-ray photoelectron spectroscopy analysis of
copper and zinc oxides and sulphides. Surf. Interface. Anal. 18, 3946.
Ekmeki, Z., Demirel, H., 1997. Effects of galvanic interaction on collectorless
otation behaviour of chalcopyrite and pyrite. Int. J. Miner. Process. 52, 3148.
Finkelstein, N.P., 1997. The activation of sulphide minerals for otation: a review.
Int. J. Miner. Process. 52, 81120.
Frew, J.A., Smart, R.S., Manlapig, E.V., 1994. Effects of ne grinding on otation
performance: generic statements. In: Fifth Mill Operators conference,
Australasian Institute of Mining and Metallurgy, Roxby Downs, South Australia.
Fullston, D., Fornasiero, D., Ralston, J., 1999. Zeta potential study of the oxidation of
copper sulde minerals. Colloid Surface. A 146, 113121.
Gardner, J.R., Woods, R., 1979. An electrochemical investigation of the natural
otability of chalcopyrite. Int. J. Miner. Process. 6, 116.
Grano, S., 2009. The critical importance of the grinding environment on ne particle
recovery in otation. Miner. Eng. 22, 386394.

Harmer, S.L., Pratt, A.R., Nesbitt, W.H., Fleet, M.E., 2004. Sulfur species at
chalcopyrite (CuFeS2) fracture surfaces. Am. Mineral. 89, 10261032.
Heyes, G.W., Trahar, W.J., 1977. The natural otability of chalcopyrite. Int. J. Miner.
Process. 4, 317344.
Heyes, G.W., Trahar, W.J., 1979. Oxidation-reduction effects in the otation of
chalcocite and cuprite. Int. J. Miner. Process. 6, 229252.
Hu, Y.H., Zhang, S.L., Qiu, G.Z., Miller, J.D., 2000. Surface chemistry of activation of
limedepressed pyrite in otation. Trans. Nonferrous Met. Soc. China (Engl. Ed.)
10, 798803.
Huang, G., Grano, S., 2005. Galvanic interaction of grinding media with pyrite and
its effect on oatation. Miner. Eng. 18, 11521163.
Johnson, N.W., 2006. Liberated 010 lm particles from sulphide ores, their
production and separation-recent developments and future needs. Miner. Eng.
19, 666674.
Lascelles, D., Finch, J.A., 2002. Quantifying accidental activation. Part I. Cu ion
production. Miner. Eng. 15, 567571.
Leppinen, J.O., 1990. FTIR and otation investigation of the adsorption of ethyl
xanthate on activated and non-activated sulde minerals. Int. J. Miner. Process.
30, 245263.
Martin, C.J., McIvor, R.E., Finch, J.A., Rao, S.R., 1991. Review of the effect of grinding
media on otation of sulphide minerals. Miner. Eng. 4, 121132.
McIntyre, N.S., Cook, M.G., 1975. X-ray photoelectron studies on some oxides and
hydroxides of cobalt, nickel, and copper. Anal. Chem. 47, 22082213.
Nakazawa, H., Iwasaki, I., 1985. Effect of pyrite-pyrrhotite contact on their
oatabilities. Miner. Metall. Proc. 2, 206211.
Peng, Y., Grano, S., 2010. Effect of iron contamination from grinding media on the
otation of sulphide minerals of different particle size. Int. J. Miner. Process. 97,
16.
Peng, Y., Grano, S., Fornasiero, D., Ralston, J., 2003a. Control of grinding conditions in
the otation of chalcopyrite and its separation from pyrite. Int. J. Miner. Process.
69, 87100.
Peng, Y., Grano, S., Fornasiero, D., Ralston, J., 2003b. Control of grinding conditions in
the otation of galena and its separation from pyrite. Int. J. Miner. Process. 70,
6782.
Smart, R.S.C., Skinner, W.M., Gerson, A.R., 1999. XPS of sulphide mineral surfaces:
metal-decient, polysulphides, defects and elemental sulphur. Surf. Interface.
Anal. 28, 101105.
Steger, H.F., Desjardins, L.E., 1980. Oxidation of sulde minerals; V, Galena,
sphalerite and chalcocite. Can. Mineral. 18, 365.
Tolley, W., Kotlyar, D., Van Wagoner, R., 1996. Fundamental electrochemical studies
of sulde mineral otation. Miner. Eng. 9, 603637.
Weisener, C., Gerson, A., 2000. Cu(II) adsorption mechanism on pyrite: An XAFS and
XPS study. Surf. Interface. Anal. 30, 454458.
Wightman, E.M., Evans, C.L., Bradshaw, D.J., 2011. Investigating the behaviour of
secondary copper sulphide minerals at plant scale. In: Flotation11, Cape Town,
South Africa.
Wills, B.A., Napier-Munn, T.J., 2006. Wills mineral processing technology: an
introduction to the practical aspects of ore treatment and mineral recovery.
Butterworth-Heinemann, Oxford.
Wong, G., Lascelles, D., Finch, J.A., 2002. Quantifying accidental activation. Part II. Cu
activation of pyrite. Miner. Eng. 15, 573576.
Yelloji Rao, M.K., Natarajan, K.A., 1989. Electrochemical effects of mineralmineral
interactions on the otation of chalcopyrite and sphalerite. Int. J. Miner. Process.
27, 279293.
Zachwieja, J.B., McCarron, J.J., Walker, G.W., Buckley, A.N., 1989. Correlation
between the surface composition and collectorless otation of chalcopyrite. J.
Colloid Interface Sci. 132, 462468.
Zhang, Q., Xu, Z., Bozkurt, V., Finch, J.A., 1997. Pyrite otation in the presence of
metal ions and sphalerite. Int. J. Miner. Process. 52, 187201.