electrochemical behaviour and surface analysis of ......alkaline solutions and are collectively...

Contents lists available at ScienceDirect

Hydrometallurgy

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

Electrochemical behaviour and surface analysis of chalcopyrite in alkalineglycine solutions

G.M. O'Connor, K. Lepkova, J.J. Eksteen⁎, E.A. OrabyWestern Australian School of Mines, Minerals, Energy and Chemical Engineering, Curtin University, GPO Box U1987, Perth, WA 6845, Australia

A R T I C L E I N F O

Keywords:ChalcopyriteGlycineElectrochemical dissolutionLeachingPassivationRamanXPS

A B S T R A C T

Electrochemical experiments with a chalcopyrite rotating disk electrode were carried out in alkaline glycinesolutions. This showed no apparent passivation behaviour during anodic dissolution that is observed in acidsolutions. The current increased with applied potential from the open circuit potential with no resemblance tothe passivation region seen in acid solutions. A loosely held porous layer developed on the surface consistinglargely of iron oxyhydroxides that had a limited effect on the anodic current. Elemental sulfur and a disulfidespecies were detected using XPS and Raman spectroscopy but did not passivate the surface as has been proposedfor acid solutions. The disulfide species is sometimes used to infer a metal deficient sulfide or polysulfide that isresponsible for passivation but in this study it had no passivating influence. Current-potential curves showedfeatures of a non-ideal semiconductor that were explained by charge transfer via surface states.

1. Introduction

Recent research at the Western Australian School of Mines has fo-cussed on alkaline leaching of gold, silver and base metals using glycineas a complexing agent (Oraby and Eksteen 2014, Tanda et al., 2019).The complexing action of glycine with copper is enhanced above pH 9.8where the glycinate anion is dominant (O'Connor et al., 2018). Theanodic dissolution of chalcopyrite in glycine solutions may form ele-mental sulfur or sulfate by the following half reactions:

+ + ⇌ + + +

+

CuFeS 2Gly 19OH Cu(Gly) Fe(OH) 2SO 8H O

17e2

‐ ‐2 3 4

2‐2

‐ (1)

+ + ⇌ + + +CuFeS 2Gly 3OH Cu(Gly) Fe(OH) 2S 5e2 ‐ ‐ 2 3 ‐ (2)

Other iron species such as FeOOH, Fe2O3 or Fe3O4 may also form inalkaline solutions and are collectively referred to as iron oxyhydroxides(Grano et al., 1997). Other sulfur species such as thiosulfate may alsoform and will be the topic of further research.

Chalcopyrite dissolves at a slower rate than supergene and copperoxide minerals in glycine solutions (Eksteen et al., 2017; Tanda et al.,2017, 2018a,b). This is also true for a variety of leaching systems in theacidic pH range and in alkaline solutions with ammonia or cyanide ascomplexing agents (Marsden, 2006; Razzell and Trussell, 1963;Stanczyk and Rampacek, 1966; Watling, 2013). Such slow rates ofdissolution have prevented the development of a financially viable

hydrometallurgical process for leaching copper from chalcopyrite.The slow rate of dissolution has been known since the early 20th

century, particularly for acid solutions (Greenawalt, 1912; Pike et al.,1930; Sullivan, 1933). Strategies developed to enhance the leach rateincluded ultra-fine grinding and using near boiling temperatures. Fromthe 1970s, the slow rate has sometimes been attributed to the formationof a layer of sulfur, jarosite or a metal deficient sulfide/polysulfide. Thisinhibits leaching and is termed “passivation” due to some similaritieswith the well-known phenomenon in corrosion science. The proposedpassivating species have been described and critiqued at length in manyreviews (Córdoba et al., 2008; Klauber, 2008; Li et al., 2013; Watling,2013). Of the proposed species, the metal-deficient sulfide or poly-sulfide Cu1-xFe1-yS2 has gained popularity, but the literature is incon-sistent with regard to its electronic and physical properties(Ghahremaninezhad et al., 2013; Hackl et al., 1995; Nicol, 2017c;Parker et al., 1981).

In alkaline systems such as ammoniacal solutions, iron oxyhydr-oxides are readily observed on the chalcopyrite surface. It is sometimessuggested that these species act as an inhibiting species (Beckstead andMiller, 1977; Guan and Han, 1997; Nabizadeh and Aghazadeh, 2015).Some authors have however noted its porous nature, suggesting that itis unlikely to be completely protective (Burkin, 1969; Cattarin et al.,1990; Warren and Wadsworth, 1984). An alternative proposal is that apassivating metal-deficient sulfide layer, such as that observed in acidsystems, underlies the iron oxyhydroxides (Buckley and Woods, 1984;

https://doi.org/10.1016/j.hydromet.2018.10.009Received 17 May 2018; Received in revised form 3 October 2018; Accepted 13 October 2018

⁎ Corresponding author.E-mail address: [email protected] (J.J. Eksteen).

Hydrometallurgy 182 (2018) 32–43

Available online 15 October 20180304-386X/ © 2018 Elsevier B.V. All rights reserved.

T

http://www.sciencedirect.com/science/journal/0304386Xhttps://www.elsevier.com/locate/hydromethttps://doi.org/10.1016/j.hydromet.2018.10.009https://doi.org/10.1016/j.hydromet.2018.10.009mailto:[email protected]://doi.org/10.1016/j.hydromet.2018.10.009http://crossmark.crossref.org/dialog/?doi=10.1016/j.hydromet.2018.10.009&domain=pdf

Cattarin et al., 1990; Chander, 1991; Smart, 1995; Warren andWadsworth, 1984; Wills and Finch, 2016; Yin et al., 2000). This layermight inhibit leaching in the same way as proposed for acid solutions.

The need for a passivation model in alkaline systems seems super-fluous when electrochemical experiments are examined. In alkalinesolutions with a complexing agent, the apparent passivating effect thatis observed in the acid range is notably absent. This can be seen in theanodic sweeps in alkaline ammonia solutions in Fig. 1 (Warren andWadsworth, 1984; Warren et al., 1982). Here, the current increaseswith applied potential. By contrast, in sulfuric acid solution little cur-rent flows up to a potential of about 0.8 V (vs Ag/AgCl). This lack of apassivating region has also been noted by other researchers in alkalineammonia solutions or at pH 13 where copper and iron are soluble (Nicoland Zhang, 2017; Yin et al., 1995).

The slower leach rate of chalcopyrite compared to other coppersulfides in alkaline solutions is likely not due to passivation, but isdependent on the presence and concentration of a suitable oxidant.Studies of oxygen reduction on chalcopyrite in alkaline solutions haveshown that it is a poor oxidant, with copper (II) being more effective inammonia solutions (Moyo et al., 2015; Nicol, 2017b). Further researchinto viable oxidants is needed for alkaline systems.

An alternative explanation for the apparent passive region in Fig. 1is that the electronic structure of chalcopyrite dictates its dissolutionbehaviour. Chalcopyrite, like many sulfide minerals, is well known forits semiconducting properties and was one of the first semiconductorsdiscovered (Braun, 1875; Shuey, 1975). The application of semi-conductor theory to mineral leaching has been described in detail bymany researchers, based on the work of Gerischer in the 1960s (Brysonet al., 2016; Crundwell, 1988; Crundwell, 2015; Gerischer, 1961;Gerischer, 1969; Hiskey, 1993; Osseo-Asare, 1992; Zevgolis and Cooke,1975).

According to this theory, electron exchange between a semi-conductor and a redox couple in solution can only occur between equalenergy levels in both species (Gerischer, 1961; Gerischer, 1969). Thatis, the potential of a redox couple in solution must correspond withpopulated energy levels of the semiconductor. There is however anunpopulated, forbidden region of energy levels in a semiconductorknown as the band gap. For chalcopyrite this corresponds to the stan-dard redox potential of couples such as the ferric-ferrous couple(Crundwell, 1988). Consequently, the exchange of electrons andtherefore the rate of dissolution is slow. A suitable redox couple wouldhave a standard potential outside the energy levels of the band gap,which might explain the higher rates of leaching with strong oxidants athigh redox potentials or when the potential is controlled at relativelylow values (Watling, 2013).

The precise energy levels of the band gap have not been directlymeasured due to the antiferromagnetic nature of chalcopyrite (Shuey,

1975). In a theoretical study the energy at the centre of the band gap,the Fermi level, was calculated from electronegativity values of theconstituent atoms to be 5.15 eV on the absolute scale, or 0.45 V (vs Ag/AgCl) (Xu and Schoonen, 2000). With a band gap of 0.6 eV spanning theFermi level, the band edges would be expected to be at 0.75 V (vs Ag/AgCl) for the valence band and 0.15 V (vs Ag/AgCl) for the conductionband. A suitable oxidant should therefore have a potential outside thisrange. This is consistent with the relative rates of reduction observedfor various oxidants, and the rapid increase in the slope of current-potential curves at about 0.75 V (vs Ag/AgCl) (Crundwell, 2015; Parkeret al., 1981).

Many electrochemical studies have shown that chalcopyrite has amuch lower open circuit potential (OCP) in alkaline compared to acidsolutions (Azizkarimi et al., 2014; Moyo et al., 2015; Nicol, 2017b; Yinet al., 2000). For the study in Fig. 1 it is −0.1 V (vs Ag/AgCl), which isless than the anticipated conduction band edge at 0.15 V (vs Ag/AgCl)and outside the band gap. Electron transfer therefore initially occursreadily with increasing potential from the OCP. However, in Fig. 1 thecurrent continues to increase beyond 0.15 V (vs Ag/AgCl) and into theband gap where electron exchange should not occur for an idealsemiconductor. If the semiconductor model is to apply, the inclusion ofa more complex model is required, such as that involving surface states.These states are energy levels that occur in the band gap at the surface,allowing electron exchange. This is well known in semiconductor sci-ence and has been applied to other mineral systems to explain non-idealbehaviour (Bryson and Crundwell, 2014; Mishra and Osseo-Asare,1992; Morrison, 1980; Springer, 1970; Tributsch and Bennett, 1981a).Other complexities to the chalcopyrite band gap may also cause de-viations from ideal behaviour, particularly with photo-processes andreflectivity (Bryson et al., 2016; Oguchi et al., 1980).

The purpose of this study was to fit the dissolution behaviour ofchalcopyrite in alkaline glycine solutions to either passivation orsemiconductor theory using electrochemical and surface analysismethods. The methods used follow closely those of other researchersand include staircase potential step voltammetry, chronoamperometryand capacitance measurements as well as the surface techniques XPSand Raman spectroscopy (Crundwell et al., 2015; Ghahremaninezhadet al., 2013; Nicol, 2017c; Parker et al., 1981).

2. Experimental

2.1. Sample details

A high purity chalcopyrite sample was obtained fromGeodiscoveries Australia. Optical microscopy at various timesthroughout the study confirmed purity at 95% with small inclusions ofquartz and feldspars identified by SEM-EDS. The sample was analysedfor stoichiometry by electron microprobe and showed a slight excess ofmetal over sulfur over 140 spot locations. This is consistent with naturalchalcopyrite showing n-type conductivity (Shuey, 1975). Metal im-purities were highly variable from 0 to 0.02% depending on the spotlocation, consistent with small mineral inclusions. The thermoelectriccurrent was measured by heating the positive electrode with a solderingiron and measuring the current with a digital multimeter. The resultingcurrent was highly variable but positive, indicating an n-type semi-conductor.

2.2. Electrochemical experiments

All electrochemical tests were carried out using a Bio-logic VMP3potentiostat. The working electrode was a chalcopyrite core embeddedin epoxy resin with an exposed surface area of 0.78 cm2. Test solutionswere made from analytical grade glycine (99%, Sigma Aldrich) andsodium hydroxide (98%, Sigma Aldrich) using Mili-Q deionised waterwith a resistivity of 18.2MΩ.cm.

The working electrode was progressively polished to a 3 μm

Fig. 1. Comparison of alkaline (A) and acid (B) current potential curves. Scanrate 30mV/min (Warren and Wadsworth, 1984; Warren et al., 1982).

G.M. O'Connor et al. Hydrometallurgy 182 (2018) 32–43

33

diamond finish, rinsed with DI water and immediately placed in the testsolution. The electrode was rotated at a set rate and the temperatureadjusted to target within±1 °C accuracy. Tests were carried out at1000 rpm and 25 °C unless otherwise specified.

The experiments were carried out in a three-electrode electro-chemical cell with a working volume of 500mL. The reference elec-trode was single junction Ag/AgCl (3.5M) held in a Luggin capillaryplaced close to the working electrode to minimize any error due to iRdrop. The same distance was maintained between the reference andworking electrodes for all tests. The counter electrode was Hastelloy Cwith a surface area of 5.5 cm2. The working electrode was allowed60min to stabilise before beginning the experiment.

A staircase potential sweep measurement was carried out with 20steps from the open circuit potential to 1.0 V (vs Ag/AgCl). Each stepwas held for 20min and the current was recorded, followed by capa-citance measurements at 1 to 100 kHz with a sinus amplitude of 17mV.Capacitance effects at 1 kHz are presented here, consistent with otherstudies (Ghahremaninezhad et al., 2010; Olvera et al., 2016; Warrenet al., 1982) Chronoamperometry measurements were performed at aplateau in the current- potential curve at 0.65 V (vs Ag/AgCl) for per-iods of 300 s with intermittent rest intervals at the open circuit poten-tial.

2.3. Surface analysis

The chalcopyrite electrode was removed from the solution after atest at a specified potential, rinsed, dried under vacuum before im-mediate analysis by XPS analysis or Raman spectroscopy. The solutionused was 0.3 M glycine at pH 10.5. For XPS a Kratos Axis Ultra DLD X-ray photoelectron spectroscope with a monochromatic Al source at1486.7 eV was used to characterise the surface species. A survey spec-trum was collected at binding energies between 0 and 1200 eV andhigh-resolution regional spectra were collected for copper, iron, sulfur,oxygen and carbon. Charge compensation was used where samplecharging occurred, typically where thick surface layers were presentwhich resulted in a reduction in spectrum resolution. For Ramanspectroscopy, a Labram 1B dispersive Raman spectrometer with a632.817 nm source and 2mW power was used to determine the sulfurspeciation.

3. Results and discussion

3.1. General features of current-voltage behaviour

Several researchers have noted that high scan rates for anodicsweeps can hide the effects of a passive layer due to insufficient time forit to form (Ghahremaninezhad et al., 2010; Viramontes-Gamboa et al.,2007). In order to allow time for the formation of a possible passivelayer at a given potential, a staircase potential step method was usedwith each step held for 20min while recording the current. Unlessotherwise specified, only the final current after 20min is presentedhere.

3.2. Effect of pH

The effect of pH on the current-potential curve is shown in Fig. 2.The 0.3 M sulfuric acid run was with no glycine. These curves show theincreased current response above pH 9, due to the higher mole fractionof glycinate anion described in other studies (Aksu and Doyle, 2001;O'Connor et al., 2018). Two plateau regions are present for all alkalinepH values. The apparent passive region for the acid conditions is ob-vious, from the open circuit potential to about 0.77 V (vs Ag/AgCl)where currents are about 0.02mA/cm2. A rapid increase in currentoccurs above this potential and continued to increase off the scale of thegraph. At alkaline pH in glycine the increase in current is more mod-erate, likely limited by the complexing ability of the glycinate anion.

The current eventually drops or forms a plateau at higher potentials asglycine is depleted, possibly due to some inhibition by an oxyhydroxidelayer on the surface.

Higher current densities were observed at pH 10.5 than 11.5 atpotentials above 0.5 V (vs Ag/AgCl), this may be due to copper oxidesbeing more stable than copper glycinate at this pH. Similar behaviourhas been observed in copper metal electrochemistry (O'Connor et al.,2018).

3.3. Effect of glycine concentration

The effect of glycine concentration can be seen in Fig. 3. The generalshape of the current-potential curves is similar at all concentrationvalues, but with a less pronounced plateau as the glycine concentrationincreases. For all concentrations, no significant effect on the currentdensity was observed up to 0.5 V (vs Ag/AgCl). In the plateau regionfrom 0.5 to 0.8 V (vs Ag/AgCl) considerably higher current densitieswere recorded as the concentration of glycine increased. This is con-sistent with limitation by transport through a porous layer as opposedto solution diffusion, which would be observed with changes in rotationspeed as discussed in Section 3.4. This plateau is unlike the passivationfor chalcopyrite in acid solutions, where the current is close to zero. Thecurrent density is eventually limited by glycine diffusion through thebulk solution at potentials greater than 0.85 V (vs Ag/AgCl) for the0.1 M and 0.3M solutions as also shown by the effect of rotation speed.

Fig. 2. Current-potential curves at different pH values in 0.3M glycine com-pared to 0.3M H2SO4 solution.

Fig. 3. Effect of glycine concentration on current potential curves at pH 10.5.Glycine concentration is 0.3M and temperature is 25 °C.


34

3.4. Effect of rotation speed

The effect of rotation speed is shown in Fig. 4. There is no significantdifference in the current potential curves at 500 rpm and above, exceptat high potentials above 0.9 V (vs Ag/AgCl). This indicates that diffu-sion through the bulk solution is not the limiting factor for most of thepotential range at 500 rpm and above. Other researchers have seen nosignificant effect of rotation rate in the alkaline studies with ammoniasolutions, although not all of these have investigated the effects at0 rpm. (Guan and Han, 1997; Reilly and Scott, 1977; Warren andWadsworth, 1984). Above 0.9 V (vs Ag/AgCl), there is a dependence ofcurrent density for rotation rates of 500 rpm and above. This suggestssolution diffusion plays a role at these potentials, but a final steadylimiting current density was not observed.

3.5. Effect of temperature

Increasing the temperature from 25° to 60 °C had a positive effect onthe current density up to about 0.77 V (vs Ag/AgCl), after which it wasslightly lower (Fig 5).The curve still shows an increase at 0.77 V (vs Ag/AgCl), but not as great as at room temperature. There is no drop incurrent at high potentials at 60 °C, which in the previous section wasattributed to the transfer of glycine to the surface

3.6. Effect of a pre-oxidised surface layer

During the test program, surface species were readily observed at allpotentials after removing the chalcopyrite electrode from the solution.Generally, these were a slightly tarnished surface at potentials less than0.4 V (vs Ag/AgCl), above this potential was a brown, loosely held layerthat is likely to be iron oxyhydroxides that are observed in other studies(Grano et al., 1997). The effect of this layer was investigated by pro-gressively stepping the potential to 0.65 V (vs Ag/AgCl) before startinga new scan. This generated a thick oxide layer, with the main effectbeing a lower current in the plateau regions and above 0.9 V (vs Ag/AgCl) as can be seen in Fig. 6. The thicker oxide layer had little effect oncurrent in the active regions around 0.3 and 0.8 V (vs Ag/AgCl).

3.7. Effect of potential step duration

The time held at each potential step was varied to determine if therewas an effect from allowing surface layers more time to thicken andimpede transport of reactants to and from the surface. No significanteffect was observed up to about 0.5 V (vs Ag/AgCl) as can be seen inFig. 7. The sample held for 60min at each step returned a lower currentin the plateau region above 0.5 V (vs Ag/AgCl).

3.8. Capacitance

Capacitance measurements have been used by several authors forelectrochemical impedance and Mott Schottky analyses of chalcopyritein an effort to understand layer formation in situ (Crundwell et al.,

Fig. 4. Effect of rotation speed on the current potential curves at pH 10.5 and0.3M glycine concentration at 25 °C.

Fig. 5. Effect of temperature at 25 °C and 60 °C. Glycine concentration 0.3Mand pH 10.5.

Fig. 6. Effect of a surface layer generated at 0.65 V (vs Ag/AgCl). Glycineconcentration is 0.3M and pH 10.5.

Fig. 7. Effect of different times at each step at pH 10.5 and 0.3M glycine.


35

2015; Ghahremaninezhad et al., 2013; Nicol, 2017c). The basis of thesemeasurements is that a mineral-solution interface is considered to act asa parallel plate capacitor where the capacitance, C, is given by Eq. (3)(Bard and Faulkner, 2001):

=C kε A/d0 (3)

where k is the relative permittivity of the dielectric material betweenthe plates, ε0 is the permittivity of free space, A is the area of thecharged plate and d is the separation of the plates. Changes in capaci-tance are therefore a function of these variables. It has been suggestedthat as a polysulfide passive layer thickens on a chalcopyrite surface,the separation of charge between the bulk mineral and solution growswider and so the capacitance decreases (Nicol, 2017c). This behaviourof chalcopyrite in acid media was said to be similar to a thickeningoxide layer on a metal. At high potentials the layer is thought to beoxidised and breaks down, so the gap narrows, and capacitance in-creases along with an increase in current. Results in acid solution on thechalcopyrite used in this study are shown in Fig. 8 and appear to beconsistent with this theory and compare well with the trends in thestudy by (Nicol, 2017c).

However, this model based on charge separation does not fit theobserved behaviour of chalcopyrite in an alkaline glycine solution. Ascan be seen in Fig. 8, the capacitance decreases with applied potentialup to 0.77 V (vs Ag/AgCl). This decrease clearly does not reflect athickening passive layer because the current increases during this time.At 0.77 V (vs Ag/AgCl) the slope of the capacitance curve changes signat the same time as an increase in current, as was observed in acidsolution. While not consistent with a passive layer formation andbreakdown, this behaviour is as expected for the inversion region of ann-type semiconductor seen in Mott Schottky studies (Crundwell, 2015).The curves show little resemblance to copper with a genuine oxidepassive layer in glycine solutions, which show a complex behaviour dueto porosity and semiconducting properties of the duplex Cu2O/CuOlayer (O'Connor et al., 2018).

3.9. Chronoamperometry

Several authors have proposed that a passivated chalcopyrite sur-face will reactivate upon removal of potential, based on interpretationof chronoamperometry (Lu et al., 2000; Nicol, 2017a; Parker et al.,1981). This has been attributed to the thermal breakdown of thepolysulfide or solid state diffusion of iron and copper in the mineral. Asimilar reactivation effect was observed for this system when an elec-trode was held at 0.65 V (vs Ag/AgCl) for five-minute periods withvarying rest times at the open circuit potential. The degree of re-activation increased with increased rest times as can be seen in Fig. 9,consistent with previous research in acid solutions (Parker et al., 1981).Copper metal showed a similar behaviour in alkaline glycine solutions,but with a strong dependence on stir rate (O'Connor et al., 2018).

A close-up view of the first two current transients in Fig. 10 shows acomplex decay curve. For the initial transient a high current drops to aminimum within 1–2 s, the current then rises to reach a maximum

before slowly decreasing again. For the second and each subsequenttransient there is a small oscillation that gradually decays, resemblingan under-damped second order system. Similar patterns have beenobserved in other alkaline studies of chalcopyrite but are not well de-scribed or understood (Azizkarimi et al., 2014; Yin et al., 2000).

A current time transient for a stepped anodic sweep over the wholepotential range is shown in Fig. 11. The most obvious feature is thechange in mechanism after 0.77 V (vs Ag/AgCl), where the currentaccelerates for several steps. Before this point the current shows a spikeand rapid decay as expected in a relaxation process (Crundwell et al.,2015). The oscillation feature observed in Fig. 10 is beyond the re-solution of this graph, but was present between 0.26 V (vs Ag/AgCl)and 0.82 V (vs Ag/AgCl). The current decay seen at each step might beargued as evidence of passivation if longer times are employed, but the

Fig. 8. Comparison of current and capacitance at 1 kHz in 0.3M H2SO4, left, and 0.3M glycine at pH 10.5, right.

Fig. 9. Chronoamperometry at 0.65 V (vs Ag/AgCl).

Fig. 10. Chronoamperometry at 0.65 V (vs Ag/AgCl). Close-up view of the in-itial and second current transients of Fig. 9.


36

results from longer step times of 1 h as shown in Fig. 7 still yield acurrent greater than that observed for the passive region in acid solu-tions.

The process of reactivation after a rest at the OCP seen in this studyand others suggests that if surface species are responsible for the currentdecay, they will not be present for analysis at a later time. Analyses ofthe surface are not likely to be of passivating species, but of other non-passivating reaction products. Alternatively they could be daughterproducts of a passive or inhibiting species that has altered upon re-moval from the system.

3.10. XPS and Raman spectroscopy analysis

Six samples for surface analysis were stepped to potentials rangingfrom the open circuit potential to 1.0 V (vs Ag/AgCl) and measured byXPS and Raman spectroscopy in solutions of 0.3M glycine at pH 10.5. Afreshly polished sample was exposed to the atmosphere for 12 h andalso measured. At potentials above 0.4 V (vs Ag/AgCl) the samples hada friable overlayer. The surface underneath was also measured in areaswhere it had detached during rinsing and handling for XPS. Completedeconvolution was only attempted for the sulfur peak. For iron, peaksdistinguishing oxide iron and lattice sulfide iron were determined.

3.11. Samples oxidised in air, at the OCP and at 0.15 V (vs Ag/AgCl)

The spectrum for an air oxidised sample showed typical features of achalcopyrite surface as can be seen by the uppermost trace in Fig. 12.Copper is present only as Cu (I). Surface iron is in oxide or hydroxideform and has a low BE shoulder at 708 eV indicating lattice iron bonded

to sulfur (Buckley and Woods, 1984; Hackl et al., 1995; Luttrell andYoon, 1984; McCarron et al., 1990). The sulfur spectrum shows typicalpeaks for a monosulfide and disulfide species. There is no appreciableindication of polysulfides or an energy loss peak.

The Raman spectrum for the air oxidised sample shown in Fig. 13shows a relatively broad peak typical of chalcopyrite with the main A1mode at 292 cm−1. A B2/E mode forms a shoulder at 320 cm−1 withanother at 353 cm−1 (Parker et al., 2008). The broad peak width in-dicates some degree of poor crystallinity compared to those observed atpotentials of 0.15 V (vs Ag/AgCl) and above. This is probably an arte-fact of sample preparation.

No significant change was observed in the XPS or Raman spectra forthe sample held at the open circuit potential for 12 h compared to theair-oxidised sample. The iron spectrum shows the same shoulder at708 eV indicating lattice iron bonded to sulfur, and sulfur shows thesame ratio of monosulfide to disulfide as the air oxidised sample. Thisratio is similar in other studies that have attributed these species to ametal-deficient sulfide (Ghahremaninezhad et al., 2013; Hackl et al.,1995). A minor sulfate peak was also detected.

The sample stepped to 0.15 V (vs Ag/AgCl) showed a tarnishedsurface typical of a weathered chalcopyrite. The overlayer showed noshoulder at 708 eV in the iron spectrum, indicating the presence of adifferent surface layer. The surface percentage copper has dropped from1.0% to 0.3% but still gives a strong signal. The sulfur peak is morecomplex, with the contribution from monosulfur dropping con-siderably. A new peak at 163.5 eV is possibly an indication of elementalsulfur. The presence of elemental sulfur in an ultra-high vacuum ispossible due to the protective nature of iron oxides (McCarron et al.,1990; Smart et al., 1999). The small peak at 167.9 eV is assigned tosulfate. The Raman spectrum showed a distinct narrowing of the mainchalcopyrite peak compared to previous samples as seen in Fig. 13. Thiswould be expected for a highly crystalline specimen, which suggests

Fig. 11. Current vs time for a potential step experiment at pH 10.5 with 0.3Mglycine at 25 °C.

Fig. 12. Copper, iron and sulfur XPS peaks of chalcopyrite. From top to bottom: (a) air oxidised, (b) oxidised at OCP, (c) stepped to 0.15 V (vs Ag/AgCl). Solution was0.3M glycine at pH 10.5.

Fig. 13. Raman spectrum of oxidised chalcopyrite. From top to bottom: (a) Airoxidised, (b) Oxidised at OCP, (c) oxidised at 0.15 V (vs Ag/AgCl). Solution was0.3M glycine at pH 10.5.


37

that disordered or amorphous chalcopyrite is formed during prepara-tion and dissolved in the initial potential steps. The B2/E modes arevisible at 320 cm−1 and 353 cm−1.

3.12. Sample stepped to 0.4 V (vs Ag/AgCl)

In this region the current increased with each potential step and aloose overlayer formed that partially detached from the surface. Thisallowed both the overlayer and the underlying surface to be measuredas shown in Fig. 14. Copper is still only present as Cu (I). For theoverlayer, no shoulder was detected at 708 eV for lattice iron on thesurface. The sulfur peak was noisy in this region due to sample char-ging, but a substantial sulfate peak and decreased monosulfide peaksare evident. Duplicate analyses showed this was a repeatable peak withabout 5% variation in calculated sulfur species from the deconvolutionprocess. The underlayer was similar to the fresh surface, with ashoulder for sulfide iron at 708 eV and a sulfur peak consisting of monoand disulfide.

Raman spectra for the overlayer showed mixed spectra of poorlycrystalline iron oxyhydroxides, elemental sulfur and chalcopyrite. Abroad oxyhydroxide peak is at 1313 cm−1 and overlapping peaks arebetween 650 and 720 cm−1. Elemental sulfur was detected in cracks inthis oxide overlayer with distinct peaks due to S-S-S bending at152 cm−1 and 219 cm−1, and SeS stretching at 473 cm−1 (Fig. 15).The surface where the overlayer had detached showed a highly crys-talline chalcopyrite peak as with the previous sample and is not re-peated here.


In this plateau region both an underlayer and overlayer were again

measured (Fig. 16). In the overlayer, again no shoulder denoting latticeiron was detected. Copper is almost completely obscured from thesurface in this region with peaks barely above noise. As for the previoussample, sulfur has a noisy peak due to sample charging which is diffi-cult to model. It can be stated that the monosulfide contribution hasdisappeared, leaving a disulfide, elemental sulfur and a sulfate. Anotherspecies appears to be present at 165.6 eV that is yet to be positivelyidentified, but is possibly sulfite. The underlayer has features of achalcopyrite surface in the XPS spectrum as described in previous sec-tions.

Raman spectroscopy showed the overlayer to be a mix of poorlycrystalline iron oxyhydroxides and elemental sulfur as shown in Fig. 17.Thick elemental sulfur was detected in cracks in the oxide overlayer asevidenced by a mixed spectrum of sulfur and chalcopyrite. Chalcopyritein the underlayer was in a highly crystalline form similar to that shownin Fig. 13.


In this active region faint copper peaks were visible in XPS for theoverlayer but no positive assignments can be made (Fig. 18). Sulfur ispresent as disulfide, elemental sulfur and sulfate. Raman spectra mostlyshow a mix of sulfur and chalcopyrite spectra in the cracks of theoverlayer (Fig. 19).

The underlayer in this region shows a significantly different sulfurXPS peak compared to other regions, with a contribution from ele-mental sulfur. Previously it was only detected as a component in theoverlayer. Iron again showed the shoulder at 708 eV attributed to ironbonded to sulfur in the lattice.

Fig. 14. Copper, iron and sulfur XPS peaks for chalcopyrite stepped to 0.4 V (vs Ag/AgCl) in 0.3M glycine at pH 10.5. From top to bottom: (a) overlayer, (b)underlayer.

Fig. 15. Left: Raman spectrum of overlayer showing mixed spectra of chalcopyrite, sulfur and oxyhydroxides. Right: elemental sulfur with minor chalcopyrite inoxide layer cracks. Sample stepped to 0.4 V (vs Ag/AgCl) in 0.3M glycine at pH 10.5.


38

3.15. Sample stepped to 1 V (vs Ag/AgCl)

The overlayer in this region is similar to that observed at 0.85 V (vsAg/AgCl) as shown in Fig. 20. Sulfur is present as disulfide, elementalsulfur and minor sulfate with no monosulfide detected. The underlayershowed a higher than usual disulfide level, but no elemental sulfur inthis case. Raman spectroscopy again showed elemental sulfur in theoverlayer cracks and crystalline chalcopyrite elsewhere in the under-layer (Fig. 21).

3.16. Summary of XPS results

The elemental distribution calculated from survey spectra is pre-sented as elemental ratios with respect to copper in Table 1. Thedominant species in most cases are adventitious carbon, oxygen andnitrogen from the atmosphere. The overlayer shows high iron valuesdue to the presence of iron oxyhydroxides. The surface is rich in sulfur(or copper deficient) except for the underlayer at 1 V (vs Ag/AgCl). Theunderlayer at 0.88 V has a higher sulfur to copper ratio than the otherunderlayers, possibly reflecting its status as a reaction product in thisregion of high oxidation rates.

Sulfur speciation is shown in Table 2. The monosulfide species isattributed to the bulk chalcopyrite lattice (Klauber et al., 2001). Itscontribution to the total sulfur peak diminishes as the overlayerthickens and is not present in this layer at 0.65 V (vs Ag/AgCl) andabove. The disulfide peak by contrast is a major sulfur species for bothlayers in all samples analysed. The disulfide has sometimes been at-tributed to a passivating metal-deficient sulfide at acidic pH, but in thiscase there is no correlation with any features resembling passivation inthe current potential curves.

4. Discussion

No evidence of the passivation effect that is observed in acid solu-tions was apparent during anodic dissolution of chalcopyrite in alkalineglycine solutions. Unlike acidic solutions, the current-potential curvesshowed no apparent passive region above the OCP, which is consistentwith other studies in alkaline solutions with glycine or ammonia (Moyoet al., 2015; Nicol, 2017b; Warren and Wadsworth, 1984). Trends withcapacitance versus potential show no resemblance to a metal with athickening passive oxide layer as has been suggested for chalcopyrite inacid solutions (Nicol, 2017c). Elemental sulfur and a disulfide speciesthat might be attributed to a metal-deficient sulfide were present insignificant amounts in plateau regions and regions of increasing cur-rent. These species are clearly not passivating at alkaline pH. This lackof a passive region in alkaline solutions has also been noted by otherresearchers (Nicol and Zhang, 2017; Yin et al., 1995).

An alternative to the passivation proposal is to apply semiconductortheory. Chalcopyrite in glycine solutions does not behave as an idealsemiconductor, but has characteristics intermediate between a semi-conductor and metal. Metal-like behaviour can be seen in semi-conductors in several ways. One is by impurities substituting into thelattice, which is utilized in the well-known doping process in thesemiconductor industry. Doping has been suggested by some authors asa reason for metal-like behaviour in chalcopyrite, the reasoning beingthat natural chalcopyrite has high impurity levels (Nicol et al., 2016).Such a heavily doped or degenerate semiconductor is often char-acterised by the absence of a thermoelectric effect and has been ob-served for synthetic chalcopyrite doped with zinc (Xie et al., 2016).However, a thermoelectric effect was observed for the sample used inthis study and in many others, so there is some doubt that doping of thechalcopyrite lattice is a cause of metal-like behaviour. Studies haveshown that impurities in chalcopyrite have little influence on the

Fig. 16. Copper, iron and sulfur XPS peaks of chalcopyrite stepped to 0.65 V (vs Ag/AgCl). “S*” denotes an unidentified species- possibly sulfite in 0.3M glycine atpH 10.5. From top to bottom: (a) overlayer, (b) underlayer.

Fig. 17. Left: Raman spectrum of mixed sulfur and oxyhydroxide in the surface overlayer. Right Elemental sulfur in cracks in the overlayer in 0.3M glycine atpH 10.5.


39

charge carrier density (Pridmore and Shuey, 1976).Another way in which metal-like behaviour can be observed in a

semiconductor is when surface states are present in high density. Thesestates occur through the termination of the lattice or by adsorbedspecies that create energy levels within the bandgap (Morrison, 1980).These allow electron exchange at energy levels within the gap, but thesurface limitation means a thermoelectric effect is still observed in thesemiconductor bulk. This concept of a surface state mechanism hasbeen used to explain the electrochemical behaviour of chalcopyrite andother minerals for many years (Bryson and Crundwell, 2014; Mishraand Osseo-Asare, 1992; Olvera et al., 2016; Springer, 1970; Tributschand Bennett, 1981b).

In addition to the presence of surface states in the band gap, metal-like behaviour can be observed when the semiconductor has an accu-mulation or inversion layer at the surface (Gomes and Cardon, 1982;Morrison, 1980). An accumulation layer is proposed here to occur onchalcopyrite in alkaline solutions from the OCP at −0.1 V (vs Ag/AgCl)

up to the conduction band edge at about 0.15 V (vs Ag/AgCl). Theseelectrons are readily removed from the conduction band when thepotential is applied, in a similar manner as for a metal (Gomes andCardon, 1982). Also, if the potential is increased beyond the valenceband edge, an inversion layer is created and current will flow via a holemechanism. For chalcopyrite this would occur at about 0.95 V (vs Ag/AgCl).

Taking these factors into account, band diagrams can be used tovisualise the mechanisms via accumulation/inversion layers and viasurface states (Fig. 22). Between the open circuit potential and 0.15 V(vs Ag/AgCl), electrons are removed from the accumulation layer in aone-step process. At potentials in the band gap between 0.15 and 0.75 V(vs Ag/AgCl), electrons tunnel from surface states to the conductionband. In the inversion region electrons tunnel from the valence bandedge at the surface to the conduction band with mobile holes generatedat the surface. This tunnelling can occur directly or via bulk defects inthe mineral.

These three mechanisms are distinguished by different behavioursshown in the current potential curves in Fig. 2 to Fig. 7. From the opencircuit potential to 0.15 V (vs Ag/AgCl), current increases with eachpotential step followed by a current decay with time typical of a re-laxation process. A small plateau is observed at 0.15 V (vs Ag/AgCl),which coincides with the potential of the conduction band edge. Fromthis point, electron exchange is via surface states and the current decaycurves resemble an under damped second order system shown in Fig. 9.This feature remains up to about 0.82 V (vs Ag/AgCl), after the appliedpotential crosses the valence band edge.

The two-step surface state mechanism described by other re-searchers is proposed for the band gap region (Crundwell et al., 2015;Gerischer, 1969; Vanmaekelbergh, 1997). First, electrons are removedfrom the surface to form holes at a rate proportional to the density ofoccupied surface states and the applied potential across the spacecharge layer. The second step is the reaction of the hole at the surface to

Fig. 18. Copper, iron and sulfur XPS peaks of chalcopyrite stepped to 0.85 V (vs Ag/AgCl) in 0.3M glycine at pH 10.5. From top to bottom: (a) overlayer, (b)underlayer.

Fig. 19. Raman spectrum of mixed sulfur, oxyhydroxide and chalcopyrite inoverlayer at 0.85 V (vs Ag/AgCl) in 0.3M glycine at pH 10.5.

Fig. 20. Copper, iron and sulfur XPS peaks of chalcopyrite stepped to 1.0 V (vs Ag/AgCl) in 0.3M glycine at pH 10.5. From top to bottom: (a) overlayer, (b)underlayer.


40

form oxidation products such as copper glycinate, iron oxides and sulfuror sulfate. The first step is rate controlling up to about 0.45 V (vs Ag/AgCl) and is characterised by an increasing current with each potentialstep. Surface layers have no effect on the rate in this region, as evi-denced by Fig. 6 and Fig. 7. As glycine is depleted at the surface by

transport through the porous surface layer, the second step is ratelimiting resulting in the plateau region.

The plateau region ends at 0.76 V (vs Ag/AgCl) which is the an-ticipated potential for the valence band edge. At this point electronexchange is no longer via surface states and so the reaction is notlimited by this mechanism. A change in the slope of the capacitance-potential curve indicates an inversion region and a p-type conductionmechanism as seen in other studies (Crundwell et al., 2015). Currentincreases with potential in this region until it is finally limited by so-lution diffusion at about 0.9 V (vs Ag/AgCl). The current potentialcurves obtained in this study closely resemble the calculated curves fora surface state mechanism published by Vanmaekelbergh (1997).

The question remains as to why the behaviour is different to that foracid systems. The surface of chalcopyrite is different under each system,so it is possible that different surface states are involved. The currentdensities observed are a function of the density and occupancy of sur-face states. At around 0.75 V (vs Ag/AgCl) both systems undergo anincrease in current and a change in slope of the capacitance curves. Inthis region the surface states would not be expected to play a role.

5. Conclusion

Anodic dissolution in alkaline glycine solutions did not show apassive region that is often seen in acidic solutions. This is despitesurface species being formed that are often attributed to passivation,such as elemental sulfur and the metal-deficient sulfide. The anodicdissolution behaviour of chalcopyrite can be attributed to it being anon-ideal n-type semiconductor, with a high density of surface states.This research is consistent with recent studies that show the semi-conducting properties should be considered in the anodic dissolution ofchalcopyrite (Bryson et al., 2016; Crundwell et al., 2015; Olvera et al.,2016; Zhao et al., 2017; Zhou et al., 2015).

Future research on leaching chemistry should focus on applying amore effective oxidant than dissolved oxygen from air. A primaryconcern is that it should not break down glycine. Some progress has

Fig. 21. Left: Raman spectrum of mixed oxide/sulfur in overlayer; right: elemental sulfur in overlayer crack in 0.3M glycine at pH 10.5.

Fig. 22. Band diagram for chalcopyrite in alkaline glycine solutions at differentapplied potentials (vs Ag/AgCl).

Table 1Surface elemental distribution from survey scans.

Holding potential Cu Fe S C O N

Air 1.0 0.6 5.2 78 13 1.6OCP 1.0 0.6 5.6 67 13 3.50.15 V 1.0 12.0 6.0 199 101 140.4 V over layer 1.0 3.5 2.6 95 35 6.40.4 V under layer 1.0 1.0 2.9 8.1 5.0 0.40.65 V over layer 1.0 14.6 6.9 152 97 150.65 under layer 1.0 1.0 2.6 4.5 1.9 0.30.88 V over layer 1.0 11.7 9.3 190 100 210.88 V under layer 1.0 1.2 5.7 153 16 51 V over layer 1.0 24.3 9.0 176 114 101 V under layer 1.0 0.7 1.8 5.8 3.5 0.4

Table 2Sulfur species on the friable overlayer and the exposed surface beneath.

Monosulfide Disulfide Elemental sulfur Sulfate Sulfite

Sample (eV) % (eV) % (eV) % (eV) % (eV) %

Air 160.7 60 162.0 40OCP 160.5 56 162.0 41 167.9 30.15 V 160.5 36 161.7 37 163.5 18 167.8 90.4 V over layer 160.6 9 161.7 40 163.4 33 167.8 170.4 V under layer 160.8 59 162.1 410.65 V over layer 161.8 29 163.9 40 167.9 19 165.7 120.65 V under layer 160.8 58 162.2 420.88 V over layer 162.1 24 163.7 59 167.2 170.88 V under layer 160.7 52 161.7 28 163.5 211 V over layer 162.0 37 163.9 50 167.7 131 V underlayer 160.9 36 162.1 64


41

already been made in this area, where Cu2+ was shown to be a moreeffective oxidant than oxygen in glycine solutions (Nicol, 2017b). Op-timisation of the process may yield more satisfactory results.

Acknowledgements

The authors acknowledge Dr. JP Veder and Mr. Peter Chapman ofCurtin University for XPS and Raman spectroscopy and Dr. MacolmRoberts of the University of Western Australia for microprobe analysis.Professor Brian Kinsella and staff of the Curtin Corrosion EngineeringIndustry Centre are thanked for the allocation of laboratory resources,technical assistance and discussions. Mr O'Connor's PhD research wasfunded by the Research Training Scheme through Curtin CurtinUniversity.

References

Aksu, S., Doyle, F.M., 2001. Electrochemistry of copper in aqueous glycine solutions. J.Electrochem. Soc. 148 (1), B51–B57.

Azizkarimi, M., Tabaian, S.H., Rezai, B., 2014. Electrochemical Investigation ofChalcopyrite Oxidation in Alkaline Solutions. Sep. Sci. Technol. 49 (16), 2595–2601.

Bard, A., Faulkner, L., 2001. Electrochemical Methods 2nd ed 2001. John Wiley & Sons,New York.

Beckstead, L.W., Miller, J.D., 1977. Ammonia, oxidation leaching of chalcopyrite-surfacedeposit effects. Metall. Trans. B 8B (1), 31–38.

Braun, F., 1875. Ueber die Stromleitung durch Schwefelmetalle. Ann. Phys. 229 (12),556–563.

Bryson, L.J., Crundwell, F.K., 2014. The anodic dissolution of pyrite (FeS2) in hydro-chloric acid solutions. Hydrometallurgy 143, 42–53.

Bryson, L.J., van Aswegen, A., Knights, B., Crundwell, F.K., 2016. Investigation of themechanism of dissolution of chalcopyrite using photo-effects. In: HydrometallurgyConference 2016: Sustainable Hydrometallurgical Extraction of Metals. SouthernAfrican Institute of Mining and Metallurgy, Cape Town.

Buckley, A., Woods, R., 1984. An X-ray photoelectron spectroscopic study of the oxidationof chalcopyrite. Aust. J. Chem. 37 (12), 2403–2413.

Burkin, A.R., 1969. Solid-State Transformations During Leaching. Miner. Sci. Eng. 1 (1).Cattarin, S., Flechter, S., Pettenkofer, C., Tributsch, H., 1990. Interfacial reactivity and

oscillating behavior of chalcopyrite cathodes duringH2O2 reduction: II.Characterization of electrode corrosion. J. Electrochem. Soc. 137 (11), 3484–3493.

Chander, S., 1991. Electrochemistry of sulfide flotation: Growth characteristics of surfacecoatings and their properties, with special reference to chalcopyrite and pyrite. Int. J.Miner. Process. 33 (1), 121–134.

Córdoba, E.M., Muñoz, J.A., Blázquez, M.L., González, F., Ballester, A., 2008. Leaching ofchalcopyrite with ferric ion. Part I: General aspects. Hydrometallurgy 93 (3–4),81–87.

Crundwell, F.K., 1988. The influence of the electronic structure of solids on the anodicdissolution and leaching of semiconducting sulphide minerals. Hydrometallurgy 21(2), 155–190.

Crundwell, F.K., 2015. The semiconductor mechanism of dissolution and the pseudo-passivation of chalcopyrite. Can. Metall. Q. 54 (3), 279–288.

Crundwell, F.K., et al., 2015. The effect of visible light on the dissolution of naturalchalcopyrite (CuFeS2) in sulphuric acid solutions. Hydrometallurgy 158, 119–131.

Eksteen, J.J., Oraby, E.A., Tanda, B.C., 2017. A conceptual process for copper extractionfrom chalcopyrite in alkaline glycinate solutions. Miner. Eng. 108, 53–66.

Gerischer, H., 1961. Electrode Processes. Annu. Rev. Phys. Chem. 12 (1), 227–254.Gerischer, H., 1969. On the role of electrons and holes in surface reactions on semi-

conductors. Surf. Sci. 13 (1), 265–278.Ghahremaninezhad, A., Asselin, E., Dixon, D.G., 2010. Electrochemical evaluation of the

surface of chalcopyrite during dissolution in sulfuric acid solution. Electrochim. Acta55 (18), 5041–5056.

Ghahremaninezhad, A., Dixon, D.G., Asselin, E., 2013. Electrochemical and XPS analysisof chalcopyrite (CuFeS2) dissolution in sulfuric acid solution. Electrochim. Acta 87(0), 97–112.

Gomes, W.P., Cardon, F., 1982. Electron energy levels in semiconductor electrochemistry.Prog. Surf. Sci. 12 (2), 155–215.

Grano, S.R., Sollaart, M., Skinner, W., Prestidge, C.A., Ralston, J., 1997. Surface mod-ifications in the chalcopyrite-sulphite ion system. I. collectorless flotation, XPS anddissolution study. Int. J. Miner. Process. 50 (1), 1–26.

Greenawalt, W.E., 1912. The Hydrometallurgy of Copper. McGraw-Hill, New York.Guan, Y.C., Han, K.N., 1997. The leaching kinetics of chalcopyrite (CuFeS2) in ammo-

nium lodide solutions with iodine. Met Trans B 28 (6), 979–985.Hackl, R.P., Dreisinger, D.B., Peters, E., King, J.A., 1995. Passivation of chalcopyrite

during oxidative leaching in sulfate media. Hydrometallurgy 39 (1–3), 25–48.Hiskey, J.B., 1993. Chalcopyrite semiconductor electrochemistry and dissolution. In:

Reddy, R.G., Weizenbaeh, R.N. (Eds.), Extractive Metallurgy of Copper, Nickel andCobalt Volume I: Fundamental Aapeds. The Minerals, Metals & Materials Society.

Klauber, C., 2008. A critical review of the surface chemistry of acidic ferric sulphatedissolution of chalcopyrite with regards to hindered dissolution. Int. J. Miner.Process. 86, 1–4): 1-17.

Klauber, C., Parker, A., van Bronswijk, W., Watling, H., 2001. Sulphur speciation of

leached chalcopyrite surfaces as determined by X-ray photoelectron spectroscopy.Int. J. Miner. Process. 62 (1–4), 65–94.

Li, Y., Kawashima, N., Li, J., Chandra, A.P., Gerson, A.R., 2013. A review of the structure,and fundamental mechanisms and kinetics of the leaching of chalcopyrite. Adv.Colloid Interf. Sci. 197–198 (0), 1–32.

Lu, Z.Y., Jeffrey, M.I., Lawson, F., 2000. An electrochemical study of the effect of chlorideions on the dissolution of chalcopyrite in acidic solutions. Hydrometallurgy 56 (2),145–155.

Luttrell, G.H., Yoon, R.H., 1984. Surface studies of the collectorless flotation of chalco-pyrite. Colloids Surf. 12 (C), 239–254.

Marsden, J.O., 2006. Chemistry of Gold Extraction. Chemistry of Gold Extraction. SME,Littleton.

McCarron, J.J., Walker, G.W., Buckley, A.N., 1990. An X-ray photoelectron spectroscopicinvestigation of chalcopyrite and pyrite surfaces after conditioning in sodium sulfidesolutions. Int. J. Miner. Process. 30 (1–2), 1–16.

Mishra, K.K., Osseo-Asare, K., 1992. Fermi level pinning at pyrite (FeS2)/electrolytejunctions. J. Electrochem. Soc. 139 (3), 749–752.

Morrison, S.R., 1980. Electrochemistry at Semiconductor and Oxidized Metal Electrodes.Moyo, T., Petersen, J., Franzidis, J.P., Nicol, M., 2015. An electrochemical study of the

dissolution of chalcopyrite in ammonia–ammonium sulphate solutions. Can. Metall.Q. 54 (3), 269–278.

Nabizadeh, A., Aghazadeh, V., 2015. Dissolution study of chalcopyrite concentrate inoxidative ammonia/ammonium carbonate solutions at moderate temperature andambient pressure. Hydrometallurgy 152 (0), 61–68.

Nicol, M.J., 2017a. The anodic behaviour of chalcopyrite in chloride solutions: Overallfeatures and comparison with sulfate solutions. Hydrometallurgy 169, 321–329.

Nicol, M.J., 2017b. A comparative assessment of the application of ammonium chlorideand glycine as lixiviants in the heap leaching of chalcopyritic ores. Hydrometallurgy175, 285–291.

Nicol, M.J., 2017c. The use of impedance measurements in the electrochemistry of thedissolution of sulfide minerals. Hydrometallurgy 169, 99–102.

Nicol, M., Zhang, S., 2017. The anodic behaviour of chalcopyrite in chloride solutions:Potentiostatic measurements. Hydrometallurgy 167, 72–80.

Nicol, M.J., Tjandrawan, V., Zhang, S., 2016. Cathodic reduction of iron(III) and copper(II) on various sulfide minerals in chloride solutions. Hydrometallurgy 166(Supplement C), 113–122.

O'Connor, G., Lepkova, K., Oraby, E., Eksteen, J.J., 2018. Electrochemical Behaviour ofCopper in Alkaline Glycine Solutions. Hydrometallurgy 181, 221–229.

Oguchi, T., Sato, K., Teranishi, T., 1980. Optical Reflectivity Spectrum of a CuFeS2 SingleCrystal. J. Phys. Soc. Jpn. 48 (1), 123–128.

Olvera, O.G., Rebolledo, M., Asselin, E., 2016. Atmospheric ferric sulfate leaching ofchalcopyrite: Thermodynamics, kinetics and electrochemistry. Hydrometallurgy 165(Part 1), 148–158.

Oraby, E.A., Eksteen, J.J., 2014. The selective leaching of copper from a gold–copperconcentrate in glycine solutions. Hydrometallurgy 150, 14–19.

Osseo-Asare, K., 1992. Semiconductor electrochemistry and hydrometallurgical dissolu-tion processes. Hydrometallurgy 29 (1), 61–90.

Parker, A.J., Paul, R., Power, G., 1981. Electrochemical aspects of leaching copper fromchalcopyrite in ferric and cupric salt solutions. Aust. J. Chem. 34 (1), 13–34.

Parker, G.K., Woods, R., Hope, G.A., 2008. Raman investigation of chalcopyrite oxidation.Colloids Surf. A Physicochem. Eng. Asp. 318 (1–3), 160–168.

Pike, R.D., West, G.H., Steck, R., Cummings, R., Little, B.P., 1930. Electrolytic iron fromsulfide ores trans. A.l.M.E. 90, 15.

Pridmore, D.F., Shuey, R.T., 1976. The electrical resistivity of galena, pyrite, and chal-copyrite. Am. Mineral. 61 (3), 248–259.

Razzell, W.E., Trussell, P.C., 1963. Microbiological leaching of metallic sulfides. Appl.Microbiol. 11 (2), 105–110.

Reilly, I.G., Scott, D.S., 1977. The leaching of a chalcopyrite concentrate in ammonia.Can. J. Chem. Eng. 55 (5), 527–533.

Shuey, R.T., 1975. Semiconducting ore minerals. Elsevier Scientific Pub. Co, Amsterdam.Smart, R.S.C., 1995. Surface Anlaytical Studies of Oxidation and Collector Adsorptio in

Sulfide Mineral Flotation.Smart, R.S.C., Skinner, W.M., Gerson, A.R., 1999. XPS of sulphide mineral surfaces: metal-

deficient, polysulphides, defects and elemental sulphur. Surf. Interface Anal. 28 (1),101–105.

Springer, G., 1970. Observations on the Electrochemical Reactivity of SemiconductingMinerals Transactions of the Institution of Mining and MetallurgyTransactions of theInstitution of Mining and Metallurgy, Section C. vol. 79. pp. 5.

Stanczyk, M.H., Rampacek, C., 1966. Oxidation Leaching of Copper Sulfides inAmmoniacal Pulps at Elevated Temperatures and Pressures. Report of Investigations/United States. vol. 6808 U.S. Dept. of the Interior, Bureau of Mines, Washington, D.C.(13 p).

Sullivan, 1933. Chemical and Physical Features of Copper Leaching. Transact. Am. Inst.Min. Metal. Eng. 1.

Tanda, B.C., Eksteen, J.J., Oraby, E.A., 2017. An investigation into the leaching behaviourof copper oxide minerals in aqueous alkaline glycine solutions. Hydrometallurgy 167,153–162.

Tanda, B.C., Oraby, E.A., Eksteen, J.J., 2018a. Kinetics of malachite leaching in alkalineglycine solution. In: Mineral Processing and Extractive Metallurgy (TIMM C), 1–10.Taylor and Francis. https://doi.org/10.1080/25726641.2018.1505211. Online.

Tanda, B.C., Oraby, E.A., Eksteen, J.J., 2018b. Kinetics of chalcocite leaching in oxyge-nated alkaline glycine solutions. Hydrometallurgy 178, 264–273.

Tanda, B.C., Eksteen, J.J., Oraby, E.A., O'Connor, G.M., 2019. The kinetics of chalcopyriteleaching in alkaline glycine/glycinate solutions. Min. Eng (Under Review).

Tributsch, H., Bennett, J.C., 1981a. Semiconductor-electrochemical aspects of bacterialleaching - 2. Survey of rate-controlling sulphide properties. J. Chem. Technol.


42

http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0005http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0005http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0010http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0010http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0015http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0015http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0020http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0020http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0025http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0025http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0030http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0030http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0035http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0035http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0035http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0035http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0040http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0040http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0045http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0050http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0050http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0050http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0055http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0055http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0055http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0060http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0060http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0060http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0065http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0065http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0065http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0070http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0070http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0075http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0075http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0080http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0080http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0085http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0090http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0090http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0095http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0095http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0095http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0100http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0100http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0100http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0105http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0105http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0110http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0110http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0110http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0115http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0120http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0120http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0125http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0125http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0130http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0130http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0130http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0135http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0135http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0135http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0140http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0140http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0140http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0145http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0145http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0145http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0150http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0150http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0150http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0155http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0155http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0160http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0160http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0165http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0165http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0165http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0170http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0170http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0175http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0180http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0180http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0180http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0185http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0185http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0185http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0190http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0190http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0195http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0195http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0195http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0200http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0200http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0205http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0205http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0210http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0210http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0210http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0215http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0215http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0220http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0220http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0225http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0225http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0225http://refhub.elsevier.com/S0304-386X(18)30373-6/rf1005http://refhub.elsevier.com/S0304-386X(18)30373-6/rf1005http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0230http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0230http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0235http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0235http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0240http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0240http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0245http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0245http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0250http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0250http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0255http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0255http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0260http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0260http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0265http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0270http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0270http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0275http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0275http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0275http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0280http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0280http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0280http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0285http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0285http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0285http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0285http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0290http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0290http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0295http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0295http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0295https://doi.org/10.1080/25726641.2018.1505211http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0305http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0305http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0310http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0310http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0315http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0315

Biotechnol. 31 (10), 627–635.Tributsch, H., Bennett, J.C., 1981b. Semiconductor-electrochemical aspects of bacterial

leaching. I. Oxidation of metal sulphides with large energy gaps. J. Chem. Technol.Biotechnol. 31 (1), 565–577.

Vanmaekelbergh, D., 1997. Direct and surface state mediated electron transfer at semi-conductor/electrolyte junctions—I. A comparison of steady-state results.Electrochim. Acta 42 (7), 1121–1134.

Viramontes-Gamboa, G., Rivera-Vasquez, B.F., Dixon, D.G., 2007. The active-passivebehavior of chalcopyrite: comparative study between electrochemical and leachingresponses. J. Electrochem. Soc. 154 (6), C299–C311.

Warren, G.W., Wadsworth, M.E., 1984. The electrochemical oxidation of chalcopyrite inammoniacal solutions. Met. Trans. B 15 (2), 289–297.

Warren, G.W., Wadsworth, M.E., El-Raghy, S.M., 1982. Passive and transpassive anodicbehavior of chalcopyrite in acid solutions. Met. Trans. B 13 (4), 571–579.

Watling, H.R., 2013. Chalcopyrite hydrometallurgy at atmospheric pressure: 1. Review ofacidic sulfate, sulfate–chloride and sulfate–nitrate process options. Hydrometallurgy140 (0), 163–180.

Wills, B.A., Finch, J.A., 2016. Chapter 12 - Froth Flotation, Wills' Mineral ProcessingTechnology (Eighth Edition). Butterworth-Heinemann, Boston, pp. 265–380.

Xie, H., et al., 2016. The Role of Zn in Chalcopyrite CuFeS2: Enhanced ThermoelectricProperties of Cu 1- x Zn x FeS 2 with In Situ Nanoprecipitates. Adv. Energy Mater. 7,1601299.

Xu, Y., Schoonen, M., 2000. The absolute energy positions of conduction and valencebands of selected semiconducting minerals. Am. Miner. 543–556.

Yin, Q., Kelsall, G.H., Vaughan, D.J., England, K.E.R., 1995. Atmospheric and electro-chemical oxidation of the surface of chalcopyrite (CuFeS2). Geochim. Cosmochim.Acta 59 (6), 1091–1100.

Yin, Q., Vaughan, D.J., England, K.E.R., Kelsall, G.H., Brandon, N.P., 2000. Surface oxi-dation of chalcopyrite (CuFeS2) in alkaline solutions. J. Electrochem. Soc. 147 (8),2945–2951.

Zevgolis, E.N., Cooke, S.R.B., 1975. Electrochemical properties of the semiconductormineral chalcopyrite. In: di Cagliari, m. Università, di arte, Istituto (Eds.),Proceedings: Eleventh International Mineral Processing Congress. Cagliari, Sardinia.Università, Istituto di arte mineraria e preparazione dei minerali, Cagliari, Sardinia.

Zhao, H., et al., 2017. Comparison of bioleaching and dissolution process of p-type and n-type chalcopyrite. Miner. Eng. 109, 153–161.

Zhou, S., et al., 2015. Catalytic effect of light illumination on bioleaching of chalcopyrite.Bioresour. Technol. 182, 345–352.


43

http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0315http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0320http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0320http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0320http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0325http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0325http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0325http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0330http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0330http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0330http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0335http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0335http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0340http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0340http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0345http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0345http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0345http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0350http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0350http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0355http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0355http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0355http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0360http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0360http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0365http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0365http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0365http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0370http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0370http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0370http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0375http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0375http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0375http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0375http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0380http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0380http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0385http://refhub.elsevier.com/S0304-386X(18)30373-6/rf0385

Electrochemical behaviour and surface analysis of chalcopyrite in alkaline glycine solutionsIntroductionExperimentalSample detailsElectrochemical experimentsSurface analysis

Results and discussionGeneral features of current-voltage behaviourEffect of pHEffect of glycine concentrationEffect of rotation speedEffect of temperatureEffect of a pre-oxidised surface layerEffect of potential step durationCapacitanceChronoamperometryXPS and Raman spectroscopy analysisSamples oxidised in air, at the OCP and at 0.15 V (vs Ag/AgCl)Sample stepped to 0.4 V (vs Ag/AgCl)Sample stepped to 0.65 V (vs Ag/AgCl)Sample stepped to 0.85 V (vs Ag/AgCl)Sample stepped to 1 V (vs Ag/AgCl)Summary of XPS results

DiscussionConclusionAcknowledgementsReferences

electrochemical behaviour and surface analysis of ......alkaline solutions and are collectively...

Documents