Analysis of iron and copper speciation and activities in chloride leaching solutions of high ionic strength

Journal Pre-proof Analysis of iron and copper speciation and activities in chloride leaching solutions of high ionic strength

Mohsen Hashemzadeh, Wenying Liu PII:

S0304-386X(19)30772-8

DOI:

https://doi.org/10.1016/j.hydromet.2020.105262

Reference:

HYDROM 105262

To appear in:

Hydrometallurgy

Received date:

31 August 2019

Revised date:

10 January 2020

Accepted date:

17 January 2020

Please cite this article as: M. Hashemzadeh and W. Liu, Analysis of iron and copper speciation and activities in chloride leaching solutions of high ionic strength, Hydrometallurgy(2019), https://doi.org/10.1016/j.hydromet.2020.105262

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof

Analysis of iron and copper speciation and activities in chloride leaching solutions of high ionic strength Mohsen Hashemzadeh and Wenying Liu Department of Materials Engineering, University of British Columbia, 309-6350 Stores Road,

of

Vancouver, BC V6T 1Z4, Canada

ro

Abstract

In light of the recent increased use of seawater for copper heap leaching, we analyzed the

-p

speciation and activities of iron and copper in chloride media, with a focus on chalcocite

re

leaching. Compared with other published work, the present study carried out such analysis at high solution ionic strength up to 3 M. For this purpose, the SIT (Specific ion Interaction Theory)

lP

activity model was used in the thermodynamic calculations performed by the STABCAL software. The speciation results obtained from the software were validated by the experimental

na

measurements of solution potential, which were performed in a sealed jacketed reactor under an ultra- high purity nitrogen environment. The speciation results showed that FeCl2+ and hydrated

ur

Fe2+ were the dominant Fe(III) and Fe(II) species in the chloride concentration range 0.5 M – 3 M. In the case of copper speciation, hydrated Cu2+ and CuCl2 – dominated at chloride

Jo

concentrations less than 0.5 M. In the chloride concentration range 0.5 M – 3 M, CuCl+ and CuCl3 2– were the dominant copper species. Based on the speciation and activities results, we determined the actual cathodic and anodic half reactions responsible for copper leaching from chalcocite in chloride media. The proposed cathodic half reactions pinpoint ways of manipulating solution potentials, which is vital for determining the kinetics of the second stage of chalcocite leaching.

Keywords: Seawater heap leaching, Copper chloride speciation, Iron chloride speciation, Chloride leaching

1

Journal Pre-proof 1

Introduction

The use of seawater in mining operations is being increased as an alternative to freshwater (Liu et al., 2013, 2011). One application of seawater is for heap leaching of secondary copper sulfide minerals, mainly chalcocite. In such applications, the chloride concentration of the process water could be much higher than that of seawater (0.55 M) (Kennish, 2000) due to evaporation and continuous recirculation of process water. It is very likely that the ionic strength of the process water reaches as high as 3 M. Therefore, to understand the thermodynamics and kinetics of such

of

chloride leaching processes, ionic strength should be taken into consideration as a key process variable. Our study emphasized thermodynamic analyses of iron and copper speciation and

ro

activities performed at high solution ionic strength up to 3 M, with a focus on chalcocite heap

-p

leaching.

In a chemical chloride leaching process, both cupric and ferric ions can be used as the oxidant.

re

The use of cupric chloride as the oxidant for copper sulfides leaching has been studied for many

lP

years (Herreros et al., 2006; Herreros and Viñals, 2007; Ruiz et al., 1998, 1997). Cuprous ions are stabilized in the presence of chloride ions by forming chloro-complexes, which in turn participate in the electrochemical oxidation of copper sulfides. The fast oxidation of cuprous to

na

cupric by oxygen can offer an alternative to the bacterial assisted oxidation of ferrous to ferric (Parker et al., 1981). Chalcocite leaching process in chloride media, similar to sulfate media, is

ur

reported to occur in two stages (Hashemzadeh et al., 2019a; Miki et al., 2011). The second stage

Jo

of leaching, i.e., the leaching of a covellite- like material produced in the first stage, is particularly sensitive to leaching solution potential. Miki et al. (2011) reported that the covellitelike phase cannot be leached at a potential of 500 mV vs SHE or less. To control potential in a desired range, it is essential to determine the predominant species participating in the redox reactions and their activities at varied leaching conditions. Many studies have been conducted to identify the speciation of copper and iron in chloride media (Hubli et al., 1997; Kim et al., 1986; Kimura et al., 1984). However, in these studies either the effect of ionic strength has been ignored, or the speciation studies were conducted using activity models suitable only for low ionic strength (< 0.5 M). For example, Kim et al. (1986) investigated the speciation of iron in chloride concentration range 1 – 5 M using the Davies model, which is suitable for low ionic strength (< 0.5 M). Kimura et al. (1984) also employed 2

Journal Pre-proof Davies model to study the speciation in the FeCl3 –FeCl2 –CuCl2 –CuCl–NaCl–HCl–H2 O system in the ionic strength range 0.1 – 7 molal. They showed that when the concentration of all components added to solution was 0.01 molal, the dominant species were Fe3+, Fe2+, Cu2+, CuCl2 – , while FeCl3 , FeCl2 , CuCl2 , and CuCl3 2– became dominant when the concentration of all components increased to 3 molal. This study performed thermodynamic calculations to identify the speciation of iron and copper at conditions that might be encountered in heap leaching of chalcocite using seawater as the process

of

water. Note that the actual measurement of the concentrations of various species is possible, such as the method described by (Zhao et al., 2013), who determined the copper species concentration

ro

over a wide range using voltammetric techniques. Such measurements were not performed in the present study. The SIT activity model, which is appropriate for high ionic strength, was applied

-p

to calculate the activity coefficients of iron and copper species. Based on the thermodynamics

re

calculations, the cathodic reduction of ferric and cupric species and the anodic reaction of chalcocite leaching in chloride media were determined. To validate the accuracy of the proposed

lP

cathodic reactions of ferric and cupric, the ORP values calculated were compared with those measured experimentally. The findings of this study provide ways of estimating the ORP of

na

chloride leaching solutions and measures of achieving sufficiently high solution potentials for the

Experimental

2.1 2.1.1

Solution potential measurements

Jo

2

ur

second stage of chalcocite leaching.

Experimental design

The experimental objective was to measure the solution potential at varied ferric/ferrous ratio and chloride concentration, the results of which were used to verify the ORP values calculated by the STABCAL software. Table 1 shows the list of the tests carried out at different nominal ferric/ferrous ratio and chloride concentration at pH 1.5 and 25 C. Nominal [Fe3+]/[Fe2+] ratio was defined as the ratio of total Fe(III) to total Fe(II) added to the reactor. The nominal [Fe3+]/[Fe2+] ratios were chosen in reference to the practical ORP values for heap leaching (Liu and Granata, 2018). The total chloride concentration was varied in the range of 0.5 to 3 M. This range was chosen in an attempt to cover all possible concentrations that may be relevant to a 3

Journal Pre-proof heap leaching process, from lower than in a typica l seawater (0.55 M) (Kennish, 2000) to 3 M that represents concentrated leaching solutions due to evaporation and recirculation of process water. In addition, chloride media has been investigated as an alternative to sulfate media to leach copper sulfides due to its benefits over sulfate media. In these investigations, various chloride concentration up to 5 M have been applied (Cheng and Lawson, 1991; Fisher et al., 1992; Hashemzadeh et al., 2019a, 2019b; Nicol et al., 2017). The total chloride concentration was calculated as the summation of the chloride from the hydrochloric acid used for pH

of

adjustment, ferric chloride, ferrous chloride, and sodium chloride. Table 2 shows the list of the tests carried out for measuring solution ORP at different nominal

ro

cupric/cuprous ratio and chloride concentration at pH 1.5 and 25 C. The nominal [Cu2+]/[Cu+] ratios were selected based on the practical heap leaching ORP values (Liu and Granata, 2018).

-p

The total chloride concentration was calculated as the summation of the chloride from the

re

hydrochloric acid used for pH adjustment, cupric chloride, cuprous chloride, and sodium

lP

chloride.

It is noteworthy that the activity of proton is sensitive to chlo ride concentration of the solution.

na

Proton activity increases with increasing chloride concentration (Muir, 2002). Therefore, at a lower chloride concentration, a higher amount of HCl was added to adjust the pH value to 1.5 than the case of higher chloride concentrations. In the present study, the total iron concentration

ur

was fixed at 100 mM (5.6 g/L) while varying ferric/ferrous ratio. Another way of designing the

Jo

experiment is fixing ferric concentration, while varying ferric/ferrous ratio. For example, at a fixed ferric concentration of 50 mM and the same ferric/ferrous ratios shown in Table, 1, the total iron concentration is in the range of 3.1 – 11.3 g/L. Practical heap leach operations have achieved successful copper extraction at total iron levels in excess of 10 g/L to levels below 2 g/L (Schnell, 1997). Given the wide range of total iron concentration and the more important effect of ferric concentration, the latter design is probably more appropriate.

4

Journal Pre-proof Table 1. The experimental design for ORP measurements in ferric/ferrous chloride media at pH 1.5, total iron concentration of 0.1 M, and 25 C [Fe3+], mM 25.0 50.0 80.0 90.1 25.0 50.0 80.0 90.1 25.0 50.0 80.0 90.1 25.0 50.0 80.0 90.1

[Cl– ], M 0.5 0.5 0.5 0.5 1.0 1.0 1.0 1.0 2.0 2.0 2.0 2.0 3.0 3.0 3.0 3.0

of

[Fe2+], mM 75.0 50.0 20.0 9.9 75.0 50.0 20.0 9.9 75.0 50.0 20.0 9.9 75.0 50.0 20.0 9.9

ro

[Fe3+]/[Fe2+] 0.33 1.00 4.00 9.10 0.33 1.00 4.00 9.10 0.33 1.00 4.00 9.10 0.33 1.00 4.00 9.10

-p

Test # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Jo 2.1.2

lP

[CuCl2 ], mM 31.442 31.162 28.612 15.737 31.442 31.162 28.612 15.737 31.442 31.162 28.612 15.737 31.442 31.162 28.612 15.737

na

Cu 2+/Cu + 1000 100 10 1 1000 100 10 1 1000 100 10 1 1000 100 10 1

ur

Test # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

re

Table 2. The experimental design for the ORP measurements in cupric/cuprous media at pH 1.5, total copper concentration of 32 mM, and 25 C [CuCl], mM 0.031 0.312 2.861 15.737 0.031 0.312 2.861 15.737 0.031 0.312 2.861 15.737 0.031 0.312 2.861 15.737

[Cl– ], M 0.5 0.5 0.5 0.5 1.0 1.0 1.0 1.0 2.0 2.0 2.0 2.0 3.0 3.0 3.0 3.0

Materials and methods

The reagent grade chemicals used in the ORP measurement experiments include ferrous chloride (FeCl2 .4H2 O, 98%, Alfa Aesar), hydrochloric acid (HCl, 36.5-38%, VWR) for pH adjustment, ferric chloride (FeCl3 , 97%, Acros) as the oxidant, cupric chloride (CuCl2 , 99%, Acros), cuprous chloride (CuCl, 97%, Sigma-Aldrich), saturated potassium chloride (KCl, LabChem) for refilling pH electrodes, 4M potassium chloride saturated with silver chloride (KCl, Fisher Scientific) for

5

Journal Pre-proof refilling the ORP electrodes, and sodium chloride (NaCl, Ward’s Science) for adjusting chloride concentration. The ORP measurements were carried out in 500 mL jacketed reactors, where the stirring was provided by means of a magnetic stirrer and temperature was maintained constant using a water bath to circulate water through the jacket of the reactor (Fig. 1). The procedure to prepare the chemical solutions involved introducing hydrochloric acid to deionized water to reach pH around 1.8, adding desired amounts of ferric ions, and making up the leaching solution to 250 mL by

of

adding deionized water. After adding ferric chloride salt, the pH value dropped to 1.6 due to the hydrolysis of ferric ions.

ro

The reactor was then sealed using a rubber stopper. An ultra-purity nitrogen gas was sparged into

-p

the reactor through a glass gas sparger for 1 h. Ferrous chloride was then added through a syringe to avoid air leakage into the reactor. Solution ORP was measured using an Accumet Platinum

re

ORP Electrode from Fisher Scientific. In our experiments, 4 M KCl solution saturated with silver

lP

chloride was used as the filling solution of the Ag/AgCl reference electrode, which gave a +200 mV potential against SHE (McDermand, 2015). Therefore, 200 mV was added to the measured ORP values to convert them to the SHE scale. All redox potentials were reported with respect to

na

the SHE at 25 °C in the present study unless otherwise stated. Oxygen content was randomly measured in some experiments to ensure that the solution contained no oxygen. In the case of

ur

cupric/cuprous solution, the procedure was the same except that the initial pH was 1.8 instead of

Jo

1.6 due to the insignificant hydrolysis of cupric ions.

6

Journal Pre-proof Syringe ORP Probe

Ultra Pure N₂

pH Probe

Water Bath

ro

of

Pump

-p

Magnetic bar

re

Magnetic Stirrer

2.2

lP

Fig. 1. A schematic of the ORP measurement apparatus

Activity models

na

A number of models have been developed to predict the activity of species in aqueous electrolyte solutions. These models include Debye‐ Huckel, Davies, B-dot, Brønsted–Guggenheim-

ur

Scatchard Specific Interaction Model (SIT), Bromley–Zaemaitis, Electrolyte NRTL, Mixed

Jo

Solvent Electrolyte (MSE), Meissner, Three-Characteristic-Parameter Correlation (TCPC), Binding Mean Spherical Approximation (BIMSA), and Pitzer model (Ge et al., 2008; Kusik and Meissner, 1975; Lin et al., 1998; Meissner and Tester, 1972; Pitzer and Silvester, 1977; Renon, 1986; Scatchard, 1976). The Pitzer model is the most comprehensive model to describe the non- ideal behavior of ionic solutions. The Pitzer equations calculate both the long-range forces between ionic species and the short-range forces between all species. However, the model generally requires a large number of coefficients, and the complete set of the Pitzer coefficients is only available for simple salt-water systems. Such data are usually scarce for complex systems, especially systems containing complexes such as copper and iron chloro-complexes (Wang et al., 1997; Yue et al., 2014). Meissner model can also be applied for high ionic solution (up to 20 molal). However, 7

Journal Pre-proof this model is suitable for salt activity calculation in aqueous solutions when the ion pairing and complex formation are negligible (Renon, 1986). The SIT model is a simplified version of the Pitzer model, ignoring the triple interactions, which become important only in very concentrated solutions, and the interactions between ions of the same charge, which are very small (Wang et al., 1997). The SIT activity model has been successfully applied for ionic strength up to 4 M and for NaCl solution up to saturation concentration at 25 ºC (Xiong, 2006). Eq. (1) represents the SIT model. (

)

of

∑

(1)

ro

where A is the Debye‐ Huckel parameter (at 25 °C and 1 atm, A = 0.51), m j is the concentration

-p

of ion j in mol/kg, (i,j) is the SIT coefficient, and the summation covers the interactions of all ions of opposite charge and neutral species. The first term in SIT model calculates the long-range

re

electrostatic interactions of ions, and the second term represents the short-range interactions of

lP

ion- ion, ion-neutral, and neutral- neutral. The (i,j) has a value of zero if the charge of ions are identical. For completely dissociated electrolytes, the accuracy is believed to be above 90% for

na

ionic strength as high as 6-10 molal (Wang et al., 1997). In the present work, the SIT model was applied to calculate the activity coefficients of iron and copper species in chloride media. Computer program

ur

2.3

Jo

STABCAL (hereafter also referred to as “the software”) is a computer program developed to perform thermodynamic calculations for constructing different types of stability diagrams, such as speciation diagram, Eh-pH diagram, and Eh-Ligand diagram (Gow et al., 2016; Huang and Young, 1996; Huang et al., 2005). In the software, one can choose Gibbs free energies of formation for chemical species from different databases, including NBS, MINTEQ, Naumov, NIST, LLNL, HSC, Woods, WatEq4F, and Helgeson. The program applies the Extended Debye‐ Huckel equation for ionic strength <0.1, the Davies model for ionic strength <0.5, and the SIT model for ionic strength in the range of 0.5 to 5.0. In the present study, STABCAL was applied to calculate the speciation and activities of iron and copper species possibly present in chloride leaching system. The SIT model was selected for calculating the activity coefficients.

8

Journal Pre-proof 3

Results and discussion

3.1 3.1.1

Ferric/ferrous chloride system Verification of the STABCAL reliability

Prior to applying the software to calculate the speciation of iron in chloride media, the reliability of the software was evaluated by comparing the ORP values calculated by the software with those measured experimentally. The ORP measurements were conducted at varied nominal [Fe3+]/[Fe2+] ratios and varied chloride concentrations according to Table 1. All available

of

databases in the software were tested to calculate the ORP values and the best values were

ro

obtained by applying the Woods database, which gave an average error of 0.56% between the calculated and the measured ORP values. Fig. 2 shows that using this database, the software was

-p

capable of satisfactorily calculating the ORP of the acidic ferric/ferrous solutions at varied nominal [Fe3+]/[Fe2+] ratio and varied chloride concentration. The equation the software uses to

re

calculate ORP is shown as Eq. (2), where E0 is the standard reduction potential and

and

Jo

ur

na

lP

are the activities for free ferric and free ferrous ions.

9

(2)

Journal Pre-proof 780

[Cl– ] 0.5 M

760

ORP vs SHE , mV

ORP vs SHE, mV

780

740

720

y = 59.270x + 713.86 R² = 0.9995

700 680

[Cl– ] 1 M

760 740

y = 59.628x + 708.19 R² = 0.9996

720

700 680

660

660 -0.5

0

0.5

1

-0.5

1.5

0

780

1

1.5

740 y = 59.561x + 701.78 R² = 0.9999

720

740 720 700

y = 59.097x + 698.03 R² = 0.9999

-p

700

[Cl– ] 3 M

760

ro

2M

ORP vs SHE , mV

760

of

780

[Cl– ]

680

680 660

re

ORP vs SHE , mV

0.5

Log([Fe3+]/[Fe2+])

Log([Fe3+]/[Fe2+])

660 -0.5

0

0.5

1

1.5

0

0.5

1

1.5

Log([Fe3+]/[Fe2+])

lP

Log([Fe3+]/[Fe2+])

-0.5

na

Fig. 2. Comparison of the ORP values calculated by STABCAL with those measured experimentally under the following conditions: pH 1.5, total iron of 0.1 M, and 25 ˚C, the squares denote the measured values and triangles denote the calculated values

Because we generally only know the total concentration of an element, the activity terms in Eq.

ur

(2) were expressed as the total ferric and ferrous concentration multiplied by their respective

Jo

activity coefficient in Eq. (3). Due to iron complexation with chloride, the two activity coefficients in Eq. (3),

and

, should be called the total activity coefficient of ferric

and ferrous, each representing the product of the percent free ions and the free ion activity coefficient. [Fe3+] and [Fe2+] are the total (nominal) concentration of ferric and ferrous. Eq. (3) was further rearranged as Eq. (4), in which the total activity coefficients were combined with the standard reduction potential E0 into a new standard potential E0, termed the formal potential. E0 can be obtained by measuring the ORP of the solutions with varied nominal [Fe 3+]/[Fe2+] ratio and chloride concentration (Hiroyoshi et al., 2000). The value of the standard reduction potential for the Fe3+/Fe2+ couple (0.770 V) was calculated from the free Gibbs energies in the Woods database.

10

Journal Pre-proof (3) [

]

[

]

(4)

In Fig. 2, the intercept of the straight lines with the y axis represents the formal potential (E0) for that specific solution composition. Straight lines implied that E0 value was not affected by the nominal [Fe3+]/[Fe2+] ratio. However, the E0 values decreased by approximately 16 mV when chloride was increased from 0.5 M to 3 M. The slope of the linear lines was approximately 59

of

mV/decade, which is in agreement with the theoretical value calculated by the Nernst equation at

ro

25 C.

-p

To more clearly show the effect of chloride on the formal potential, Fig. 3 plotted the formal potential E0 as a function of logarithmic chloride concentration. The linear line with a negative

re

slope represents a negative correlation between the two variables, suggesting that increasing

lP

chloride concentrations decreased the formal potential. Because -

is related to

, mathematically, the negative slope means that the ratio of the total activity

na

coefficient for ferric to that for ferrous is less than 1, i.e.,

, suggesting that ferric has a

higher tendency to complex with chloride than ferrous, which agrees with the results published

ur

by other researchers (Miki et al., 2011; Muir, 2002). This ratio decreases with increasing chloride

Jo

concentration in a power- law relationship with an exponent of -0.345 (-20.4/59.2). Combining results presented in Fig. 2 and Fig. 3, we derived Eq. (5), which calculates solution potential when the two easily measurable parameters, the nominal [Fe3+]/[Fe2+] ratio and chloride concentration, are given. Compared with Eq. (2) that involves activity terms, the experimentally derived Eq. (5) is a straightforward empirical equation for calculating solution potential without resorting to a seemingly complicated software. [

]

11

(5)

Journal Pre-proof

Formal potential, mV

715

y = -20.41x + 707.9 R² = 0.9991

710

705 700 695 -0.4

-0.2

0 0.2 Log[Cl– ], M

0.4

0.6

Activities of iron species in chloride media

ro

3.1.2

of

Fig. 3. Formal potential versus logarithmic chloride concentration under the following conditions: pH 1.5, total iron of 0.1 M, and 25 ˚C

In order to determine the actual cathodic reaction re sponsible for chalcocite leaching in ferric

-p

chloride media, the species possibly present in the system were identified using the software for

re

the conditions shown in Table 1. Fig. 4 shows the activities of Fe(III) species as a function of chloride concentration at varied nominal Fe3+/Fe2+ ratios. FeCl2+ has a higher activity than

lP

hydrated Fe3+ and FeCl2 + at all nominal [Fe3+]/[Fe2+] ratios. This suggests that the Fe(III) species responsible for chalcocite dissolution is most likely to be FeCl2+. In the case of Fe(II) species, as

na

shown in Fig. 5, hydrated Fe2+ was the dominant ferrous species. This agrees with the findings by Zhao and Pan (2001), who showed that in the temperature range 10 – 100 C and at chloride

ur

concentrations less than 3 molal, Fe2+ was the predominant species. In another study, Heinrich and Seward (1990) investigated the speciation of ferrous in a system with 0.01 to 3.4 molal of

Jo

chloride, 0.005 to 0.025 molal of Fe(II), and at different temperatures. They showed that at temperatures lower than l00 °C, only Fe2+ and FeCl+ were dominant, whereas at higher temperatures a minor concentration of FeCl2 started to emerge. Because FeCl2+ and Fe2+ were identified as the Fe(III) and Fe(II) species of the highest activities in the chloride concentration range of 0.5 M to 3 M, the actual cathodic reaction occurring in the leaching system and the corresponding Nernst equation were proposed as Eqs. (6) and (7). By using Eq. (7), we calculated the solution potentials with relevant thermodynamic data obtained from the software. The calculated values were compared with those measured experimentally to validate the proposed cathodic reaction. Fig. 6 shows that the difference between the calculated and the experimentally measured values was rather small, with the largest difference being 8.7 12

Journal Pre-proof mV and the average error being 0.21%. More importantly, this average error was lower than the value of 0.56%, which represents the average error between what the software calculated using Eq. (2) and the experimentally measured values (shown in Fig. 2). Therefore, it is concluded that Eq. (6) is likely to be the actual cathodic reaction responsible for chalcocite leaching. The value of the standard reduction potential for the FeCl2+/Fe2+ couple (0.704 V) was calculated from the free Gibbs energies in the Woods database. FeCl2+ + e– = Fe2+ + Cl–

(6)

25

FeCl2+

Activity of Fe(III) species×103

8

-p

15

FeCl2 +

6

FeCl2+

20

FeCl2 +

FeCl3 Fe3+

2

0 0

1

2 [Cl– ], M

[Fe3+]/[Fe2+]: 4

3

Activity of Fe(III) species×103

FeCl2+

30

FeCl2 +

ur

20

0 0

Jo

10

1

2 [Cl– ], M

FeCl3 Fe3+

3

5

FeCl3 Fe3+

0

4

na

40

re

10

4

lP

Activity of Fe(III) species×103

[Fe3+]/[Fe2+]: 1

[Fe3+]/[Fe2+]: 0.33

10

Activity of Fe(III) species×103

ro

12

of

(7)

0

1

2 [Cl– ], M

4

40 [Fe3+]/[Fe2+]: 9

FeCl2+

30

FeCl2 +

20 10

FeCl3 Fe3+

0 4

3

0

1

2 [Cl– ], M

3

4

Fig. 4. The activities of Fe(III) species calculated by the STABCAL software under the following conditions: total iron of 0.1 M, pH 1.5 and 25 C

13

Journal Pre-proof [Fe3+]/[Fe2+]: 0.33

16

Fe2+ Activity of Fe(II) species×103

15 10

5

0 6

1

2 [Cl– ], M

3

[Fe3+]/[Fe2+]: 4

8

4

4

0 3

Fe2+

1

FeCl+ 2 [Cl– ], M

3

-p

1

4

3

4

Fe2+

0

1

FeCl+ 2 [Cl– ], M

3

4

re

0

0

2 [Cl– ], M

[Fe3+]/[Fe2+]: 9

2

2

1

of

4

0

FeCl+

0

Activity of Fe(II) species×103

Activity of Fe(II) species×103

12

FeCl+

0

Fe2+

[Fe3+]/[Fe2+]: 1

ro

Activity of Fe(II) species×103

20

lP

Fig. 5: The activities of Fe(II) species calculated by the STABCAL software under the following conditions: total iron of 0.1 M, pH 1.5 and 25 C 780

[Cl– ] 0.5 M

na

760

ORP, mV

720

y = 60.423x + 740.36

700

Measured

680

Jo

-1

-0.5

0

0.5

y = 60.253x + 721.09

720 700

Measured 680

Theoretical

660

[Cl– ] 1 M

760

740

ur

ORP, mV

740

780

Theoretical

660 1

-1

-0.5

Log(aFeCl 2+/aFe2+)

780

[Cl– ]

760

2M

1

[Cl– ] 3 M

760 740

y = 59.893x + 700.21

720 700 Measured

680

0.5

780

ORP, mV

ORP, mV

740

0

Log(aFeCl 2+/aFe2+)

y = 59.228x + 687.05

720 700 Measured 680

Theoretical

660

Theoretical

660

-0.5

0

0.5

1

1.5

-0.5

Log(aFeCl 2+/aFe2+)

0

0.5

1

1.5

Log(aFeCl 2+/aFe2+)

Fig. 6. Comparison between the ORP values calculated by Eq. (7) and those measured experimentally under the following conditions: total iron of 0.1 M, pH 1.5 and 25 C

14

Journal Pre-proof 3.2 3.2.1

Cupric/cuprous chloride system Verification of the STABCAL reliability

The reliability of the software was assessed by comparing the ORP values calculated by the software with those measured experimentally (Table 2). Among the databases available in the software, the HSC database provided the lowest average error, 2.09%, between the calculated and the measured values. Fig. 7 shows that using the HSC database, the software was able to calculate the ORP of the acidic cupric/cuprous solutions at varied nominal [Cu2+]/[Cu+] ratio and

of

varied chloride concentration, but the discrepancy between the calculated and the measured values became larger at higher chloride concentration. The software uses Eq. (8) to calculate the

ro

solution ORP, where E0 is the standard reduction potential and

are the activities

-p

for free cupric and cuprous ions.

and

re

(8)

lP

Eq. (9) is the Nernst equation for the Cu2+/Cu+ couple that converts the respective activity term of free cupric and cuprous ions in Eq. (8) to the multiplication of the total concentration of cupric

na

or cuprous and the total activity coefficient. To be more specific, the total concentration of cupric and cuprous, [Cu2+] and [Cu+], also defined as the nominal concentration, referred to the total and

,

ur

amount of Cu(II) and Cu(I) added into the reactor. The total activity coefficient,

is a combination of two terms: the percentage of the total concentration of cupric or cuprous that

Jo

is free and the activity coefficient of free cupric or cuprous ions. Because of the much stronger complexation of cuprous with chloride compared with that of cupric, the percentage of the total cuprous that is free is extremely small. For example, calculations with STABCAL showed that at 3 M chloride and a nominal [Cu2+]/[Cu+] of 1, the percentage of the total cuprous that is free was 1.2 × 10-5 %, whereas the percentage of the total cupric that is free was 60.8%; the activity coefficient of free cuprous and cupric ions was 0.645 and 0.173, respectively. These calculations implied that the ratio of the total activity coefficient,

, is large, the magnitude of which

could reach 106 . Eq. (9) was further rewritten as Eq. (10), in which the total activity coefficients of cupric and cuprous ions were combined with the standard reduction potential E0 into the formal potential E0. 15

Journal Pre-proof (9)

(10) In Fig. 7, the formal potential (E0) for any specific solution composition is obtained as the intercept of the straight lines with the y axis. Straight lines imp lied that E0 value was not affected by the nominal [Cu2+]/[Cu+] ratio. However, the E0 values increased by approximately 120 mV when chloride was increased from 0.5 M to 3 M. The slope of the linear lines in Fig. 7 is 53

of

mV/decade on average, which is lower than 59 mV/decade according to the Nernst equation.

ro

This may be attributed to the rapid oxidation of cuprous ions by oxygen in the process of transferring it into the reactors, though measures were strictly taken to prevent oxidation. This

-p

might have affected the nominal Cu2+/Cu+ ratio and thereby developed discrepancy between the

re

ORP values measured experimentally and those calculated by the software. 750

0.5 M

lP

[Cl– ]

na

550

450 350 -0.5

0.5

1.5

2.5

[Cl– ] 1 M

650

y = 54.676x + 429.15 R² = 0.9969

ur

ORP, mV

650

ORP, mV

750

550 y = 54.605x + 466.28 R² = 0.9966

450

350 -0.5

3.5

0.5

Jo

750

2.5

3.5

750

[Cl– ] 2 M

[Cl– ] 3 M 650

550

ORP, mV

650

ORP, mV

1.5

Log([Cu 2+]/[Cu +])

Log([Cu 2+]/[Cu +])

y = 55.683x + 506.36 R² = 0.9965

450

550

y = 57.483x + 532.16 R² = 0.9991

450

350

350 -0.5

0.5

1.5

2.5

3.5

-0.5

Log([Cu 2+]/[Cu +])

0.5

1.5

2.5

3.5

Log([Cu 2+]/[Cu +])

Fig. 7. Comparison of the ORP values calculated by STABCAL with those measured experimentally under the following conditions: pH 1.5 and 25 C, total copper of 32 mM, the squares denote the measured values and triangles denote the calculated ORP values by the software

16

Journal Pre-proof To more clearly show the effect of chloride on the formal potential, Fig. 8 plotted the formal potential as a function of logarithmic chloride concentration. The linear line with a positive slope represents a positive correlation between the two variables, suggesting that increasing chloride concentrations raised the formal potential. As opposed to the ferric/ferrous system, the positive slope means that cupric has a lower tendency to complex with chloride than cuprous, which agrees with the results published by other researchers (Muir, 2002). Based on the experimental results shown in Fig. 8, an empirical equation for estimating solution potential for cupric/cuprous system was shown as Eq. (11).

of

]

-p

540 520

re

500

480

lP

ORP, mV

(11)

ro

[

460

420

400

-0.2

ur

-0.4

na

440

0

y = 132.01x + 467.74

0.2

0.4

0.6

Log[Cl– ], M

3.2.2

Jo

Fig. 8. Formal potential versus logarithmic chloride concentration under the following conditions: pH 1.5, total copper of 32 mM, and 25 ˚C

Activity of copper species in chloride media

In order to determine the actual cathodic and anodic reactions for chalcocite leaching in cupric chloride media, the speciation and activities of copper species in chloride media were determined using the software for the conditions shown in Table 2. In the case of Cu(II), Fig. 9 shows that the activity of CuCl+ is higher than that of hydrated Cu2+, except at low chloride concentrations (<0.5 M), where the hydrated Cu2+ has a higher activity. Collings et al. (2000) showed that below 75 C, only hydrated Cu2+ was present in solutions with lower than 2.2 M chloride and CuCl+ species was predominant at higher chloride concentration (5 M NaCl) between 25 C and 75 C.

17

Journal Pre-proof In the case of Cu(I) speciation, Fig. 10 shows that CuCl3 2– has a higher activity at chloride concentrations higher than 0.5 M, but at chloride concentrations lower than 0.5 M, CuCl2 – has a higher activity than CuCl3 2– . These results agree with those reported in the literature that in solutions containing less than 6 M chloride and at room temperature only CuCl2 – and CuCl3 2– were predominant (Ahrland and Rawsthorne, 1970; Fritz, 1981; Sharma and Millero, 1990; Sugasaka and Fujii, 1976). Brugger et al. (2007) concluded that only CuCl3 2– predominated at room temperature and high salinity (LiCl > 3 M) and Sherman (2007) observed a stable CuCl3 2–

80

160

[Cu 2+]/[Cu +]: 10

120

CuCl2

0 1

2 [Cl– ], M

160 [Cu 2+]/[Cu +]: 100

CuCl+

ur

80

0

0

Jo

40

1

2 [Cl– ], M

3

Cu 2+ CuCl2

CuCl2

0

4

na

120

3

lP

0

Cu 2+

40

re

Cu 2+

20

80

-p

40

CuCl+

ro

60

Activity of Cu(II) species×104

CuCl+

0

1

2 [Cl– ], M

3

[Cu 2+]/[Cu +]: 1000

CuCl+

120 80

Cu 2+

40

CuCl2

0

4

4

160

Activity of Cu(II) species×104

Activity of Cu(II) species×104

[Cu 2+]/[Cu +]: 1

Activity of Cu(II) species×104

of

complex in 1 M CuCl and 4 M chloride solution at room temperature.

0

1

2 [Cl– ], M

3

4

Fig. 9. The activities of Cu(II) species calculated by the STABCAL software under the following conditions: total copper 32 mM, pH 1.5, 25 C

18

Journal Pre-proof 40

8 [Cu 2+]/[Cu +]: 1 Activity of Cu(I) species×104

Activity of Cu(I) species×104

[Cu 2+]/[Cu +]: 10

30 CuCl3 2–

20 10 CuCl2 – 0

6 CuCl3 2–

4 2

CuCl2 –

0 0

1

2 [Cl– ], M

3

4

0

0.8

0.08

CuCl2 –

0.2

CuCl3 2–

0.06 0.04 0.02

CuCl2 –

0

1

2 [Cl– ], M

3

4

0

1

2 [Cl– ], M

3

4

re

0

-p

0

4

of

0.4

3

ro

Activity of Cu(I) species×104

Activity of Cu(I) species×104

CuCl3 2–

2 [Cl– ], M

[Cu 2+]/[Cu +]: 1000

[Cu 2+]/[Cu +]: 100

0.6

1

lP

Fig. 10. The activities of Cu(I) species calculated by the STABCAL software under the following conditions: total copper 32 mM, pH 1.5, 25 C

According to the speciation results, the cathodic reaction and the corresponding Nernst equation

na

were proposed as Eqs. (12) and (13) for chloride concentrations lower than 0.5 M and as Eqs. (14) and (15) for chloride concentrations in the range of 0.5 M – 3 M. The standard reduction

ur

potentials were calculated from the Gibbs free energies obtained from the HSC database. Similar equations were proposed by Muir (Muir, 2002). The appropriateness of these two equations in

Jo

describing the cathodic process was tested by comparing the ORP values calculated by these two equations with those measured experimentally. Fig. 11 shows that the proposed equations can satisfactorily describe the reduction of cupric species in the chloride concentration range investigated. Cu2+ + 2Cl– + e– = CuCl2 –

[Cl– ] < 0.5

(12) (13)

⁄

CuCl+ + 2Cl– + e – = CuCl3 2–

0.5 < [Cl-] < 3

(14) (15)

⁄

19

Journal Pre-proof 650

650

[Cl– ] 0.5 M ORP, mV

500

y = 60.267x + 417.48

550 y = 57.734x + 453.75

500 450 400

400 0

1

2

Log(aCu

3

2+/a

CuCl 2

0

4

1

–)

2

3

4

Log(aCuCl +/aCuCl 3 2–) 750

700

2M

[Cl– ] 3 M

700

600

y = 57.447x + 489.32

550

of

650

ORP, mV

[Cl– ]

650

ro

ORP, mV

550

450

ORP, mV

[Cl– ] 1 M

600

600

y = 58.939x + 511.78

600 550

-p

500

500

450

1

2

3

0

4

re

0

1

2

3

4

Log(aCuCl +/aCuCl 3 2–)

lP

Log(aCuCl +/aCuCl 3 2–)

3.3

na

Fig. 11. Comparison between the ORP values calculated by Eqs. (13) and (15) and those measured experimentally under the following conditions: total copper 32 mM, pH 1.5, 25 C, the squares denote the measured values and triangles denote the calculated ORP values by the software

Integration of the ferric/ferrous and cupric/cuprous couple in chalcocite chloride leaching

ur

To better understand the effect of chloride concentration on chalcocite leaching, the anodic reactions and the corresponding Nernst Equations are shown in Table 3. The standard reduction

Jo

potentials were calculated from the Gibbs free energies of all species obtained from the HSC database. The anodic reaction omits the intermediate products that could be formed in the first stage of leaching, because the thermodynamic data for these intermediates is unavailable (Miki et al., 2011). Fig. 12 shows that increasing chloride concentration reduces the solution potentials of the anodic half reaction for both stages of chalcocite dissolution. More importantly, the difference between the cathodic and anodic potentials represents the thermodynamic driving force for chalcocite leaching to occur. Overall, increasing chloride concentration leads to a greater difference, thereby rendering chalcocite leaching thermodynamically more favorable. Our previous publications (Hashemzadeh et al., 2019b, 2019a) on the kinetics of chalcocite leaching in ferric chloride and cupric chloride media provide evidence for the leachability of chalcocite mineral in the conditions presented in Fig. 12. 20

Journal Pre-proof Table 3. The anodic half reactions and the corresponding Nernst equations for chalcocite leaching under the following conditions: varied chloride concentrations, total copper of 32 mM, pH 1.5, 25C [Cl-] 0.5 M st

1 stage

–

Cu 2 S + 2Cl = CuCl2 – + CuS + e–

(16) (17)

2nd stage

CuS + 2Cl– = CuCl2 – + S0 + e–

(18) (19)

0.5 M < [Cl – ] < 3 M 1st stage

Cu 2 S + 3Cl– = CuCl3 2– + CuS + e–

(20)

CuS + 3Cl– = CuCl3 2– + S0 + e–

ro

2nd stage

of

(21) (22) (23)

-p

In the case of using cupric as the oxidant, Fig. 12 shows that when chloride concentration is lower than 0.5 M, it is likely that the second stage of chalcocite leaching cannot be initiated due

re

to insufficient potential difference between the cathodic and anodic processes. Miki et al. (2011)

lP

showed that above 20 g/L chloride, the reduction potential of Cu(II)/Cu(I) was higher than that of the second stage of chalcocite leaching. However, they did not provide sufficient information

na

about the activity model and the ionic strength range they used in their thermodynamic calculations.

ur

900

Fe(III)/Fe(II)

700

CuS/S, Cu(I)

300

Cu 2 S/CuS, Cu(I)

ORP, mV

Jo

Cu(II)/Cu(I)

500

100 0.5

1

1.5

2

[Cl– ], M

2.5

3

Fig. 12: Comparison of reduction potentials of cathodic and anodic reaction of the first and second stage of chalcocite leaching under the following conditions: [Cu 2+] 31.44 mM, [Cu +] 31.10×10–3 mM, [Fe3+] 0.05 M, [Fe2+] 0.05 M, pH 1.5, 25 ºC, the symbols denote the experimentally measured ORP and the dashed lines denote the calculated ORP

Fig. 12 also shows that in the chloride concentration range investigated, the predominant oxidizing couple is ferric/ferrous species. However, when chloride concentration is elevated

21

Journal Pre-proof beyond a certain level, the reduction potential of cupric/cuprous couple may exceed that of the ferric/ferrous couple, especially when ferric/ferrous ratio is low. In this case, the oxidation of ferrous species by cupric species becomes thermodynamically possible and cupric species may become increasingly important as the oxidants. This hypothesis requires further experimental proof. 4

Conclusions

Determination of iron and copper speciation and activities of various species in chloride media is

of

essential for estimating solution potential, which is a key factor determining the kinetics of

ro

copper sulfide leaching. In this study, using the STABCAL software, we determined the speciation of iron and copper in chloride media at high solution ionic strength (< 3 M) that might

-p

be encountered in heap leaching using seawater as process water. The SIT activity model, which is appropriate for high solution ionic strength (up to 3 M), was applied to calculate the activity

re

coefficients of various iron and copper species. To verify the reliability of the software, we

lP

measured solution potential at various nominal ferric/ferrous and cupric/cuprous ratio and different chloride concentration and then compared the measured values with those calculated by

na

the software. After verification of its reliability, the software was used to determine the speciation and activities of various iron and copper species. Then the actual cathodic and anodic

ur

reactions responsible for chalcocite leaching in chloride media were proposed. In general, the potentials calculated by the software matched those measured exper imentally,

Jo

even though discrepancies were observed in the cupric/cuprous system. As chloride concentration was increased, the potential of the ferric/ferrous couple slightly decreased, while the potential of the cupric/cuprous couple markedly increased. Empirical equations (Eq. 5 and 11) were derived from the experimental data that allow the calculation of solution potential without resorting to a thermodynamic software package. The speciation results show that the dominant species of ferric and ferrous were FeCl2+ and hydrated Fe2+. At [Cl– ] < 0.5 M, hydrated Cu2+ and CuCl2 + were the predominant cupric and cuprous species. At 0.5 M < [Cl– ] < 3 M, CuCl+ and CuCl3 2– were the dominant species of cupric and cuprous ions. Based on the speciation results, the cathodic reaction of ferric and cupric reduction and the anodic reaction of chalcocite leaching were proposed. According to the 22

Journal Pre-proof proposed anodic reaction, increasing the chloride concentration reduces the solution potential of chalcocite dissolution. Overall, increasing chloride concentration increases the potential difference between the cathodic and anodic processes, thereby rendering chalcocite leaching thermodynamically more favorable.

Author Statement Mohsen Hashemzadeh carried out this work as his PhD thesis project under the supervision of Wenying Liu.

Reference

re

5

-p

ro

of

Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ahrland, S., Rawsthorne, J., 1970. The stability of metal halide complexes in aqueous solution

lP

VII. The chloride complexes of copper(I). Acta Chem. Scand. 24, 157–172.

na

Brugger, J., Etschmann, B., Liu, W., Testemale, D., Hazemann, J.L., Emerich, H., van Beek, W., Proux, O., 2007. An XAS study of the structure and thermodynamics of Cu(I) chloride

71, 4920–4941.

ur

complexes in brines up to high temperature (400 °C, 600 bar). Geochim. Cosmochim. Acta

Jo

Cheng, C.Y., Lawson, F., 1991. The kinetics of leaching chalcoc ite in acidic oxygenated sulphate-chloride solutions. Hydrometallurgy 27, 249–268. Collings, M.D., Sherman, D.M., Ragnarsdottir, K.V., 2000. Complexation of Cu2+ in oxidized NaCl brines from 25 o C to 175 o C: Results from in situ EXAFS spectroscopy. Chem. Geol. 167, 65–73. Fisher, W.W., Flores, F.A., Henderson, J.A., 1992. Comparison of chalcocite dissolution in the oxygenated, aqueous sulfate and chloride systems. Miner. Eng. 5, 817–834. Fritz, J.J., 1981. Representation of the solubility of CuCl in solutio ns of various aqueous chlorides. J. Phys. Chem. 85, 890–894. 23

Journal Pre-proof Ge, X., Zhang, M., Guo, M., Wang, X., 2008. Correlation and prediction of thermodynamic properties of some complex aqueous electrolytes by the modified three-characteristicparameter correlation model. J. Chem. Eng. Data 53, 950–958. Gow, R.N., Huang, H., Young, C., 2016. Utility of mass-balanced EH-pH diagrams I – Applications of Gibbs’ Phase Rule. Miner. Metall. Process. 33, 58–67. Hashemzadeh, M., Dixon, D.G., Liu, W., 2019a. Modelling the kinetics of chalcocite leaching in acidified cupric chloride media under fully controlled pH and potential. Hydrometallurgy

of

189, 105114.

ro

Hashemzadeh, M., Dixon, D.G., Liu, W., 2019b. Modelling the kinetics of chalcocite leaching in acidified cupric chloride media under fully controlled pH and potential. Hydrometallurgy

-p

189, 105114.

re

Heinrich, C.A., Seward, T.M., 1990. A spectrophotometric study of aqueous iron (II) chloride

lP

complexing from 25 to 200 °C. Geochim. Cosmochim. Acta 54, 2207–2221. Herreros, O., Quiroz, R., Longueira, H., Fuentes, G., Viñals, J., 2006. Leaching of djurleite in

na

Cu2+/Cl− media. Hydrometallurgy 82, 32–39. Herreros, O., Viñals, J., 2007. Leaching of sulfide copper ore in a NaCl–H₂ SO₄ –O₂ media

ur

with acid pre-treatment. Hydrometallurgy 89, 260–268.

Jo

Hiroyoshi, N., Miki, H., Hirajima, T., Tsunekawa, M., 2000. A model for ferrous-promoted chalcopyrite leaching Naoki. Hydrometallurgy 57, 31–38. Huang, H.H., Twidwell, L.G., Young, C.A., 2005. Speciation for aqueous systems - Equilibrium calculation approach, in: International Conference on Computational Analysis in Hydrometallurgy, 35th Annual Hydrometallurgy Meeting, CIM. Calgary, Alberta, canada, Alberta, canada. Huang, H.H., Young, C. a., 1996. Mass-balanced calculations of Eh-pH diagrams using STABCAL, in: Woods, R., Richardson, P.E., Doyle, F.M. (Eds.), Fourth International Symposium on Electrochemistry in Mineral and Metal Processing. Electrochemical Society, Inc., New Jersey, p. 460. 24

Journal Pre-proof Hubli, R.C., Mkhlerjee, T.K., Venkatachalam, S., Bautista, R.G., Gupta, C.K., 1997. A chemical model to characterize the dissolution of sulfides in CuCl₂ , CuCl, FeCl₃ , FeCl₂ , NaCl, HCl, H₂ O system: Cupric chloride leaching of synthetic NiS. Miner. Met. Mater. Soc. 119–138. Kennish, M.J., 2000. Practical Handbook of Marine Science. Kim, S.H., Henein, H., Warren, G.W., 1986. An investigation of the thermodynamics and kinetics of the ferric chloride brine leaching of galena concentrate. Metall. Trans. B 17, 29–

of

39.

ro

Kimura, R.T., Haunschild, P. a, Liddell, K.C., 1984. A mathematical model for calculation of equilibrium solution speciations for the FeCl₃ -FeCl₂ -CuCl₂ -CuCI-HCI- NaCI-H₂ O

-p

system at 25 °C. Metall. Trans. B 15, 213–219.

lP

Int. J. Miner. Process. 2, 105–115.

re

Kusik, C.L., Meissner, H.P., 1975. Calculating activity coefficients in hydrometallurgy-review.

Lin, C.L., Tseng, H.C., Lee, L.S., 1998. A three-characteristic-parameter correlation model for

na

strong electrolyte solutions. Fluid Phase Equilib. 152, 169–185. Liu, W., Granata, G., 2018. Temperature control in copper heap bioleaching. Hydrometallurgy

ur

176, 26–32.

Jo

Liu, W., Moran, C.J., Vink, S., 2013. A review of the effect of water quality on flotation. Miner. Eng. 53, 91–100.

Liu, W., Moran, C.J., Vink, S., 2011. Quantitative risk-based approach for improving water quality management in mining. Environ. Sci. Technol. 45, 7459–7464. McDermand, L., 2015. Measuring ORP on YSI 6-series sondes: Tips, cautions and limitations. Ohio, USA. Meissner, H.P., Tester, J.W., 1972. Activity coefficients of strong electrolytes in aqueous solutions. Int. J. Miner. Process. 11, 128–133. Miki, H., Nicol, M., Velásquez-Yévenes, L., 2011. The kinetics of dissolution of synthetic 25

Journal Pre-proof covellite, chalcocite and digenite in dilute chloride solutions at ambient temperatures. Hydrometallurgy 105, 321–327. Muir, D.M., 2002. Basic principles of chloride hydrometallurgy, in: Chloride Metallurgy 2002: Practice and Theory of Chloride/Metal Interaction, 32nd Annual Hydrometallurgy Meeting. Canadian Institute of Mining, Metallurgy and Petroleum, Montreal, pp. 759–792. Nicol, M., Miki, H., Zhang, S., 2017. The anodic behaviour of chalcopyrite in chloride solutions :

of

Voltammetry. Hydrometallurgy 171, 198–205. Parker, a. J., Paul, R.L., Power, G.P., 1981. Electrochemical aspects of leaching copper from

ro

chalcopyrite in ferric and cupric salt solutions. Aust. J. Chem. 34, 13–34.

-p

Pitzer, K.S., Silvester, L.F., 1977. Thermodynamics of electrolytes. XI. Properties of 3-2, 4-2,

re

and other high-Valence types. J. Phys. Chem. 1–15.

lP

Renon, H., 1986. Electorlyte solutions. Fluid Phase Equilib. 30, 181–195. Ruiz, M.C., Honores, S., Padilla, R., 1998. Leaching kinetics of digenite concentrate in

na

oxygenated chloride media at ambient pressure. Metall. Mater. Trans. B 29, 961–969. Ruiz, M.C., Honores, S., Padilla, R., 1997. Leaching of a chalcocite concentrate with cupric

ur

chloride-oxygen. TMS Annu. Meet. 139–147.

Jo

Scatchard, G., 1976. Equilibrium in Solutions and Surface and Colloid Chemistry. Harvard University Press, Cambridge. Schnell, H.A., 1997. Theory, Microbes and Industrial Processes, in: Rawlings, D.E. (Ed.), Biomining: Bioleaching of Copper. Springer, Berlin Heidelberg, pp. 21–43. Sharma, V.K., Millero, F.J., 1990. Equilibrium constants for the Formation of Cu(I) Halide Complexes. J. Solution Chem. 19, 375–390. Sherman, D.M., 2007. Complexation of Cu+ in hydrothermal NaCl brines: Ab initio molecular dynamics and energetics. Geochim. Cosmochim. Acta 71, 714–722. Sugasaka, K., Fujii, A., 1976. A spectrophotometric study of copper (I) chloro-complexes in 26

Journal Pre-proof aqueous 5M Na(Cl, ClO₄ ) solutions. Bull. Chem. Soc. Jpn. 49, 82–86. Wang, M., Zhang, Y., Muhammed, M., 1997. Critical evaluation of thermodynamics of complex formation of metal ions in aqueous solutions I. A description of evaluation methods. Hydrometallurgy 45, 21–36. Xiong, Y., 2006. Estimation of medium effects on equilibrium constants in moderate and high ionic strength solutions at elevated temperatures by using specific interaction theory (SIT): Interaction coefficients involving CI, OH− and Ac− up to 200 °C and 400 bars. Geochem.

of

Trans. 7.

ro

Yue, G., Zhao, L., Olvera, O.G., Asselin, E., 2014. Speciation of the H₂ SO₄ -Fe₂ (SO₄ )₃ FeSO₄ -H₂ O system and development of an expression to predict the redox potential of the

-p

Fe3+/Fe2+ couple up to 150 °c. Hydrometallurgy 147–148, 196–209.

re

Zhao, H., Chang, J., Boika, A., Bard, A.J., 2013. Electrochemistry of high concentration copper

lP

chloride complexes. Anal. Chem. 85, 7696–7703.

Zhao, R., Pan, P., 2001. A spectrophotometric study of Fe(II)-chloride complexes in aqueous

ur

na

solutions from 10 to 100°C. Can. J. Chem. 79, 131–144.

Jo

Highlights Fe and Cu species were identified using activity models for high ionic strength. Empirical equations for calculating solution potential in chloride were derived. Cu2+, CuCl+, CuCl2–, and CuCl32– were identified as the predominant copper species. The predominant iron species were identified as FeCl 2+ and Fe2+. Reactions responsible for copper leaching in elevated chloride were proposed.

27