~

'ZT ~.

!~r'!ll
-jr, 2.:,

hydrometallurgy
Hydrometallurgy 46 (1997) 157-169

ELSEVIER

Electro-oxidation of pyrite in sodium chloride solutions

Fatma Arslan a,,, Paul F. Duby b
a Istanbul Technical University, Mining Engineering Department, Mineral and Coal Processing Division, 80626 Maslak, lstanbul, Turkey b Columbia University, Henry Krumb School of Mines, New York, NY 10027, USA

Received 9 August 1996; accepted 16 December 1996

Abstract
The mechanism of the electro-oxidation of pyrite in NaCI solutions was investigated. Laboratory work was carried out on the production of hypochlorite and the oxidation of pyrite in an electrochemical cell with a 10% sodium chloride electrolyte at 35-40C. The production rate of hypochlorite was higher when the pH was not controlled than when it was maintained at the initial value of about 6.5. Pyrite was oxidized by the electrochemically generated hypochlorite, with an overall current efficiency of about 97%. The oxidation rate was directly related to the current and it was not affected by the pulp density.

1. Introduction Pyrite is a major sulphur-bearing mineral in some refractory gold ores and in coal. The cyanide leaching o f some refractory gold ores is incomplete because a large portion o f the fine gold remains locked within the host mineral and never comes in contact with the cyanide solution. Finely disseminated gold can be found in sulphides (pyrite and arsenopyrite) or siliceous minerals. In several ore deposits a significant portion of gold occurs in submicroscopic form, and even fine grinding fails to liberate the gold. Therefore, it is necessary to break down the crystal lattice o f the host mineral in order to liberate the gold. The Carlin-type N e v a d a ores, for instance, contain up to 3% pyrite [1].

* Corresponding author. Tel.: [email protected]

+90

212 285 6357. Fax:

+90

212 285 6863. E-maih

0304-386X/97/$17.00 1997 Elsevier Science B.V. All rights reserved, PII S0304-386X(97)00009- l

158

F. A rslan. P.F. Duby / HydrometallLt rgy 46 (1997) 157-169

In the desulphurization of coal it is generally difficult to remove sulphur when it is in original crystalline FeS 2 form. Therefore, the conversion of pyrite to other more easily removable forms of sulphur constitutes a key issue in coal desulphurization [2]. Chlorination is one of the chemical desulphurization processes. Dissolution of pyrite in acidic chlorine solution was studied as early as 1957 [3]. Chlorine leaching of pure pyrite mineral has been studied by a number of investigators [2,4]. Chlorination is applied by sparging CI 2 gas through the solution, by using hypochlorite as an oxidant, or by producing hypochlorite ions electrochemically in saline solutions. Electro-oxidation was investigated by the U.S. Bureau of Mines, treating an ore pulp prepared from finely ground ore and NaC1 brine [5,6]. Initial tests were conducted on Carlin-type carbonaceous ores and several parameters, such as: salt concentration, electrolysis time, temperature, current density, type of electrodes, electrode spacing, particle size of the ore, and pulp density were investigated. In the studies the optimum conditions were found to be: temperature around 40C, salt concentration 10%, pulp density 40%, and current density 775-1163 A / m 2 of anode surface. Although the chlorination of pyrite has been studied earlier, there has been no work on the electrooxidation of pyrite or pyritic ores. In our studies, an electro-oxidation process, which has not been applied to pyrite before, was used in the oxidation of pyrite. The purpose of this investigation is to achieve a better understanding of the electro-oxidation mechanism of pyrite for the pre-treatment of refractory gold ores and for the desulphurization of coal. Therefore, the electrochemical production of hypochlorite, an important part of oxidation, and the electro-oxidation of pyrite were investigated separately.

2. Experimental procedure
The experimental arrangement is shown in Fig. 1. The cell consisted of separate anolyte and catholyte compartments (500 ml each) connected through a porous glass frit. Both compartments were closed to the atmosphere. A circulation of nitrogen gas removed the reaction products of the anodic process. The outgoing gas stream was sparged through a 0.5 M sodium hydroxide solution to absorb the chlorine and the carbon dioxide. A current in the range 0.5-1.5 A (400-1200 A / m 2 of anode surface) was applied to the cell using a D.C. power supply. A copper coulometer was placed in series with the cell. A P t / I r (70/30) coated (5 g / m 2) titanium plate was the anode and a spiral-shaped platinum wire served as cathode. In all experiments 10 wt% NaC1 solutions were used and in the case of pyrite oxidation, 1-5 wt% pyrite in the range - 2 7 5 + 325 mesh (from Huanzala, Peru) was added to the anolyte. The temperature was measured during the experimental runs. It was found that in 1 h the anolyte temperature went up to about 40C and remained constant. Since the optimum temperature has been found to be in the range 35-40C by the USBM and others, there was no need to control the temperature during the tests [2,5]. Cell potentials, anodic potentials (versus a saturated calomel electrode, SCE), and anolyte pH were also recorded during the runs. Anolyte pH was adjusted with 5.0 M

F. Arshm, P.F. Duby / Hydrometallurgy 46 (1997) 157-169

159

Fig. 1. Experimental arrangement. C = cathode; A = anode; 1 = copper coulometer; 2 = nitrogen gas tank; 3 = saturated calomel electrode; 4 = pH electrode; 5 = sampling; 6 = 0.5 M NaOH solution; 7 = catholyte; 8 = glass frit; 9 = anolyte; 10 = saturated CaO solution; 11 = thermometer. N a O H whenever required. Anolyte and catholyte solutions were stirred in order to provide good mixing. The cell was designed in such a way that 5 ml of sample solution could be withdrawn for analysis at known time intervals. Anolyte, catholyte and N a O H solutions were analyzed for chloride by potentiometric titration with silver nitrate, and for hypochlorite by iodimetry. In the case of pyrite oxidation the anolyte and catholyte solutions were analyzed for total iron using a spectrophotometer (Hach, D R / 2 0 0 0 ) at a wavelength of 510 nm. The solid residue at the end of each run was weighed.

3. Experimental results and discussions

Preliminary experiments with the one-compartment cell were carried out to test three different sets of a n o d e - c a t h o d e : graphite-copper, graphite-graphite, and P t / I r - c o a t e d t i t a n i u m - c o p p e r [7]. Hypochlorite production was almost the same in all cases. When a graphite anode was used it was oxidized with production o f CO 2 and mechanical degradation occurred with time. A precipitate was observed during the experiments with a copper cathode. The P t / I r - c o a t e d Ti anode was found to be chemically and dimensionally stable in this system and was used in subsequent tests. Experimental results are summarized in two parts: hypochlorite production and pyrite oxidation.

160

F. Arslan, P.F. Duby / Hydrometallurgy 46 (1997) 157-169

3.1. Hypochlorite production

The chemical reactions which occur during the electrolysis of NaC1 solution can be expressed as follows. The chloride ion is converted to chlorine at the anode: 2C1 - --->C12 q- 2eAt the cathode, water is hydrolyzed to produce hydroxyl ion and hydrogen: 2H20 + 2 e - ~ H 2 + 2 O H (2) ( 1)

At the anode, hypochlorite is produced according to the following reactions. Chlorine reacts with water to produce hypochlorous acid:
Cl 2 + H 2 0 ~ HC10 + H + + C1-

(3)

then, the dissociation of hypochlorous acid to hypochlorite and hydrogen ions depends on the solution pH: HC10 + C 1 0 - + H + (4)

and the hydroxyl ions produced at the cathode react with hydrogen ions to form water: H++ OH---) H20 (5)

After a certain time, hypochlorite production levels off and this is attributed to the production of chlorates [8] either chemically: 2HCIO + C10- --->C10 3 + 2H + + 2C1or electrochemically: 6C10- + 3H20 --->2CIO 3 + 6 H + + 4C1-+ 702 + 6eto: 3

(6)

(7)

Electrochemical reduction of hypochlorites at the cathode is also possible according C 1 0 - + 2 H + + 2e---) H 2 0 q- C I (8)

The anodic oxidation of water consumes part of the charge passing through the cell: 2H20 --> 4 H + + 02 + 4 e (9)

Analysis of catholyte at the end of each run shows some hypochlorite, which diffused from the anodic compartment. The production of hypochlorite can be expressed as:

( M / n ) . I. t
m F ~ (10) and the hypochlorite production rate per unit volume is given as: a =

( M / n ) . I. e F'V

(11)

where: m = total amount of hypochlorite produced (g); M = molar mass (g mol - l ); n = electrons exchanged per mole (n---2); F = Faraday constant (96500 C mol-1);

F. Arshm. P.F. Duby/Hydrometallurgy 46 (1997) 157-169

161

V = anolyte volume (500 cm3); 1 = applied current (A); t = time (s); e = current efficiency. The rate of hypochlorite loss due to chlorate formation, diffusion to catholyte, and cathodic reduction is assumed proportional to concentration: /3C. The change in hypochlorite concentration in the electrolyte (or anolyte) is equal to the difference between the production rate and the rate of hypochlorite loss, and is given as: dC
dt

-~-/3C

(12)

Assuming further that the parameters c~ and /3 do not depend on time and remain constant for a given experiment, integration of this gives:

C=~(1-e -t~')

OL

(13)

Fig. 2 illustrates the hypochlorite production versus time. Lines in the figure are drawn according to Eq. (13). When the anolyte pH was controlled at about 6.5, the hypochlorite production increased sharply with time at the beginning but levelled off after 6 h. In the case of pH control, the hypochlorite production increased gradually with time and started to level off after 12 h, reaching higher values than in the pH controlled case. Curve 2 in Fig. 2 represents four different runs under the same experimental conditions, showing good reproducibility. At the end of the runs the hypochlorite production was twice that of the case with pH control. When solution pH was not controlled, Eq. (13) fits poorly to the experimental data (dotted line). This may be due to the continuous change in the solution pH. Therefore, the values of c~ and /3 are not constant with time. In the pH-controlled case, Eq. (13) fits the experimental data very well. Since pH remains constant throughout the run, c~

0.20

1. pH controlled

(ovg

6.5)

xJ

E
0.15

2. pH not controlled"

E
c0

o.lo

/~ I I / " 0//

am ,~El 3 13 []

co9 o

8
o
!

0.05

1 u
2 0 ~ = A 0.00 5
I

II0
Time

I 15

I 20

25

/ h

Fig. 2. Hypochlorite production vs time. * pH change with time is given in Fig. 3a.

162

F. Arslan, P.F. Duby / Hydrometallurgy 46 (1997) 157-169

Table 1 Parameters for hypochlorite production

(107/gcm 3 s - l )

(105/s-I) 10.64 1.60

(%) 75 34

pH control no pH control

3.972 1.796

and /3 values do not change with time. Table 1 summarizes a and /3 values for both cases. The current efficiency was found to be 75% in the pH-controlled case, compared to 34% when the solution pH was not controlled. Current efficiency may be lost due to the anodic oxidation of hypochlorite (production of C103), the anodic evolution of oxygen from water and the diffusion of hypochlorite through the frit to the catholyte (Eqs. (6)-(9)). An analysis of the catholyte at the end of each run shows some hypochlorite that diffused from the anolyte, supporting the assumption of hypochlorite reduction at the cathode. The pH change in the anolyte with time is shown in Fig. 3a. As soon as the current was started the anolyte pH decreased from about 5.5 to 2.5 in 1 h; it increased gradually to the initial pH at the end of 10 h and stayed there for the remainder of the run. The catholyte pH was about 13 at the end of the run in both cases. According to Eq. (2), C12

5 I o. ~ o c .< 4 3 2

(o)
0

I 5

110

I 15

2fO

25

Time / h
>

bJ C) b3

1.6
1.5

-6 1.4
tO "~ 1.3 a_ -~ 1.2 o

(b)
I 5 I 10 [ 15 I 20 25

Time /

Fig. 3. (a) Anolyte pH and (b) anode potential changes with time during the production of hypochlorite.

F. Arslan, P.F. Duby / Hydrometallurgy 46 (1997) 157-169

163

produced at the anode reacts with water, resulting in the production of HCIO along with H + and C1-. This explains the pH decrease when the current was switched on, as seen in Fig. 3a. As soon as O H - ions (produced at the cathode) start to diffuse toward the anolyte, they react with H + ions, as in Eq. (5), and the anolyte pH starts to rise until it reaches the initial pH value and stays constant until the end. The anode potential change with time during the hypochlorite production is shown in Fig. 3b. It decreases from about 1.5 V (vs. SCE) to 1.4 V and remains approximately constant. Anode potential and anolyte pH measurements show that hypochlorite is produced according to the reaction in Eq. (3).

3.2. Electro-oxidation of pyrite

The oxidation of pyrite with hypochlorite generated electrochemically in the cell can be expressed as follows: 2FeS_~ + 15C10- + H 2 0 + 2H + ~ 2Fe 3+ + 4HSO 4- + 15C12FeS 2 + 15HC10 + H 2 0 ~ 2Fe 3+ + 4HSO4- + 15C1- + 13H (14) (15)

Pyrite oxidation with hypochlorite is a chemical reaction, as given in reactions (14) and (15), but the production of chlorine, resulting in hypochlorite production, is an electrochemical process. When the overall reaction is considered, 7.5 moles of HCIO are required for the oxidation of 1 mole of pyrite, corresponding to 15 electrons. In all experiments there was a linear relationship between the total iron in the solution and time (Figs. 4 and 7). It can be expressed as follows: C = c~. t + C 1 dC
- = a

(16)
(17)

dt

(M/n). I F.V

(18)

where: c~=oxidation rate (g cm -3 s - l ) ; C - t o t a l iron concentration (g c m - 3 ) ; Cj = integration constant; M = molar mass (g m o l - ~); F = Faraday constant (96500 C m o l - l ) ; n = electrons exchanged per mole of pyrite oxidation (n = 15); V = anolyte volume (500 cm3); I = applied current (A); t = time (s); ep = current efficiency for pyrite oxidation. Theoretically, C 1 should be equal to zero at t = 0, but it was found experimentally to be slightly higher than zero (less than 0.05). This may be due to the possible existence of pyrrhotite in the supposedly pure pyrite. Oxidation of pyrrhotite with hypochlorite could be stated as: 2FeS + 9HC10 ~ 2Fe 3+ + 9C1- + 2HSO 4 + H 2 0 + 5H (19)

Fig. 4 shows the change in iron concentration in the solution with time at different applied currents. Iron concentration in the solution increased linearly with time at all currents and all lines were drawn according to Eq. (16). This indicates that pyrite is

164

F Arslan, P F Dub},/f4ydrometallurgy 46 (1997) 157-169 4.0

7 E ~.s "{2

I/A
z, 0.5 o 1.0

3.0

._o 2.5 t- 2.0

0
0 0 tO
{-

t-

1.5

p 1.0
.~ 0.5 0 F-0.0

4 Time

6 / h

10

12

Fig. 4. The effect of the applied current on the change in total iron in solution with time

0.05

?
"0

oO.5A

E E
tO

"5

0.04 0.03
0.02

ml.0A t, 1.5 A

k=

E
tO

O tO

0.01z
pr ,

x 2
j i 0 i I

o I

o.0o ~

8 8 10
I t I I I

12

Time /

Fig. 5. Residual hypochlorite production vs. time.

Table 2 Pyrite oxidation results Current (A) 0.5 1.0 1.5 a (108/g cm- 3 s3.5 7.5 10,6

1)

ep (%) 91 97 91

Weight loss (070) 34.4 50.0 74.0

F Arslan. P.F. Duby/Hydrometallurgy 46 (1997) 157-169

1 gO

165

80

6O o d

.Y:
~

40

//
p

20
p / / F

.0

0.5
Total

1.0

1.5

2.0

Current /

Fig. 6. Weight loss vs. time during pyrite oxidation.

10

?
"ID

E
8

1 wt 5 wt~ wt~

Pyrite

o
tLD

6
/

~Y

c(J O O C / /

p.o /

--

-6
-g

0~

1~0
Time

I
15 2~0

25

Fig. 7. Effect of pulp density on the pyrite oxidation with time.

Table 3 Parameters for residual hypochlorite production Current ~ fl ~R

(A)
0.5 1.0 1.5

(108/g cm-3 s - l )
3.16 4.98 7.40

(105//s - 1)
3.17 2.48 2.65

(~7o)
12 9 9

166

F. Arslan, P.F. Duby / Hydrometallurgy 46 (1997) 157-169

6 ]

(o)

4 I o.

<

-2

I I0

12

Time / h
> 1,8 r,i 0 > L~ 1.7 1.6

~. 0 . 5 A

(b)

~.5
1.4 1.3 1.2
) 2 I 4 f 6 I 8 T 10

-o 0 m 0) o .~

Time / h

Fig. 8. (a) Anolyte pH and (b) anode potentialchanges duringpyrite oxidation.

oxidized at a constant rate. As the current increased concentration lines were shifted to higher concentrations. The oxidation rates ( a ) and current efficiencies that were calculated from c~ are given along with the weight losses in Table 2. A current of 1 A gave the highest current efficiency. Weight loss increased linearly in proportion to the current applied, as shown in Fig. 5. There was a good agreement between solution analysis and weight loss (about + 2%). The analysis of anolyte showed that the HCIO produced was not completely used during the electro-oxidation of pyrite. The concentration of residual HCIO was plotted against time at all currents in Fig. 6. Curves through the experimental points were drawn according to Eq. (13), as explained in the section dealing with hypochlorite production. Values of a, /3, and current efficiency (~R) for residual hypochlorite are summarized in Table 3. As seen from Fig. 6, the residual hypochlorite concentration started to level off after 8 h, which shows that the hypochlorite produced after that point is completely consumed by pyrite. Eq. (13) fits well the residual hypochlorite concentration change with time but it fits poorly in the production of hypochlorite itself when the solution pH was not controlled. The total current efficiency (Tables 2 and 3) was about 100% in all cases. These results indicate that there was no current efficiency loss, in contrast to the hypochlorite production itself. This means that the hypochlorite produced in the cell was

 Arslan, P.F. Duby/Hydrometallurgy 46 (1997) 157-169

167

Fig. 9. SEM micrographs of pyrite particles before oxidation.

consumed continuously for pyrite oxidation. This is an advantage over chlorination and hypochlorite leaching processes. As seen in Fig. 7, the pulp density has no effect on the pyrite oxidation. Oxidation rates and current efficiencies were almost the same in all cases. These results indicate that pyrite oxidation is limited by the production of hypochlorite. Anolyte pH change with time during the pyrite oxidation is shown in Fig. 8a. Anolyte pH decreased very sharply from 5 to 1 in 1 h and continued to decrease slowly until the end. It followed the same trend in all cases. This is due to the continuous production of H +, according to Eq. (15). Anode potentials were recorded versus time as shown in Fig. 8b. As current increased the potential curves were shifted to higher values. Increasing pulp density did not affect the anode potentials. Anolyte pH and anode potential measurements are consistent with the oxidation of pyrite to Fe 3+ according to Eq. (15). Analysis of anolyte also showed no Fe 2 in the solution, in agreement with the electrochemical potential measurements. Fig. 9 shows scanning electron micrographs (SEMs) of the pyrite used in the preliminary studies before the oxidation. Most of the small particles had disappeared

168

F. Arslan, P.F. Duby / Hydrometallurgy 46 (1997) 157-169

Fig. 10. SEM micrographs of pyrite particles after oxidation.

after the oxidation, as can be seen in Fig. 10. Some surface etching was also observed on the surface of particles after the oxidation.

4. Conclusions Electrochemical generation of hypochlorite in 10 wt% NaC1 solution is possible. Although the initial current efficiency is found to be higher (75%) when the solution pH is controlled at about 6.5, the total production of hypochlorite at the end of 24 h is lower than that when pH is not controlled. It was shown that pyrite can be oxidized by hypochlorite generated electrochemically in the cell, with up to 97% current efficiency. The oxidation of pyrite is dependent on the production of hypochlorite and is a function of current. Increasing pulp density has no effect on the oxidation. Results indicate that pyrite can be efficiently oxidized by electro-oxidation. This process could be used efficiently and economically for the treatment of refractory gold ores or for the desulphurization of coals.

F. Arslan, P.F. Duby / Hydrometallurgy 46 (1997) 157-169

169

Acknowledgements
F. Arslan is grateful to the T u r k i s h G o v e r n m e n t for financial support. This research has been partially supported by the D e p a r t m e n t of the I n t e r i o r ' s Mineral Institutes Program, administered by the B u r e a u of M i n e s under a l l o t m e n t grant n u m b e r G 1184136.

References
[1] D.M. Hausen, Process Mineralogy of Auriferrous Pyritic Ores at Carlin, Nevada, in: D.M. Hausen, W.C. Parks (Eds.), Process Mineralogy, 1981, TMS-AIME, Warrendale, PA, 1981, pp. 271-289. [2] S. Bayrakceken, S. Yasar, S. Colak, Kinetics of the chlorination of pyrite in aqueous suspension, Hydrometallurgy 25 (1990) 27-36. [3] M.I. Sherman, J.D.H. Stnckland, Dissolution of pyrite ores in acid chlorine solutions, J. Met. (1957) 1386-1388. [4] N.P. Vasilakos, C. Gage, Slurry-phase chlorination of iron pyrite in aqueous and organic media, Fuel 65 (1986) 593-594. [5] B.J. Scheiner, R.E. Lindstrom, T.A. Henrie, Processing refractory carbonaceous ores for gold recovery, J. Met. (1971) 37-40. [6] B.J. Scheiner, R.E. Lindstrom, T.A. Henrie, Process for recovering gold from carbonaceous ores, U.S. Pat. No. 3,639,925 (1972). [7] F. Arslan, Electrooxidation of refractory gold ores, Ph.D. Thesis, Columbia Univ., 1991. [8] L.P. Eblin, Chemistry - - A Survey of Fundamentals, Harcourt, Brace and World, 1968, pp. 430-431.