Top Banner
hydrometallwgy Hydrometallurgy 38 (1995) 125-147 Kinetics of the aqueous chlorination of gold in suspended particles J. Vifials”, C. N6fiezb,t, 0. Herrerosc,l “Department of Chemical Engineering and Metallurgy, Universify of Barcelona, Martii FranquZs I, 08028, Barcelona, Spain “E.T.S. de Ingenieros Industriales, Universidade da Coruria, Mendizabal s/n, Esteiro, 1.5403 Ferrol, La Coruti, Spain ‘Department of Mining Engineering, Universidad de Antofagasta, Antofagasta, Chile Received 14 January 1994; revised version accepted 10 June 1994 Abstract The effects of the particle geometry and size, stirring speed, pH, chlorine and chloride concentrations and temperature on the kinetics of the aqueous chlorination of gold were studied. Surface examinations of gold samples leached in different conditions were carried out by SEM/EDS and XPS. The kinetics appear to be chemically controlled in conditions of full suspension of the particles. The dependence of pH and the chlorine and chloride concentrations indicate that the rates were essentially determined by the concentration of the trichloride ion. The expression obtained for the specific rate was: Rate (mol cm _ * min -I) =4.25 lo5 [Cl; ] exp (-43.5/RT) A discussion of the possible mechanism is included, based on the kinetic dependence, previously reported electrochemical studies and mixed potential measurements. 1. Introduction Aqueous chlorination was the first hydrometallurgical process for the recovery of gold [ I] but it was rapidly displaced because of the obvious advantages of the cyanidation process for the treatment of non-refractory gold ores. Thus, gold+hloride hydrometallurgy is mainly restricted to the electrorefining of gold [2] and numerous studies have been reported on the electrochemical behaviour of gold in chloride media [ 3-71. Leaching of gold in chlorine solutions, however, has been used for cyanicide-containing products, such Present address: Department of Chemical Engineering and Metallurgy, University of Barcelona, Marti i Franquks 1.08028, Barcelona, Spain. 0304-386X/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDlO304-386X(94)9999-2
23
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

hydrometallwgy

Hydrometallurgy 38 (1995) 125-147

Kinetics of the aqueous chlorination of gold in suspended particles

J. Vifials”, C. N6fiezb,t, 0. Herrerosc,l “Department of Chemical Engineering and Metallurgy, Universify of Barcelona,

Martii FranquZs I, 08028, Barcelona, Spain “E.T.S. de Ingenieros Industriales, Universidade da Coruria, Mendizabal s/n,

Esteiro, 1.5403 Ferrol, La Coruti, Spain ‘Department of Mining Engineering, Universidad de Antofagasta, Antofagasta, Chile

Received 14 January 1994; revised version accepted 10 June 1994

Abstract

The effects of the particle geometry and size, stirring speed, pH, chlorine and chloride concentrations and temperature on the kinetics of the aqueous chlorination of gold were studied. Surface examinations of gold samples leached in different conditions were carried out by SEM/EDS and XPS.

The kinetics appear to be chemically controlled in conditions of full suspension of the particles. The dependence of pH and the chlorine and chloride concentrations indicate that the rates were essentially determined by the concentration of the trichloride ion. The expression obtained for the specific rate was: Rate (mol cm _ * min -I) =4.25 lo5 [Cl; ] exp (-43.5/RT) A discussion of the possible mechanism is included, based on the kinetic dependence, previously reported electrochemical studies and mixed potential measurements.

1. Introduction

Aqueous chlorination was the first hydrometallurgical process for the recovery of gold [ I] but it was rapidly displaced because of the obvious advantages of the cyanidation process for the treatment of non-refractory gold ores. Thus, gold+hloride hydrometallurgy is mainly restricted to the electrorefining of gold [2] and numerous studies have been reported on the electrochemical behaviour of gold in chloride media [ 3-71. Leaching of gold in chlorine solutions, however, has been used for cyanicide-containing products, such

’ Present address: Department of Chemical Engineering and Metallurgy, University of Barcelona, Marti i Franquks 1.08028, Barcelona, Spain.

0304-386X/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDlO304-386X(94)9999-2

Page 2: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

126 .I. Virials et al. / Hydrometallurgy 38 (1995) 125-147

as pyrite cinders [ 8,9] and is an alternative for the processing of a variety of refractory gold ores [ 10,111.

The kinetics of leaching of gold in aqueous chlorine was mainly studied in the 1970s by Russian authors [ 12-151, who basically concluded that the reaction is controlled by trans- port but with a complex dependence of the chloride concentration, acidity and temperature. The activation energy values reported were between 5.944.8 kJ/mol ( 1.4-10.7 kcal/mol), depending on the experimental conditions. None of these studies, however, correlated the kinetic data with an accurate determination of the distribution of the chlorine species and gold surface studies in support of some formulations of the mechanism are not reported. On the other hand, most of these studies were carried out using the rotating disc technique, which has the advantage of a well defined surface area and flux pattern [ 161 but has the disadvantage that the mass transfer coefficients in the Levich regime are relatively small (oftheorderof lo-‘ems-‘at500min-’ for aqueous solutions at 25°C) and, consequently, some leaching systems tend to exhibit transport or mixed control kinetics under these conditions. In these cases the study of the leaching systems in stirred reactors with particle suspension could have the advantage of showing chemical kinetics, since the mass transfer coefficients for very dense, suspended particles [ 171 are almost one order of magnitude higher than in the Levich regime.

The need to re-examine the fundamental aspects of halogen leaching in gold hydromet- allurgy seems evident. Pesic and Sergent [ 181 have recently published the kinetics and mechanism of the aqueous bromation of gold and Angelidis et al. [ 191 the kinetics of dissolution of gold in iodine/iodide solutions. The present paper is a kinetic study on the aqueous chlorination of suspended particles with the aim of clarifying the type of reaction control in this leaching regime and correlating the kinetic data with the solution chemistry.

2. Materials and experimental procedure

2.1. Materials

Pure gold (99.9%) was used in all the leaching experiments. In order to determine the possible effect of the particle geometry, two extreme types were used: (1) flat-packs measuring 1 mm X 1 mmX 16 pm, which were obtained by cutting gold sheets; and (2) spheres which were obtained by cutting gold wires (0.2 mm diameter) in a microtome (5- 15 /*rn aperture) and placing the resulting cylinders in a coupelle type crucible at temper- atures just above the melting point of gold. The resulting spheres were sieved in order to obtain families of particles of set sizes: 180-200; 100-140; 80-100; and 60-80 pm.

2.2. Leaching experiments

Leaching experiments were carried out in a spherical glass reactor immersed in a ther- mostatically controlled water bath equipped with a constant temperature circulator. The reactor was equipped with a stirrer of teflon coated steel with a pitched blade turbine (5.5 cm diameter) connected to a variable speed motor. Except when indicated, 390 and 900 min-’ were used for the suspension of the flat-packs and spheres, respectively. The potential

Page 3: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

.I. Virials et al. / Hydrometallurgy 38 (1995) 125-147 121

of the solution was measured continuously by a platinum electrode and a reference electrode of Ag/AgCl (3 M KCl) . The reference electrode was separated from the solution by a salt bridge. The pH of the solution was also measured. The reactor was hermetically closed by a cover supporting the electrodes, stirrer, salt bridge, thermometer and the sampling device. A small volume of air ( - 50 cm”) remained over the solution ( 1200 cm”).

Leaching experiments were designed for a constant concentration of total chlorine in solution. This was obtained by generating Cl, ‘in situ’ from the reaction between commercial sodium hypochlorite and hydrochloric acid:

NaClO + NaCl + 2HCl-+ 2NaCl+ Cl, + H, 0 (1)

A measured volume of a stock solution of dilute hypochlorite (6.34 X 10e2 M in total Cl,) was placed in the reactor and a controlled volume of dilute hydrochloric acid ( 1.3 M) was added up to the pH selected. Distilled water was added to complete the solution volume to I200 ml. In some experiments NaCl was added to increase the total chloride concentration. The initial pH, potential and concentration of total chlorine (determined by the iodide/ thiosulphate method) was measured. A series of preliminary experiments were performed to contrast the stability of the chlorine concentration in the ranges of temperatures and concentrations used. It was observed that in the first 5 min there was a loss of chlorine of 4-15%, due to saturation by the gas layer over the solution. Thereafter, however, the concentration of chlorine was practically constant (loss < 5%) during the times involved in this investigation ( - 1 h) and the potential also remained constant to + 3 mV.

Leaching experiments were therefore initiated 5 min after preparation of the solution. A total of 6 mg of Au was added and 5 ml samples of solution were withdrawn at selected times and analyzed for gold concentration by atomic absorption spectrophotometry using matrix-matched standards. The fraction reacted, LY, was determined by the ratio of gold concentration at time t to the concentration when it was completely dissolved.

In order to minimize the formation of Cl< (see section 3) some experiments were performed in the same way but using solutions obtained by bubbling Cl, (gas) through dilute H,SO,.

2.3. SEMEDS examination

The morphology of attack and possible formation of passivating layers of reaction prod- ucts in some leaching conditions were examined by conventional scanning electron micros- copy (SEM) coupled to an X-ray energy dispersive spectrometer (EDS).

2.4. XPS analysis

The original gold samples (flat-packs) as well as those obtained in different attack media were studied in a surface analysis chamber (Perkin Elmer, PHI 5500) configured for X-ray photoelectron spectroscopy. The excitation source was Al K, (1486 eV) and the pass energy in the spectrometer was 93.9 eV for the survey analysis and 11.75 eV for individual peak analysis. Sputtering deep profiles were also recorded using a beam of Ar ions (4 KeV, 6.7 PA/mm’) with an incident angle of 45”. The time of sputtering was 0.5 min in each cycle. Peaks were assigned by reference to the base data in the Handbook of X-ray Photo-

Page 4: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

128 J. Virkls et al. /Hydrometallurgy 38 (1995) 125-147

Fig. I. Distribution of chlorine species as a function of pH. Conditions: 18°C. Cc,? = I .27 X 10 m-2 M, C,, _ = 0.48 M.

electron Spectroscopy [ 201. Quantifications were performed using the sensitivity factors of the Au4,., C,,, Or, and Cl,, peaks [ 211.

The samples of original gold were not preserved from the atmospheric environment in the same way as those used in the kinetic experiments.

Leached samples were extracted from the solution and placed in the chamber, after draining the liquid film retained by placing one corner in contact with a filter paper. All samples were examined at an analyzing pressure of lop7 Pa and re-examined 24 h later, after the samples had been maintained in the chamber at this pressure. No significant change was observed in the intensity or position of any of the peaks.

3. Distribution of chlorine species

The distribution of chlorine species in the range of concentrations, pH and temperatures used in this study was computed through of the equilibria involved [ 22,231:

Cl, +H,O=H+ +Cl- +HClO K,,(29to =3.94x 1O-4 (2)

HClO=H+ +ClO- Kd(z98j =7x lo-” (3)

c1,+c1-=c1; K,,,,,, =1.95x10-’ (4)

The mass balance equations for total chlorine and total chloride can be written as:

C,,z = [Cl21 + [Cl,] + [HClO] + [ClO-] (5)

Cc,- = [Cl-] + [HClO] - [Cl,] (6)

where Cc12 is the total chlorine concentration determined iodometrically and Cc,- is the

Page 5: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Viiials et al. /Hydrometallurgy 38 (1995) 125-147 129

s ; lo-3_ z % E

-4 ZlO -

r ”

Fig. 2. Distribution of chlorine species as a function of total chloride. Conditions: I8”C, Cc,, = 1.27X IO -’ M, pH 1.04.

total chloride added to the solution as HCI and NaCl. If the [H] + of the solution is fixed and Cc,? and C,,- are known, the system can be solved considering that the activity coefficients are close to 1.

Fig. 1 shows the distribution of chlorine species as a function of pH for a solution of Cc,, = 1.27 X 10 ~’ M and Cc, - = 4.8 X 10 - ’ M. Molecular chlorine is the predominant species at low pH, but hydrolysis is already appreciable at pH > 2 and HClO becomes the

24

Temperature (‘C)

Fig. 3. Distribution of chlorine species as a function of the temperature. Conditions: pH = 1.07, Cc,, = 1.27X IO --? M, c,.,- = 1.30x lo-’ M.

Page 6: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

130 J. Vin’als et al. / Hydrometallurgy 38 (1995) 125-147

predominant species for pH > 3. There is an important difference with the solutions of bromine [ 181, in which the formation of HBrO (K,, = 5.8 X 10m9) is negligible until pH > 5. The dissociation of hypochlorous acid is significant only at pH.6. Trichloride ion is a minor but significant species at low pH and values decrease as pH increases.

Fig. 2 shows the distribution as a function of total chloride for a solution of pH = 1.04 and C,-,2 = 1.27 X 10 * M. Under these conditions the formation of HClO is negligible. Trichloride ion increases when the total chloride increases to values of 23% of total chlorine at Co, = 1.4 M. There is another important difference with the bromine solutions (K, = 1.59 X lo’), in which tribromide ion becomes the predominant species at moderate and high bromide concentrations.

The effect of temperature for a solution of Cciz = 1.27 X 10 -* M, Cc, = 1.3 X 10 - ’ and pH = 1.07, is shown in Fig. 3. The constants were taken from Connick and Chia [ 241 for the hydrolysis reaction (Ki, = 4.05 exp( -2774/T)) and from Camacho et al. [ 251 for the association reaction (K, = 3.6 X lop3 exp( 1190/Z’)). The variation of molecular chlorine is not significant but trichloride ion decreases about 30% from 6” to 30°C and hypochlorous acid increases two-fold in the same range.

4. Results and discussion

4.1. Effect of the stirring speed

For flat-packs, the applied kinetic model was [ 16,261:

a=kexpt

where the specific rate can be computed as:

r= $cxppAueo

For spheres, the appropriate dissolution model was:

l-(l-cz)t=k~xpt

and:

(7)

(8)

(9)

r = LpPAu ro (10)

These models were consistent with the experimental results (Figs. 4, 5, 8, 11, 13 and 16).

The effect of the stirring speed on the specific rates is shown in Fig. 6. In the reactor used, the flat-packs became well suspended at > 390 min -’ and the largest spheres ( r, = 95 & 5 pm) at >, 800 min- ’ . In both cases the specific rates were independent of the stirring speed in conditions of full suspension. This effect is not conclusive for chemical reaction control, since in stirred reactors the mass transfer coefficients for very dense suspended particles are virtually independent of the agitation intensity [ 17,22,27,28]. How- ever, the fact that the specific rate is practically independent of the particle geometry (see also Figs. 12, 14, 17) in conditions of full suspension is strongly indicative of chemical (or electrochemical) control.

Page 7: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. ViCals et al. /Hydrometallurgy 38 (199s) 125-147

20 40 60 60 t (min)

0

Fig. 4. Effect of the stirring speed on the leachiq of flat-packs and spheres. Conditions: 18°C. C,.,, = 1.27X 10 e-Z M,pH=1.04.C ,.,. =1.3X10-‘M.

On the other hand, the experimental dissolution rates can be compared with those obtained

131

from appropriate models for mass transfer. For the global process:

Au,,, +3/2Cl,(,,, +CI,,, +AuC&,,,

: / Cl0 r0

0 =llZpm 0 = 95 I4

/, ,

A = 45 ”

_- _- 10 20

i/r0 (l/mm)

(11)

3

Fig. 5. Effect of the particle size for spheres in full suspension. Conditions: 18°C. Cc,? = 1.27 X IO-’ M,

C~.,-=0.13M.pH=1.04,900n~in~‘.

Page 8: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

132 J. Virials et al. / Hydrometallurgy 38 (1995) 125-147

0 flat-packs

0 Spheres

&3--e -o-o-o-

cl

0

0

1 1 1

250 350 450 /e

rpm

Fig. 6. Plot of specific rates as a function of the stirring speed. Conditions as in Fig. 4.

the kinetic expression for spherical particles, if the rate is limited by transport of aqueous chlorine, can be written as:

ID dr

I

bC

k At -=-

m(r) PA” r

(12)

where a reasonable approach for k,,(,., in the case of very dense, suspended spheres is [ 22,271:

k mCrj =Dlr+0.31[Apglv]tDi (13)

The substitution of Eq. ( 13) in Eq. ( 12) and subsequent integration makes it possible to predict the time conversion curves. Fig. 7 (solid line) is a plot for r,=45 pm and Cell = 1.27 X lo-* M, taking 8.58 X lop4 cm*/min for DCh and 0.6 cm*/min for the kinematic viscosity, v [ 291. The experimental data (dots) have been included in the same figure. The rate observed was 1 order of magnitude slower than that predicted for mass transfer, which was consistent with chemical reaction control.

4.2. Effect of pH

The effect of pH on gold dissolution is shown in Fig. 8. Fig. 9 shows the specific rates obtained from data in Fig. 8. For pH < 1.5 the rates were insensitive to pH but for pH > 1.5 the rates decrease progressively until the pH is -4. At higher pH values no significant dissolution of gold takes place. This effect can be attributed to the change in the distribution of chlorine species (Fig. 1) , indicating that HClO and ClO- do not have a significant role in the gold dissolution rate.

Page 9: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Virials et al. /Hydrometallurgy 38 (1995) 125-147 133

- transport model

?? experimental

. . .

1 1 0 10 20

t tmim

I

Fig. 7. Comparison of the experimental rate with that expected for transport control. Conditions: 18°C. r, = 45 ym,CC,2=1.27X10~~M,CC.,-=1.3X10~’M,pH=1.07,900min~‘.

Gubailovskii et al. [ 121 have postulated the formation of a surface film of AuCI,,, during the gold dissolution. The pH effect (as HCl addition) has been explained by the HCl effect on solubilizing this film. In the present study, the surfaces of original gold and those obtained at different pH values were examined by SEM/EDS. For pH < 5 a corrosion pattern was observed (Fig. 10). At higher pH values, however, the surface morphology was identical to that of the original gold and no sign of attack was observed. At all pH values no formation

1.0

LI

u al 4-I ” m 0) L

4= 1.53 5 = 1.97 6 ??2.70 7 = 3.18 81: 3.90

20 40 t Imin)

b

Fig. 8. Dissolution curves as a function of pH. Conditions: flat-packs, 18°C. Cc12 = 1.27X 10 -’ M, C,,, - = 0.48 M.

Page 10: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

134 .I. Virials et al. /Hydrometallurgy 38 (1995) 125-147

5 7 DH

Fig. 9. Effect of pH on the specific rare. Conditions: I ST, Cv,2 = I .27 X 10 -’ M, Cc, - =0.48 M.

of AuC1 films or possible products of their disproportionation or reaction (such as pulver- ulent gold or chloride-containing species) was detected at high SEM magnifications. This suggests that, if a passivating layer is formed at high pH values, this could be restricted at levels of molecular thickness. This aspect was investigated by XPS (see section 5).

4.3. Effect of chlorine and chloride concentrations

The effect of the total chlorine concentration at pH 1.04 and C,-, - = 0.48 M is shown in Fig. 11. Fig. 12 is a plot of the specific rates versus the total chlorine concentration. The rate (when pH and total chloride arc constant) is proportional to the total chlorine concen-

Fig. IO. Surface of the flat-packs of gold, attacked at pH I .04, Cc,, = 1.27 X 10 -’ M, C,., - = 0.48 M, 5 min.

Page 11: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Viiials et al. /Hydrometallurgy 38 (1995) 125-147 135

In

I , -

0 20 40 60 80 t Imln)

Fig. 1 I. Effect of the total chlorine concentration. Conditions: 18°C. C,., -= 0.48, pH = 1.04, spheres, r, = 45 pm.

tration. This effect has been reported by Gubailovskii et al. [ 121 and Nikolaev et al. [ 151. However, this apparent effect does not indicate the possible role of the molecular chlorine

and trichloride ion in the gold dissolution rates. According to the equilibrium 4, for a pH < 1.5 and fixed total chloride, an increase in the total chlorine concentration tends to increase almost proportionally both molecular chlorine and trichloride ion concentrations.

The effect of the total chloride is shown in Fig. 13 and Fig. 14 is a plot of the specific rates. An approximate order of 0.5 was found. A significant difference with the behaviour

5

log ccl2 (molll)

Fig. 12. Plot of the specific rates versus total chlorine concentration. Conditions: 18”C, pH = 1.04, Cc, - = 0.48 M.

Page 12: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

136 J. Virials et al. /Hydrometallurgy 38 (1995) 125-147

l= l.CO 2=1.00 3=0.75 6=O.C0 s-0.13 6= 0.005

t tmln)

Fig. 13. Effect of the total chloride concentration. Conditions: 18°C. pH = 1.04, C,-,> = I .27 X 10 -’ M, spheres, t-,=45 jm~. 6 = a calculated equilibrium concentration for a solution of chlorine in dilute sulphuric acid at the concentrations and pH value indicated.

of bromine [ 181 is that the rates in chlorine solutions increase even at high chloride addition. However, the main effect observed is that in the absence of chloride addition (that is, in the presence only of the chloride ion generated by hydrolysis) the rate was extremely slow.

E -5.0 ._ ?? flat - packs o/ E

“E 0 Spheres

z -5.6 -

E

,I

o”- 6.0 -

-6.5 r 0

I / I I I

2.0 1.5 -1.0 -0.5 0

,

log cc,-(molll)

Fig. 14. Plot of the specific rates versus total chloride concentration. Conditions: 18°C. pH = 1 .O4, C<.,,=1,27XlO-‘M.

Page 13: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Virials et al. /Hydrometallurgy 38 (1995) 125-147 137

[cl;]-10~ Imot/t)

Fig. 15. Plot of the specific rates against trichloride concentration. Conditions as Figs. 4,6,9, 12 and 14.

The dependence of pH, total chlorine and total chloride are only apparent. All the effects studied are consistent with the fact that hypochlorite, hypochlorous acid and also molecular chlorine do not significantly affect the gold dissolution rates. Fig. 15 shows all the data in Figs. 5, 6, 9, 12 and 14, replotted against the trichloride ion concentration. The first order obtained and the practical intercept of the line at the origin confirm the conclusion that the trichloride ion is the active species in the aqueous chlorination of gold.

Pesic and Sergent [ 181 report a similar conclusion on the role of tribromide in the kinetics of the bromation of gold in conditions of an excess of bromide ions.

4.4. Temperature effect

Kinetic data for the temperature effect are shown in Fig. 16. The Arrhenius plot (Fig. 17) was made through In i/Cl; versus 1 / Tto correct the effect of [Cl; ] on the temperature dependence (section 3). An activation energy of 43.5 kJ/mol (10.4 kcal/mol) was obtained, which confirms the chemical reaction control. The kinetic equations can be written as: Flat-packs:

(Y= 2X4.25X lO’[Cl,]exp( -43.5/R7’)t

(14) PAU eo

Spheres:

l-(J-a)t= 4.25X lO’[CI,]exp( -43.5/RT)t

PAL, ro (15)

Page 14: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

138 J. Viii& et al. / Hydrometallurgy 38 (1995) 125-147

- 1.0 y” : a

m L

Temperature-X

flat-packs spheres

l= 6 6= 6

I=18

8=24

9= 30

t (min)

Fig. 16. Dissolution curves as a function of the temperature. Conditions: pH= 1.04, Cc,, =1.27X IO-’ M, Cc, = 0. I3 M, spheres, r. = 45 pm.

4.5. XPS analysis

The survey spectra of the original gold in the range O-600 eV show, essentially, Ols, AQ, C,, and Au4[ peaks. Similar spectra, plus a relatively small Cl*,, peak, were recorded

Cl flat-packs

0 spheres

a3 3.4 3.6 3.6

tooolT(iIK)

Fig. 17. Arrhenius plot for the dissolution of gold in t&Cl- acidic media. Conditions: Cr,, = 1.27 X 10 -’ M, Cc.,- =0.13M.pH=l.04.

Page 15: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Virials et al. /Hydrometallurgy 38 (1995) 125-147

10

9

8

7

6 Y G z

5

4

3

2

1

0 T : : : : : : : : I : : I

600.8 500.0 400.0 300.0 200.0 loo. 0 0.0 BIKtIffi MRW, eV

Fig. 18. Survey spectra of: 1 = original gold; 2 = sample leached at pH 7, Cc,2 = I .27 X 10 -’ M, Cc, - = 0.48 M, IS min; 3 = sample leached at pH 1, Cc,, = I .27 X 10 -’ M, C,., - = 0.48 M, 5 min.

for the leached samples. Fig. 18 shows the survey spectra for three selected samples: original gold, a sample leached at pH 7 and a sample leached at pH 1.

The quantification performed using the different sensitivity factors is shown in Table 1. In all the samples a high concentration of carbon and oxygen and a small concentration of gold were detected on the surface. It was observed, however, that atomic ratio of Au/Cl

Table I Relative atomic concentration on the surface

Element Area Sensitivity factor (cts-eV/s)

Sample 1 Cl, 8049 0.296 01, 9153 0.711 Au,, 13890 62.50 Cl?, 0.891 .kq’le 2 C,, 9503 0.296 01, 8878 0.711 AU41 II614 6.2.50 CL,, 1374 0.891 Sumnj,le 3 C,\ I1897 0.296 O,, 5306 0.71 I Au,, 9815 6.250 CIZ,~ 1533 0.891

” Relative to %C + %O + ?&Au + %CI = 100.

Concentration”

(%I

63.04 31.80

5.16

66.90 26.02

3.87 3.21

78.89 14.65 3.08 3.38

Page 16: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

140 J. Virials et al. /Hydrometallurgy 38 (1995) 125-147

) i ‘8

\ ,:’ v ; i, ,I

\,&‘.’ .,.....’ , \.

. r /...-.-‘.. ..,(. L.- :

.-..

‘-. .c._.i.

CIS 3

2 I.. ‘L f > !A .I , \ : , : ’ ::1

: I :: ( > .’ I : (

,i ’ : ( ‘; I

93.8 85.8 80.0 e 288.8 282.8 288.8 BDODlG EERGV, eV BDiEihd~;i

Fig. 19. Individual Au,, and C,, for: I = original gold; 2 = sample leached at pH 7; 3 = sample leached at pH 1. Other conditions as in Fig. 18.

close to 1 appeared in the leached samples, which could suggest that during the leaching AuCI,, was formed on the gold surface. The extensive adsorption of oxygen and carbon compounds makes it difficult to conclude, however, whether during the leaching this species appears in localized sites or as a monolayer.

Fig. 19 shows the individual Au4r and C,, peak analysis for the three samples. The C,, peaks show maximum intensity at 284-285 eV, typical values for -( CH2)- hydrocarbons [ 201. Au4r spectra show doublets with Gaussian-shaped 5 12 and 712 components at binding energies of 87.5 and 83.7 eV for the three samples. These energies are those reported [20] for gold in the elemental state. Binding 4f,,, energy for AuCl is displaced at 86.2 eV, whereas for a gold(II1) species such as NaAuCl, the displacement occurs at 87.3 eV. Unfortunately, these results only show that the gold leached samples were essentially free of gold compounds in the XPS chamber but a compound such as AuCl,,, could possibly be reduced to the elemental state as a consequence of the hydrocarbon adsorption when the sample was extracted from the solution.

Sputter deep profiles for Au,,. peaks show a substantial increase in the intensity after the first two sputtering cycles, due to the elimination of the C, 0 and Cl adsorbed species, but there was no change in the peak positions corresponding to the elemental state. In contrast, sputter deep profiles for Cl, 0 and C decrease in intensity and any layer of gold compound overlapped by the outermost layers was detected.

4.6. Rates and mechanism

From the kinetic dependence and previously reported electrochemical studies, the pos- sible mechanism for the aqueous chlorination of gold in the presence of chloride ions can be discussed.

Page 17: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Virials et al. / Hydrometallurgy 38 (1995) 125-147 141

In this formulation, it was considered that the adsorption of trichloride ion in cathodic sites was the precursor and rate-limiting step for the electrochemical reaction sequence:

Au k’6 ,_8,+ClGq-

Au . Cl;,d,

0, slow

‘ k- 1.s (16)

Once adsorbed, Cl, is reduced to chloride:

Au.Cl,,, k17

& +2e-Au+3Cll =i (17)

k-17

where Blc is the fraction of cathodic sites covered by Cl,,,. Since Cl, is necessarily a minor solution species and the process in Bq. (16) was slow, it was assumed that 1 % 0,. Under these conditions, the cathodic current density can be written throughout the rate of adsorption as:

i, =2Fk,,[Cl,] (18)

For the oxidation of gold in anodic sites the following pathway proposed by Diaz et al. [ 71 was assumed:

Au%, k*’ AuCl, ads + 2e

0 +2c1- -

IR 8 k-21 2a

AuCI, ads kzz

(19)

(21)

(22)

where 8,, and 0,, are the fraction of anodic sites covered by AuCl,,, and AuCl, adsr respec- tively. According to this pathway, the dissolved gold can proceed, in principle, from the processes in both (20) and (22).

Assuming that, for an excess of chlorine in solution (Es,,> 1.1 V versus AgCl/Ag (KCl,,,) ) , the back reaction kinetics are negligible, the anodic current density at the mixed potential can be written through the Butler-Volmer equation as:

i, =F( 1 - 0,, - &,)k,,exp(FE,/2RT) [Cl-]

+2FB,,k,,exp(FE,,,/RT)[C11]2 (23)

In Eq. (23)) the transfer coefficients were taken to be equal to 0.5 and it was considered that reaction (2 1) occurs as two single-electron steps [ 71.

Applying the stationary state to the AuCl,,,, it can be written:

(1 - 19,~ - B,,,)k,gexp(FE,,,/2RT) [Cl-l - &,k20[C111

-8,,k,,exp(FE,lRT)[C1-]2=0 (24)

Page 18: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

142 .I. Viiials et al. / Hydrometallurgy 38 (1995) 125-147

Considering that the mixed potentials developed in the leaching system under the con- ditions studied (see Fig. 21) are close to N 0.9 V (vs SCE) and that, under these conditions, process (20) is slow compared with ( 19) and (21) [ 71, this results in:

i;, z2F( 1 -H,, - Bza)k,9exp(FE,/2RT) [Cl-]

~3F8,,k,,exp(FE,lRT)[ClP]* (25)

Eq. (25) involves two limiting cases. If 1 X- BIa + eza, it becomes:

i, z2Fk,,exp(FE,I2RT) [Cl-] (26)

Eq. (26) establishes a Tafel slope equal to 2RT/F and a first-order rate anodic process with respect to [Cl-].

Since, at the mixed potential it can be written:

i, 4;iA = i, &_A

combining Eqs. ( 18), (26) and (27)) results in:

(27)

FE, Abs[Cl:i 1 eXG=g,k,,[C1-] (33)

in which E,, should depend of the [Cl, J / [Cl- ] ratio with an slope of 118 mV decade- ’ at 25°C:

E, E 0.11 Slog [Cl; ] / [Cl - ] + constant, (29)

The other limiting case is when 8,, + 1, which results in:

iz,~3Fk,,exp(FE,,,/RT)[C1-]2 (30)

Eq. (30) establishes a Tafel slope of RT/F and a second-order rate anodic process with respect to [Cl- 1. Combining Eqs. ( 18)) (27) and (30), gives:

FEm = exp RT WcMCl, 1 3&k2,[Cl-]*

(31)

in which E,,, should depend on the [Cl; ] / [Cl- I2 ratio with a slope of 59 mV decade-’ at 25°C:

E,, z 0.0591og [ Cl; ] / [Cl- ] 2 + constant, (32)

A series of potentiometric measurements were performed for determining the dependence of the mixed potentials. A stationary gold electrode, consisting of a 0.2 mm diameter gold wire mounted in an unreactive epoxy resin was used. Mixed potentials were measured against an AgCl/Ag ( KCl,,,) electrode separated from the solution by a salt bridge placed 1 cm away from the gold surface.

Fig. 20 shows the effect of the stirring speed of the solution. At < 300 mitt-‘, mixed potential increases with the stirring speed. Under these conditions the reaction is probably controlled by cathodic diffusion [ 30,3 11. At > 300 rpm, however, mixed potential was not dependent on the agitation intensity, which was indicative of chemical/electrochemical control.

Page 19: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Virials et al. / Hydrometallurgy 38 (1995) 125-147 143

5 8ool , , , , , 0 100 200 300 kO0 500

r.p.m. Fig. 20. Effect of stirring speed on mixed potential. Conditions: 25°C Cc,, = 1.06X IO-* M, Cc, - = 1.4 M, pH= 1.04.

Fig. 21 shows that the mixed potentials increase with the trichloride concentration and decreases with the chloride concentration. Fig. 22 shows the data in Fig. 21 replotted against [Cl; ] / [Cl- ] *. A dependence of 52 mV decade- ’ was obtained, which is in reasonable agreement with the value expected from Eq. (32).

Thus, a high degree of coverage by Au&,, in anodic sites was consistent with the model and with the mixed potential dependence; in spite of the fact that the detection of this reaction intermediate by XPS was unsuccessful. The value of 52 mV decade-’ is also consistent with the Tafel slopes reported (56 mV decade-‘) by Horikoshi et al. [ 31. The second-order dependence of the mixed potential with respect to the chloride concentration was also consistent with those reported for the anodic intensities by Horikoshi et al. [ 31 ( 1.9 order) and Dfazet al. [7] (1.7 order).

920 /r 52mV

A /

0

900 - 8’O

51mV

880 -

/

.A0

/r

o l.LM CI-

860 - A 0.5M CI- 00

840 + - 3.5 -3 -2.5

Fig. 2 I. Effect of trichloride concentration on mixed potential. Conditions: 25°C. 300 min-‘, pH = 1.04

Page 20: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

144 J. Virials et al. / Hydrametallurgy 38 (1995) 125-147

940 A/

+ 5: 920 -

G Y if

a/

A 52mV

a, 900 -

3 G .B’O ul 880 1 0 1.4M CI- Q

0’

In /

A 0.5M CI- >

- E 860 0 W /

840 I I I I I -4.0 -3.6 -3.2 -2.6 -2.4 -2.0

Fig. 22. Correlation between the mixed potentials and the trichloride and chloride ion concentrations. Conditions: 2X, 300 min- ‘, pH = I .04.

On the other hand, the rate expression can be written as:

Rate = i, $J 3F

and, combining Eqs. (30), (31) and (33) results in:

Rate= (2/3)&,k,,[Cl,]

in agreement with the kinetic dependence of the reaction.

(33)

(34)

5. Conclusions

From the above, the following conclusions can be drawn: ( 1) The aqueous chlorination of gold in suspended particles is chemically controlled.

Specific rates are independent of the particle geometry, size and agitation speed. An acti- vation energy of 43.5 kJ/mol (10.4 kcal/mol) was found.

(2) For pH < 1.5 there is no pH effect on the leaching rate. Above this value, the rates decrease progressively until the pH is * 4. At higher pH no significant dissolution of gold takes place.

(3) Leaching rates exhibit an apparent order of one with respect to the total chlorine concentration and an apparent order of one half with respect to the total chloride concentra- tion.

(4) The dependence of pH, total chlorine and total chloride indicate that the rate of the aqueous chlorination is basically determined by the concentration of the trichloride ion.

The kinetic equation can be written as:

Page 21: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. Vin*als et al. / Hydrometallurgy 38 (1995) 125-147 14s

Rate( mol cm- 2 min -‘)=4.25X lO’[Cl;]exp( -43.5/R7’)

(5) Gold surfaces not preserved from the atmospheric environment show a high coverage of adsorbed oxygen and hydrocarbon. An atomic ratio of Au/Cl close to 1 was observed in the leached samples. However, gold appeared in the XPS chamber in the elemental state.

(6) A possible mechanism for the aqueous chlorination of gold was discussed. It was assumed that the adsorption of trichloride ion was the precursor and rate-limiting step of an electrochemical reaction sequence in which the anodic pathway was consistent with previous electrochemical studies. The measured mixed potential dependence:

E, =O.O591og [ Cl, ] / [Cl- ] 2 + constant

was also in agreement with the leaching model.

Acknowledgements

The authors wish to thanks Dr. J. Portillo of the Serveis Cientifico-Tecnics de la Univ- ersitat de Barcelona for his assistance in the XPS analysis. 0. Herreros thanks the Agencia de Cooperation International and the Universidad de Antofagasta (Chile) for his research studentship.

Nomenclature

A = surface area

b” = fraction reacted = stoichiometric factor 2/3 (mole solid dissolved/mole leaching agent)

C, = concentration of reactant (mol 1-l) D = diffusion coefficient (cm2 min- ‘) E = flat-pack thickness (cm)

E,, = mixed potential F = Faraday’s constant 4,. & = fraction of total surface which exhibit cathodic and anodic behaviour

g = gravitational acceleration (cm min-‘)

i;, = anodic current density i, = cathodic current density k cxp = experimental rate constant (min-‘) ki, k_i = rate constants for the direct and reverse reactions

K, = association equilibrium constant

Kd = dissociation equilibrium constant Kh = hydrolysis equilibrium constant k,,,(,, = mass transfer coefficient (cm s- ‘) I1 = reaction order R = universal gas constant r = particle radius (cm)

Page 22: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

146 J. Virials et al. /Hydrometallurgy 38 (1995) 125-147

t-0

r

PA”

Pf

Yip T

8 = fraction of surface covered by adsorption sites V = kinematic viscosity (cm* min-‘)

= initial particle radius (cm) = specific rate (mol Au cm- * min- ’ ) = molar density of gold (0.09858 mol Au cmd3) = fluid density (g cm- 3, = solid density (g cm- “) = ( ps - pflpf) relative phase density difference = temperature (K) = time (min)

References

[ 1 1 Habashi, F., Principles of Extractive Metallurgy. Gordon and Breach, New York, Vol. 2,2nd ed. ( 1980). p. 39.

[ 21 Stanley, G.G. (Editor), The extractive metallurgy of gold in South Africa. S. Afr. Inst. Min. Metall. Monogr. Ser. M7.2 (1987): 627.

131 Horikoshi, T., Yoshimura, S., Kubota, N. and Sate, E., The anodic dissolution behavior of gold in sulfuric acid solutions containing chloride ions. Nippon Kagaku Kaishi, 8 ( 1983): 1118-l 123.

141 Nicol, M., An electrochemical and kinetic investigation of the behaviour of gold in chlorine solutions. Il. The anodic dissolution of gold. Nat. Inst. Metall., NIM-1844 (1981), p. 23.

[ 5 1 Herrera-Gallego, J., Castellano, C.E., Calandra, A.J. and Arvia, A.J., The electrochemistry of gold in acid aqueous solutions containing chloride ions. J. Electroanal. Chem., 66 (1975): 207-230.

161 Lovrecek, B., Moslavac, K. and Matic, D., Anodic dissolution and passivation of gold, particularly in presence of chlorine. Electrochem. Acta, 26(8) (1981): 1087-1098.

[7] Diaz, M.A., Kelsall, G.H. and Welham, N.J., Electrowinning coupled to gold leaching by electrogenerated chlorine. I. Au( Ill)-Au(l) /Au kinetics in aqueous Cl,/Cl- electrolytes. J. Electroanal. Chem., 361 ( 1993): 25-38.

181 Habashi, F., The recovery empire built on fool’s gold. Eng. Min. J., 170 ( 1969): 59-64. [9] Nuiiez, C. and Vifials, J., Study of Spanish pyrite cinders: Cl,HCl leaching of fluidized-bed arsenical cinders

I. Trans. Inst. Min. Metall. (Sect. C: Mineral Process. Extr. Metall.), 93 (1984): 162-172. I 101 John, N.G., Glenn, R.P. and William, W.W., The recovery of gold from refractory ores by the use of carbon-

in-chlorine leaching. J. Met., 42(9) (1990): 12-14. I I I 1 Ximing, L., Jiajun, K., Xinhui, M. and Bin, L., Chlorine leaching of gold-bearing sulphide concentrate and

its calcine. Hydrometallurgy, 29 ( 1992): 205-215. [ 12 1 Gubailovskii, V.V., Kakovskii, LA. and Khalezov, B.D., Kinetics of the dissolution of gold in chlorine

containing aqueous solutions. lzv. Akad. Nauk SSSR, Met., 6 (1973): 106-l 11. [ 13 1 Chekushin, VS, Fomina, T.E. and Solozhenkin, P.M., Selective dissolution of precious metals. Dokl. Akad.

Nauk Tadzh. SSR, 26( 1) ( 1983) : 35-38. ( 14 J Volkova, G.V. and Filippov, A.A., Kineticsofgoldoxidation bychlorinein sodiumchlorideand hydrochloric

acid solutions. lzv. Sib. Otd. Akad. Nauk SSSR, 2 (1974): 81-84. [ 151 Nikolaev, A.V., Peshchevitskii, B.I., Zemskov, S.V., Krivospitskii, 0.1. and Galkin, P.S., Kinetics of

hydrochlorination of metallic gold in sodium chloride-sulfuric acid solutions. Deposited Dot., VINITI 1084. 75 (1975),28pp.

( 161 Sohn, H.Y. and Wadsworth, M.E., Rate Processes of Extractive Metallurgy. Plenum Press, New York (1979).

I 17 I Levins, D.M. and Glastonbury, J.R., Particle-liquid hydrodynamics and mass transfer in a stirred vessel. Part Il. Mass Transfer. Trans. Inst. Chem. Eng., 50 (1972): 123-146.

[ 18 1 Pesic. B. and Sergent, R.H., Reaction mechanism of gold dissolution with bromine. Metall. Trans. B, 24 (1993): 419431.

Page 23: CINÉTICA DE LAS PARTICULAS EN EL PROCESO DE CLORACIÓN DEL ORO

J. VKaials et al. / Hydrometallurgy 38 (1995) 125-147 14-l

I 19 1 Angelidis, T.N., Kydros, K.A. and Matis, K.A., A fundamental rotating disk study of gold dissolution in iodine-iodide solutions. Hydrometallurgy, 34 ( 1993): 49-64.

1201 Muilenberg, GE. (Editor), Handbook of X-ray photoelectron spectroscopy, a reference book of standard data for use in X-ray photoelectron spectrometry. Perkin-Elmer Corp., Physical Electronics Division, Minnesota, USA (1978), pp. 39,43,59, 155, 188.

[ 2 I ] Powell, C.J. and Seah, M.P., Precision, accuracy and uncertainty in quantitative surface analysis by auger- electron spectroscopy and X-ray photoelectron spectroscopy. J. Vat. Sci. Technol. A., 8(2) ( 1990): 73S- 763.

[ 22 1 Viiials, J., Oliveras, J. and NuAez, C., Leaching kinetics of natural cobalt triarsenide in chlorine solutions. Metall. Trans. B, 17 (1986): 629637.

[23 1 Majima, H., Peters, E., Awakura, Y. and Tsugui, K., Fundamental studies on chlorine behavior as related to zinc electrowinning from aqueous choride electrolytes. Metall. Trans. B, 21 ( 1990): 251-258.

[24 1 Connick, R.E. and Chia, Y.T., The hydrolysis of chlorine and its variation with temperature. J. Am. Chem. Sot., 81 (1951): 1280-1284.

125 ] Camacho, F., Diaz, F., Moreno, V. and Clemente, J.A., Solubilidad de1 cloro en agua y disoluciones de electrolitos. An. Quim., 76 (1980): 397-400.

[ 26 1 Levenspiel, O., Ingenieria de las Reacciones Qufmicas. Revert& Barcelona ( 1979). [ 27 1 Vu, C. and Han, K.N., Effect of system geometry on the leaching behavior of cobalt metal: mass transfer

controlling case. Metall. Trans. B, 10 (1979): 57-61. [ 28 1 Sareyed-Dim, N.A. and Lawson, F., Cementation onto particulates. Trans. IMM, 85 ( 1976): Cl-6. 1291 Lide, D.R. (Editor), CRC Handbook of Chemistry and Physics. CRC Press, Boca Raton, Fla., USA, 71st

Ed. (1990-1991). [ 301 Power, G.P. and Ritchie, I.M., Mixed potential measurements in the elucidation of corrosion mechanisms-

I. Introductory theory. Electrochim. Acta, 26(8) (1981): 1073-1078. 13 1 1 Power, G.P., Staunton, W.P. and Ritchie, I.M., Mixed potential measurement in the elucidation of corrosion

mechanisms-II. Some measurements. Electrochim. Acta, 27( 1) (1982): 165-169.