Method Carbon

Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

CHARACTERIZATION OF ACTIVATED

CARBONS FOR THE EXTRACTIVE


METALLURGY OF GOLD
ALTININ METALURJK KAZANIMINDA AKTF
KARBONUN KULLANILMASI
Alfredo Romero *
Carlos H. Sampaio **
Keywords : Surface Area, Adsorption, Aurocyanide Anion, Loading
Capacity, Poluant Ions.

ZET
Bu almada Au (CN)7 kompleksinin sourulmas iin aktif karbon
kullanlmtr.
Al ayn aktif karbon numunesi kullanlarak bunlarn sourma
performanslar tespit edilmi, sourma kapasitesi ve kinetii her nu
munenin fiziksel ve kimyasal zellikleriyle karlatrlmtr.

ABSTRACT
This work presents a study of characterization of activated carbons
for the adsorption of Au (CN^ complex.
A comparative study of six samples of activated carbons, three
Brazilian and three produced out of Brazil was made in order to determine
their qualification for a performance in adsorption circuits in the extractive
metallurgy of gold. The adsorptive capacity and the kinetics of adsorption
were compared to the physical and chemical properties of each one. The
influence of poluant ions on the kinetics and loading capacity of
aurocyanide anion was also determined.

Metallurgical Engineer, M.Sc. student PPGEMM / UFRGS


Mining Engineer, M. Sc., Dr. Eng., Professor

TRKYE X M L M A D E N C L K K O N G R E S , 1 9 9 3

427

1.

INTRODUCTION

Among the most important application of activated carbon,


the extractive metallurgy of gold in the stage of
aurocyanide complex Au(GN) 2 ~ adsorption can be mentioned.
The surface area, that depends on the diameter of the pore,
is the most important property of the activated carbon.
The activated carbon has been used for a long time in the
recovery
of
gold
through
cyanide
solutions
from
lixiviation, but the mechanism with which the carbon
adsorbs gold is still undetermined and there are many
theories and different opinions about it (QUISPE and
SAMPAIO, 1992). The goal of this work was to characterize
3 brazilian carbons and three foreign ones, aiming their
utilization in adsorption circuits in the extractive
metallurgy of gold and to number the individual ejects on
the loading capacity of gold brought about by t - most
common ionic poluants which belong to cyanide lixi /.Mon
industrial solutions, such as silver, copper, iron,
thiocyanate, cyanide, etc.
The ionic interation was
considered a significative factor in the kinetics of g^'d
adsorption.
After testing 6 samples of activated carboa
one of them was chosen to measure:
the effects of
poluants, the effects of stabilizing cations (Ca2+, Na + ,
and K + ) , pH, temperature and velocity of agitation on the
kinetics and loading capacity.
2.

EXPERIMENTAL PROCEDURE

2.1
2.1.1

Materials
Samples of Activated Carbon

The 6 samples of activated carbon were identified by the


letters from A to F, being A, B and C foreign carbons and
D, E and F brazilian ones.
The granulometric fractions
provided by manufacturers in ASTM mesh were: A, B and E:

428

8 x 16 (2,36 x 1,18 mm); C: 6 x 12 (3,35 x 1,70 mm); D and


F: 6 x 16 (3,35 x 1,18 mm).
2.1.2

Reagents

All the reagents utilized in order to prepare solutions


were of analytical grade, deionized water was used and as
pH modifiers HCl and NaOH 0.1 N were used as well.
2.2

Methodology

The activated carbons utilized


in the tests were
homogenized and quartered until aliquotes of 20 g were
obtained.
For moisture analysis, proximate analysis and
pH, these aliquotes of 20 g were ground to 100% -50 mesh
(ASTM, 1991) . For Iodine Number and Loading Capacity the
carbons were ground to 100% -325 mesh (ASTM, 1991; CALGON,
1983). In the remaining tests the carbons were used in the
size they were provided by the manufacturers. In order to
determine the resistance of the carbons to abrasion the
method of the "Companhia Vale do Rio Doce" (CVRD) was
applied.
According to this method the loss caused by
abrasion should be less than 2%.
The degree of gold
adsorption occurring at equilibrium was measured using of
the empirical Freunlich Isotherm (CHEREMISINOFF P. N. , et
al, 1980). For all the tests of kinetics of adsorption 1
g of dry sample of granular activated carbon was mixed
during 8 hours with 1700 ml of a solution with 5mg/l of
gold in a batch reactor with baffles.
The agitation was
adjusted to 900 rpm. After 2, 4, 6 and 8 hours, aliquotes
of 10 ml were taken and the amount of gold was analysed by
atomic absorption.
The ionic effect on the aurocyanide
adsorption
caused by the most common poluants ions from
industrial solutions of gold cyanidation was measured. The
variables which were mantained constant for each test were:
volume of the solution, weight of the activated carbon,
temperature and agitation.
The ionic solutions were
prepared with deionized water. The used solution of gold

429

was a buffer of borate in pH 10, with different initial


concentrations of gold.
3.
3.1

RESULTS AMD DISCUSSIONS


Physical Properties and Chemical Characteristics

Table I. Physical properties and chemical characteristics


of adsorption of the activated carbons (*)
Activated FC
Carbon
%
A
B
C
D
E
F

VM
%

ASH MOIST.
X
X

AP. DENS. HARDNESS ABRASION IODINE No.


g/cc
X
X
mg 12/g AC

5,79
6,34

1,72
3,26

4,02
2,83

0,50
0,50

91,70
98,00

91,88 6,36
82,99 10,55
86,48 7,86
88,21 9,64

1,76
6,46
5,66
2,15

3,65
4,29
8,55
6,74

0,47
0,50
0,50
0,45

98,00
98,00

92,49
90,40

95,00
97,00

2,43
1,88
1,60
2,65
3,80
2,82

pH VALUE
suspension filtrate

1110
1200
1230

850
1000

881

9,50
9,60
9,50

9,00
9,30

9,00
6,50
9,30

8,40
6,40
8,90

9,10

(*) the proximate analysis was carried out on dry basis.


3.2

Loading Capacity

Table II.

pH
10

Values of the Isotherm of the Activated Carbon A

Grams of Carbon Cone, of gold


mg Adsorbed
in eguilib. (C)
(M)
mg/1
mg
(X)
Soluc. cabea
0,4000
0,3000
0,2000
0,1003

100,00
0,90
3,62
14,20
45,00

10,000
0,090
0,362
1,420
4,500

9,910
9,638
8,580
5,500

Carbon
Loading
X/M
24,78
32,13
42,90
54,8

The isotherms of the other activated carbons are figured


out in the same way. Table III shows the coordinates of the
experimental and calculated isotherms, fitted by the leastsquares method. The fitted isotherms are presented in Fig.
1 and table IV presents the results of loading capacity.

430

Tabela I I I .
Activated
Carbon

Isotherm Coordinates P l o t t e d in F i g . 1

[Au]eq.
C

mg Au a d s . / g a. c .
X/M e x p .
X/M c a l c .

0,90
3,62
14,20
45,00

24,78
32,13
42,90
54,84

24,60
32,49
42,70
53,78

0,999

0,006

1,40
4,52
16,00
43,00

24,65
31,83
42,00
56,38

24,27
32,15
43,55
55,21

0,995

0,014

2,30
7,00
23,50
51,00

24,43
31,00
38,25
49,00

24,52
31,33
40,89
48,49

0,987

0,018

3,20
11,20
29,00
65,50

24,20
29,60
35,50
34,50

24,87
29,27
33,12
36,82

0,893

0,031

9,00
16,20
34,50
67,00

22,75
27,93
32,75
33,00

24,06
26,90
31,06
35,23

0,876

0,033

9,50
21,00
37,50
69,00

22,63
26,33
31,25
31,00

22,71
25,99
28,68
31,81

0,901

0,026

Table IV.
Activated
Carbon

Results of the Loading Capacity


K-value

1/n

Maximum Capacity
of the Adsorbent
(X/M) co

0,20
0,24
0,22
0,13
0,19
0,17

63
68
56
39
38
34

KgAu/t
A
B
C
D
E
F

R
Standard e r r o r
Squared
e s t . de X/M

25
22
20
21
16
15

431

3.3

Kinetics of Adsorption

It has been showed in tests of batch adsorption that the


rate in which the gold gets out of the solution can be
described as a first order process for short periods of
time, less than 12 hours, (NICOL et al, 1984; FLEMING et
al, 1984).
The kinetics model is described by:
[1]
In which: Cc and Cs are the respective concentrations of
gold in the carbon and in the solution, at time t in mg/kg
and K is the kinetics constant of first order. From [1]
and from the mass balance equation at any time in the batch
reactor we have:
[2]
in which C s is the concentration of gold in the solution
at time 0 in mg/kg. All the kinetics tests were analysed
through the equation [2]. Table V presents a comparison of
the results of loading capacity and kinetics. Fig. 2 shows
the kinetics results of the tested carbons.
"able V.

Comparison of the Re-.ults of Loading and Kinetics

Carvo Ativado
A
B
C
D
E
F

432

Cte.

Cinttica
h-1
590
828
776
200
147
212

K-value
kgAu/t
25
22
20
21
16
15

1/n
0.20
0.24
0.22
0.13
0.19
0.17

4. EFFECT OF THE POLUANT IONS


Metalic copper and silver were individually dissolved in
cyanide of potassium until solutions with different initial
concentrations of cyanide, copper and silver were obtained.
Then a buffer solution, pH 10, with initial concentration
of 10 mg Au/1 was added. Fig. 3 shows that gold adsorption
is extremely diminished in 200 ppm of Cu, whereas there is
almost no reduction in gold adsorption in solutions with
100 ppm of copper. It can be concluded, as it was shown by
FLEMING and NICOL, that in low ratios of free cyanide and
copper, the Cu(CN) 2 ~ complex is adsorbed by carbon, whereas
in high ratios the Cu(CN) 3 2 - and Cu(CN) 4 3 - complexes are
formed and the adsorption on carbon is not significative.
Fig. 4 presents the results, where the adverse effect of
silver in gold adsorption is shown. It can be also noticed
that a silver/gold ratio of 2:1 has a pronounced effect on
decreasing gold adsorption.
The individual dissolution of ferrocyanide of potassium and
thiocyanate of potassium in deionized water produced iron
and thiocyanate poluant solutions. From Fig. 5 it can be
concluded that the iron poluant does not practically
present any noticeable effects on gold loading for the
established levels; the curves are similar.
Fig. 6
presents the results and shows that the SCN~ is not as
hazardous as the copper and silver poluants, although some
decrease in the loading happened. Fig. 7 shows that the
high concentration of cyanide inhibits the adsorption and
in Fig. 8 it can be seen that the addition of 0,067 M of
calcium ions has great influence on the kinetics of gold
adsorption, if it is compared to the presence of threetime-higher concentrations of sodium and potassium ions,
which do not improve the loading.
The solution without
ions has a slower kinetics, what leads to the conclusion
that these cations, especially the calcium, make the
complex Ca[Au(CN)2]2 stable (formation of pair ion)
promoting the kinetics and the loading. Fig. 6 shows that
the adsorption capacity diminishes with the pH increase,

433

despite the presence of


cloride.
The decrease
the competition between
adsorption on the carbon

200 ppm de Na added as sodium


of the k-value can be inferred to
the OH" and the Au(CN) 2 ~ in the
surface area.

The loading capacity at 30, 40 and 60 C was measured. The


results are shown in Fig. 10, where the adsorption's
tendence of increasing with the decrease of the temperature
in a linear variation can be observed, proving that the
adsorption is an exotermic reaction.
Fig. 11 clearly
presents that the kinetics constant increases notoriously
with the increase of the agitation for diluted solutions.
The same does not occur for high gold concentrations.
5.

CONCLUSIONS

The characterization of the carbons showed that the


brazilian ones present minor indices than the imported
carbons, therefore, their use get compromised for CIP and
CIL adsorption circuits. However they could be used in CIC
circuits.
Considering the typical conditions of the industry such as
dilute
solutions
and
alkaline
environments,
the
significative presence of copper and/or silver in the
cyanide solutions from gold ores have remarkable influence
on the rate of adsorption and in the final extraction of
gold. The remaining poluants have minor effect.
The effect of the pH, free cyanide and temperature are the
most important variables which affect gold adsorption and
loading.
ACKNOWLEDGEMENTS
The authors would like to thank Ana Maria dos S. da Silva,
Daniela Veek dos Santos e Luciane F. de Souza for their
helping in the laboratory tests. Acknowledgement is also

434

led to Angela R. Stainki for her helping in the translation


of this article into English.
REFERENCES
ASTM DESIGNATION:
D 2866-83 (Reapproved 1988), Standard
Test Method for Total Ash Content of Activated Carbon,
Annual Book of ASTM Standards, Section 15, p. 343 (1991)
ASTM DESIGNATION:
D 2867-83 (Reapproved 1988), Standard
Test Methods for Moisture in Activated Carbon, Annual Book
of ASTM Standards, Section 15, p. 344-345 (1991)
ASTM Designation:
D 3838-80 (Reapproved 1990), Standard
Test Method for pH of Activated Carbon, Annual Book of ASTM
Standards, Section 15, p. 386-387 (1991)
ASTM DESIGNATION:
D 4607-86 (Reapproved 1990), Standard
Test Method for Determination of Iodine Number of activated
Carbon, Annual Book of ASTM Standards, Section 15, p. 397401 (1991)
CALGON CORPORATION, Determination of the Gold Adsorption
Capacity (K-value) of Activated Carbon, p. 1-13 (1983).
CHEREMISINOFF PAUL N., ELLERBUSCH FRED, Carbon Adsorption
Handbook, USA, Ann Arbor Science Publishing Inc., p. 56-83
(1980) .
CVRD, Ensaios de Qualificaco de Amostras de Carves
Ativados, Superintendncia de Tecnologia, Departamento de
Desenv. tecnolgico, p.8.
FLEMING C. A. and NICOL M. J., The Absorption of Gold
Cyanide Onto Activated Carbon. III.
Factors Influencing
the Rate of Loading and the Equilibrium Capacity, J. S.
Afr. Inst. Min. Metall., p. 85-93 (1984).
NICOL M. J., FLEMING C. A., CROMBERGE G., The Absorption of
Gold Cyanide Onto Activated Carbon. I.
The Kinetics of
Absorption from Pulps, J. S. Afr. Inst. Min. Metall., p.
50-54 (1984).
QUISPE A. R., SAMPAIO C. H., Gold Recovery from Cyanide
Solutions Using Activated Carbon, Proceedings of the III
Meeting of the Southern Hemisphere on Mineral Technology /
XV
Brazilian
Meeting
on
Mineral
Processing
and
Hydrometallurgy, So Lourenco, Minas Grais, Brazil, p.
112-131 (1992).

435

Fig. 1: Isotherms of Aurocyanide


Adsorption on Activated Carbon

Fig. 3: Effect of Copper on


Aurocyanide Adsorption

436

Fig. 2: Kinetics of Adsorption

Fig. 4: Effect of Silver on


Aurocyanide Adsorption

Fig. 5: Effect of Iron on


Aurocyanide Adsorption

Fig. 7: Effect of Cyanide on


Aurocyanide Adsorption

Fig. 6: Effect of Thiocyanate on


Aurocyanide Adsorption

Fig.8: Effect of Ca, K, and Na on


Aurocyanide Adsorption

437

Fig. 9: Effect of pH on Gold Adsorption


In Presence of 2000 ppm Na

Fig. 10: Effect of Temperature


initial Buffer Solution 100 ppm Au

Fig. 1 1 : Effect of stirring on the


Kinetics of Adsorption

438

You might also like