See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/250153874

Gold Cementation from Ammonium Thiosulfate Solution by Zinc, Copper and

Aluminium Powders

Article  in  MATERIALS TRANSACTIONS · March 2002

DOI: 10.2320/matertrans.43.485

CITATIONS READS

39 1,593

3 authors, including:

Toyohisa Fujita
Guangxi University
405 PUBLICATIONS   4,163 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Utilization of Natural Resources View project

Book project on the various developments in this area View project

All content following this page was uploaded by Toyohisa Fujita on 05 October 2014.

The user has requested enhancement of the downloaded file.

Materials Transactions, Vol. 43, No. 3 (2002) pp. 485 to 493

2002
c The Mining and Materials Processing Institute of Japan

Gold Cementation from Ammonium Thiosulfate Solution by

Zinc, Copper and Aluminium Powders
Harunobu Arima1 , Toyohisa Fujita2, ∗ and Wan-Tai Yen3
1
Graduate Student, Department of Mining Engineering, Queen’s University, Kingston, Ontario, K7L 3N6, Canada
2
Professor, Faculty of Engineering and Resource Science, Akita University, Akita 010-8502, Japan
3
Professor, Department of Mining Engineering, Queen’s University, Kingston, Ontario, K7L 3N6, Canada

Gold cementation test was conducted without de-aeration by using zinc, copper and aluminium powders from an ammonium thiosulfate
solution contained 8 mg/l Au. The amount of metal powder was varied in the range of 30–450 Metal/Gold mass ratio. The solution composition
was 1–5 mol/l NH4 OH, 0.01–0.05 mol/l CuSO4 ∗ 5H2 O, 0.2–0.4 mol/l (NH4 )2 S2 O3 and pH 9.5–10.5. The results indicated that the gold was
effectively recovered from a solution of lower ammonia and copper concentrations and higher thiosulfate concentration. The optimum reagent
composition for the gold cementation from the ammonium thiosulfate solution was founded to be 1 mol/l NH4 OH, 0.01 mol/l CuSO4 ∗ 5H2 O
and 0.4 mol/l (NH4 )2 S2 O3 at pH 9.5. 100% of gold was recovered by zinc and aluminium powders at a Metal/Gold mass ratio of 30. Copper
powder recovered 93% of gold at a Metal/Gold mass ratio of 50. Zinc might re-generate thiosulfate concentration and precipitate most of
copper in the solution. Aluminium precipitation might recover gold with less amount of copper deposition and some thiosulfate reduction.
Copper precipitation reduced a small amount of thiosulfate concentration and greatly increased copper concentration. Ammonia concentration
stayed constant during cementation process.

(Received May 1, 2001; Accepted November 20, 2001)

Keywords: gold, zinc, copper, aluminium, thiosulfate, cementation

1. Introduction to recover gold from thiosulfate solution,18–20) but the effect

of copper, thiosulfate and ammonia concentration on the re-
Ammonium thiosulfate leaching is an alternative gold ex- quired amount of metal powder was not reported. The ob-
traction method to cyanidation because of its low toxicity.1–7) jective of this investigation was to find the relationship be-
However, the gold recovery method from the leach solution tween the solution composition and the amount of metal pow-
was not widely studied. In our study of gold leaching,8) an der required, and to establish the optimum condition for gold
optimum condition to extract gold from Hishikari ore was de- cementation.
veloped. In this paper, the cementation method to recover
gold from the leach solution is presented and discussed. 2. Chemistry
CIP or CIL and Merill Crowe processes are the standard
gold recovery process in cyanidation.9–14) It was reported that 2.1 Leach solution
aurous thiosulfate complex has no affinity to activated car- In the ammonium thiosulfate solution, gold and copper oc-
bon.15) Therefore, CIP or CIL processes is not applicable in curred in the form of amine or thiosulfate complexes, such
the ammonium thiosulfate leaching process. Marchbank et al. as [Au(NH3 )]+ , [Cu(NH3 )4 ]2+ , [Cu(NH3 )2 ]+ , [Au(S2 O3 )2 ]3−
suggested the addition of cyanide to thiosulfate solution fol- and [Cu(S2 O3 )2 ]3− . According to the reaction chemistry of
lowed by carbon adsorption.16) However, this method cannot gold leaching as shown in eq. (7),5–8) it is assumed that most
be applied in the wholly non-cyanide process. of copper should form amine complex, and gold should form
Gold cementation by metal powders may be considered as thiosulfate complex:
one of the best gold recovery methods. The use of zinc, cop-
3 + [Cu(NH3 )4 ]
Au + 2S2 O2− 2+
per and aluminium could be the candidate for gold cementa-
tion because of their lower reduction potentials than gold as → [Au(S2 O3 )2 ]3− + [Cu(NH3 )2 ]+ + 2NH3 (7)
shown in eqs. (1)–(6).17)
Au+ + e− → Au E0 = 1.68 V (1) 2.2 Zinc precipitation
The main cathodic reaction of gold cementation by zinc
4H2 O + O2 + 4e− → 4OH− E 0 = 1.23 V (2) powder in the ammonium thiosulfate solution may be ex-
Cu2+ + 2e− → Cu E 0 = 0.34 V (3) pressed in eq. (8):

2H+ + 2e− → H2 ↑ E0 = 0 V (4) [Au(S2 O3 )2 ]3− + e− → 2S2 O2−

3 + Au ↓ (8)
The anodic reaction of zinc precipitation is shown in eq.
Zn2+ + 2e− → Zn E 0 = −0.76 V (5)
(9):21–24)
Al3+ + 3e− → Al E 0 = −1.66 V (6) Zn + 2S2 O2− + 2e−
3 → [Zn(S2 O3 )2 ]
2−
(9)
The use of zinc, copper and aluminium has been suggested Since the pregnant solution contained ammonia, thiosulfate
and cupric amine complex, following reactions may also oc-
∗ Corresponding author: E-mail: [email protected]
486 H. Arima, T. Fujita and W.-T. Ye

cur:21–24) (20).18, 23)

Zn + 4NH3 + 2H2 O → [Zn(NH3 )4 ]2+ + 2OH− + H2 ↑ Al + 2S2 O2− −
3 → [Al(S2 O3 )2 ] + 3e
−
(20)
(10)
Also, aluminium will be oxidized in the ammonia solution
Zn + 2S2 O2−
3 + 2H2 O → [Zn(S2 O3 )2 ]
2−
+ 2OH− + H2 ↑ into aluminium hydroxo complex as shown in eq. (21):21, 24)
(11) 2Al + 2NH4 OH + 6H2 O → 2(NH4 )[Al(OH)4 ] + 3H2 ↑
Zn + [Cu(NH3 )4 ] 2+
→ [Zn(NH3 )4 ]
2+
+ Cu ↓ (12) (21)

The combination of eqs. (8), (9) and (10) is the overall re- The combination of eqs. (8), (20) and (21) is the over all
action of gold cementation by zinc powder in ammonium reaction of gold cementation with aluminium:
thiosulfate solution: 3Al + 3[Au(S2 O3 )2 ]3− + 2NH4 OH + 6H2 O
2Zn + 2[Au(S2 O3 )2 ]3− + 4NH3 + 2H2 O → [Al(S2 O3 )2 ]− + 2(NH4 )[Al(OH)4 ]
→ [Zn(S2 O3 )2 ]2− + [Zn(NH3 )4 ]2+ + 2S2 O2− + 4S2 O2−
3 3 + 3H2 ↑ +3Au ↓ (22)
−
+ 2OH + H2 ↑ +2Au ↓ (13) Equation (22) indicates that one mole of gold can be precipi-
Equation (13) indicates that one mole of zinc can precipitate tated by one mole of aluminium. Also, cupric amine complex
one mole of gold. Also, copper precipitation by zinc powder in the solution could be reduced by aluminium as shown in
could be occurred in the sum of eqs. (11) and (12): eq. (23):

2Zn + [Cu(NH3 )4 ]2+ + 2S2 O2− 2Al + 3[Cu(NH3 )4 ](OH)2 + 12H2 O

3 + 2H2 O
→ 2(NH4 )[Al(OH)4 ] + 10NH4 OH + 3Cu ↓ (23)
→ [Zn(NH3 )4 ]2+ + [Zn(S2 O3 )2 ]2− + 2OH− + H2 ↑ +Cu ↓
(14) Thus, the copper content in the barren solution could be di-
Equation (14) indicates that two moles of zinc could precipi- minished slightly:21, 22)
tate one mole of copper in the solution.
3. Experimental
2.3 Copper precipitation
Table 1 shows the ammonium thiosulfate leaching con-
In gold cementation by copper powder, cathodic reaction
ditions and the final compositions of the pregnant solu-
(8) will take place along with the following anodic reaction
tions obtained from a silicate type Hishikari ore.8) The
(15):19, 23)
range of reagent concentration were 1–5 mol/l NH4 OH, 0.01–
Cu + 2S2 O2−
3 → [Cu(S2 O3 )2 ]
3−
+ e− (15) 0.05 mol/l CuSO4 and 0.2–0.4 mol/l, (NH4 )2 S2 O3 and pH
9.5–10.5. The pH was adjusted by adding 10 vol% H2 SO4
Also, copper powder will be oxidized to cupric amine com-
or NaOH solution. The gold concentration in the pregnant
plex by ammonia as shown in eq. (16):21–24)
solution was 8 mg/l Au.
2Cu + 8NH3 + 2H2 O + O2 → 2[Cu(NH3 )4 ]2+ + 4OH− Average particle size of metal powder used for the gold ce-
(16) mentation was 45 µm. The amount of metal powders required
The combination of eqs. (8), (15) and (16) is the over all was calculated and added in a 100 ml pregnant solution. The
reaction of gold cementation with copper: sample was gently agitated by a magnetic stirrer for 60 min at
ambient temperature without de-aeration.
3Cu + [Au(S2 O3 )2 ]3− + 8NH3 + 2H2 O + O2 After the gold cementation, the pulp was filtered with mi-
cro filter immediately. The filtrate was assayed for gold, cop-
→ [Cu(S2 O3 )2 ]3− + 2[Cu(NH3 )4 ]2+ + 4OH− + Au ↓
(17)
Equation (17) demonstrates that one mole of gold needs three Table 1 Composition of the Initial and the Pregnant Solutions.8)
moles of copper in copper precipitation. Also, more copper
NH4 OH CuSO4 (NH4 )2 S2 O3 Au
powder could be oxidized as cuprous amine and thiosulfate (mol/l) (mol/l) (mol/l)
pH
(mg/l)
complexes as shown in eqs. (18) and (19):
Initial Preg. Initial Preg. Initial Preg. Initial Preg. Preg.
3 + 2H2 O + O2 → 4[Cu(S2 O3 )2 ]
4Cu + 8S2 O2− 3−
+ 4OH− 3 3 0.03 0.026 0.3 0.18 10.0 10.0 8.17
(18) 1 1 0.03 0.027 0.3 0.16 9.5 9.5 8.41
4Cu + 8NH3 + 2H2 O + O2 → 4[Cu(NH3 )2 ]+ + 4OH− 5 5 0.03 0.026 0.3 0.19 10.2 10.2 8.43
3 3 0.01 0.007 0.3 0.23 10.0 10.0 8.05
(19)
3 3 0.05 0.043 0.3 0.18 10.0 10.0 8.18
3 3 0.03 0.026 0.2 0.10 10.0 10.0 8.13
2.4 Aluminium precipitation
3 3 0.03 0.027 0.4 0.28 10.0 10.0 8.21
In gold cementation by aluminium powder, cathodic re-
3 3 0.03 0.026 0.3 0.24 9.5 9.5 8.40
action (8) will take place along with the anodic reaction 3 3 0.03 0.022 0.3 0.17 10.5 10.5 8.18
1 1 0.01 0.008 0.4 0.33 9.5 9.5 8.37
 Gold Cementation from Ammonium Thiosulfate Solution by Zinc, Copper and Aluminium Powders 487

120 0.06 0.12

100
100
80 0.09
80 0.04

Cu [mol/l]

Cu [mol/l]
60
60 0.06

40 0.02 40
0.03
20 20

0 0.00
0 0.00
0 50 100 150 200 250 300
0 100 200 300 400 500
Zn/Au+ Mass Ratio +
Cu/Au Mass Ratio
Au recovery: NH4OH 1mol/l 3mol/l 5mol/l
Au recovery: NH4OH 1mol/l 3mol/l 5mol/l
Cu conc. : NH4OH 1mol/l 3mol/l 5mol/l Cu conc. : NH4OH 1mol/l 3mol/l 5mol/l
Fig. 1 Effect of Zn/Au+ mass ratio and ammonia concentration on gold
Fig. 2 Effect of Cu/Au+ mass ratio and ammonia concentration on gold
recovery and copper concentration. 8 mg/l-Au pregnant solution with
recovery and copper concentration. 8 mg/l-Au pregnant solution with
0.03 mol/l CuSO4 and 0.3 mol/l (NH4 )2 S2 O3 at pH 9.5–10.2, 60 min agi-
0.03 mol/l CuSO4 and 0.3 mol/l (NH4 )2 S2 O3 at pH 9.5–10.2, 60 min agi-
tation.
tation.

per, thiosulfate and ammonia. Gold and copper concentra- alytic action of cupric amine complex as shown in eq. (7).
tions were measured by atomic adsorption spectrophotome-
The final copper concentrations at the optimum Cu/Au+ mass
ter. The thiosulfate concentration was measured by idiomet- ratio in the 1, 3 and 5 mol/l NH4 OH solutions were 0.045,
ric titration.25) The ammonia concentration was measured by 0.068 and 0.072 mol/l respectively. Additional 0.015, 0.038
Nessler’s method.26) The sensitively of Nessler’s method is at and 0.042 mol/l of copper was dissolved in the solution. The
semi-quantitative level, and the standard deviation is around
large consumption of copper powder at the higher ammonia
±0.5 mol/l NH+ 4 . The effects of ammonia, copper and thio- concentration may be attributed to the formation of cuprous
sulfate concentrations and pH on the gold recovery were in- amine and thiosulfate complexes as shown in eqs. (18) and
vestigated. To investigate the feasibility of recycling the am- (19). During the gold cementation, the dark blue solution was
monium thiosulfate solution, the total reductions of ammonia,
gradually faded away to light blue. The evidence demon-
copper and thiosulfate due to the gold cementation process strates that cupric amine complex was reduced to cuprous
were evaluated. amine complex by the copper powder dissolution as shown
in eq. (24):19)
4. Result and Discussion
Cu + [Cu(NH3 )4 ]2+ → 2[Cu(NH3 )2 ]+ (24)
4.1 Effect of ammonia and metal/gold mass ratio When the light blue-barren solution was left for a couple of
Figure 1 shows the effect of Zn/Au+ mass ratio and am- hours, the colour was back to dark blue gradually. This fact
monia concentration on gold recovery and final copper con- indicates that the large amount of cuprous amine complex was
centration. The results show that the optimum Zn/Au+ mass gradually oxidized to cupric amine complex as shown in eq.
ratio to recover 100% of gold in the presence of 1, 3 and (25):8, 19)
5 mol/l NH4 OH were 50, 150 and 150 respectively. The fi-
nal copper concentrations at the optimum Zn/Au+ mass ratio 4[Cu(NH3 )2 ]+ + 8NH3 + 2H2 O + O2
in the 1, 3 and 5 mol/l NH4 OH solutions were 0.010, 0.014
→ 4[Cu(NH3 )4 ]2+ + 4OH− (25)
and 0.010 mol/l respectively. It may be concluded that larger
amount of zinc powder was required at higher ammonia con- The results may be concluded that a lower ammonia concen-
centration to recover gold. The cause of zinc consumption tration is required to recover the gold with smaller copper
may be due to the oxidation reaction and the copper precipita- powder consumption.
tion as shown in eqs. (10), (11) and (12). Thus, low ammonia Figure 3 shows the effect of Al/Au+ mass ratio and ammo-
concentration could minimize the required zinc powder for nia concentration on gold recovery and final copper concen-
gold recovery. tration. The minimum Al/Au+ mass ratio to recover more
Figure 2 shows the effect of Cu/Au+ mass ratio and am- than 90% of gold in the 1, 3 and 5 mol/l NH4 OH solu-
monia concentration on gold recovery and final copper con- tions were 100, 350 and 150 respectively. The final cop-
centration. The optimum Cu/Au+ mass ratio to recover more per concentrations at the optimum Al/Au+ mass ratio in the
than 90% of gold in the 1, 3 and 5 mol/l NH4 OH solutions 1, 3 and 5 mol/l NH4 OH solutions were 0.025, 0.022 and
were 200, 350 and 400 respectively. The maximum gold re- 0.022 mol/l respectively. Less than 0.005 mol/l of copper
covery could not achieve 100% but 95%. The results may was co-precipitated with gold. The 3 mol/l NH4 OH solu-
be due to the re-dissolution of precipitated gold by the cat- tion consumed a larger amount of aluminium powder than the
488 H. Arima, T. Fujita and W.-T. Ye

120 0.06 120 0.06

100 100

80 0.04 80 0.04

Cu [mol/l]
Cu [mol/l]
60 60

40 0.02 40 0.02

20 20

0 0.00 0 0.00
0 100 200 300 400 500 0 100 200 300 400 500
+
Al/Au+ Mass Ratio Zn/Au Mass Ratio
Au recovery: NH4OH 1mol/l 3mol/l 5mol/l Au recovery: CuSO4 0.01mol/l 0.03mol/l 0.05mol/l
Cu conc. : NH4OH 1mol/l 3mol/l 5mol/l Cu conc. : CuSO4 0.01mol/l 0.03mol/l 0.05mol/l

Fig. 3 Effect of Al/Au+ mass ratio and ammonia concentration on gold Fig. 4 Effect of Zn/Au+ mass ratio and copper sulphate concentration on
recovery and copper concentration. 8 mg/l-Au pregnant solution with gold recovery and copper concentration. 8 mg/l-Au pregnant solution with
0.03 mol/l CuSO4 and 0.3 mol/l (NH4 )2 S2 O3 at pH 9.5–10.2, 60 min agi- 3 mol/l NH4 OH and 0.3 mol/l (NH4 )2 S2 O3 at pH 10.0, 60 min agitation.
tation.
0.25
100
solutions of 1 and 5 mol/l NH4 OH. This may be explained
0.20
that 1 mol/l of ammonia was not enough to stabilize cop-
80
per ion as cupric amine complex. In our previous study,8)
the thiosulfate consumption was at lowest between 3 and

Cu [mol/l]
0.15
60
5 mol/l NH4 OH. When the NH4 OH concentration decreased
to 1 mol/l, the thiosulfate consumption increased. Therefore, 0.10
it is assumed that the stability of [Cu(NH3 )4 ]2+ , which is the 40
catalyst for gold extraction, was greatly increased with the
increase of ammonia concentration from 1 mol/l to 3 mol/l, 20 0.05
and the Al/Au+ mass ratio required was increased. Further
increase of NH4 OH to 5 mol/l, the stability of [Al(OH)4 ]− 0 0.00
became higher than that of [Cu(NH3 )4 ]2+ due to the increase 0 100 200 300 400 500
of OH− and reduced the Al/Au+ mass ratio required. The +
Cu/Au Mass Ratio
consumption of aluminium may be due to the production of Au recovery: CuSO4 0.01mol/l 0.03mol/l 0.05mol/l
[Al(OH)4 ]− as shown in the eq. (21) and the formation of Cu conc. : CuSO4 0.01mol/l 0.03mol/l 0.05mol/l
Al2 O3 ∗ H2 O (boehmite) as shown in eq. (26):22)
Fig. 5 Effect of Cu/Au+ mass ratio and copper sulphate concentration on
2Al + 4H2 O → Al2 O3 ∗ H2 O + 3H2 ↑ (26) gold recovery and copper concentration. 8 mg/l-Au pregnant solution with
3 mol/l NH4 OH and 0.3 mol/l (NH4 )2 S2 O3 at pH 10.0, 60 min agitation.
The copper content was slightly diminished at 0.005–
0.008 mol/l. The results could be due to the production of alu-
minium hydroxo complex by copper reduction as explained in Thus, a low copper concentration could result in a high gold
eq. (23). recovery with a low copper powder consumption.
Figure 5 shows the effect of Cu/Au+ mass ratio and cop-
4.2 Effect of copper sulphate and metal/gold mass ratio per sulphate concentration on gold recovery and final copper
Figure 4 shows the effect of Zn/Au+ mass ratio and cop- concentration. The optimum Cu/Au+ mass ratio to recover
per sulphate concentration on gold recovery and final copper more than 90% of gold in the 0.01 and 0.03 mol/l CuSO4
concentration. The optimum Zn/Au+ mass ratio to recover solutions were 150 and 350 respectively. The final copper
100% of gold in the 0.01, 0.03 and 0.05 mol/l CuSO4 solu- concentrations at the optimum Cu/Au+ mass ratio in the 0.01
tions were 30, 150 and 250 respectively. The correspondent and 0.03 mol/l CuSO4 solutions were 0.021 and 0.068 mol/l
final copper concentrations at the optimum Zn/Au+ mass ratio respectively. The amount of dissolved copper concentration
in the 0.01, 0.03 and 0.05 mol/l CuSO4 solutions were 0.003, was double. The 0.05 mol/l CuSO4 solution could not recover
0.014 and 0.023 mol/l. The total copper loss at the optimum gold more than 65% in the range of 0 to 450 Cu/Au+ mass ra-
Zn/Au+ mass ratio in the 0.01, 0.03 and 0.05 mol/l CuSO4 tio. The results indicate that high copper sulphate concentra-
solutions were 0.007, 0.016 and 0.027 mol/l, which indicated tion could require more copper powder. The large consump-
that about half of copper was precipitated from the initial con- tion of copper powder at high copper sulphate concentration
centration. The precipitation of copper may be attributed to may be attributed to the production of cuprous thiosulfate and
the co-precipitation with gold by zinc as shown in eq. (14). amine complexes as shown in eqs. (18), (19) and (24). In or-
 Gold Cementation from Ammonium Thiosulfate Solution by Zinc, Copper and Aluminium Powders 489

120 0.09 120 0.06

100 100

80 0.06 80 0.04

Cu [mol/l]

Cu [mol/l]
60 60

40 0.03 40 0.02

20 20

0 0.00
0 0.00
0 100 200 300 400 500
0 50 100 150 200 250 300
+
Al/Au Mass Ratio
Zn/Au+ Mass Ratio
Au recovery: CuSO4 0.01mol/l 0.03mol/l 0.05mol/l
Au recovery: (NH4)2S2O3 0.2mol/l 0.3mol/l 0.4mol/l
Cu conc. : CuSO4 0.01mol/l 0.03mol/l 0.05mol/l
Cu conc. : (NH4)2S2O3 0.2mol/l 0.3mol/l 0.4mol/l
Fig. 6 Effect of Al/Au+ mass ratio and copper sulphate concentration on
gold recovery and copper concentration. 8 mg/l-Au pregnant solution with Fig. 7 Effect of Zn/Au+ mass ratio and ammonium thiosulfate concen-
3 mol/l NH4 OH and 0.3 mol/l (NH4 )2 S2 O3 at pH 10.0, 60 min agitation. tration on gold recovery and copper concentration. 8 mg/l-Au pregnant
solution with 3 mol/l NH4 OH and 0.03 mol/l CuSO4 at pH 10.0, 60 min
agitation.

der to minimize the copper powder consumption, it is better 0.12

to minimize the initial copper sulphate concentration at both
100
leaching and cementation stages.
Figure 6 shows the effect of Al/Au+ mass ratio and cop-
80 0.09
per sulphate concentration on gold recovery and final copper
concentration. The optimum Al/Au+ mass ratio to recover

Cu [mol/l]
more than 90% of gold in the 0.01 and 0.03 mol/l CuSO4 so- 60
0.06
lutions were 30 and 350 respectively. The final copper con-
centrations at the optimum Al/Au+ mass ratio in the 0.01 and 40
0.03 mol/l CuSO4 solutions were 0.000 and 0.022 mol/l re-
0.03
spectively. The 0.05 mol/l CuSO4 solution could not recover 20
gold in the range of 0 to 450 Al/Au+ mass ratio. As the cop-
per concentration increased, the more aluminium powder was
0 0.00
consumed. The aluminium consumption may be attributed to
0 100 200 300 400 500
the oxidation reaction of Al to [Al(OH)4 ]− or Al2 O3 ∗ H2 O as
indicated in eqs. (19) and (24). Thus, a low copper concen- Cu/Au+ Mass Ratio
tration is preferred for minimizing the consumption of alu- Au recovery: (NH4)2S2O3 0.2mol/l 0.3mol/l 0.4mol/l
minium powder. Cu conc. : (NH4)2S2O3 0.2mol/l 0.3mol/l 0.4mol/l

Fig. 8 Effect of Cu/Au+ mass ratio and ammonium thiosulfate concen-

4.3 Effect of ammonium thiosulfate and metal/gold mass tration on gold recovery and copper concentration. 8 mg/l-Au pregnant
ratio solution with 3 mol/l NH4 OH and 0.03 mol/l CuSO4 at pH 10.0, 60 min
Figure 7 shows the effect of Zn/Au+ mass ratio and am- agitation.
monium thiosulfate concentration on gold recovery and final
copper concentration. The optimum Zn/Au+ mass ratio to re- Figure 8 shows the effect of Cu/Au+ mass ratio and thio-
cover 100% of gold in the 0.2, 0.3 and 0.4 mol/l (NH4 )2 S2 O3 sulfate concentration on gold recovery and final copper con-
solutions were 200, 150, 100 respectively. The correspondent centration. The optimum Cu/Au+ mass ratio to recover more
final copper concentrations at the optimum Zn/Au+ mass ra- than 90% of gold in the 0.2, 0.3 and 0.4 mol/l (NH4 )2 S2 O3 so-
tio in the 0.2, 0.3 and 0.4 mol/l (NH4 )2 S2 O3 solutions were lutions were 400, 350, 350 respectively. The final copper con-
0.011, 0.014 and 0.016 mol/l respectively. Higher thiosul- centrations at the optimum Cu/Au+ mass ratio in the 0.2, 0.3
fate concentration reduced both the zinc powder required and and 0.4 mol/l (NH4 )2 S2 O3 solutions were 0.067, 0.068 and
the loss of copper ion in gold recovery. The results indi- 0.055 mol/l respectively. The large amount of copper powder
cate that zinc powder was oxidized easily as [Zn(S2 O3 )2 ]2− was required at the low thiosulfate concentration. The result
with reducing [Au(S2 O3 )2 ]3− at a higher thiosulfate concen- suggests that copper powder was oxidized to [Cu(S2 O3 )2 ]3−
tration. It also confirms the results of Fig. 1, which indi- at a higher thiosulfate concentration. Thus, the condition of
cated that the lower ammonia concentration recovered more high thiosulfate concentration could result in a high gold re-
gold at the lower Zn/Au+ mass ratio due to the increase of covery with a lower copper consumption.
S2 O2−
3 /NH3 (aq) mole ratio. Thus, higher thiosulfate concen- Figure 9 shows the effect of Al/Au+ mass ratio and am-
tration would minimize the amount of zinc powder required. monium thiosulfate concentration on gold recovery and final
490 H. Arima, T. Fujita and W.-T. Ye

120 0.06 120 0.06

100 100

80 0.04 80 0.04

Cu [mol/l]

Cu [mol/l]
60 60

40 0.02 40 0.02

20 20

0 0.00 0 0.00
0 100 200 300 400 500 0 50 100 150 200 250
Al/Au+ Mass Ratio
Zn/Au+ Mass Ratio
Au recovery: (NH4)2S2O3 0.2mol/l 0.3mol/l 0.4mol/l
Au recovery pH9.5 pH10.0 pH10.5
Cu conc. : (NH4)2S2O3 0.2mol/l 0.3mol/l 0.4mol/l
Cu conc. pH9.5 pH10.0 pH10.5
Fig. 9 Effect of Al/Au+ mass ratio and ammonium thiosulfate concen-
tration on gold recovery and copper concentration. 8 mg/l-Au pregnant Fig. 10 Effect of Zn/Au+ mass ratio and pH on gold recovery and cop-
solution with 3 mol/l NH4 OH and 0.03 mol/l CuSO4 at pH 10.0, 60 min per concentration. 8 mg/l-Au pregnant solution with 3 mol/l NH4 OH,
agitation. 0.03 mol/l CuSO4 and 0.3 mol/l (NH4 )2 S2 O3 , 60 min agitation.

0.12
copper concentration. The maximum gold recovery in the 100
0.2 mol/l (NH4 )2 S2 O3 solution was 61% at 450 of Al/Au+
mass ratio. The optimum Al/Au+ mass ratio to recover more 80 0.09
than 90% of gold in the 0.3 and 0.4 mol/l (NH4 )2 S2 O3 solu-

Cu [mol/l]
tions were 350 and 100 respectively. The final copper con- 60
centrations at the optimum Al/Au+ mass ratio in the 0.3 and 0.06
0.4 mol/l (NH4 )2 S2 O3 solutions were 0.022 and 0.024 mol/l.
40
The results indicate that thiosulfate may oxidize aluminium
as a product of [Al(S2 O3 )2 ]− as shown in eq. (27): 0.03
20
− −
2Al + 2S2 O2−
3 + 6H2 O → 2[Al(S2 O3 )2 ] + 6OH + 3H2 ↑
(27)
0 0.00
Compared to the result of zinc precipitation as shown in 0 100 200 300 400 500
Fig. 7, the diminished copper concentration was low at around +
Cu/Au Mass Ratio
0.007 mol/l, which support the results that aluminium was
Au recovery pH9.5 pH10.0 pH10.5
oxidized to thiosulfate complex rather than amine complex. Cu conc. pH9.5 pH10.0 pH10.5
In conclusion, 0.4 mol/l (NH4 )2 S2 O3 should be the optimum
concentration for the gold recovery by aluminium precipita- Fig. 11 Effect of Cu/Au+ mass ratio and pH on gold recovery and cop-
per concentration. 8 mg/l-Au pregnant solution with 3 mol/l NH4 OH,
tion.
0.03 mol/l CuSO4 and 0.3 mol/l (NH4 )2 S2 O3 , 60 min agitation.

4.4 Effect of pH and Metal/Gold mass ratio

Figure 10 shows the effect of Zn/Au+ mass ratio and pH Copper ion is stable as amine complex rather than thiosul-
on gold recovery and final copper concentration. At the fate complex around pH 9.5,8, 27) and zinc has an affinity
reagent combination of 3 mol/l NH4 OH, 0.03 mol/l CuSO4 for both ammonia and thiosulfate to form [Zn(NH3 )4 ]2+ and
and 0.3 mol/l (NH4 )2 S2 O3 , the optimum Zn/Au+ mass ratio [Zn(S2 O3 )2 ]2− as explained in eqs. (10) and (11). This evi-
to recover 100% of gold was 150 at the pH range of 9.5–10.5. dence also supports the eq. (28). Thus, pH 9.5 could be the
The copper concentrations at the optimum Zn/Au+ mass ratio optimum pH for zinc precipitation.
in the pH 9.5, 10.0 and 10.5 solutions were 0.021, 0.014 and Figure 11 shows the effect of Cu/Au+ mass ratio and pH
0.016 mol/l respectively. The result indicates that lower pH on gold recovery and final copper concentration. At the
might prevent the copper precipitation. The cause of lower reagent combination of 3 mol/l NH4 OH, 0.03 mol/l CuSO4
copper consumption at pH 9.5 is assumed that cupric amine and 0.3 mol/l (NH4 )2 S2 O3 , the optimum Cu/Au+ mass ratio
worked as an oxidant for zinc powder to produce zinc thiosul- to recover 90% of gold in the pH 9.5, 10.0 and 10.5 solutions
fate complex and assisted to reduce aurous thiosulfate com- were 300, 350 and 400 respectively. The copper concentra-
plex as shown in eq. (28): tions at the optimum Cu/Au+ mass ratio in the solutions at
Zn + 2S2 O2− pH 9.5, 10.0 and 10.5 were 0.059, 0.068 and 0.066 mol/l re-
3 + 2[Cu(NH3 )4 ]
2+
spectively. The lower the pH was, the less the copper was
→ [Zn(S2 O3 )2 ]2− + 2[Cu(NH3 )2 ]+ + 4NH3 (28) consumed. The small amount of copper powder required at
 Gold Cementation from Ammonium Thiosulfate Solution by Zinc, Copper and Aluminium Powders 491

120 0.06 120 0.025

100 100
0.020

80 0.04 80

Cu [mol/l]
Cu [mol/l]
0.015
60 60
0.010
40 0.02 40

20 0.005
20

0 0.00 0 0.000
0 100 200 300 400 500 0 50 100 150 200 250
Al/Au+ Mass Ratio +
Metal/Au Mass Ratio
Au recovery pH9.5 pH10.0 pH10.5
Cu conc. pH9.5 pH10.0 pH10.5 Au recovery Zn Cu Al
Cu conc. Zn Cu Al
Fig. 12 Effect of Al/Au+ mass ratio and pH on gold recovery and cop-
per concentration. 8 mg/l-Au pregnant solution with 3 mol/l NH4 OH, Fig. 13 Effect of Metal/Au+ mass ratio on gold recovery and copper
0.03 mol/l CuSO4 and 0.3 mol/l (NH4 )2 S2 O3 , 60 min agitation. concentration. 8 mg/l-Au pregnant solution with 1 mol/l NH4 OH and
0.01 mol/l CuSO4 and 0.4 mol/l (NH4 )2 S2 O3 at pH 9.5, 60 min agitation.

the pH 9.5 may be attributed to the activation of cupric amine

complex as the oxidant for copper powder to produce cuprous The pH for the optimum combination at 0.01 mol/l CuSO4 ,
thiosulfate complex. In the study of gold leaching,8) it was 1 mol/l NH4 OH and 0.4 mol/l (NH4 )2 S2 O3 is 9.5. At this
confirmed that Cu2+ is stable as [Cu(NH3 )4 ]2+ rather than lower ammonia and copper concentrations and higher thio-
CuO at pH around 9.5. These facts suggest that copper pow- sulfate concentration, pH adjustment would not be required
der was oxidized to [Cu(S2 O3 )2 ]3− readily by the catalytic for gold cementation.
action of [Cu(NH3 )4 ]2+ as pH decreased from 10.5 to 9.5 in
eq. (29): 4.5 Final evaluation for Zn, Cu and Al precipitation
4.5.1 Gold recovery and copper concentration
Cu + 2S2 O2−
3 + [Cu(NH3 )4 ]
2+
From the results of Figs. 1–12, the optimum condition
→ [Cu(S2 O3 )2 ]3− + [Cu(NH3 )2 ]+ + 2NH3 (29) could be concluded as 1 mol/l NH4 OH, 0.01 mol/l CuSO4 and
0.4 mol/l (NH4 )2 S2 O3 at pH 9.5. Figure 13 confirms the ef-
Thus, pH 9.5 could be the optimum condition to cement more fect of Metal/Gold mass ratio on gold recovery and final cop-
gold with less copper powder consumption. per concentration at this optimum condition. The optimum
Figure 12 shows the effect of Al/Au+ mass ratio and Zn/Au+ and Al/Au+ mass ratio to recover 100% of gold were
pH on gold recovery and final copper concentration at the 30. The optimum Cu/Au+ mass ratio to recover 90% of gold
reagent combination of 3 mol/l NH4 OH, 0.03 mol/l CuSO4 was 50. To achieve 100% of gold recovery by copper pow-
and 0.3 mol/l (NH4 )2 S2 O3 . The results showed that the so- der, a Cu/Au+ mass ratio of 200 was required. The final
lution at pH 9.5 could not recover gold by aluminium in the copper concentrations at the optimum Metal/Gold mass ratio
range of 0 to 450 Al/Au+ mass ratio. The optimum Al/Au+ for Zn, Cu and Al were 0.001, 0.009 and 0.003 mol/l respec-
mass ratio to recover more than 80% of gold in the solutions tively. Copper is not only the most expensive metal among
at pH 10.0 and 10.5 was 350 for both cases. The final cop- the used metal powders but also required a larger Metal/Au+
per concentrations at the optimum Al/Au+ mass ratio in the mass ratio than that of zinc and aluminium. The total reduc-
solutions at pH 10.0 and 10.5 were 0.022 and 0.000 mol/l re- tion of copper concentration by zinc and aluminium powders
spectively. The solution at pH 10.5 precipitated entire cop- was respectively 0.009 mol/l and 0.007 mol/l. The replen-
per in the ammonium thiosulfate solution. The large copper ished amount of copper in zinc precipitation is more than that
consumption may be attributed to the increase of aluminium in aluminium precipitation process. The excess addition of
solubility at high pH. Aluminium cannot form amine com- copper powder increased the copper dissolution. For exam-
plex. It means that [Cu(NH3 )2 ]2+ cannot work as an oxi- ple, at the Cu/Au+ mass ratio of 200, final copper concentra-
dant for aluminium oxidation to form thiosulfate complex like tion was 0.003 mol/l higher than the initial copper concentra-
Au, Zn and Cu as shown in eqs. (7), (28) and (29). Alu- tion. In conclusion, for recycling these barren solutions at the
minium is less soluble as [Al(OH)4 ]− at a lower pH.21, 22) optimum Metal/Gold mass ratio to the gold leaching stage,
These facts suggest that the increase of pH without the use it could be required to make up the copper concentration for
of ammonia causes the large copper consumption due to the both zinc and aluminium precipitations and not required for
increase of the [Al(OH)4 ]− stability against the decrease of copper precipitation.
the [Cu(NH3 )4 ]2+ stability as explained in eq. (23). The poor 4.5.2 Ammonia and thiosulfate concentration
gold recovery at pH 9.5 could be attributed to the increase Under the selected optimum condition (Section 4-5-1), the
of [Cu(NH3 )4 ]2+ stability and the poor [Al(OH)4 ]− stability. effects of Metal/Gold mass ratio on the final ammonia and
492 H. Arima, T. Fujita and W.-T. Ye

Table 2 Comparison of Three Metal Powders in Gold Cementation at the Optimum Reagent Combination of 1 mol/l NH4 OH, 0.01 mol/l
CuSO4 ∗ 5H2 O and 0.4 mol/l (NH4 )2 S2 O3 at pH 9.5.

Zn Cu Al
Metal/Au+ mass ratio
30 50 30
(8 mg/l Au)
Gold recovery 100% 93% 100%
Thiosulfate consumption Increase Decrease Slightly decrease
Copper concentration Decrease to near zero Increase greatly Decrease slightly
Need to replenish copper Need to replenish thiosulfate Need to replenish both copper and
Recycling barren solution
concentration concentration thiosulfate concentration

3 0.45 pregnant solution were 1–5 mol/l NH4 OH, 0.01–0.05 mol/l
CuSO4 , 0.2–0.4 mol/l (NH4 )2 S2 O3 and pH 9.5–10.5. The re-
2
sults demonstrated that the lower ammonia and copper con-
0.40
centrations and higher thiosulfate concentration could min-
1

S2O32- [mol/l]
NH4+ [mol/l]

imize the required Metal/Gold mass ratio to recover more

0 0.35 gold. It was concluded that the optimum reagent combination
for gold cementation was 1 mol/l NH4 OH, 0.01 mol/l CuSO4
-1 and 0.4 mol/l (NH4 )2 S2 O3 at pH 9.5. Under this optimum
0.30 condition, the following results were concluded.
-2 NH4+: Zn Cu Al (1) Zinc and aluminium precipitations recovered 100% of
S2O32-: Zn Cu Al gold with Metal/Gold mass ratio of 30. The copper consump-
-3 0.25 tion was 0.009 mol/l by zinc precipitation and 0.007 mol/l by
0 50 100 150 200 250 aluminium precipitation.
+
Metal/Au Mass Ratio
(2) Copper precipitation was the lowest copper loss with
93% of gold recovery at Cu/Au+ mass ratio of 50. The excess
Fig. 14 Effect of Metal/Au+ mass ratio on thiosulfate and ammonia addition of copper powder increased the copper concentra-
concentrations. 8 mg/l-Au pregnant solution with 1 mol/l NH4 OH and
tion.
0.01 mol/l CuSO4 and 0.4 mol/l (NH4 )2 S2 O3 at pH 9.5, 60 min agitation.
(3) At Zn/Au+ mass ratio of 100, the thiosulfate concen-
tration increased 0.01–0.02 mol/l at zinc precipitation. On
thiosulfate concentrations are shown in Fig. 14. the other hand, the thiosulfate concentration decreased 0.01–
As the Zn/Au+ mass ratio was higher than 50, the final thio- 0.03 mol/l at aluminium and copper precipitations.
sulfate concentration increased 0.01–0.02 mol/l. This may be (4) Ammonia concentration did not changed significantly
attributed to the formation of amine complex as shown in eqs. during the gold cementation process.
(13) and (30):
Acknowledgements
[Zn(S2 O3 )2 ]2− + 4NH3 → [Zn(NH3 )4 ]2+ + 2S2 O2−
3 (30)
At copper precipitation, 0.01–0.03 mol/l S2 O2− was con- Authors would like to acknowledge that Sumitomo Metal
3
sumed at the Cu/Au+ mass ratio of 200. At aluminium pre- Mining Co., Ltd. provided the gold ore samples and the fi-
cipitation, the Al/Au+ mass ratio did not significantly change nancial support from MMAJ, Metal Mining Agency of Japan,
the thiosulfate concentration. However, excess addition of and CANMENT, Canada Centre for Mineral and Energy
aluminium powder to the Al/Au+ mass ratio of 150 caused Technology.
about 0.01 mol/l S2 O2−
3 reduction. The thiosulfate consump-
tion may be attributed to the formation of [Al(S2 O3 )2 ]− as REFERENCES
shown in eqs. (22) and (27). 1) R. D. De Voto and T. P. McNulty: Mining Engineering, V 52, No. 12
Ammonia concentration maintained from the beginning to (2000) 19–27.
the end of gold cementation process for all three metal pow- 2) “Cambier forces second suit over cyanide spill at mine”, National Post,
ders. Thursday, August 19 (1999).
3) R. J. Lewis: “Hazardous Chemicals Desk Reference Forth Edition”
The comparison of three metal powders in gold cementa- (1989).
tion is shown in Table 2. The choice of metal powder de- 4) P. L. Breuer and M. I. Jeffrey: Minerals Engineering, 13 (2000) 1071–
pends upon the efficiency of gold recovery and the feasibility 1081.
of recycling the barren solution. 5) W. T. Yen, K. Storogan and T. Fujita: Resources Treatment Technology,
V 43–2 (1996) 83-87.
6) W. T. Yen, M. Aghamirian, G. Deschenes and S. Theben: Interna-
5. Conclusion tional Symposium on gold recovery, CIM, Montreal Quebec, Canada,
Paper#THAM61.1, May 3–7, (1998).
Gold cementation test was carried out from an 8 mg/l 7) W. T. Yen, H. Guo and G. Deshenes: “Development in percolation
leaching with ammonium thiosulfate for gold extraction of a mild re-
Au ammonium thiosulfate solution by using 45 µm of zinc, fractory ore”, EPD Congress, ed. B. Mishra (The Minerals, Metals &
copper and aluminium powders. The composition of the
 Gold Cementation from Ammonium Thiosulfate Solution by Zinc, Copper and Aluminium Powders 493

Materials Society, 1999) pp. 441–455. in aqueous solutions, (Prentice-Hall, Inc., 1959).
8) H. Arima and W. T. Yen: “Development of ammonium thiosulfate 18) A. B. Bjerre and E. Sorensen: Trans. Instn. Min. Metall. 98 (1989) C84-
leaching process for gold extraction from a Hishikari ore”, Queen’s C87.
University, unpublished data (2001). 19) E. Guerra and D. B. Dresinger: Hydrometallurgy, 51 (1999) 155–172.
9) J. D. Miller, R. Y. Wan and J. R. Parga: Hydrometallurgy, 24 (1990) 20) R. Ravaglia and O. Barbosa-Filho: Hydro and Biohydrometallurgy,
373–392. A6-41-A6-47, Rome, Italy, (2000).
10) C. C. Kenna, I. M. Ritche and P. Singh: Hydrometallurgy, V 23 (1990) 21) Y. Hoshino: Photo Science—KAGAKU ZUROKU, (SUKEN Publish-
263–279. ing, Co., Ltd., 1999).
11) V. M. Torres, A. P. Chaves and J. A. Meech: Minerals Engineering, 12 22) M. Pourbaix: Atlas of Electrochemical Equilibria in Aqueous Solution,
No. 3 (1999) 245–254. (Oxford University Press, U.K., 1966).
12) A. R. Heath and J. A. Rumball: Minerals Engineering, 11 (1998) 999– 23) M. G. Aylmore and D. M. Muir: Minerals Engineering, 14 (2001) 135–
1010. 174.
13) W. Stange: The journal of The South African Institute of Mining and 24) R. M. Smith, A. E. Martell and R. J. Motekaitis: “Critically Selected
Metallurgy, January/February (1999) 13–25. Constants of Metal Complexes Database, Ver. 5.0”, (National Institute
14) S. V. Carbajal, I. Gonzalez and G. T. Lapidus: J. Appl. Electrochem., of Standards and Technology, 1998).
30 (2000) 217–229. 25) W. H. Skoog: 7th Edition, Saunders College Publishing (1996).
15) R. Y. Wan and J. A. Brierley: Mining Engieering, August (1997) 26) JIS 0102, Testing Methods for Industrial Waste Water, Japanese Indus-
pp. 76–80. trial Standards Committee, July 17 (1971).
16) A. R. Marchbank, K. G. Thomas, D. Dreisinger and C. Fleming: U.S. 27) K. Tozawa: “Leaching”, Non-Ferrous Metallurgy, The Japan Institute
Patent 5, 536, 297, (1996). of Metals, Vol. 2, pp. 174, (1980).
17) W. M. Latimer: The oxidation states of the elements and their potentials

View publication stats