THE FERRIC ION -GOD’S GIFT TO

HYDROMETALLURGISTS TO KEEP ‘EM HUMBLE

By Chris Fleming

SGS Lakefield Research Ltd.

2

TOPICS

Ø The first SEx war

Ø There’s nothing basic about basic iron sulphate

3

THE FIRST SEx WAR

History§ Great interest in solvent extraction in the 1960’s, initially

for uranium and then for copper.

§ Low grade oxide copper ores were being processed in the USA and Chile by acidic heap leaching (small scale).

§ Copper sulphate in solution was recovered by cementation onto scrap iron, and then smelted and refined.

§ Most leach liquors contained ferric ions, which reacted with the scrap iron wastefully, and the process was:

(i) Expensive(ii) Yielded an impure copper product

4

THE FIRST SEx WAR

History§ Early economic projections predicted the cost of SEx/EW

would be half the cost of cementation/smelting/refining.§ The “perfect” reagent had to be able to extract copper (II)

from weak acid leach solution (pH 1-2), be strippable in strong acid (50 -100g/L H2SO4) to be compatible with EW, and be very selective for copper – particularly versus the ferric ion.

§ Copper SEx reagent development was spearheaded by two US companies, General Mills, who produced the LIX reagents, and Ashland Chemical, who produced Kelex100.

5

H

O

R N

HYDROXYQUINOLINE

KELEX

C = N

R

HO

LIX

HYDROXYOXIME

OH

COPPER SEx REAGENTS

6

THE FIRST SEx WAR

History§ The first copper SEx plant was built at the Blue Bird

Mine in Arizona in 1968 (6000 tpa Cu), soon followed by a much bigger plant at Nchanga mine in Zambia (65,000 tpa Cu).

§ More plants followed, and 3% of world copper production was via SEx/EW by 1975. By 2007, this had grown to 22% of annual Cu production, (3.5M tons of cathode copper). This was being produced in 70 SEx/EW plants in 16 countries (60% in Chile).

§ But who was making the SEx reagents and who was winning the reagent war?

7

Copper extraction with LIX and KELEX reagents as a function of pH

-10

10

30

50

70

90

0 1 2 3 4 5 6

pH

Copper Extraction

(%)

LIX 63 4.8LIX 64 3.3LIX 64N 2.9LIX 70 2.6KELEX 100 1.8

0

Initial Rate of Copper Extraction (g/L/min)KELEX 100 0.98LIX 64N 0.11

8

KINETIC AND STABILITY CONSTANTSFOR THE REACTION OF Cu(II) AND Fe(III)

WITH HYDROXYQUINOLINES

Initial Rate of Extraction with Kelex

100 (g/L/min)

III

Metal

Cu( )Fe( II)

0.980.067

Stability with 8 Hydroxy Quinoline

Log β2 = 23.0Log β3 = 36.9

9

RATES OF EXTRACTION OF Cu(II) AND Fe(III) BY KELEX 100

0

25

50

75

100

0 50 100 150 200Stirring time, min

Extraction with 10% KELEX 100

(%)Cu2+ from pH 1 solutionCu2+ from pH 2 solution

0

25

50

75

100

0 50 100 150 200Stirring time, min

Extraction with 10% KELEX 100

(%)Cu2+ from pH 1 solutionCu2+ from pH 2 solutionFe3+ from pH 1 solutionFe3+ from pH 2 solution

10

6157591000555250802049483940604038200100CuCuCu

432 h24 h1 hmol %mol %

Metal Extracted (%)Salicyl-aldoxime

LIX65N

RATES OF EXTRACTION OF Cu(II) AND Fe(III) BY LIX65N AND ITS PRECUSOR,

SALICYLALDOXIME

Fe2 1 2 1

Fe2 1 2 2

Fe2 1

0.30.1

11

THE MORAL OF THE STORY

If you want to win a SEx war, it is better to be slow and selectivethan to be fast and flirtatious.

THERE’S NOTHING BASIC ABOUT

BASIC IRON SULPHATE

13

BACKGROUND

§ Basic iron sulphate (BFS) is a solid compound that is formed under certain conditions during the oxidation of pyrite or other iron sulphide minerals with oxygen at high temperatures in an autoclave.

§ Iron sulphide minerals are oxidized in an autoclave to produce ferric sulphate and sulphuric acid in solution. The ferric sulphate then hydrolyzes slowly, precipitating back out of solution as hematite and/or BFS.

14

OXIDATION

2FeS2 + 702 + 2H2O → 2FeSO4 + 2H2SO4

4FeSO4 + 2H2SO4 + O2 → 2Fe2(SO4)3 + 2H2O

Overall:4FeS2 + 1502 + 2H2O → 2Fe2 (SO4)3 + 2H2SO4

15

HYDROLYSIS

Ferric sulphate hydrolyzes to hematite at higher temperatures and lower acidity

Fe2(SO4)3 + 3H2O → Fe2O3 + 3H2SO4

and it hydrolyzes to BFS at lower temperatures and higher acidity

Fe2(SO4)3 + 2H2O⇌ 2Fe(OH)SO4 + H2SO4

16

OXIDATION AND HYDROLYSIS

Overall reaction for the oxidation of pyrite to ferric sulphate followed by hydrolysis to hematite

4FeS2 + 15O2 + 8H2O → 2Fe2O3 + 8H2SO4

Overall reaction for the oxidation of pyrite to ferric sulphate followed by hydrolysis to BFS

4FeS2 + 15O2 + 6H2O → 4Fe(OH)SO4 + 4H2SO4

17

Stability domains of compounds of the ferric ion in water as a function of temperature and pH

20

60

100

140

180

220

260

0 2 4 6 8 10 12

pH

Temp (ºC)

Fe 3+

Fe (OH)3

Goethite FeO.OH

Hematite Fe2O3

Basic Iron SulphateFe(OH)SO4

18

WHY IS BFS BAD NEWS IN A CYANIDATION CIRCUIT?

§ BFS is not basic, it is actually acidic…..and it must be neutralized before cyanidation

§ The rate of release of acid by BFS is extremely slow in weakly acidic solution (pH <7). This means BFS cannot be neutralized with a cheap alkali such as limestone.

Fe(OH)SO4 + Ca(OH)2pH>7 Fe(OH)3 + CaSO4

pH3.5very slowFe(OH)SO4 + CaCO3 + H2O Fe(OH)3 + CaSO4 + CO2

19

WHY IS BFS BAD NEWS IN A CYANIDATION CIRCUIT?

§ Even under alkaline conditions, the rate of release of acid by BFS is quite slow – but it is persistent at the pH needed for cyanide leaching (pH ~10)

§ Consequently, the pH constantly drifts downwards under normal cyanide leach operating conditions, into the pH region where cyanide is converted to HCN gas.

20

WHY IS BASIC IRON SULPHATE BAD NEWS IN CYANIDATION CIRCUITS

Ø For health and safety reasons related to HCN formation, the BSF must be fully neutralized prior to cyanidation. This will take 12-24 hours and add significantly to plant capital cost.

Ø As a result, most of the sulphate generated in the autoclave has to be neutralized with hydrated lime, rather than limestone. Lime can be at least 10 times the price of limestone.

Ø If not dealt with appropriately, the increased capex and opex associated with BFS formation could eliminate POX from consideration for many refractory gold projects

21

WHAT IS THE BEST SOLUTION?

THE “HOT CURE” PROCESS

22

The basis of the hot cure process is the fact that the hydrolysis reaction that produces BFS in the autoclave is reversible at lower temperatures:

BFS Formation

BFS Decomposition

THE HOT CURE PROCESS

Fe2(SO4)3 + 2H2O Fe(OH)SO4 + H2SO4T>150ºC

Fe(OH)SO4 + H2SO4 Fe2(SO4)3 + 2H2O90-140ºC

fastRT

very slow

23

2Fe(OH)SO4 + H2SO4 Fe2(SO4)3 + 2H2O

THE HOT CURE PROCESS

20

60

100

140

180

220

260

0 2 4 6 8 10 12

pH

Temp (ºC)

Fe3+

Fe (OH)3

Goethite FeO.OH

Hematite Fe2O3

BFS

24

THE HOT CURE PROCESS

Once the basic iron sulphate has decomposed to ferric sulphate, it can be separated from the solids by CCD or filtration, and neutralized with limestone

Fe2(SO4)3 + 3CaCO3 + 3H2O 2Fe(OH)3 + 3CaSO4 +3CO2

25

NEUTRALIZATION OF THE ACID AND SULPHATE WITH LIMESTONE

(1) Fe2(SO4)3 + 3CaCO3 + H2O 2FeO.OH + 3CaSO4 + 3CO2

(2) Fe2(SO4)3 + 3CaCO3 + 3H2O 2Fe(OH)3 + 3CaSO4 + 3CO2

20

60

100

140

180

220

260

0 2 4 6 8 10 12 14pH

Temp (ºC)

Fe3+

Hematite Fe2O3

BFS

Goethite FeO.OH

Fe(OH)3

(1)

(2)

26

A REFRACTORY GOLD POX FLOWSHEET INCORPORATING HOT CURING

Oxygen

Concentrate

Pressure Oxidation

Hot Cure

Solid/Liquid Separation

SolidLiquid

Solid/Liquid Separation

Neutralisation

Solid

CaSO4

Fe(OH)3

Cyanide Leach

Gold Recovery

CyanideDestruction

Tailings

Steam90 – 100 °C 4 to 12 hours

CO2

NaCNCa(OH)2

Liquid

Base Metal Recovery ?

CaCO3

27

QUIMSACOCHA PROJECT, ECUADOR (IAMGOLD CORPORATION)

∆Fe (mass/mass) = 18.1%∆SO4 (mass/mass) = 32.0%

Fe/SO4 = 0.57

0

10

20

30

40

-3 0 3 6

Time at 90ºC (hr)

Fe

SO4

ACD

Conc. In AutoclaveDischarge

Solids(%)

28

QUIMSACOCHA PROJECT

0

10

20

30

40

50

60

70

-3 0 3 6Time (hr)

Autoclave Discharge Solution

(g/L)

Fe

H2SO4

ACD

29

QUIMSACOCHA PROJECT

Concentrate head grade: 24 g/t Au, 104 g/t Ag

PRODUCT RECOVERY ALKALI CONSUMED APPROX. COST

Au % Ag % CaCO3 kg/t Ca(OH)2 kg/t $/t

Autoclave Discharge

99.6 94.8 370 260 43

Hot Cure Discharge

99.4 91.9 704 15 9

30

31