Geochemical Controls On Acid Mine Drainage Generation, Neutralisation Treatment and Attenuation at Iron Duke Mine

Geochemical Controls on Acid Mine Drainage Generation,
Neutralisation Treatment and Attenuation at Iron Duke Mine

C Magombedze1, K L Sandvik2, B Manda3 and K Musiwa4
1. GAusIMM, Norwegian University of Science and Technology – NTNU, Department of

Geology and Mineral Resources Engineering, 7491 Trondheim, Norway.
Email: [email protected]
2. Norwegian University of Science and Technology – NTNU, Department of Geology and
Mineral Resources Engineering, 7491 Trondheim, Norway.
3. Iron Duke Pyrites (Pvt) Limited, Private Bag 2003, Glendale, Zimbabwe.
4. University of Zimbabwe, Department of Mining Engineering, PO MP167, Mount Pleasant,
Harare, Zimbabwe.
ABSTRACT
Mining and processing of pyrite ore at Iron Duke Mine, north east of Zimbabwe, generates large
quantities of acidic mine water as a result of the oxidation of iron sulfides on exposure to air and
water used in the drilling and blasting operations. The acid mine drainage (AMD) contains elevated
levels of dissolved Fe, Al, Mn and sulfate and has characteristic pH values ranging from about 0.50 to
3.00. Handling and disposal of such a drainage places significant impacts on the economics of the
mining operation due to the corrosive nature of the drainage on infrastructure, the limitations placed
on water reuse and discharge to the environment under a new law governing effluent discharge.
Treated sludge at Iron Duke is chemically unstable, changing pH from ~7 to ~3 within four days after
deposition in the evaporation ponds. The effluent does not meet the permitted level of neutral pH, 500
mg/l sulfate and less than 2000 mg/l total dissolved solids. Excessive acid seepage from the
evaporation ponds has resulted in contamination of the ground water below the ponds area, which
shows high mean annual levels of iron (368.67 mg/l), sulfate (1702.89 mg/l) and low pH (2.86) in
addition to other pollutants. This poor quality seepage infiltrates into the nearby Yellow Jacket River,
contaminating the public water source. Furthermore, a deterioration of the land area and vegetation
adjacent to the tailings dump is also observed.
Understanding the controls on the neutralisation of acidic drainage and long-term sludge stability after
placement in evaporation ponds is essential to minimise costs of closure requirements and for the
protection of environment and human health. This study examined the effects of the master variables
pH, redox potential, acidity and alkalinity on acid neutralisation. The results indicate that ferrous
hydroxides are not as stable as ferric hydroxides and that proper oxidation of iron also leads to better
sludge stability, higher treatment efficiency and a manageable sludge viscosity. Metal hydrolysis and
secondary mineral formation at the evaporation ponds under a wide range of pH and redox conditions
is responsible for equilibrium shifts in metal reactions, resulting in high potential for metal leaching
and mobility. Mechanisms of metal attenuation are also discussed.
INTRODUCTION
The mining and processing of pyrite ore at Iron Duke Mine generates large quantities of acid mine
drainage. The main flows of water into the underground workings include cooling water used in the
drilling and blasting operations and groundwater percolating through fissures. These streams are
mixed, gravitate to the lowest level and pumped to the surface where they combine with spillage from
the jig processing plant and collected drainage from the plant tailings. The combined acid stream of
about 180 m3/day and an annual average acidity level of 17563.84 mg/l CaCO3 is neutralised with
lime before discharge into evaporation ponds.
Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 440

18 - 19 September 2006 Perth, WA
However, several problems were observed with the operation of the neutralisation plant and sludge
disposal in the evaporation ponds during the study at Iron Duke.
a) The pH of the neutralised sludge does not remain stable in the 6.5 to 7 range after discharge to
the ponds. It was observed that within 2 to 3 days after disposal, the clean overflow drops in
pH from 6.5 to <3.5.
b) Groundwater on the down gradient of the evaporation ponds is very acidic with pH values in
the range 0.9 to 3.5, and contains extremely high sulfate, metal ions and total dissolved solids.
c) The Yellow Jacket River is polluted by acidic seepage from the contaminated ground water
and a stream from the old waste rock dump.
d) The precipitate formed during neutralisation is very voluminous, contains unstable chemical
constituents and leads to difficulties in pumping.
The chemical instability of the neutralised sludge is a cause for concern as the resulting effluent does
not meet regulated levels for permissible discharge. The objectives of this study were to evaluate the
processes controlling the behaviour of major and trace components in acidic drainage, and to use
chemical equilibrium analysis methods to interpret the chemical evolution of AMD and the effect of
more than six years of pump, treatment and sludge disposal practices on hydro-geochemical
parameters.
THE CHEMISTRY OF ACID MINE DRAINAGE GENERATION
Acid mine drainage (AMD) is a low pH, sulfate and metal bearing solution usually formed when
rocks containing sulfide minerals (essentially pyrite and pyrrhotite) are exposed to the atmosphere or
an oxidising environment, and subsequently leached by water. It ranges widely in quality from
alkaline (pH ~10) to strongly acidic (pH ~ <3), usually corrosive, and contains dissolved solids
ranging from about 100 to 100 000 mg/l (Rose and Cravotta, 1998; Nordstrom et al, 2000). AMD is a
major issue affecting the metal mining and coal industry throughout the world.
The problem of acid mine drainage begins with sulfide oxidation (mainly pyrite oxidation) and is
accompanied by the dissolution and precipitation processes of metals and minerals. It has been a
major focus of investigation over the last 60 years (Sato, 1960; Bryner et al, 1967; Nordstrom, 1977
and 1982; Nordstrom et al, 1979; Ritcey, 1989; Jambor and Blowes, 1994 and 1998; Alpers and
Blowes, 1994; Evangelou, 1995; Morin and Hutt, 1997; Nordstrom and Alpers, 1999). The problem is
multifaceted as AMD forms within a complex environmental system where several factors need to be
considered. Furthermore, factors such as microbial catalysis, neutralization reactions, sorption
reactions, and climatic effects have an important influence on pyrite weathering.
The acid production processes can be split into three categories. The first is the oxidation of sulfide
minerals by oxygen and ferric iron. The second involves bacteria assisted oxidation and the third is
hydrolysis of mainly Fe(III) and subsequent precipitation of ferric oxyhydroxides or oxyhydroxide
sulfates. The most common sulfide mineral is pyrite (FeS2). A complex interaction of physical,
chemical and microbiological processes can influence the generation of acidic drainage and typically
these factors vary on a site specific basis. The initiator reaction in AMD generation from mine sites is
the oxidation of pyrite, (Singer and Stumm, 1970; Forstner and Wittmann, 1979).

1) Pyrite oxidation by oxygen
When pyrite is exposed to water and oxygen, it decomposes into water soluble components, acid is
produced. Oxidation of the disulfide according to Equation 1 releases Fe2+ and two protons.
7
FeS 2 + O 2 + H 2O Fe 2+ +2 SO 4 2− +2 H + (1)
2
The relatively reduced water soluble components (ferrous iron) are further oxidised at low and
circum-neutral pH, and acid is consumed, (Equation 2), (Moses and Herman, 1991).
1 1
Fe 2+ + O 2 + H + Fe 3+ + H 2O (2)
4 2
The conversion of ferrous (Fe2+) by oxidation to ferric (Fe3+), reaction 2, is followed by hydrolysis of
Fe3+ to form ferric hydroxide (insoluble at pH>3.5) releasing 3 more protons in the process, (Equation
3).
Fe 3+ +3H 2O Fe(OH ) 3 +3H + (3)
Thus, in sufficiently aerated/oxidised environments, the formation of ferric iron in water resulting in
the precipitation of ferric hydroxide is a key acid producing stage. Once sulfides have been oxidised,
it is extremely difficult to avoid ferric hydroxide precipitation.
2) Pyrite oxidation by ferric iron (Fe3+)
With the formation of soluble ferric iron (Fe3+) in the presence of fresh iron sulfide, further sulfide
oxidation is accelerated, as represented in Equation 4, (Nordstrom and Alpers, 1999).
FeS 2 +14 Fe 3+ +8H 2O 15Fe 2+ +2 SO 4 2− +16 H + (4)
Equation (1) describes the initial step of pyrite oxidation in the presence of atmospheric oxygen. Once
ferric iron is produced by oxidation of ferrous iron, Fe3+ will be the primary oxidant of pyrite
(Equation 4) (Nordstrom, 1979; Moses et al, 1987; Ehrlich, 1996). Under abiotic conditions the rate
of oxidation of pyrite by ferric iron is controlled by the rate of oxidation of ferrous iron (Reaction 2),
which decreases rapidly with decreasing pH. Therefore pyrite continues to oxidize as long as ferric
iron is regenerated. The ferric iron is reduced by the pyrite itself (Singer and Stumm, 1970) according
to reaction (4).
MATERIALS AND METHODS
Processing of pyrite at Iron Duke
The geology of the area around Iron Duke has been described by Ferguson and Wilson (1964), and
Bleknisop, et al, (2000). The Iron Mask mountain range occurs in the Zimbabwe Archean craton and
is characterised by stratiform Fe mineralization, metasediments and metavolcanics. These rocks are
locally succeeded by Archean greenstones of the Acturas Formation and metagreywackes of the
Shamvain Group to the north east.
The main orebody at Iron Duke comprises massive pyrite, with minor pyrrhotite, chalcopyrite, and
scheelite. Traces of gold, arsenopyrite and other sulfides also occur. The gangue minerals comprise
mainly chert (silica and manganese). The mine started in 1914 as an open cast gold operation,
(Smythe, 1995). Pyrites mining of the sulfide facies started in 1940. The mine targets to extract

12 500 t per month of Run of Mine (ROM) ore with a minimum grade of 30%S using an underground
stoping method. The ore is mined with little waste that is stock piled on the surface. Wet drilling is
used, and approximately 180m3/day of water is drawn from nearby Yellow Jacket River for this
purpose and to suppress dust and fumes during the blasting operations. The treatment flowsheet is
shown in Figure 1. The ROM ore is stage crushed from <300 mm to +/-12 mm. A small gravity
separation plant utilising two Velco jigs in parallel circuit (5 t/hr each), is used to upgrade the minus
12 mm fraction of the pyrites from 30 to 36%S for the export market in Zambia, South Africa and
Europe for applications in metallurgical fluid bed roasters and Foundry Industries. The bulk of the
product output around 30%S is used in the production of phosphate based local fertilisers.
Grizzly
Primary jaw
crusher
Sorted Waste
<4% Sulphur
Single Deck
Vibrating Screen
Circulating
Secondary jaw crusher
Gyrosphere
Crusher
70mm
Zimphos Product Double Deck +12mm

Jig Vibrating Screen
(@30.0% S)
Concentrate
12mm
Jigs (3) Silo (Ore
14%SJig
Fig. 1: Iron Duke Plant flow sheet.
The mine water handling process
Introduction of water to the drilling faces with sulfidic rock chips in the drill holes forms sludge in a
perfect mixture for acid formation. Another source of acid drainage underground is associated with
seepage of ground waters into old workings. The oozing takes place in stoped out areas. The
reticulation is such that the acid water (pH ~2) gravitates to the bottom of the mine, 9th level (270 m
below ground level), is collected in sumps and pumped in two stages to the surface as shown in Figure
2. Special precautions are needed for equipment which includes stainless steel pumps, acid resistant
paints, extra greasing and PVC piping. Due to intermittent drilling and blasting, the acidic mine water
is pumped to a holding tank in the neutralisation plant in a batch wise operation lasting 8 - 10 hours.

Pump
Pollution Ponds/
Jig Process Water
Evaporation
Neutralization Ponds
Plant Discharge pump
Underground water
Level 3
Sump pump
Level 6
Sump pump
Level 9
Sump pump
Fig. 2: Water Reticulation Process at Iron Duke.
Figure 3 shows the process flow diagram for neutralisation treatment of the acidic mine effluent at
Iron Duke. The neutralisation plant consists of six tanks fitted with agitators, each with a volume of
32.5 m3. The first two are used as lime mixing and storage units. The lime is simply added into water
with continuous stirring to produce a paste. The lime paste from the dosing tank gravitates at a
controlled rate to the buffer tank or first neutralising stage where it’s continuously mixed with raw
mine effluent, pH ~2.5, at predetermined rates. It is then allowed to overflow into the second stage
tanks which provide for residence time. When pH reaches the desired range (6.5 to 7) the second stage
tanks are drained through a pump and the neutralised discharge is stored in holding ponds. The ponds
are designed so as to allow for solids settlement in the receiving compartment with the clean overflow
collecting and draining into a separate pond where it is expected to evaporate.
From underground
and pollution ponds
Water receiving tank

Lime hopper
Main tank 1
1 5
4
To evaporation ponds
pump
Fig.3: Neutralisation plant flow sheet.

Monitoring and sampling locations
Treatment of water samples
A field reconnaissance and drainage monitoring at Iron Duke was carried out once every month, from
May 2004 to April 2005. Surface water samples were collected from the nearby Mazowe and Yellow
Jacket rivers, evaporation ponds, seeps, mine water and drainage locations in proximity to tailings and
waste rock piles. Ground water samples were collected from monitoring boreholes shown in Figure 4
and domestic boreholes. The monitoring boreholes were flushed out by removing 20 litres of water
and given 30 minutes to recharge before collecting the water sample. Where possible, photographs of
contaminated locations within sites were taken for recording observations of colour, texture and
physical appearance.
Fig.4: Sampling and monitoring points

(adapted from Ravengai et al, 2005).
Temperature, pH, ORP, EC, and TDS were measured in the field using the Myron ultra meter Model
6P. The quick analysis is done by just filling the cell cup, rinse three times with sample, pushing a
parameter key and taking the reading. Dissolved oxygen was determined by means of a membrane
based oxygen meter, after calibration and allowing for temperature correction. The performance of the
probes in the reading cell was confirmed using the pH 4 and pH 7 buffer standards, and also calibrated
using the standards for TDS and EC. All water samples collected for trace metal analysis were filtered
at the site using a 0.45 μm membrane filter. The filtered solution (1litre) was stored in polyethylene
containers which were used exclusively throughout the sampling campaign. The sample containers
were first rinsed with an aliquote of the filtered sample, then filled to the brim to exclude oxygen and

acidified with 65% nitric acid solution to pH<2. Samples for major anion analysis were also filtered
through 0.45 μm membrane filter but were not acidified. Samples were kept in a cool place until
returned to the Department of Metallurgical Engineering, University of Zimbabwe, where they were
analysed within seven days.
Concentrations of Ca, Cu, Fe, K, Mg, Mn, Na, Ni and Zn were determined by atomic absorption
spectrometry. Cl-, CN-, NO3- and SO42- were determined by titration and gravimetric analysis.
Treatment of solid samples
i) Stream sediments
Stream sediment samples were composited by collecting sediment increments from several places at
the sample site, generally along a 100 m stretch of the stream. Attempts were made to collect 20 to 30
increments in order to improve sample representativeness. About 1 kg of sediment sample sieved on
site through a 2mm stainless steel screen was collected in plastic sample bags and then air dried.
Sediment samples were predominantly stream alluvium but a few were collected from ponds and
others from runoff gullies below the jig tailings dump and old waste rock pile.
In the laboratory, samples were oven dried at 50oC to prevent baking and sieved through 180 μm. The
coarse fraction was discarded and the finer fraction was random sampled in 30 small increments and
pulverised in preparation for chemical analysis.
ii) Precipitate samples
Secondary minerals and precipitate samples of flocs precipitating out of acidic metaliferous waters
were also collected by the grab sampling technique. The samples were composited from wet areas
such as ponds and tailings drainage gullies and placed in sealed plastic bags. The samples were
bagged in plastic sample carriers and retained to the laboratory where they were oven dried at 40oC,
riffle split, pulverised without sieving and then prepared for chemical analysis by XRF.
RESULTS AND DISCUSSION
Variations in mine drainage composition
Variation of heavy metals with pH
Figure 5 shows the sum of base metal cations concentration in mili-equivalents per litre as a function
of pH for mine waters that drain from the different mineral deposit types in the Mazowe area in NE
Zimbabwe, compared to the massive pyrite ore body at Iron Duke. In general, metal content increases
with decreasing pH reflecting greater amounts of pyrite and other sulfide minerals coupled with lesser
amounts of acid buffering minerals. The massive pyrite ore body at Iron Duke has extreme acid
generating capacity, (up to 463.75 kgH2SO4/t). As a result, the acidic drainage mine water contains
extreme concentrations of Fe, Mn, and Al and moderate amounts of Cu, Ni and Zn. In contrast,
carbonate containing tailings deposits and olivine rich waste rock piles most commonly generate
drainage with near neutral pH values and associated moderate quantities of Ni (as high as 7.52 mg/l)
and lesser copper (maximum 1.06 mg/l).

1000
Fe+Mn+Al+Ni+Cu+Zn (meq/l)
100
Massive pyrite
10 Pyrite poor gold telluride veins with chalcopyrite & low carbonate contents
Low iron copper-nickel sulf ides rich in magnesium basalts
Pyrite poor pentlandite high in pyroxene and olivine
0.1
0.01
0 2 4 6 8 10 12
pH
Fig.5: Variations in base metal concentrations draining
from different rock types.
The importance of geologic controls can be illustrated in other mine waste components of
environmental concern such as tailings drainage, cyanide processing solutions, waste rock, etc. It can
be seen that environmental signature of mine site drainage is a predictable function of mineral deposit
geology, deposit size, host rocks, ore and gangue mineralogy. The scale of environmental effects
generally increases with increasing deposit size. Host rock influences factors such as mineral textures,
trace element geochemistry and the rate at which minerals weather and oxidise.
As weathering and erosion of mineral deposits and waste piles take place, secondary minerals in the
form of new, more chemically stable mineral suites develop and exert a predominant controlling
effect on the environmental behaviour of acid mine drainage.
The effects of dry and wet cycles on drainage chemistry
Amounts of rainfall and prevailing temperatures influence the amount of water available as surface
runoff, the level of the water table, rates of reaction, amounts of organic material, and other
parameters that affect the oxidation of sulfide minerals. Water tables are shallow in wet climates and
deep in semi arid climates. However, depth to the water table can be highly variable across short
distances within a mine site as observed at Iron Duke where water tables vary from 2.25 m to 60 m.
Drinking water borehole levels are in excess of 45 m, whereas a shallow aquifer with water levels
from 2.25 m to 6.56 m exists on the down gradient of the evaporation ponds.

SO4 Dry
1000
TDS Dry TDS Wet

SO4 Wet
Mn Dry Mn Wet
100
mg/l
10
Fe Dry Fe Wet
0.1
Acidity, metal concentration and levels of other pollutants generally tend to increase in the dry season
with lower values observed in the wetter months from December to March due to increased
percolation of ground water to the mining level. This implies that pollution by AMD has a significant
impact on water quality during the dry period. The increased volume of mine water has ramifications
in terms of water handling procedures for the mine in that it means more pumping and treatment
costs. The seasonal variations in TDS, sulfate, iron and manganese concentrations are shown in Figure
6.
a) pH and Trace metal fluctuations
The range of pH fluctuation throughout the year for various drainage waters at Iron Duke is best
illustrated by the box plot in Figure 7. As expected, neutral treated sludge and surface water along
Yellow Jacket River, upstream of mine site, have the least range (difference between minimum and
maximum pH) throughout the year. In contrast, the pH in the evaporation ponds after sludge disposal
shows a wide range, with minimum non-outlier pH values around 3.0 observed in the drier months of
the year. The same applies to ground water from the contaminated shallow aquifer, which shows an
outlier minimum pH of 1.1 recorded in October (the hottest and driest month just before the rainy
season starts in November). Drainage from the jig tails dump is most acidic as well during the drier
months, but the flow is only a trickle, greatly reducing pollution impacts. Surface water at the seepage
point is greatly influenced by recharge along the Yellow Jacket River and mixing with fresh streams
due to rainfall in the wet season resulting in a wide range between the 25 and 75% quartiles. Non
outlier maximum pH is also high at 5.5 in the wet season, compared to 2.7 in the drier months.

Seepage
Point H
Upstream of
Jig Tails point H
Ground Water
Evaporation
Ponds
Neutral
Sludge
Mine water
0 2 4 6 8 10
pH
Fig.7: Summary statistics of drainage pH.
b) Contaminant fluxes and flushing model
The release of acidity from the jig plant tailings is controlled to varying degrees by the dry and wet
cycle changes throughout the year. It has been observed that:
i) During the dry spells, base flow conditions develop, a small proportion of the reactive
surfaces are leached which allows oxidation products to build up in some sections of the
dump. During this period, drainage flow decreases markedly, sometimes even drying up
completely. However, intermediate oxidation products (sulfate and hydroxysulfate
minerals) accumulate in the dump as oxidation proceeds.
ii) When rainfall first comes, infiltration into the dump occurs, resulting in leaching and
rinsing of more reactive surface areas. The accumulated secondary products are dissolved
and mobilised, resulting in a highly potent ‘first flush’ of concentrated contaminant load
in the jig tailings drainage water, Equation 5.
KAl 3 ( SO 4 ) 2 (OH ) 6 +3H 2O ⇔ 3 Al (OH ) 3 + K + +2 SO 4 2− +3H + (5)
iii) As wet conditions persist, the high initial acidity and metal concentrations of the first
flush decay, leading to reduced pollutant loadings and diluted drainage flow. The most
abundant dissolved components in mine water and tailings drainage at Iron Duke are
sulfate and iron, followed by manganese and aluminium. Figure 8 shows the decay of iron
concentrations in the jig tailings drainage following a series of rainfall events between
November 2004 and February 2005. The dotted line is the predicted non – linear
exponential decay for the iron concentration (contaminant load, y = [Fe]) given by:
[Fe] = 1906.5e −0 0708t (6)
where t=x is the time taken for exponential decay flushing.

y = 1906.5e -0.0708x
R2 = 0.8558
3500 60
3000 A ctual Fe
50
Rainfall (mm)
2500 Daily Rainfall 40
Fe (mg/l)
2000
30
1500
1000 20
500 10
0 0
04
04
04
04
04
04
05
05
05
04
04
4
5
00
00
00
1/
1/
2/
2/
2/
2/
1/
1/
2/
20
20
/2
/2
/2
/1
/1
/1
/1
/1
/1
/0
/0
/0
2/
2/
12
12
07
22
28
14
22
25
30
16
27
25
/1
/1
6/
8/
1/
10
12
Fig.8: Contaminant loading in jig tails drainage following rainfall events.
iv) When dry conditions are re-established, contaminant loads are generally lower than wet
conditions due to reduced flow. However, Fe concentrations begin to increase again, but
with trickle flow, reach a steady peak around 1100 mg/l during the cool and hot dry
months from May to October. There is a sudden jump in concentration when rainfall
starts again in November, and the rise in Fe concentration reaches a level around 1800
mg/l, somewhat lower than the level witnessed at the onset of rainfall in the previous
season. With increased rainfall, concentration falls exponentially, and the cycle is
repeated.
Mechanisms of metal attenuation
The principal objectives in treating AMD at Iron Duke are to neutralize free acidity and reduce the
concentrations of Fe, Al, Mn and sulfate to very low levels so that the effluent quality is acceptable
for recycle or release. The active chemical treatment of acidic drainage is typically accomplished
using lime to raise the pH of the acidic stream and remove dissolved metals through precipitation and
co-precipitation. These processes result in the formation of a gypsum/metal and/or
hydroxide/carbonate sludge which requires disposal in evaporation ponds.
The overall picture for the mobilisation, dispersion and attenuation of AMD pollutants at Iron Duke
can be summarised in Figure 9. About 3 t/day of acid are neutralised and deposited as sludge in the
evaporation ponds. Acid and metal fluxes have been monitored through the flow pattern and the
attenuation takes place through four major stages as depicted in the diagram. The mechanisms of
metal attenuation are discussed in the sections that follow.

Mine Water Nuetralisation Plant 1
3t/d acid
Collected Sludge
Drainage
Jig Tailings Evaporation Ponds 2

0.6t/d acid
Seepage
Fugitive Seepage Perched Aquifer 3
Seepage
Yellow Jacket River Mazowe River

6t/d acid 4
Fig.9: Schematic overview of drainage flow pattern at Iron Duke.
Stage 1: Equilibrium Controls on Acid Neutralisation with Lime
The most common by – product of lime neutralisation is gypsum, which is precipitated when the acid
mine water comes into contact with the added lime according to Equation 7.
Ca (OH ) 2 + H 2SO 4 CaSO 4 .2 H 2O = Ca 2+ + SO 4 2− +2 H 2O (7)
This reaction is responsible for scale formation and excessively large volumes of sludge observed in
the treatment plant. It is presumed that the large amount of sulfate in the feed water (~9700 mg/l)
most likely pushes the solubility product well above saturation. The approximate concentrations of Ca
and SO42- that may be expected upon gypsum saturation can be estimated from the Law of Mass
action and hydrochemical data tables (Appelo and Postma, 1996), for Equation 8, where Keq is 10-4 58
at 25oC.
(Ca 2+ ).( SO 4 2− ).( H 2O) 2

−4 58
mol 2 (8)
K eq = = (Ca 2+ ).( SO 4 2− ) = 10
(CaSO 4 .2 H 2O) dm 6
Thus, as (Ca2+) = (SO42-), the activity of both Ca and sulfate must be 10-2 29 mol/dm3 for dissolution of
gypsum at 25oC. Assuming that the activities of Ca and sulfate approximate to concentrations, and
that Ca weighs 40 080 mg/mol and sulfate weighs 96 064 mg/mol, gives Ca = 206 mg/l and sulfate =
493 mg/l upon gypsum saturation. Hence a sulfate concentration above 9700 mg/l in the feed water to
the neutralisation plant will precipitate a lot of gypsum. The Saturation Index, (SI) is defined as the
logarithm to base 10 of the ratio of the product of the activities of the relevant ions to the Keq.
§ (Ca 2+ ).( SO 4 2− ) ·
SI = log¨ ¸ (9)
¨ K ¸
© eq ¹

If SI = 0, equilibrium exists, SI < 0, the solution is under saturated with respect to gypsum and
gypsum will dissolve and when SI > 0, the solution is oversaturated and gypsum will precipitate. The
major components in mine water before and after treatment by the normal plant configuration are
presented in Table 1. The SI of gypsum for treated water is 1.66, suggesting that it will tend to
precipitate even after deposition. However, this is not the case as explained below.
Table 1: Chemical composition of mine water before and after treatment.
Parameter Feed ± standard Treated Treatment Target

deviation efficiency (%)
pH 2.84 ± 0.48 6.75 100temporary 6.5 – 7.0
Eh (mV) 67.13 ± 27.53 207 - -
Acidity (mg/l CaCO3) 17 538.42 ± 658.67 370.16 97.89 temporary < 50
Sulfate (mg/l) 9705.77 ± 1476.81 3108.53 67.97 < 500
Iron (mg/l) 5379.18 ± 1143.32 3564.96 33.73 < 10
Aluminium (mg/l 146.03 ± 24.36 33.08 77.35 <5
Manganese (mg/l) 96.52 ± 9.33 74.18 23.15 <5
Calcium (mg/l) 1.30 ± 0.87 1486.75 - -
Although titrated acidity is greatly reduced (97.89%) by neutralisation, it has been observed that the
pH of treated sludge shifts markedly from within the target range of 6.5 - 7.0 to less than 3.0 in the
evaporation ponds just after three days deposition. Except for aluminium, the removal of iron and
manganese is not efficient and way off target. One possible explanation for this is that while the
influent to the evaporation ponds is a weak acid semi neutral sludge, what evaporates is pure water.
This means that acid and metal concentrations rise as a function of time. The ponds have reached
supersaturation with respect to jarosite and other iron hydroxides which form a thick yellowish to
brown red sludge at the bottom of the pond. The pH of the ponds has dropped to around 3.0 and acid
is seeping to the ground water down gradient of the ponds area.
Stage 2: pH - Eh Controls on Metal Precipitation and Hydrolysis in the Evaporation Ponds
a) Precipitation and co-precipitation of hydroxides and oxyhdroxides of Iron, Al and Mn
When the mine water is pumped to the neutralisation plant, increased access to oxygen and increase in
pH on mixing with lime causes oxidation, hydrolysis and precipitation of metal oxyhydroxides.
Ferrous iron oxidises and precipitates as an orange ferric oxyhydroxide (ochre) as depicted in
Equations 10 and 11.
4 Fe 2+ +O 2 +4 H + ⇔ 4 Fe 3+ +2 H 2O (10)
Fe 3+ +3H 2O ⇔ 3H ++ Fe(OH ) 3 (11)
The overall ochre precipitation reaction is proton generating and this may further explain why the
evaporation ponds have become ‘acid generators’, and hence pH of neutral sludge does not remain
stable after deposition. However, the reaction may become potentially self limiting in that generation
of protons may lower the pH to a point where ferric ions no longer precipitate as hydroxide unless
adequate neutralization capacity (alkalinity) is present in the pond. However, with deposition of
freshly neutralised sludge, the alkalinity is made available and ferric hydroxide precipitates generating
more acid.
The bottom of the ponds is characterised by reddish brown and brownish yellow (orange) sediments
(precipitates) suggesting domination of the composition by ferrihydrite and schwertmannite,

respectively. More than a dozen compounds fit within the designations of iron oxides, hydroxides, and
oxyhydroxides (Schwertmann and Cornell, 1991). Some of the most common are ferrihydrite
[Fe5O7(OH)nH2O], goethite (α-FeOOH), lepidocrocite (γ-FeOOH), hematite (α-Fe2O3), and
amorphous ferric hydroxide Fe(OH)3. This is supported by the evidence of other researchers (Webster
et al, 1998; Carlosn et al, 2002; Kim and Chon, 2002) who performed XRD analysis on precipitates
from similar samples and found out that the two minerals commonly precipitate from acid sulfate rich
waters. Table 2 shows that the reddish brown precipitate consists approximately of 68 - 77% Fe, and
7.8 - 9.5% CaO, with small amounts of 0.8 - 1.7% MnO2, 2.5 - 3.6% Al2O3, and 0.9 - 1.4% SiO2. The
balance (LOI – Loss on Ignition) is probably mostly chemisorbed water. The brownish yellow
precipitate exhibited Fe:S ratios from 4.72 - 6.31 (deduced from sulfate content by the barium
chloride precipitation and gravimetric method). Although Fe:S ratio for the ideal schertmannite
[Fe8O8(OH)6SO4] is 8, the lower values are usually found and are attributed to adsorption of excess
sulfate on the scheertmannite surface, (Ghiorse and Ehrlich, 1992; Schwertmann et al, 1995).
Additional minerals form when the sludge comes into contact with acidic/sulfate water in the
evaporation ponds. Bigham et al, (1994), reported that such minerals may include jarosites
[(H,K,Na)Fe3(OH)6(SO4)2], melanterite [Fe(SO4)7H2O] and ferric sulfate [Fe2(SO4)3]. These minerals
differ in important characteristics, including degree of crystallinity, solubility, and thermodynamic
stability. Amorphous ferric hydroxide exhibits no long range order (and thus no X-ray diffraction
pattern with conventional methods), and appears to be the most common initial solid formed when
iron(III) precipitates. With time and loss of water, amorphous solids tend to convert to the crystalline
forms of goethite and/or to hematite (Schwertmann and Cornell, 1991). These minerals are important
because of their common occurrence in mine waste environments and demonstrated ability to adsorb
or form co-precipitates with organic compounds and ions of a wide variety of elements (Jambor,
1994; Jambor and Dutrizac, 1998).
Since the efficiency of technology would be limited by the effective solubility of iron as solution
moves through the treatment system, an understanding of iron solubility fundamentals is important.
Ferric iron is quite insoluble compared to ferrous iron.
Table 2: Chemical composition of ponds precipitates by XRF.
Sample Fe2O3 CaO Al2O3 MnO2 SiO2 P2O5 LOI

ID
RB - 1 67.84 7.81 3.58 0.76 0.88 0.06 17.14
RB - 2 75.53 9.47 2.50 1.24 1.37 0.08 15.82
RB - 3 76.92 8.05 3.27 1.65 1.04 0.11 16.07
BY - 1 48.26 9.37 4.06 1.14 0.82 0.31 28.11
BY - 2 58.19 12.18 3.25 0.96 1.26 0.18 33.29
BY - 3 52.06 10.94 2.93 1.42 1.19 0.07 27.95
RB – reddish brown (ferrihydrite); BY – brown yellowish (schwertmannite)
The Ksp for ferrous hydroxide is 4.87 x 10-17, while the Ksp for ferrihydrite is 2.64 x 10-39, and the Ksp
for amorphous ferric hydroxide is often taken to be 10-38. While there is some dispute among
researchers as to the precise Ksp for each of the oxyhydroxide forms of Fe(III), there is no
disagreement that ferric oxyhydroxides are extremely insoluble, with minerals of increasing
crystallinity being more insoluble. The hydrolysis of ferric iron in the presence of sulfate is strongly
influenced by pH. Figure 10 shows how pH controls the solubility of ferric aluminium hydroxide
species in water. Appelo and Postma (1996) suggest a value for Keq = 104 = (Fe3+)/(H+). Thus for
systems in equilibrium with Fe(OH)3:
log(Fe3+) = 4 - 3pH (12)

Equation 12 approximates to the line marked Fe3+ in Figure 10. At pH = 6, (Fe3+) = 10-14 mol/dm3,
when pH = 2, (Fe3+) = 10-2 mol/dm3, which explains the very low solubility of ferric iron at high pH
and appreciable solubility at very low pH. At pH values more than eight, the solubility increases as a
result of complexation with the OH ligand. In a similar way dissolved aluminium precipitates as a
white to yellowish hydroxide commonly observed at the seepage point D along Yellow Jacket River.
Fig. 10: pH dependence of ferric and aluminium hydroxide

species in water. Dashed lines represent total element solubility
(after Stumm and Morgan, 1996, and Younger et al, 2002).
b) Redox and stability of iron phases
The redox potential describes the reduction – oxidation state of a solution and controls the oxidation
state of elements that can exist in several oxidation states, hence element mobility. Some metals are
more mobile in one oxidation state than the other. The lower the value, the more reducing the
solution, and the higher the value, the more oxidizing the solution. The Eh for the mine water before
treatment is very low averaging 67.13 mV but rising to 207 mV after agitated lime treatment. The
stable phases for a combination of Fe – S – C – H2O species in solution under a variety of Eh – pH
values at temperature 25oC, is shown in Figure 11, together with that of manganese superimposed
(modified after Appelo and Postma, 1996).
The figure shows the following:

• That pyrite is stable in circum – neutral, reducing conditions, ie the subsurface of the
geosphere.
• By increasing Eh and thus moving vertically from the pyrite field, it is seen that pyrite
becomes unstable and gives way to siderite or ferric oxyhdroxide domains. As a result, in
acidic mildly reducing conditions in the untreated mine water ferrous ions are the most stable,
while ferric ions become more stable in strongly acidic and oxidizing conditions in the
evaporation ponds.
• Mn(OH)2 is precipitated at very high pH >9.5 under reducing conditions and this explains the
very low removal efficiency (23.15%) observed for manganese during treatment.

Fig 11 Eh – pH diagram for iron and manganese systems
The implications of these deductions on the technology of AMD control at Iron Duke are as follows:
i) Aeration – Oxidation
Ferrous hydroxides (Fe2+) precipitate at a much higher pH, and are not as stable as ferric hydroxides
when the sludge is exposed to acidic waters. This implies more lime consumption and higher
operating costs. As a result, aeration should be applied to oxidize the iron to the stable form according
to Equation 13.
Fe(OH ) 2 + 1 H 2O + 1 O 2 Fe(OH ) 3 (13)

2 4
Other observed problems with ferrous hydroxides are that they do not settle as well as ferric
hydroxides and they can create a highly viscous sludge. There are three major advantages for proper
oxidation of iron, namely sludge stability, better treatment efficiency and sludge viscosity. Therefore,
it appears that the inclusion of an aeration step in the lime neutralization circuit will be advantageous.
ii) Formation of calcite (Calcium Carbonate)
Calcium carbonate is also a common by-product of lime neutralization, Equation 14.
Ca 2+ +CO 3 2− CaCO 3 (14)
The inorganic carbon for this reaction comes from the AMD itself or the carbon dioxide in air, which
is dissolved during agitation. This carbon dioxide converts to bicarbonate and then partially to
carbonate at high pH (Figure 12), and the carbonate fraction will precipitate with the high calcium
content of the slurry to form calcite. This calcite can play an important role in the stability of the final
sludge product as it provides neutralising potential to the sludge when deposited in the evaporation
ponds.

Fig.12: Inorganic
carbon equilibrium
(after Langmuir,
1997).
As shown in Figure 12, the carbonate species (CO32-) is only prevalent at pH values more than
approximately 10.3 and is practically non existent at pH values less than 8.5. Therefore, calcite
precipitation at pH values below 8.5 is not expected unless the total inorganic carbon concentration is
very high. The implication is that the treatment pH target range of 6.5 to 7.0 is too low for the low
density sludge process and at least 8.5 should be the target if manganese and aluminium are also to be
precipitated.
Stage 3: Adsorption and ion exchange in the perched aquifer
The water chemistry of the shallow aquifer located on the down gradient of the evaporation ponds
was monitored on a monthly basis from May 2004 to April 2005, and the summary statistics are
presented in Table 3. The ground water is heavily impacted by acidic seepage from the ponds and
show low pH (1.21 – 3.70), high levels of iron (265.11 - 406.71) mg/l, Mn (149.23 – 245.03) mg/l, Al
(0.46 – 3.56) mg/l and considerable amounts Ni, Zn and Cu. Iron values are negatively skewed, while
Ni values are positively skewed, suggesting poor distribution around the mean for these elements.
The migration of contaminants through the aquifer (vadose zone) takes place through advection,
mechanical dispersion and molecular diffusion. These mechanisms take place on the dissolved and
suspended contaminants in response to gravity flow, pressure and density difference, and chemical
kinetic activity over a concentration gradient as the water containing contaminants flows around the
solids in the aquifer media. The fate of contaminants in the vadose zone is such that their transport or
mobility may be retarded and attenuation may take place yielding lower concentrations in the
liquid/mobile phase. This is due to processes such as adsorption and ion exchange, where ions are
transferred from water to the soil. Heavy metals are adsorbed to the soil particles by either cation
exchange or chemisorption. Patrick and Verloo (1998) reported that heavy metals are adsorbed to the
clay and organic matter by electrostatic attraction and will remain attached as metal atoms. Their
speciation may change as aquifer geochemical conditions such as pH and Eh change, (Groudev et al,
1999; Batty and Younger, 2002; Weibner et al, 2005).

Table 3: Site 3 Statistics of Ground Water Chemistry
X = 294798 Y = 8071046 Altitude = 1177m
Count = 12 Max Min Mean Median Geomean Sta.Dev Skewness Kurtosis
Well Depth m 6.56 2.25 4.47 4.47 4.23 1.46 -0.05 -0.83
o
Temperature C 24.10 22.30 22.95 22.65 22.94 0.65 0.62 -1.28
pH 3.70 1.21 2.86 2.95 2.74 0.77 -0.77 0.11
Eh mV 243.00 221.00 235.48 236.90 235.41 6.14 -1.16 1.58
Acidity mg/lCaCO3 4324.06 603.45 1179.43 820.47 983.62 1026.32 3.07 9.90
Alkalinity mg/lHCO3 0.00 0.00 0.00 0.00 #NUM! 0.00 #DIV/0! #DIV/0!
DO mg/l 5.74 3.82 4.67 4.68 4.64 0.56 0.35 -0.20
TDS mg/l 8338.00 2765.20 5076.62 4761.15 4765.55 1878.39 0.49 -1.03
EC uS/cm 10043.20 4128.60 6764.56 6862.10 6471.99 2079.78 0.33 -1.10
2-
SO4 mg/l 2007.74 1301.30 1702.89 1689.69 1693.56 182.50 -0.39 1.56
NO3- mg/l 98.33 42.04 56.28 48.87 54.43 16.80 1.80 2.85
-
Cl mg/l 308.42 122.89 218.50 216.81 205.69 75.78 -0.06 -1.95
Fe mg/l 406.71 265.11 368.67 381.84 366.08 42.66 -1.80 2.62
Mn mg/l 245.03 149.23 193.44 193.37 191.18 31.03 0.24 -1.19
Al mg/l 3.56 0.46 1.57 1.19 1.35 0.90 0.96 0.58
Ni mg/l 3.41 0.87 1.58 1.16 1.41 0.87 1.53 1.19
Cu mg/l 0.53 0.31 0.41 0.41 0.41 0.06 0.38 -0.15
Zn mg/l 2.04 1.06 1.40 1.23 1.36 0.35 0.74 -1.01
Ca mg/l 268.44 178.93 236.97 248.63 235.00 30.51 -1.12 0.07
Mg mg/l 42.65 14.04 27.30 26.27 26.22 8.02 0.56 0.30
K mg/l 8.76 3.05 6.33 6.40 6.13 1.51 -0.61 0.93
Na mg/l 35.21 16.18 26.22 27.94 25.64 5.42 -0.58 0.08
Clay mineralogy influences attenuation capacity as the mass of metal which can be electrostatically
bound to a unit mass of clay increases with specific surface area, (Young et al, 1996). Additional
compositional control on attenuation of metals in the aquifer is exerted by secondary minerals such as
metal oxides and hydroxides. Both Mn and Fe oxides, hydroxides and oxyhydroxides are well known
to precipitate from mine drainage waters and to have high levels of substitution of heavy metal cations
for other metal ions within their respective crystal lattices, (Randall et al, 1974; Gong and Donahoe,
1997; Lin and Herbert, 1997;). The ‘contaminant front’ has reached breakthrough point, and leakage
of metal laden seepage into the Yellow Jacket River is testimony that the sorption capacity of the
perched aquifer is now exceeded.
Stage 4: Dilution and immobilisation in the Yellow Jacket River
The Yellow Jacket River is the principal pathway of AMD related pollutants at Iron Duke. The river
receives direct seepage from the contaminated perched aquifer and also fugitive seepage from the jig
tailings dump. River waters in the seepage zone exhibit elevated levels of several contaminants and
trace metals (Al, Cu, Ni, Mn, and Zn) in comparison with background values, Table 4.

Table 4: Surface water concentrations (mg/l) of dissolved and particulate trace metals for the
dry (August 2004) and wet season (February 2005) along Yellow Jacket River.
*Background Dry Season Wet Season

Total Dissolved Total Dissolved Total Dissolved
Al 0.12 0.07 12.85 12.34 10.82 9.30
Mn 4.36 2.89 127.06 115.67 164.17 124.08
Fe 1.05 0.67 5.16 2.66 7.42 3.97
Cu 0.02 0.01 0.22 0.13 0.35 0.23
Ni 0.18 0.06 1.58 1.44 0.95 0.43
Zn 0.03 0.02 3.17 1.03 1.83 0.34
* Upstream of mining influence on the Mazowe River
In the dry season, Al, Mn, Cu and Ni were primarily present as dissolved species (<45 μm), while Fe
and Zn were largely hosted in particulate phases. Increased river flow caused by rainfall in the wet
season results in erosion of weathered minerals around the mine site leading to increased
concentrations of both particulate and dissolved metals in the river. This explains the high levels of
Mn (164.17 mg/l) and Fe (7.42 mg/l) observed during the wet season, and it results in greater loadings
of metals to the Mazowe River sub-catchment. The high particulate Fe concentrations during the wet
season suggest that trace metal associations with Fe oxyhydroxides may be significant.
Rapid attenuation of acidic drainage input is demonstrated by the changing aqueous concentrations of
metals and sulfate along a 1.5 km stretch downstream from the seepage point as shown in Figure 13.
In general all metal concentrations rise above background values as the river passes through the mine
site, reaching a peak and then decrease rapidly more or less to background levels just after the
confluence with Mazowe River mostly due to dilution. The pattern of changing concentrations of
dissolved sulfate parallels that of the metals and can be explained by the formation of Al, Fe, or Mn –
oxyhydroxides and scavenging of sulfate and trace metals from water to these minerals. Alternatively,
sulfate can also be removed through the formation of Fe and Al – hydroxysulfate precipitates, (Ali
and Dzombak, 1996; Munk et al, 2002).
Sorption is possibly the most important mechanism that can explain the observed changes in metal
concentration along Yellow Jacket River. Sorption is a chemical reaction in which metals are
transferred between stream water and the surfaces of streambed materials, such as sediments, rocks
and aquatic plants. During desorption, metals are detached from streambed materials and added to
stream water, thereby increasing dissolved metal concentrations in stream water. During adsorption,
metals are transferred from stream water to streambed material, thereby decreasing metal
concentrations.
Fig.13: Pollutant Profile along Yellow Jacket River
1000.00
Concentration mg/l
pH
100.00
DO
10.00
SO4
1.00
Mn
0.10
Al
0.01
0
5
0
60
25
45
52
60
65
75
83
95
15
Distance (m )

Sorption of a specific metal is affected by its charge, temperature and pH of stream water. Water
temperature and pH commonly increase in streams during the day and decrease during the night in
response to the daily cycles of daylight and darkness. Negatively charged ions may desorb while
positively charged ions are adsorbed under the same conditions of temperature and pH, (Nimick et al,
2003). These changes are key factors in determining the amount of each metal that is adsorbed or
desorbed.
Adsorption mechanisms are generally categorized as either physical adsorption, chemisorption, or

electrostatic adsorption. Weak molecular forces, such as Van der Waals forces, provide the driving
force for physical adsorption, while a chemical reaction forms a chemical bond between the
compound and the surface of the solid in chemisorption. Electrostatic adsorption involves the
adsorption of ions through coulombic forces, and is normally referred to as ion exchange.
Stage 5: Dispersion of heavy metals in stream sediments
The concentrations of Fe, Mn, Al, Cu, Ni and Zn in the stream sediments of Yellow Jacket River
exhibit high enrichment along 960 m downstream of the seepage point in comparison to background
values. This means that the sediments are serving as a diffusive sink for the dissolved metals. These
metals are most likely fixed as authigenic sulfide minerals as evidenced by the smell of rotten eggs
characteristic of hydrogen sulfide. Figure 14 shows that Fe, Mn and Zn exhibit maximum
concentrations further away from the seepage zone. Conversely, Cu, Ni and Al show higher levels
near the seepage zone. The contrasting behaviour of these elements suggests that their distributions
may be governed by different controls.
A number of physical, chemical and biological controls may explain the patterns observed in Figure
14. These mechanisms include:
• sorption of metals to the surfaces of streambed material,
• formation and dissolution of minerals containing metals,
• preferential uptake of metals by growing aquatic plants,
• changes in the input of metals from perched aquifer seepage/recharge or upstream source, and
• changes of geochemical conditions within the streambed and variation of streamflow.
Fig.14: Sediment Composition along Yellow Jacket River
30
25 Fe
20 Mn
Al
wt%
15
Cu
10
Ni
5 Zn
0
E
G
d
F
C
D
B
n
ou
gr
ck
Ba
Sam ple points dow stream

CONCLUSION
This study investigates the geochemical controls on the evolution of acid mine drainage in a semi arid
climate. The paper shows that pH and redox potential are the master variables because they control
the solubility and oxidation state of metal ions, the mobile phases of these elements, and also control
much of the hydrochemistry of water samples, the mineralogical reactions which the water can
undergo and therefore the character and signature of acid mine drainage. Pollution loads depend
critically on heavy metal mobility, which is to a large extent determined by pH and redox conditions.
Precipitation and adsorption of metals onto sediments in the evaporation ponds and stream bed of the
Yellow Jacket River are identified as the primary processes leading to metal accumulation within the
sediments. Iron precipitates from mine waters forming hydroxides which may help to sorb other
metals such as Mn, Al and Zn through ion exchange in the perched aquifer. These elements are
desorbed upon acidification and seep into the Yellow jacket River contaminating the water source.
Dilution with fresh stream water is a major process influencing toxicity and metal attenuation.
Understanding the processes controlling metal behaviour in the natural environment through time is
critical to any meaningful assessment of the impact and remediation of acid mine drainage.
ACKNOWLEDGEMENTS
The authors wish to thank the management of Iron Duke Pyrites for permission to publish the
technical data from their operation and the technical staff in the Health & Environment section that
helped in the collection of the relevant data.
REFERENCES
Appelo, C A J and Postma, D (1996). Geochemistry, Groundwater and Pollution. A.A.Balkema,

Rotterdam, pp 275.
Batty, L C and Younger, P L (2002). Critical role of macrophytes in achieving low iron
concentrations in mine water treatment wetlands. Environmental Science and Technology, 36, pp
3997-4002.
Ali, M A. and Dzombak, D A (1996). Interactions of Copper, Organic Acids, and Sulfate in Goethite
Suspensions, Geochimica et Cosmochimica Acta, 60(12): pp 5045-5053.
Alpers, C N and Blowes, D W (eds), 1994. Environmental Geochemistry of Sulfide Oxidation, in
American Chemical Society Symposium Series 550 (Washington, D.C.).
Evangelou, V P, 1995. Kinetics of Pyrite Oxidation and its control, pp 34 – 89 (J. Willey & Sons:
New York).
Bigham J M, Carlson I and Murad E (1994). Schwertmannite, a new iron oxyhydroxysulphate from
Pyhasalmi, Finland, and other localities. Mineralogical Magazine 58, pp 641–648.
Bryner, L C, Walker R B and Palmer, R (1967). Some factors influencing the biological oxidation of
sulfide mineral. Transactions of AIME, 238: pp 56-62.
Blekinsop, T G, Oberthu, T and Mapeto, O (2000). Gold mineralization in the Mazowe area, Harare-
Bindura-Shamva greenstone belt, Zimbabwe: I. Tectonic controls on mineralization. Mineralium
Deposita (2000) 35: pp 126 – 137.
Carlson I, Bigham, J M, Schwertmann, U, Kyek, A and Wagner F (2002). Scavenging of As from
acid mine drainage by schwertmannite and ferrihydrite: A comparison with synthetic analogues.
Environmental Science and Technology 36, pp 1712–1719.
Ehrlich, H L (1996). Geomicrobiology (Dekker, New York, ed. 3, 1996), pp 330-332.
Fergusson, J C and Wilson, T H ( 1964). The geology of the country around the Jumbo Mine,
Mazowe District. Bulletin of the Geological Survey of Rhodesia, v33, pp 137.
Ghiorse, W C and Ehrlich, H L (1992). Microbial biomineralization of iron and manganese. In
Skinner, H C W and Fitzpatrick, R W (Eds.), Biomineralization: Processes of iron and
manganese, pp 75–99, Catena Verlag, Cremlingen, Germany.

Jambor, J L (1994). Mineralogy of Sulfide-rich Tailings and Their Oxidation Products. In
Environmental Geochemistry of Sulfide Mine-wastes, Mineralogical Association of Canada Short
Course v22, pp 59-102.
Jambor, J L and Blowes, D W, 1994. The Environmental Geochemistry of Sulphide mine wastes, in
Mineralogical Association of Canada Short Course Handbook (22): pp 438.
Jambor, J L and Blowes, D W (1998). Theory and Application of Mineralogy in Environmental
Studies of Sulfide-Bearing Mine Wastes. In Modern Approaches to Ore and Environmental
Mineralogy (Cabri, L.J. and Vaughan, D.J., Eds.), Mineralogical Association of Canada Short
Course, v27, pp 367-402.
Jambor, J L and Dutrizac, J E (1998). Occurrence and constitution of natural and synthetic
ferrihydrite, a widespread iron oxyhydroxide: Chemical Reviews, v 98, pp 2549-2585.
Förstner U and Wittmann G T W (1979). Metal Pollution in the Aquatic Environment, Springer-
Verlag, Berlin, pp 486.
Gong, C and Donahoe, R J (1997). An experimental study of heavy metal attenuation and mobility in
sandy loam soils. Applied Geochemistry, 12(3), pp 243-254.
Groudev, S N, Bratcova, S G, and Komnitsas, K (1999). Treatment of waters polluted with
radioactive elements and heavy metals by means of a laboratory passive system. Minerals
Engineering 12, pp 261-270.
Kim, J Y and Chon, H T (2002). Pollution of a water course impacted by acid mine drainage in the
Imgok creek of the Gangreung coal field, Korea. Applied Geochemistry, 16, pp 1387-1396.
Lin, Z and Herbert, R B J (1997). Heavy metal retention in secondary precipitates from a mine rock
dump and underlying soil, Dalama, Sweden. Environmental Geology, 33(1), pp 1 -12.
Langmuir, D (1997). Aqueous Environmental Geochemistry. Prentice Hall, Inc., New Jersey.
Moses, C O, and Herman, J S (1991). Pyrite oxidation at circumneutral pH. Geochimica et
Cosmochimica Acta 55, pp 471-482.
Moses, C O, Nordstrom, D K, Herman, J S and Mills, A L (1987). Aqueous pyrite oxidation by
dissolved oxygen and by ferric iron. Geochim. Cosmochim. Acta, 51, pp 1561-71.
Morin, K A and Hutt, N M, 1997. Control of acidic drainage in layered waste rock at the Samatosum
Minesite: Laboratory studies and field monitoring, in Canadian Mine Environment Neutral
Drainage (MEND) Report 2.37.3.
Munk, L, Faure, G, Pride, D E. & Bigham, J M (2002). Sorption of trace metals to an aluminum
precipitate in a stream receiving acid rock-drainage; Snake River, Summit County, Colorado.
Applied. Geochemistry 17, pp 421–30.
Nimick, D A, Gammons, C H, Cleasby, T E, Madison, J P, Skaar, D and Brick, C M (2003). Diel
cycles in dissolved metal concentrations in streams - Occurrence and possible causes, Water
Resources Research, 39: pp 1247-1264.
Nordstrom, D K and Alpers, C N, 1999. Geochemistry of acid mine waters, in The Environmental
Geochemistry of Mineral Deposits (eds: Plumlee, G S and Logsdon, M J), Part A. Processes,
methods, and health issues: Reviews in Economic Geology, (6) pp 27.
Nordstrom, D K, 1977. Thermochemical redox equlibria of ZoBell’s solution, in Geochim.
Cosmochim., Acta (52): pp 1077 – 1085.
Nordstrom, D K, 1982. The effect of sulphate on aluminium concentrations in natural waters: some
stability relations in the system Al2O3-SO3-H2O at 298 K, in Geochim. Cosmochim., Acta (46):
pp 681 - 92.
Nordstrom, D K, 2000. Advances in the hydrochemistry and microbiology of acid mine waters, in
International Geology Review (42): 499–515.
Nordstrom, D K, Jenne, E A and Ball, J W (1979). Redox equilibria of iron in acid mine waters. In
Chemical Modeling in Aqueous Systems. Jenne, E A (ed). American Chemistry Society
Symposium Series 93, pp 51- 80.

Nordstrom, D K, Alpers, C N., Ptacek, C J, and Blowes, D J, 2000. Negative pH and extremely acidic
mine waters from the Iron mountain, California, in Environmental Science Technology (34): pp
254 – 258.
Patrick Jr, W H and Verloo, M, 1998. Distribution of soluble heavy metals between ionic and
complexed forms in saturated sediments as affected by pH and redox conditions, in Water Science
and Technology 37 (6 – 7) pp 165 – 172.
Randall J M, Bermann R L, Garrett V, Waiss A C Jr. (1974). Use of bark to remove heavy metal ions
from waste solutions. Forest Products Journal 24: 74–84.
Ravengai, S, Love, D, Love, I, Gratwicke, B, Mandingaisa, O, and Owen, O R J, 2005. Impact of Iron
Duke Pyrite Mine on water chemistry and aquatic life – Mazowe Valley, Zimbabwe, in Water SA,
pp 219 – 228.
Ritchey, G M, 1989. Tailings Management: Problems and Solutions in the mining industry, pp 69 –
345 (Elsevier, Amsterdam).
Rose, A W and Cravotta, C A, 1998. Geochemistry of Coalmine Drainage, in Coal Mine Drainage
Prediction and Pollution Prevention in Pennsylvania (Eds: Brady, K B C, Smith, M W, and
Schueck, J), pp 1.1–1.22.
Sato, M., 1960, Oxidation of sulfide ore bodies, Ll, Oxidation mechanism of sulfide minerals at 25oC,
Economic Geology, 55:1202-1231.
Schwertmann, U and Cornell R M (1991). Iron oxides in the laboratory: Preparation and
characterization. VCH, Weinheim, Germany.
Schwertmann, U, Murad, E, Bigham, J M, and Carlson, L, 1995. Mineralogical characteristics of
poorly crystallized precipitates formed by oxidation of Fe2+ in acid sulfate waters, in
environmental geochemistry of sulfide oxidation (eds. Alpers, C.N., and Blowes, D B,): American
Chemical Society Symposium Series (550) pp 190-200.
Singer, P C and Stumm, W, 1970. Acidic mine drainage: The rate-determining step in Science, (167)
pp 1121-1123.
Smythe, G R, 1995. Iron Duke Acid mine drainage, in Proceedings of Mining and Environment in
Zimbabwe, 1995, pp 92 – 98 (The Institute of Mining and Metallurgy in Zimbabwe).
Stumm, W S and Morgan, J J, 1996. Aquatic Chemistry, pp 788 – 791 (J. Willey & Sons: New York).
Young, G K, Stein, S, Cole, P, Kammer, T, Graziano, F and Bank F (1996). Evaluation and
Management of Highway Runoff Water Quality. FHWA-PD-96-032. Federal Highway
Administration, Office of Environment and Planning.
Younger, P L, Banwart, S A and Hedin, R S (2002). Mine water hydrology, pollution, remediation.
Kluwer Academic Publ., Boston.
Webster, J G, Swedlund P J and Webster K S (1998). Trace metal adsorption onto an acid mine
drainage iron(III) oxyhydroxy sulphate. Environmental Science and technology, 32, pp 1361 –
1368.
Wiebner, A, Kappelmeyer, U, Kuschk, P, Kastner, M, 2005. Influence of the redox condition
dynamics on the removal efficiency of a laboratory scale constructed wetland, in Water Research
(29), pp 248 – 256.


Geochemical Controls On Acid Mine Drainage Generation, Neutralisation Treatment and Attenuation at Iron Duke Mine

Uploaded by

Copyright:

Available Formats

Geochemical Controls On Acid Mine Drainage Generation, Neutralisation Treatment and Attenuation at Iron Duke Mine

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Geochemical Controls On Acid Mine Drainage Generation, Neutralisation Treatment and Attenuation at Iron Duke Mine

Uploaded by

Copyright:

Available Formats

Geochemical Controls on Acid Mine Drainage Generation,

Neutralisation Treatment and Attenuation at Iron Duke Mine

1. GAusIMM, Norwegian University of Science and Technology – NTNU, Department of

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 440

THE CHEMISTRY OF ACID MINE DRAINAGE GENERATION

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 441

Fe 3+ +3H 2O  Fe(OH ) 3 +3H + (3)

2) Pyrite oxidation by ferric iron (Fe3+)

FeS 2 +14 Fe 3+ +8H 2O  15Fe 2+ +2 SO 4 2− +16 H + (4)

MATERIALS AND METHODS

Processing of pyrite at Iron Duke

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 442

Zimphos Product Double Deck +12mm

Jigs (3) Silo (Ore

The mine water handling process

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 443

Fig. 2: Water Reticulation Process at Iron Duke.

Water receiving tank

Fig.3: Neutralisation plant flow sheet.

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 444

Treatment of water samples

Fig.4: Sampling and monitoring points

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 445

Treatment of solid samples

ii) Precipitate samples

RESULTS AND DISCUSSION

Variations in mine drainage composition

Variation of heavy metals with pH

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 446

The effects of dry and wet cycles on drainage chemistry

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 447

TDS Dry TDS Wet

a) pH and Trace metal fluctuations

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 448

Fig.7: Summary statistics of drainage pH.

b) Contaminant fluxes and flushing model

KAl 3 ( SO 4 ) 2 (OH ) 6 +3H 2O ⇔ 3 Al (OH ) 3 + K + +2 SO 4 2− +3H + (5)

[Fe] = 1906.5e −0 0708t (6)

where t=x is the time taken for exponential decay flushing.

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 449

Fig.8: Contaminant loading in jig tails drainage following rainfall events.

Mechanisms of metal attenuation

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 450

Jig Tailings Evaporation Ponds 2

Fugitive Seepage Perched Aquifer 3

Yellow Jacket River Mazowe River

Fig.9: Schematic overview of drainage flow pattern at Iron Duke.

Stage 1: Equilibrium Controls on Acid Neutralisation with Lime

Ca (OH ) 2 + H 2SO 4  CaSO 4 .2 H 2O = Ca 2+ + SO 4 2− +2 H 2O (7)

(Ca 2+ ).( SO 4 2− ).( H 2O) 2

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 451

Table 1: Chemical composition of mine water before and after treatment.

Parameter Feed ± standard Treated Treatment Target

Stage 2: pH - Eh Controls on Metal Precipitation and Hydrolysis in the Evaporation Ponds

a) Precipitation and co-precipitation of hydroxides and oxyhdroxides of Iron, Al and Mn

Fe 3+ +3H 2O ⇔ 3H ++ Fe(OH ) 3 (11)

Metallurgical Plant Design and Operating Strategies (MetPlant 2006) 452

Table 2: Chemical composition of ponds precipitates by XRF.

Sample Fe2O3 CaO Al2O3 MnO2 SiO2 P2O5 LOI

Fe 3+ +3H 2O Fe(OH ) 3 +3H + (3)

FeS 2 +14 Fe 3+ +8H 2O 15Fe 2+ +2 SO 4 2− +16 H + (4)

Ca (OH ) 2 + H 2SO 4 CaSO 4 .2 H 2O = Ca 2+ + SO 4 2− +2 H 2O (7)

Fe(OH ) 2 + 1 H 2O + 1 O 2 Fe(OH ) 3 (13)

Ca 2+ +CO 3 2− CaCO 3 (14)