Martin Fahey

1
Engineering with Tailings
Some Environmental Aspects of Tailings Management
Keynote Lecture4th International Congress on
Environmental Geotechnics Rio de Janeiro,
Brazil, 13 August 2002

• Martin Fahey

Department of Civil Resource EngineeringThe
University of Western Australia
2
Acknowledgements

• Co-Authors
• Dr Tim Newson Department of Civil Engineering,
  University of Dundee, UK(formerly Research
  Fellow, The University of Western Australia)
• Dr Yoshimasa Fujiyasu Department of Civil and
  Environmental Engineering, University of South
  Carolina (formerly PhD student, The University
  of Western Australia
• Some people who provided slides or other
  information
• Hugh Jones (Golder Associates)
• David Cooling (Alcoa)
• Ian Piggott (Collie Coal)
• John Carras (CSIRO)

3
There are many arts and sciences of which a
miner should not be ignorant De Re
MetallicaGeorgius Agricola (1556) (Translated
by Herbert Hoover, mining engineer, later
President of the United States, 1929-1933)
4
Vale of Avoca, Ireland Idyllic Beauty Spot
Avonmore Avonbeg meet to form the Avoca
River The Meeting of the Waters There is not
in this wide world a valley so sweetAs that vale
in whose bosom the bright waters meet Thomas
Moore, 1779 - 1852
5
Reality Acid Drainage since 18th Century

• Mining for copper at Avoca since 2nd century
• Revived in about 1720
• A thriving salmon fishery downstream of the mine
  was wiped out shortly afterwards by acid drainage
• This stretch of the Avoca River has been dead
  ever since, and is one of the most polluted in
  Ireland

6
Tailings Storage Facilities (TSF)

• The design life of a TSF is, effectively,
  perpetuity. A TSF could be considered to have two
  phases to its life a depositional phase with
  active human involvement, followed by an erosion
  free, environmentally benign, stage with no
  further human intervention, forever
• Guidelines on the Safe Design and Operating
  Standards for Tailings Storages. Department of
  Minerals and Energy, Western Australia

7
Requirements for TSFs

• The stability of the tailings and the retaining
  structure must be guaranteed in the long-term
  forever.
• In the long-term, there must be no escape of
  tailings to the environment through wind erosion
  or water erosion.
• Any seepage or water release that occurs from the
  tailings must not contain any deleterious
  products (acid, cyanide, heavy metals, etc).
• The TSF must be rehabilitated to whatever level
  is required this will vary with location and
  local regulation.

8
Requirements for TSFs

• The stability of the tailings and the retaining
  structure must be guaranteed in the long-term
  forever.
• In the long-term, there must be no escape of
  tailings to the environment through wind erosion
  or water erosion.
• Any seepage or water release that occurs from the
  tailings must not contain any deleterious
  products (acid, cyanide, heavy metals, etc).
• The TSF must be rehabilitated to whatever level
  is required this will vary with location and
  local regulation.

9
Stability is easily guaranteed
Then why so many failures?
10
(No Transcript)
11
(No Transcript)
12
22 February 1994, the Merriespruit (South Africa)
tailings dam failed with catastrophic results 17
people were killed and many houses were
devastated as a result of the extensive mudflow
13
Golden Cross Mine, New Zealand
14
Overtopping - prime cause of failure
50 m high tailings storage filled to the brim
(Hopes Hill, Western Australia, 1994)
freeboard 0 A good example of worst practice
15
Baia Mare, Romania Overtopping
30 January 2000 almost 100,000 m3 tailings water
with high concentration of cyanide was spilled
into the Zazar and Lápos water courses that
belong to the catchment area of river Szames,
which flows into the river Tisza.
16
Baia Mare Cyanide concentrations
Cyanide quickly breaks down on exposure to
atmosphere, and hence does not persist in
tailings
17
Fish kills, Baia Mare disaster, Hungary

• Cyanide quickly breaks down on exposure to
  atmosphere, and hence does not persist in
  tailings (hmmmph!)
• In this case, environmental conditions
  (temperature, ice on water etc) contributed to
  persistence, resulting in fish kills many km
  downstream

18
Los Frailes, Aznacollar, Southern Spain
19
Requirements for TSF

• The stability of the tailings and the retaining
  structure must be guaranteed in the long-term
  forever.
• In the long-term, there must be no escape of
  tailings to the environment through wind erosion
  or water erosion.
• Any seepage or water release that occurs from the
  tailings must not contain any deleterious
  products (acid, cyanide, heavy metals, etc).
• The TSF must be rehabilitated to whatever level
  is required this will vary with location and
  local regulation.

20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
Requirements for TSF

• The stability of the tailings and the retaining
  structure must be guaranteed in the long-term
  forever.
• In the long-term, there must be no escape of
  tailings to the environment through wind erosion
  or water erosion.
• Any seepage or water release that occurs from the
  tailings must not contain any deleterious
  products (acid, cyanide, heavy metals, etc).
• The TSF must be rehabilitated to whatever level
  is required this will vary with location and
  local regulation.

27
(No Transcript)
28
nor for next 400 years!
29
(No Transcript)
30
(No Transcript)
31
Below-ground leakage of pollutants

TSF
Ground surface
Pervious soil
Groundwater
flow
Impervious Aquitard
32
Requirements for TSF

• The stability of the tailings and the retaining
  structure must be guaranteed in the long-term
  forever.
• In the long-term, there must be no escape of
  tailings to the environment through wind erosion
  or water erosion.
• Any seepage or water release that occurs from the
  tailings must not contain any deleterious
  products (acid, cyanide, heavy metals, etc).
• The TSF must be rehabilitated to whatever level
  is required this will vary with location and
  local regulation.

33
(No Transcript)
34
Engineering with Tailings

• TSF engineers are building structures that have
  design life of hundreds (thousands?) of years
• Two approaches
• Engineering the environment to store the tailings
  as received from the mill
• accept what you are given, and make the most of
  it
• Engineering the tailings to produce the most
  favourable outcome
• try to modify/improve what you are given

Hugh Jones (Golder Associates) coined the term
designer tailings to indicate the requirement
to engineer the tailings to suit the long-term
performance of the TSF
35
Some issues

• Design for stability
• the upstream method of construction ???
• Acid drainage prevention and remediation
• Modification of tailings by thickening
• Reduction of tailings hazard by changing
  chemistry
• Dust suppression by tailings modification

36
Upstream Construction Method
37
Upstream Construction
38
Upstream Method of Construction

• According to the WISE (World Information Service
  on Energy) websitethe upstream technique is no
  longer regarded an acceptable option for tailings
  dam construction in Chile, where several
  liquefaction-type failures have occurred with
  tailings dams located in steep mountain valleys
  in areas of high seismic activity (Supreme
  Decree 86 Regulation on the Construction and
  Operation of Tailings Dams, 1970).
• Question Is upstream method worth the cost??
• very low initial cost, but potentially very high
  long-term cost

39
Discussion on Stability Issues

• Stability in seismically-active areas discussed
  by Professor Ramon Verdugo
• Static liquefaction of tailings dealt with by
  Professor Andy Fourie

40
Some issues

• Design for stability
• the upstream method of construction ???
• Acid drainage prevention and remediation
• Modification of tailings by thickening
• Reduction of tailings hazard by changing
  chemistry
• Dust suppression by tailings modification

41
Acid Drainage from Tailings

• Production of Sulphuric Acid when sulphur-bearing
  minerals (e.g. pyrite) are exposed to oxygen and
  water
• Pyrite is often associated with coal and
  metal-ore deposits
• Oxidation of pyrite is complex, involving
  chemical, biological and electrochemical
  reactions, and hence varies with environmental
  conditions
• Acid drainage frequently contains Fe, Al, SO4 and
  heavy metals such as Cu, Pb, Hg, Cd, etc
• Of a total of about 7 billion tonnes of
  metal-mine and industrial minerals tailings in
  Canada, it is estimated that 1.9 billion tonnes
  is acid generating

• Parker, G. Robertson, A. (1999). Acid Drainage.
  Australian Minerals Energy Environment
  Foundation, Occasional Paper No. 11

42
Metal Precipitates, Mt Lyall, Tasmania
43
Engineering to Prevent Acid Generation
(After Marszalek, 1996)
44
Sub-aqueous Disposal to prevent AMD
Extensive work carried out under the MEND program
in Canada on subaqueous tailngs disposal to
prevent AMD
45
Remediation of Acidic Waters from AMD
www.wvu.edu/agexten/landrec/passtrt/passtrt. htm
46
Anoxic Limestone Drain (ALD)
Problems Consumption of the limestone.
Precipitation of metals/sludge on limestone
results in armouring of the limestone,
preventing further reactions (anoxic conditions
meant to prevent this)
47
Wetlands for AMD remediation

• Wetlands (natural or man-made) useful for final
  polishing of effluent from anoxic limestone
  drain (or other) neutralisation system

May have a number of stages
48
Reactive Wall (U. of Waterloo)
Benner et al (1999) For a reactive wall they
installed at the Nickel Rim mine site near
Sudbury, Ontario, the reactive component
consisted of 20 municipal compost, 20 leaf
compost and 10 wood chips, mixed with 50 pea
gravel
49
Reactive Wall (U. of Waterloo)
50
Muja Open Cut Coal Mine, Collie, WA
Perth
Muja Open Cut Coal Mine Part of the Collie
Basin Only coal-mining area in WA. Owned by
Griffin Energy Supplies coal directly to Muja
Power Station
Kwinana
Mandurah
200 km
WA
200 km
Indian Ocean
Waroona
Muja
Bunbury
Collie
51
Muja Coal, WA Chicken Creek wetlands
overflow weir
weir
area 4 lake
passive organic treatment
picnicarea
52
Muja Coal Chicken Creek wetlands
53
Fluidised limestone bed (FLB) system

• pH too low (3.0) for wetlands
• Fluidised Limestone Bed (FLB) reactor used first
• Acidic water (pH 3.0) pumped into base of
  container
• Limestone fluidised by upward flow
• tumblingand abrasion of grains together
  prevents buildup of armouring
• Reasonable quality limestone required
• not so friable as to disintegrate in fluidisation
  process

f 1 m
Out
Fine limestone gravel
3 m
Low-pH water
54
FLB at Muja Part of two-stage process
Settling pond - after 1st stage, heavy metals
precipitated
55
Performance of FLB at Muja

• 45 litre/s (4000 m3/day)
• Stage 1 pH increases from 2.0 to 4.5
• Settling pond - for precipitation of some heavy
  metals
• Stage 2 pH increases from 4.5 to 5.8
• Discharge to wetlands for final polishing
• Running cost about 100 per week
• Limestone 10 per tonne
• Main cost is for transport of the limestone to
  the site
• (Note performance results by personal
  communication - not yet confirmed)

56
Some issues

• Design for stability
• the upstream method of construction ???
• Acid drainage prevention and remediation
• Modification of tailings by thickening
• Reduction of tailings hazard by changing
  chemistry
• Dust suppression by tailings modification

57
Engineering the Tailings Thickening

• Many of the problems associated with tailings
  storages revolve around the high water volumes
  pumped out with the tailings
• Thickening involves removing some of this water
  prior to deposition in the TSF
• slight dewatering - thickened tailings
• greater dewatering - paste
• extensive dewatering - tailings cake
• Properly engineered, thickening can produce
  significant benefits
• up front costs more than compensated for by lower
  long-term costs

58
Some Benefits of Thickening

• Enhanced water recovery (less water available for
  evaporation and seepage loss in the TSF) - water
  may be costly in arid areas.
• If the tailings are thickened sufficiently,
  segregation may not occur on the tailings beach
  this is important for the central thickened
  discharge method of tailings disposal.
• The higher deposited solids content means that
  consolidation and evaporation outcomes are
  enhanced.
• Dusting may be less of a problem (if there is
  sufficient clay content to act as a binder for
  the silt sized and fine sand sized fractions, and
  segregation does not occur).
• Having less water on the TSF will often have
  benefits for stability of the walls, and also for
  reduced base seepage.
• Thickening is usually required for tailings to be
  used for underground fill.

59
Engineering Required

• Thickener design evolving rapidly
• mechanical design, flocculants
• now a specialist industry
• Pumping characteristics must be understood
• the rheology of the thickened tailings must be
  fully understood
• does material show shear thinning or shear
  thickening behaviour?
• Transport method critical (centrifugal pumps,
  positive displacement pumps, conveyor belt,
  trucking?)
• The flow behaviour on tailings beach (or
  underground) must be undersood (rheology)

60
Alcoas WA Operating Locations
Perth
Booragoon
Mandurah
Dwellingup
Indian Ocean
Waroona
Alcoas WA Operations supply around 15 of the
world market. Very high volumes of red mud
tailings - very high pH - potential major
environmental problem
Collie
61
Schematic of Alcoas Dry Stacking Process
FLOCCULENT
CYCLONES
SUN
CLOUD
EVAPORATION
DECANT
SPRINKLERS
RUNOFF
DRY DISPOSAL AREA
UNDER DRAINAGE
62
Kwinana Residue Storage Areas
63
Alcoas Pinjarra Refinery Thickener
64
Measuring Red Mud Slurry Properties
Vane Shear Testing
Slump Testing (slumps at different solids)
65
Rheology of Red Mud
Initial
Sheared

• Red mud behaviour is well described by the
  Herschel-Buckley model ???y?k?n
• n and k do not vary significantly in the
  range of concentrations Alcoa normally deal with
• The yield stress is the parameter most indicative
  of the shear history dependence.

Shear Stress ? D?P/4L (Pa)
Initial
Sheared
50Solids
47Solids
Shear Rate ? 8V/D (sec-1)
66
Rheology of Red Mud
47 Superthickener Underflow Effect of mixing
time
100
Shear Stress ? D?P/4L (Pa)
initial state
180 min
1500 min
4080 min
8040 min
12660 min
15450 min
10
1
10
100
1000
Shear Rate ? 8V/D (sec-1)
67
Final strength gain from evaporation

• Remaining alumina forms surface crust due to
  drying
• reduces drying efficiency dramatically

68
Mud Farming Ploughing to enhance drying
69
Use of an Amphirol to aid ploughing
70
Alcoas Perceived Benefits of Dry Stacking

• The height of the deposit can be increased
  economically
• Less land is required
• Significantly reduced risks to the ground water
• Safety hazards are significantly reduced
• Deposit can be readily accessed to rehabilitate
• Overall reduction in costs

71
Some issues

• Design for stability
• the upstream method of construction ???
• Acid drainage prevention and remediation
• Modification of tailings by thickening
• Reduction of tailings hazard by changing
  chemistry
• Dust suppression by tailings modification

72
Alcoa Carbonation of Red Mud

• High pH (gt13) of red mud is major part of the
  long-term hazard (caustic soda NaOH used in
  refining process)
• Carbonation involves the addition of CO2 to
  thickened residue slurry
• NaOH is converted to carbonate
• alumina is precipitated as dawsonite
• CO2 reaction with the solids (conversion of
  residual lime to calcite and conversion of TCA to
  calcite and gibbsite)
• The resulting liquor associated with the residue
• reduction in pH
• reduction total dissolved solids
• reduction in trace metal concentrations (As, V,
  Ga, Se)

73
Carbonation prototype plant at Kwinana
CO2 storage and vaporiser units
CO2 mixing tanks
74
Trial drying beds, Kwinana refinery
Untreated
Carbonated
75
Drainage water quality monitoring
76
Drying bed monitoring Shear Strength
Strength Target (20 kPa)
77
Comparison of drying rates
78
Overall Benefits of Carbonation

• Reduced risk to clay and synthetic seals
• reduced potential for ground water contamination
• Improved quality of run-off and drainage water
• reduced pH, Al and trace metal concentrations in
  leachate
• Reduced Drying Area Costs
• reduction in
• drying area
• potential for dust
• reliance on mechanical ploughing equipment
• Avoid future classification as a Hazardous Waste
• Basel convention has adopted hazardous criterion
  of pH 11
• EU considering imposing classification and taxing
  system
• Greenhouse benefit
• use of residue as a sink for CO2

79
Neutralisation Combine waste streams

• Many mine wastes can generate acid (previous
  discussion)
• Alumina refining produces high-pH residue (red
  mud
• What about combining the two waste streams to
  neutralise the two?
• The Bauxsol technology (a proprietary process
  owned by Virotec International Ltd)
• understood to involve using processed red mud
  to neutralise acid drainage or acidic tailings
  waters (http//www.virotec.com/)

80
Producing Bauxsol
According to Virotech, the processing required to
produce Bauxsol involves pH neutralisation of
the red mud without reducing its acid
neutralising capacity. After this is done,
reaction conditions are created to form brucite,
calcite, aragonite. The new carbonate,
hydroxide and hydroxycarbonate minerals that are
formed preserve the acid neutralising capacity of
the solids. Much of the sodium originally present
in the red mud is released during this treatment
and stabilisation process
81
Applying Bauxsol to tailings water
82
Spraying Bauxsol, Mt Carrington Toe Dam
83
Results, Mt Carrington
84
Results, Mt Carrington
85
Challenges

• Modify chemistry of tailings to reduce or
  eliminate the hazards due to toxicity
• pH modification prior to disposal (for acidic or
  basic tailings)
• combining tailings streams to provide in-built
  buffering capacity against acid generation
• cyanide removal ?
• Think laterally to consider if waste streams from
  different industries can be combined to
  reduce/eliminate the hazards associated with one
  or both

86
Some issues

• Design for stability
• the upstream method of construction ???
• Acid drainage prevention and remediation
• Modification of tailings by thickening
• Reduction of tailings hazard by changing
  chemistry
• Dust suppression by tailings modification

87
Dust from Tailings Dams
88
Dust generation from tailings

• Tailings have often very narrow grading - often
  fine sand and silt sizes - very susceptible to
  dusting once they are dry
• Tailings dust can contain harmful materials
• heavy metals, arsenic, radioactivity, asbestos.
• Dust blowing from tailings dams is a very
  powerful means of generating community concern,
  anger, opposition.
• Short term as well as long term issue.

89
Dust suppression

• During operations
• watering (e.g. Alcoas sprinkler systems)
• keep submerged (not always possible or desirable)
• Permanent
• capping with topsoil, re-vegetation
• expensive
• in arid or semi-arid climates, difficult to
  maintain vegetation cover in the long term
• may need capillary breaks to isolate cover from
  tailings
• spraying with synthetic membrane of some sort
• permanence questionable
• Engineering the tailings prior to deposition

90
Sprinklers for dust suppression at Alcoa
91
Surface re-vegetation (KCGM, Kalgoorile)
Trials using different thicknesses of waste rock
cover as capillary barrier over high salt
content tailings Surface rehabilitation of this
type is very expensive!
92
Tailings modification for dust suppression

• Tailings with some clay content (say gt 10)
  appear less susceptible to dusting
• e.g. in WA, in many open-pit gold-mining
  operations, the upper levels are often oxide ore,
  with significant clay content in the tailings
• schedule mining, or stockpile some oxide ore, so
  that the last layer of tailings (1 m or so?) has
  clay content
• thickening may be required to prevent segregation
  of this layer, otherwise only central area of the
  TSF will contain a clayey cap.
• Tailings with high salt content appear to be
  resistant to dusting
• in WA, high-salinity groundwater used for gold
  processing
• salt causes many problems - but prevents dust

93
Salt crusting efficient dust suppression
94
Micro Wind Tunnel (MWT)

• MWT
• Developed by CSIRO Australia
• Tests the resistance of tailings surfaces to
  dusting

95
Results of MWT Testing

• Results expressed as Dust Yield Potential (DYP)
• Freshwater tailings surface
• DYP 32.9 g/m2/sec
• Hypersaline tailings surface (4 months to 2 years
  after completion of deposition)
• DYP 0 to 1.2 g/m2/sec

96
Additives to prevent dusting ?

• Addition of saline water to final tailings layer
  not suggested as general solution to dusting
• causes problems of reduced evaporation, toxic to
  plants etc
• But this evidence suggests that additives to
  final layer might be a viable solution to dusting
  problem
• Challenge to find suitable cheap non-toxic
  additive(s) to produce durable non-dusting final
  layer

97
Requirements for Engineering

• All of the engineering with tailings examples
  presented require detailed characterisation of
  the tailings materials (preferably well before
  mining starts in earnest)
• mineralogy, grading, rheology, chemistry,
  acid-generating potential (acid buffering
  potential)
• Past practice has often neglected much of this
  characterisation work
• Even the most basic tailings storage scheme
  (pouring slurried tailings into impoundment)
  requires detailed geotechnical testing,
  understanding, and modelling of basic processes

98
Modelling Tailings Deposition

• 2nd half of paper deals with some of the detailed
  aspects of tailings deposition
• sedimentation and beaching in sub-aerial
  deposition
• consolidation under self-weight
• effect of base drainage
• effect of evaporation
• factors affecting evaporation (salinity salt
  crusting)
• numerical modelling of filling / consolidation /
  evaporation behaviour
• Illustrates the attention to detail required in
  even this limited aspect
• Same attention to detail is required in all other
  areas

99
Conclusions

• There is a history of long-term problems from
  tailings storages, often lasting long after
  mining activity has ceased.
• Tailings storages worldwide continue to
  experience full scale failures, escape of
  material by wind (dust) and water erosion, acid
  drainage, cyanide escape, etc.
• Tailings management must be given at least as
  much attention as any other part of the mining
  process right from the start of the project.

100
Conclusions (Cont.)
Quote from the report of the tribunal set up to
investigate the causes of the Aberfan disaster in
South Wales in 1966, where children attending the
local school were killed by failure of a coal
mine waste tip
As we shall hereafter seek to make clear, our
strong and unanimous view is that the Aberfan
disaster could and should have been prevented.
But the Report that follows tells not of
wickedness, but of ignorance, ineptitude and a
failure in communication.

• Ignorance on the part of those charged at all
  levels with the siting, control and daily
  management of tips

• bungling ineptitude on the part of those who had
  the duty of supervising and directing them, and

• failure on the part of those having knowledge of
  the factors that affect tip safety to communicate
  that knowledge and see that it was applied.

101
Conclusions (Cont.)

• There have been many advances since 1967 in
  knowledge of how to manage tailings.
• Much of this knowledge is readily available (much
  of the material for this paper came from the
  internet).
• many useful guides, codes of best practice, case
  studies, are available
• Much still to be done to improve our ability to
  engineer tailings to achieve benign long-term
  outcomes.
• Improvement is not an optional luxury - the
  future of the mining industry requires it.

102
There are many arts and sciences of which a
miner should not be ignorant De Re
MetallicaGeorgius Agricola (1556)