Modeling AMD Geochemistry in Underground Mines

1
Modeling AMD Geochemistry in Underground Mines
Bruce Leavitt PE PG, Consulting Hydrogeologist
James Stiles PhD PE, Limestone Engineering Raymond
Lovett PhD, Shipshaper LLC
2

• Limitations of existing AMD Prediction Methods
• Only considers Acid and Base Potential
• Does not consider Latent Acidity
• Does not consider Oxygen Depletion
• Does not consider Solute Transport
• Does not consider Recharge Water Chemistry and
  Volume

3
Study Purpose

• To investigate the suitability of the model to
  underground mine discharges.
• To determine the appropriate mineral assemblage
  and mass concentration.
• To compare the model in different hydrologic
  settings.
• To evaluate the sensitivity of the model to
  variations in input values comparable to typical
  field variations.

4
Three Hydrologic Settings

5
Effect of Flooding on Mine Water Chemistry

• Rapid dissolution of acidic salts
• Exclusion of oxygen from the mine
• Chemical reaction with recharging ground water.

6
TOUGHREACT Earth Sciences Division, Lawrence
Berkeley National Laboratory

• TOUGHREACT was designed to solve the coupled
  equations of sub-surface multi-phase fluid and
  heat flow, solute transport, and chemical
  reactions in both the saturated and unsaturated
  aquifer zones. This program can be applied to
  many geologic systems and environmental problems,
  including geothermal systems, diagenetic and
  weathering processes, subsurface waste disposal,
  acid mine drainage remediation, contaminant
  transport, and groundwater quality.

7
Model Configuration
8
Mineral Assemblage
Mineral Volume Concentration K25 (mol/m2/s) Ea (kJ/mol)
calcite 0.001 equilibrium equilibrium
gypsum 0.0001 equilibrium equilibrium
melanterite 0.002 equilibrium equilibrium
rhodochrosite 0.010 3.55x10-6 40.0
illite 0.400 6.9185x10-13 22.2
jarosite 0.001 6.9185x10-13 22.2
Al(OH)3 (amorphous) 0.001 6.9185x10-13 22.2
gibbsite 0.001 6.9185x10-13 22.2
pyrolusite 0.001 6.9185x10-13 22.2
9
Mineral Assemblage cont.
Mineral Volume Concentration K25 (mol/m2/s) Ea (kJ/mol)
ferrihydrite 0.001 6.9185x10-13 22.2
jurbanite 0.001 1.0233x10-14 87.7
quartz 0.001 1.0233x10-14 87.7
kaolinite 0.500 Neutral 6.918x10-14 Acid 4.898x10-12 Base 8.913x10-18 22.2 65.9 17.9
chlorite 0.001 Neutral 3.020x10-13 Acid 7.762x10-12 Base N/A 88.0 88.0 N/A
pyrite 0.0015 Neutral 2.818x10-6 Acid 3.020x10-9 Base N/A 56.9 56.9 N/A
siderite 0.001 Neutral 1.660x109-9 Acid 2.570x10-4 Base N/A 62.76 36.1 N/A
magnetite 0.001 Neutral 1.260x109-11 Acid 6.457x10-9 Base N/A 18.6 18.6 N/A
10
Archetype pH
11
Archetype Iron
12
Model Results pH
13
Model Results Iron
14
Pyrite Kinetic Data

• Neutral 2.818 x 10-6 mol-m-2-s-1 McKibben and
  Barnes (1986a)
• Neutral 3.167 x 10-10 mol-m-2-s-1 McKibben and
  Barnes (1986b), Nicholson (1994), and Nicholson
  and Sharer (1994)
• Acidic 3.020 x 10-9 mol-m-2-s-1
• Acidic 1.553 x 10-8 mol-m-2-s-1 McKibben and
  Barnes (1986b), Brown and Jurinak (1989), and
  Rimstidt, et al. (1994)
• Acidic 6.0 x 10-10 mol-m-2-s-1 Calibrated

15
Ferrous Ferric Oxidation

• Fe2 1/4O2 H gt Fe3 1/2 H2O
• Oxidation rate is pH dependant.
• Model holds ferrous and ferric iron in
  equilibrium.
• Model overstates ferric iron concentration
  leading to excess pyrite oxidation.

16
High Dilution pHYear 5
17
High Dilution pHYear 10
18
High Dilution pHYear 15
19
High Dilution pHYear 20
20
High Dilution IronYear 5
21
High Dilution IronYear 10
22
High Dilution IronYear 15
23
High Dilution IronYear 20
24
Modeling Difficulties

• Ferrous iron oxidation
• Insufficient aluminum production
• CO2 partial pressure spikes at full mine flooding
• Mine complexity is limited by computational
  capacity
• Homogeneous mineral distribution
• Mine atmosphere composition

25
Other Results

• Gypsum precipitation / dissolution in the mine
• Goethite precipitation in the mine.
• Elimination of pryhotite and the reduction of the
  pyrite kinetic rate has reduced the observed
  difference in water pH and iron between the high
  dilution and low dilution cases.

26
Future Work

• Resolve the iron oxidation issue
• Closed mine atmosphere sampling.
• Sensitivity analysis of input parameters
  including recharge chemistry, mine geometry,
  initial melanterite and calcite concentrations.
• Testing of in situ remedial options.

27
Conclusions

• The TOUGHREACT program allows chemical and
  hydrodynamic interaction in a flooded and
  unflooded underground mine environment.
• TOUGHREACT is able to emulate the change in
  discharge chemistry with time.
• It is a useful tool in understanding acid
  formation, solute transport, and discharge
  relationships.
• Due to the extensive number of assumptions it is
  not, at this time, a suitable permitting tool.