🎓 Lesson 13 D5

Predicting Long-Term Leachate Quality Using PHREEQC-Based Reactive Transport Models

It’s a computer method that simulates how water moving through mine waste will dissolve and carry chemicals over decades — helping engineers predict if harmful pollution (like acid or metals) will leak out long after the mine closes.

🎯 Learning Objectives

  • Explain the conceptual linkage between mineral assemblage, redox conditions, and long-term leachate pH using PHREEQC speciation outputs
  • Design a minimal but defensible reactive transport model setup in PHREEQC for an ARD-prone waste rock pile, including selection of key phases, kinetics, and boundary conditions
  • Analyze simulated leachate concentration–time profiles to identify critical control points (e.g., peak acidity, neutralization inflection, metal breakthrough)
  • Apply charge-balance error and saturation index diagnostics to assess model reliability and constrain uncertainty

📖 Why This Matters

When a mine closes, waste rock and tailings don’t stop reacting — they can generate acidic, metal-laden leachate for 100+ years. Regulatory agencies (e.g., EPA, BC Ministry of Environment) now require predictive evidence that closure plans will protect groundwater and surface water over the full post-closure period. Traditional static geochemical testing (e.g., ABA, NAG) only gives snapshots; PHREEQC-based reactive transport models provide dynamic, mechanistic forecasts — turning qualitative risk assessments into quantifiable, auditable performance predictions.

📘 Core Principles

Reactive transport modeling rests on three pillars: (1) Thermodynamic equilibrium — governing aqueous speciation and mineral saturation via mass-action laws; (2) Kinetic constraints — controlling rates of slow-reacting minerals (e.g., pyrite oxidation, carbonate dissolution) where equilibrium is not achieved; and (3) Transport physics — describing how water movement carries dissolved species and controls reactant supply. PHREEQC handles the geochemistry; coupling it with transport requires discretizing space (e.g., 1D column) and time, then iterating chemistry at each node per time step. Critical inputs include initial mineralogy (XRD-derived), water composition, oxygen diffusion profile, and hydraulic conductivity — all calibrated against field lysimeter or porewater data.

📐 Pyrite Oxidation Rate (Kinetic Constraint)

While PHREEQC solves equilibrium, long-term ARD prediction relies on kinetic rate laws for pyrite oxidation — the dominant acid-generating reaction. The Butler–O’Connor rate law is widely adopted in coupled models to limit unrealistic instantaneous acid generation.

Butler–O’Connor Pyrite Oxidation Rate

R = k × [O₂] × 10^(0.19 × pH)

Empirical kinetic rate law for pyrite oxidation under near-neutral to acidic conditions, incorporating pH and dissolved oxygen dependence.

Variables:
SymbolNameUnitDescription
R Oxidation rate mol/(m²·s) Molar rate of pyrite oxidation per unit surface area
k Temperature-dependent rate constant mol/(m²·s) Pre-exponential factor calibrated to field temperature
[O₂] Dissolved oxygen concentration mol/L Aqueous O₂ activity driving oxidation
pH Solution pH unitless Controls proton-assisted oxidation pathway
Typical Ranges:
Temperate unsaturated waste rock: 1 × 10⁻¹⁵ to 5 × 10⁻¹³ mol/(m²·s)
Subsurface saturated tailings: 1 × 10⁻¹⁶ to 1 × 10⁻¹⁴ mol/(m²·s)

💡 Worked Example

Problem: Given: pyrite surface area = 0.8 m²/g, dissolved O₂ = 7 mg/L (≈ 2.2 × 10⁻⁴ mol/L), pH = 4.2, temperature = 12°C, rate constant k = 1.6 × 10⁻¹¹ mol/(m²·s) at 12°C (from Nordstrom & Munoz, 1994). Calculate oxidation rate R (mol/m²·s).
1. Step 1: Convert O₂ concentration to mol/L: 7 mg/L ÷ 32 g/mol = 2.19 × 10⁻⁴ mol/L
2. Step 2: Apply Butler–O’Connor law: R = k × [O₂] × 10^(0.19×pH) → R = (1.6×10⁻¹¹) × (2.19×10⁻⁴) × 10^(0.19×4.2)
3. Step 3: Compute exponent: 0.19 × 4.2 = 0.798 → 10^0.798 ≈ 6.29; then R = 1.6e−11 × 2.19e−4 × 6.29 ≈ 2.18 × 10⁻¹⁴ mol/m²·s
Answer: The calculated pyrite oxidation rate is 2.18 × 10⁻¹⁴ mol/m²·s — consistent with observed low-rate field weathering under mildly acidic, O₂-limited conditions.

🏗️ Real-World Application

At the Mount Polley Mine (British Columbia), PHREEQC-based reactive transport models were used in the 2018 Closure Plan to forecast leachate quality from the tailings storage facility over 200 years. The model incorporated XRD-quantified sulfide and carbonate content, measured O₂ diffusion profiles from core samples, and 10-year lysimeter data for calibration. It predicted peak sulfate at year 42 (1,850 mg/L), followed by gradual neutralization as carbonate buffering exhausted — directly informing the design of a 3-m thick alkaline cover and long-term monitoring trigger levels (e.g., pH < 6.0 or SO₄ > 1,000 mg/L). This model was accepted by the BC Environmental Assessment Office as compliant with the *Environmental Management Act* and *Mine Reclamation Code*.

📋 Case Connection

📋 Mount Polley Tailings Storage Facility Closure & Water Cover Implementation

Legacy tailings with sulfidic mineralogy requiring >100-year ARD suppression

📋 Cadia Valley Copper-Gold Mine Bio-Integrated Landform for Waste Rock Dump Closure

Steep, unvegetated waste rock dumps with acid-generating potential and high erosion risk

📋 Tunnel Ventilation Shaft Closure at Gotthard Base Tunnel (Switzerland)

Vertical shaft closure in karst terrain with unknown fracture flow paths and groundwater interaction

📚 References