QRA (Quantitative Risk Assessment) for Toxic Gas Dispersion
QRA for toxic gas dispersion is a step-by-step engineering method to figure out how far and how fast dangerous gases (like chlorine or ammonia) could spread if they leak—and how likely it is that people or equipment will be harmed.
⚠️ Why It Matters
📘 Definition
Quantitative Risk Assessment (QRA) for toxic gas dispersion is a systematic, physics-based engineering methodology used to quantify the likelihood and consequences of accidental releases of hazardous gases in process facilities. It integrates source term analysis, atmospheric dispersion modeling (e.g., Gaussian or CFD), toxicity dose–response relationships (e.g., ERPGs, AEGLs), and frequency estimation of initiating events to derive individual and societal risk metrics (e.g., fatality probability per year, F-N curves). The output informs safety system design, siting decisions, emergency planning, and ALARP justification per regulatory frameworks such as CCPS, COMAH, and OSHA PSM.
🎨 Concept Diagram
AI-generated illustration for visual understanding
💡 Engineering Insight
Never treat QRA as a 'one-time compliance exercise'—its true value emerges when dispersion results directly drive mechanical integrity programs. For example, if modeling shows >10% probability of exceeding ERPG-2 at a valve manifold within 30 s, that mandates not just faster ESD valves, but also double-block-and-bleed configuration with independent verification of seal integrity every 6 months. The model is only as good as the fidelity of its input assumptions—and those assumptions must be auditable, traceable, and updated with each MOC.
📖 Detailed Explanation
Next comes atmospheric behavior: gases disperse via wind-driven advection and turbulent diffusion. Gaussian models (e.g., ISCST3) assume passive, neutrally buoyant plumes and are fast and conservative for many vapor releases—but fail catastrophically for dense gases like chlorine or propylene, which hug the ground and pool. In those cases, dense-gas models solve conservation equations for mass, momentum, and energy, accounting for gravity-driven slumping, entrainment, and ambient temperature gradients. Model selection isn’t academic—it dictates whether your 1-km exclusion zone is 500 m or 2.3 km.
Advanced practice demands uncertainty quantification: Monte Carlo sampling of release rate (±35%), wind direction (±22.5°), and stability class (probabilistic weighting per local met data) yields confidence intervals on hazard distances. Integration with digital twin platforms now allows real-time QRA updates using live weather feeds and IoT sensor data—transforming static reports into dynamic operational risk dashboards aligned with ISA-84 and IEC 61511 functional safety lifecycles.
🔄 Engineering Workflow
📋 Decision Guide
| Rock/Field Condition | Recommended Design Action |
|---|---|
| High-pressure liquefied gas (e.g., Cl₂ @ 8 bar, 20°C) with elevated release point (>10 m) and stable atmosphere (Class F) | Use dense-gas dispersion model (e.g., SLAB, DEGADIS); apply 99th percentile wind speed ≤1.0 m/s; verify exclusion distance against ERPG-2 at 1 hr and 4 hr. |
| Low-pressure vapor-phase release (e.g., NH₃ from storage tank vent) in neutral stability (Class D) and moderate wind (2–4 m/s) | Apply Gaussian puff/plume model (e.g., ALOHA, PHAST); use 10-min average wind speed; assess both 10-min and 1-hr ERPG-2 contours. |
| Complex terrain or built environment (e.g., refinery with multiple structures within 500 m of release) | Supplement Gaussian modeling with CFD (e.g., FLACS, ANSYS Fluent); perform sensitivity analysis on building wake effects and channeling. |
📊 Key Properties & Parameters
Release Rate (ṁ)
0.1–50 kg/s (for 6–50 mm diameter orifices in pressurized chlorine/ammonia service)Mass flow rate of toxic gas escaping through a defined leak or rupture aperture, governed by fluid thermodynamics and hole geometry.
Dominates dispersion cloud size and duration—errors >2× propagate nonlinearly into consequence distances.
Atmospheric Stability Class
Class D (neutral, most common daytime urban/industrial) to Class F (strongly stable, nighttime, low wind)Empirical classification (A–F) of vertical mixing intensity based on solar radiation, wind speed, and cloud cover per Pasquill–Gifford.
Controls plume rise, vertical dilution, and ground-level concentration—Class F can increase 1-hr ERPG-2 distances by 3–5× vs. Class D.
Toxicity Threshold (ERPG-2)
0.003 ppm (phosgene) to 200 ppm (ammonia)Emergency Response Planning Guideline level at which nearly all individuals could be exposed for 1 hour without irreversible health effects.
Sets the critical concentration contour for hazard zone delineation—lower ERPG-2 values demand stricter release containment and larger exclusion distances.
Wind Speed (U)
0.5–6 m/s (site-specific 95th percentile for worst-case modeling per CCPS)Mean horizontal velocity of ambient air at 10 m height, driving advection and turbulent dilution of the gas cloud.
Low wind speeds (<1.5 m/s) maximize ground-level concentrations and extend hazard duration; high winds (>4 m/s) reduce peak concentrations but broaden lateral spread.
Source Height (Hₛ)
1.5–15 m (valve leaks, flange failures) to 30+ m (vessel vents, flare stacks)Vertical elevation of the release point above ground level, influencing initial plume buoyancy and entrainment.
Elevated sources reduce ground-level impact near origin but may increase exposure at downwind building rooftops or fence lines due to plume descent.
📐 Key Formulas
Choked Gas Flow Rate
ṁ = C_d A P_0 √(γ / (R T_0)) [2 / (γ + 1)]^{(γ + 1)/(2(γ − 1))}Mass flow rate for compressible gas under sonic (choked) conditions at an orifice.
| Symbol | Name | Unit | Description |
|---|---|---|---|
| ṁ | mass flow rate | kg/s | choked mass flow rate of gas |
| C_d | discharge coefficient | dimensionless | empirical coefficient accounting for non-ideal flow effects |
| A | flow area | m² | cross-sectional area of the orifice |
| P_0 | stagnation pressure | Pa | total pressure upstream of the orifice |
| γ | heat capacity ratio | dimensionless | ratio of specific heats, c_p/c_v |
| R | specific gas constant | J/(kg·K) | gas constant per unit mass |
| T_0 | stagnation temperature | K | total temperature upstream of the orifice |
Gaussian Ground-Level Concentration
C(x,y,0) = (ṁ / (2π U σ_y σ_z)) exp[−½(y/σ_y)²] exp[−½(Hₑ/σ_z)²]Steady-state centerline concentration at ground level for passive plume dispersion.
| Symbol | Name | Unit | Description |
|---|---|---|---|
| C(x,y,0) | Ground-level concentration | g/m³ or μg/m³ | Steady-state pollutant concentration at ground level (z = 0), at horizontal location (x, y) |
| ṁ | Emission rate | g/s or μg/s | Mass flow rate of pollutant emitted from the source |
| U | Wind speed | m/s | Average wind speed at effective stack height |
| σ_y | Horizontal dispersion coefficient | m | Standard deviation of plume concentration distribution in the crosswind (y) direction |
| σ_z | Vertical dispersion coefficient | m | Standard deviation of plume concentration distribution in the vertical (z) direction |
| y | Crosswind distance | m | Horizontal distance perpendicular to wind direction (from plume centerline) |
| Hₑ | Effective stack height | m | Sum of physical stack height and plume rise; height of emission above ground level |
🏭 Engineering Example
BASF Ludwigshafen Site, Germany — Chlorine Vaporizer Area
N/A (above-ground chemical process facility)🏗️ Applications
- Offsite consequence analysis for EPA RMP submissions
- Siting of control rooms and emergency shelters per NFPA 704 and API RP 752
- Design basis for gas detection coverage and alarm setpoints
- Justification of mitigation systems (water spray, vapor suppressants, blast-resistant enclosures)
📋 Real Project Case
Ammonia Refrigeration System HAZOP & LOPA Integration at Midwest Food Processing Plant
Retrofit of legacy ammonia chiller system serving 300k sq ft food processing facility