Leak component maximum concentration
Component in leak
Maximum concentration
Scenario 1 32 ppm-vol 33 ppm-vol 345 ppm-vol 975 ppm-vol
Scenario 2
Scenario 3
Scenario 4
Scenario 5
Hydrogen Methane Ethane Propane i-Butane n-Butane
1,532 ppm-vol 833 ppm-vol 2,982 ppm-vol 3,080 ppm-vol 843 ppm-vol 889 ppm-vol
11.1 vol% 20.0 vol% 7.4 vol% 9.2 vol%
6 ppm-vol
12 ppm-vol 90 ppm-vol 1.2 vol% 0.4 vol% 15.5 vol%
996 ppm-vol
i-Pentane n-Pentane C 6+ naphtha
137 ppm-vol 46 ppm-vol
0.1 vol%
Table 1
behaviour and flammability risk. The system includes both light hydrocarbon streams and heavier naphtha-range com- ponents, providing a representative basis for comparison. Figure 1 shows the naphtha reformer reactor and stabiliser process twin in Petro-SIM used for this case study. Five leak scenarios were evaluated at different locations within the stabiliser system separating the products from the naphtha reformer reactor. The different scenarios include:
quickly under stable atmospheric conditions. Near-source concentrations vary across scenarios, but decrease rapidly with distance, and calculated lower flammability limit (LFL) percentages remain well below critical thresholds. • Scenario 1 : Represents a light-end vapour release from the stabiliser overhead, characterised by rapid dispersion and relatively low source concentrations, as shown in Table 1 . Under the evaluated conditions, LFL values remain well below flammability thresholds, limiting the formation of sustained flammable vapour clouds. • Scenario 2 : Higher hydrogen, ethane, and propane content in the process stream increases is reflected in the maximum leak composition, as shown in Table 1. Despite elevated near-source concentrations, rapid dilution pre- vents sustained flammable vapour cloud formation, and overall flammability risk remains. • Scenario 3 : Evaluates a leak from a liquid-phase sys- tem, resulting in higher localised vapour concentrations of light hydrocarbon at the source, reported in Table 1. While phase conditions influence near-source concentration levels, dispersion behaviour remains consistent with light hydrocarbon releases, and flammability risk does not mate- rially increase under the evaluated conditions. Across all three scenarios, dispersion behaviour remains consistent despite differences in composition and release conditions. While leak location and phase influence near- source concentration levels, they do not materially change hazard classification under the evaluated conditions. This consistency across light hydrocarbon scenarios highlights an important limitation of detection-based approaches. Without composition context, similar leak sig- nals may be interpreted as equivalent risk events, despite representing different hazard profiles. The results show that while concentration levels and phase conditions can influence near-source behaviour, the overall hazard classi- fication remains governed by the volatility and dispersion characteristics of the fluid. This reinforces the value of inte- grating process-derived composition into post-detection analysis, enabling more confident differentiation between low-risk releases and conditions requiring escalation. Heavier hydrocarbon releases The final two scenarios evaluate leaks involving heavier hydrocarbon fractions associated with stabilised naphtha
Stabiliser overhead vapour line. Stabiliser gas stream control valve. Reflux pump discharge. Reformate pump discharge (naphtha). Stabiliser reboiler (naphtha).
Each scenario represents a different process stream within the stabiliser system, defined by its phase and boil- ing range. These streams range from light hydrocarbon vapours in the overhead system to heavier naphtha frac- tions in downstream liquid handling equipment. As a result, they exhibit fundamentally different vaporisation and dis- persion behaviour following a release. All scenarios assume a small but credible leak size (1 mm pinhole) under consistent meteorological conditions.1 Maintaining consistent external conditions allows observed dispersion differences to be attributed primarily to varia- tions in fluid composition. This approach isolates composition as the primary var- iable influencing dispersion. In operational settings, leak detection systems identify location but do not inherently capture fluid properties. By anchoring the analysis in process-derived composi- tion rather than assumed release characteristics, the study provides a more representative basis for evaluating leak behaviour following a loss of containment. This distinction is important for risk assessment, as it enables composition to be evaluated explicitly rather than inferred from detec- tion data alone. Light hydrocarbon releases Leaks involving light hydrocarbons (ethane, propane, and butanes) were evaluated across three scenarios. These streams are characterised by high volatility and rapid dispersion. Dispersion modelling done with Flaretot software indi- cates that vapour clouds formed from these releases dilute
38
PTQ Q3 2026
www.digitalrefining.com
Powered by FlippingBook