Skip to content
Upcoming courses

Pool fire

Tools
From Fire Safety Academy Wiki
Revision as of 22:34, 2 November 2026 by Jan (talk | contribs) (Remove links to pages that do not exist; retain displayed text (wiki-wide link audit).)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Contents
  1. 1Pools and spills
  2. 2How a pool fire burns
  3. 3Heat release rate
  4. 4Burning duration and flame height
  5. 5Thermal radiation
  6. 6Protection and firefighting
  7. 7Limits and uncertainty
  8. 8See also
  9. 9Notes
  10. 10References

A pool fire is a fire burning on a horizontal, upward-facing layer of fuel, usually a flammable or combustible liquid in a container, a bund or a spill on the ground.[1] The liquid itself does not burn: heat from the flames evaporates it, and the vapour burns above the surface.[2] The term can also cover flat slabs of solids that decompose much like liquids, such as polymethylmethacrylate and polyethylene.[1]

In United States nuclear power plants, for example, liquid spill fires are among the most commonly assumed fires of transient fuels.[3] Such pool fires can result from leaking pump seals, oil spills from transformers, fuel spray from diesel generators, and liquids used for cleaning and painting.[4] They can be large: a burning pool of gasoline of 1 m² releases about 2.5 MW, about the same as an average burning living-room sofa.[5][6] The size of the fire follows mainly from the area of the pool and the burning rate of the fuel, and the main hazard to the surroundings is thermal radiation from the large, sooty flames.[7][8] Engineers estimate the heat release rate, burning duration, flame height and radiant heat flux with simple correlations that rest on large-scale tests.[3][9]

Pools and spills

Gottuk and White distinguish two situations. A pool is a confined body of fuel, typically more than 1 cm deep, that forms when a release collects in a low spot such as a trench, or that exists in tanks and containers.[7] A spill is a thin layer of fuel from an unconfined release, and its fire is highly variable, depending on the source, the surface it spreads over and the moment of ignition.[7] Spill depths for liquid spill fires range from 0.7 mm to 4 mm, depending mainly on the quantity released.[10]

The difference matters for the burning rate. In tests with jet fuel spilled on concrete, unconfined spill fires larger than about 1 m in diameter burned at roughly one fifth of the maximum burning rate of confined pool fires.[11] The authors attribute this to greater heat loss from the thin fuel layer, about 1 mm deep, to the concrete.[11] For a given amount of fuel, a spill with a large area burns with a high heat release rate for a short time. A spill with a small area burns with a lower rate for longer.[1]

In the event tree of Beyler for a release of flammable material, a pool fire follows when a spill accumulates and is ignited locally after a delay.[9] The geometry of the containment, or the lack of it, strongly influences its characteristics.[9] A pool fire can also follow a vapour cloud fire that burns back to the release point.[9] In quantitative risk assessment it is one of the radiant heat outcomes, defined as a fire involving a pool of confined or spreading flammable material.[12]

How a pool fire burns

A liquid has to turn into vapour before it can burn.[2] Part of the heat of the flames therefore has to return to the liquid surface to keep the evaporation going. The balance between that heat and the energy needed to evaporate the fuel controls how fast the fuel burns.[13] The burning rate is usually given as the mass of fuel burned per square metre of pool surface per second.[3]

Cross-section of a burning liquid pool in a bund: vapour rises from the liquid, a tall flame burns above it with black smoke, arrows show heat back to the pool and radiation to a target.
A pool fire: heat fed back from the flame evaporates the liquid, the vapour burns above the pool, and the flame radiates heat to its surroundings. Based on Van Rossum (2027), §4.1 and §4.3, and Beyler (2002a), pp. 3-270 and 3-277.

The flash point is the lowest temperature at which a flammable mixture of vapour and air exists at the surface of a liquid.[14] At that temperature the vapour pressure is just high enough to give a mixture at the lower flammability limit.[15] In the NIST experiments for the World Trade Center investigation, for example, jet fuel with a flash point of 43 to 66 °C did not at first form a flammable vapour layer at room temperature.[16] The radiant heat of the burner first had to warm the fuel.[16] Once ignited, flames spread over flammable and combustible liquids at about 10 cm/s to 2 m/s.[4]

The way a pool fire burns changes with its diameter. NUREG-1805, after Babrauskas, distinguishes four modes: convective and laminar below 0.05 m, convective and turbulent below 0.2 m, radiative and optically thin from 0.2 to 1.0 m, and radiative and optically thick above 1.0 m.[1] In pools larger than about 1 m, radiation from the optically thick flame dominates the heat transfer to the surface.[1] Above a diameter of about 0.2 m the burning rate per unit area increases with diameter, and it approaches a maximum value for very large pools.[17] Fuel blends with components of very different volatility do not burn at a constant rate. The most volatile components burn first, and later the burning resembles that of the heavier components that remain.[17]

Heat release rate

The heat release rate of a pool fire is found from the burning rate per unit area, the effective heat of combustion and the area of the pool.[3] A correction factor accounts for the lower burning rate of smaller pools: it is close to zero for very small pools and approaches one for large ones.[3][17] NUREG-1805 gives the relation as follows.[3]

Q˙=m˙″ΔHc,effAf(1−e−kβD)

Symbol Description Unit
Q˙ heat release rate of the pool fire kW
m˙″ mass burning rate per unit area of a large pool kg/(m²·s)
ΔHc,eff effective heat of combustion kJ/kg
Af horizontal burning area of the fuel m²
kβ empirical constant of the fuel m−1
D diameter of the burning area m

For a pool that is not circular, an equivalent diameter is used: the diameter of a circle with the same area.[17] The relation applies to confined pools burning in the open, in still air, in a pan or tank without an excessive lip height.[11] The fuel data come from large-pool tests compiled by Babrauskas. For gasoline the burning rate is 0.055 kg/(m²·s), the effective heat of combustion 43,700 kJ/kg and kβ 2.1 m−1. For transformer oil the values are 0.039 kg/(m²·s), 46,400 kJ/kg and 0.7 m−1.[18] Where the constant kβ is unknown, NUREG-1805 advises a conservative value of 100 m−1, which sets the bracket to one.[4]

Two rising curves of heat release rate per square metre against pool diameter; gasoline levels off near 2,400 kW/m² by 2 m, transformer oil rises more slowly.
Heat release rate per square metre of pool surface against pool diameter for gasoline and transformer oil, from the large-pool data and the diameter correction. Based on Iqbal and Salley (2004), pp. 3-3 to 3-4.

Worked example. NUREG-1805 considers a spill of 5 US gallons (18.9 L) of lubricating oil that covers 9.0 ft² (0.836 m²) of a concrete floor. It uses the properties of transformer oil, with an effective heat of combustion of 46,000 kJ/kg.[4][19] The equivalent diameter is 1.03 m, so the bracket is 1 − e−0.7 × 1.03 ≈ 0.514. Then:

Q˙=0.039×46,000×0.836×0.514≈771 kW

This agrees with the 772 kW in the NUREG-1805 solution; the small difference comes from rounding.[20] Without the diameter correction the result would be about 1.5 MW, so for a pool of this size the correction roughly halves the heat release rate. For a large gasoline pool the burning rate and heat of combustion give about 2.4 MW per square metre, consistent with the 2.5 MW quoted for a 1 m² pool.[18][5] Inside a room, the heat release rate can differ from these free-burn values. Radiation from the hot smoke layer and the heated ceiling and walls increases the burning rate, while small openings limit the oxygen supply and reduce it.[1]

Burning duration and flame height

A fire that burns at a constant heat release rate consumes the fuel at a constant rate, so the burning duration follows from the amount of fuel and the burning rate.[21] For a liquid pool, the fuel surface drops at the regression rate, which is the mass burning rate per unit area divided by the density of the liquid.[22] In the lubricating oil example the regression rate is about 0.05 mm/s, and the 18.9 L spill burns for about 7.35 minutes.[20] The same volume spread over a larger area would burn more intensely for a shorter time.[22]

The flame height matters because it shows whether flames reach objects above the pool and because it sets the size of the radiating flame.[23] Several correlations exist, based on the pool diameter and the burning rate or the heat release rate; the Heskestad correlation is used in both NUREG-1805 and the SFPE handbook.[23][24] For the lubricating oil fire it gives a flame height of 2.31 m, and the Thomas correlation gives 2.67 m.[20] Both exceed the 8 ft (2.4 m) to a cable tray above the pool, so the example concludes that the flames impinge on the tray.[20][19] Pool fire flames fluctuate, so the flame tip moves well beyond the continuous luminous region, and flame height is defined in different ways in different studies.[23] Wind tilts the flame and changes its visible length.[23][25]

Thermal radiation

Thermal radiation is the main way in which large open hydrocarbon fires injure people and damage property.[8] The flame radiation comes mainly from water vapour, carbon dioxide and soot.[23] Fires of flammable liquids typically have higher heat release rates per unit area than fires of ordinary combustibles, taller flames and more luminous soot. An observer near a liquid fire therefore feels more heat than near an ordinary fire of similar size.[26]

Estimating the radiation involves three steps: describing the size of the fire, describing its radiative properties, and calculating the heat flux at the target.[9] The simplest method treats the flame as a point source at the centre of the fire, radiating a fraction of the heat release rate.[27] That fraction ranges from about 0.15 for low-sooting fuels such as alcohol to 0.60 for high-sooting hydrocarbons, and cold soot around large flames can reduce it considerably.[27] The point source method is accurate at large distances, but it is not a good choice where ignition of nearby combustibles must be assessed.[28]

The solid flame model represents the flame as a cylinder with the pool diameter and the flame height, whose surface radiates with an average emissive power.[29] The heat flux at a target equals that emissive power times the view factor between flame and target.[29] Shokri and Beyler found that the effective emissive power falls as the pool gets larger, because black smoke outside the flame increasingly hides the luminous flame.[30] Their correlation gives about 57 kW/m² for a 1 m pool, 48 kW/m² for a 10 m pool and 22 kW/m² for a 50 m pool.[30] Most of the uncertainty lies in the emissive power, not in the view factor.[30]

Falling curve of effective emissive power from about 57 kW/m² at 1 m pool diameter to about 22 kW/m² at 50 m.
Effective emissive power of large hydrocarbon pool fire flames against pool diameter, after the correlation of Shokri and Beyler. Based on Beyler (2002a), p. 3-277.

For quick screening, Shokri and Beyler also correlated the radiant heat flux to a vertical target at ground level with the distance from the pool centre divided by the pool diameter.[31] Above ground level the heat flux is expected to be higher than the correlation gives, because the radiation is strongest near the mid-height of the flame.[31] A safety factor of 2 is recommended for design use with this correlation, but not where a realistic estimate is needed.[31]

Protection and firefighting

Containment limits the size of a pool fire. A dike or curbing confines a spill to a known area, and the area, together with the burning rate, sets the heat release rate.[1][4][7] A pool in a bund therefore burns longer than an unconfined spill of the same volume, but with a smaller fire.[22]

Water is often a poor extinguishing agent for these fires. A liquid that is lighter than water and does not mix with it floats on the water and keeps burning.[32] Firefighting foam is therefore widely used for fires of flammable and combustible liquids: it keeps air from the fuel, cools as the water drains from it, and protects surfaces that are not yet burning.[32] It must be applied in sufficient volume and at a sufficient rate to make up for the foam destroyed by heat and flames.[33] Fixed foam systems are used in bunds and on spills, and subsurface injection of low-expansion foam into hydrocarbon storage tanks is a recognised design method in the NFPA standards.[34]

Tank fires carry a further hazard. When water penetrates a hot oil whose temperature is well above the boiling point of water, it turns into steam below the surface and causes an eruption of burning oil, called boilover.[32] NUREG-1805 describes boilover as a phenomenon of fires in open tanks containing blends such as crude oil, which requires water at the bottom of the tank.[35] Foam applied to a liquid at 100 °C or above can also cause boilover, because the steam formed expands the foam fourfold.[34]

Fire resistance tests for structures near pool fires need a different exposure. Most standard tests simulate a cellulosic fire, which grows more slowly than a hydrocarbon fire, although it may ultimately become hotter; specific hydrocarbon fire exposures for pool and jet fires were developed later.[36] In the United States, UL 1709 rates barriers for a hydrocarbon fire exposure, with H ratings for 120 minutes against the passage of flame and heat.[37] In Australia, the Fire Safety Verification Method data sheets (2020) refer to results of hydrocarbon fire tests and to a hydrocarbon heating regime alongside the standard one.[38]

Limits and uncertainty

The methods for pool fires are largely semi-empirical and rest on experimental data, so they carry uncertainties.[9] The burning rate correlations come from confined pools in the open and in still air. There are very few data for unconfined spills, and wind and lip height can change the burning rate in either direction.[11][39] The radiation methods assume a circular or nearly circular pool.[40] The emissive power correlation was developed from fires of kerosene, fuel oil, gasoline, jet fuels and liquefied natural gas, with pool diameters from 1 to 50 m.[40]

Inside a building, the free-burn values are only a starting point. Feedback from the smoke layer and the walls, and the supply of air through openings, change the burning rate.[1] The heat release rate is not a fundamental property of a fuel and is usually determined from tests.[41]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Iqbal and Salley (2004), p. 3-5.
  2. ↑ 2.0 2.1 Van Rossum (2027), §4.1.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Iqbal and Salley (2004), p. 3-3.
  4. ↑ 4.0 4.1 4.2 4.3 4.4 Iqbal and Salley (2004), p. 3-6.
  5. ↑ 5.0 5.1 Iqbal and Salley (2004), p. 3-2.
  6. ↑ Van Rossum (2027), §4.5.
  7. ↑ 7.0 7.1 7.2 7.3 Gottuk and White (2002), p. 2-297.
  8. ↑ 8.0 8.1 Beyler (2002a), p. 3-268.
  9. ↑ 9.0 9.1 9.2 9.3 9.4 9.5 Beyler (2002a), p. 3-269.
  10. ↑ Gottuk and White (2002), p. 2-299.
  11. ↑ 11.0 11.1 11.2 11.3 Gottuk and White (2002), p. 2-311.
  12. ↑ Barry (2002), p. 5-184.
  13. ↑ Van Rossum (2027), §4.3.
  14. ↑ Iqbal and Salley (2004), p. 6-1.
  15. ↑ Simmons (2002), p. 1-147.
  16. ↑ 16.0 16.1 Ohlemiller et al. (2005), p. 42.
  17. ↑ 17.0 17.1 17.2 17.3 Gottuk and White (2002), p. 2-310.
  18. ↑ 18.0 18.1 Iqbal and Salley (2004), p. 3-4.
  19. ↑ 19.0 19.1 Iqbal and Salley (2004), p. 3-22.
  20. ↑ 20.0 20.1 20.2 20.3 Iqbal and Salley (2004), p. 3-23.
  21. ↑ Iqbal and Salley (2004), p. 3-7.
  22. ↑ 22.0 22.1 22.2 Iqbal and Salley (2004), p. 3-8.
  23. ↑ 23.0 23.1 23.2 23.3 23.4 Beyler (2002a), p. 3-270.
  24. ↑ Beyler (2002a), p. 3-274.
  25. ↑ Beyler (2002a), p. 3-271.
  26. ↑ Iqbal and Salley (2004), p. 5-2.
  27. ↑ 27.0 27.1 Iqbal and Salley (2004), p. 5-3.
  28. ↑ Beyler (2002a), p. 3-275.
  29. ↑ 29.0 29.1 Beyler (2002a), p. 3-276.
  30. ↑ 30.0 30.1 30.2 Beyler (2002a), p. 3-277.
  31. ↑ 31.0 31.1 31.2 Beyler (2002a), p. 3-273.
  32. ↑ 32.0 32.1 32.2 Iqbal and Salley (2004), p. B-58.
  33. ↑ Iqbal and Salley (2004), p. B-59.
  34. ↑ 34.0 34.1 Iqbal and Salley (2004), p. B-60.
  35. ↑ Iqbal and Salley (2004), p. F-9.
  36. ↑ Iqbal and Salley (2004), p. C-8.
  37. ↑ Iqbal and Salley (2004), p. C-9.
  38. ↑ Australian Building Codes Board (2020a), p. 233.
  39. ↑ Gottuk and White (2002), p. 2-313.
  40. ↑ 40.0 40.1 Iqbal and Salley (2004), p. 5-17.
  41. ↑ Iqbal and Salley (2004), p. 3-1.

References

  • Australian Building Codes Board. (2020a). Data sheets: Fire Safety Verification Method (Handbook annex, Print version 1.1).
  • Barry, T. F. (2002). Quantitative risk assessment in chemical process industries. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 5-176–5-213). National Fire Protection Association; Society of Fire Protection Engineers.
  • Beyler, C. L. (2002a). Fire hazard calculations for large, open hydrocarbon fires. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-268–3-314). National Fire Protection Association; Society of Fire Protection Engineers.
  • Gottuk, D. T., & White, D. A. (2002). Liquid fuel fires. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 2-297–2-316). National Fire Protection Association; Society of Fire Protection Engineers.
  • Iqbal, N., & Salley, M. H. (2004). Fire Dynamics Tools (FDTs): Quantitative fire hazard analysis methods for the U.S. Nuclear Regulatory Commission Fire Protection Inspection Program (NUREG-1805). U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation.
  • Ohlemiller, T. J., Mulholland, G. W., Maranghides, A., Filliben, J. J., & Gann, R. G. (2005). Fire tests of single office workstations (NIST NCSTAR 1-5C). National Institute of Standards and Technology.
  • Simmons, R. F. (2002). Premixed burning. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 1-144–1-154). National Fire Protection Association; Society of Fire Protection Engineers.
  • Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.
Retrieved from "https://fire-safety.academy/w/index.php?title=Pool_fire&oldid=2886"