Skip to content
Upcoming courses

Explosion

Tools
From Fire Safety Academy Wiki
(Redirected from Physical explosion)
Contents
  1. 1What counts as an explosion
  2. 2How a gas or dust explosion develops
  3. 3Deflagration and detonation
  4. 4Explosion scenarios in fire safety
  5. 5Effects on people and buildings
  6. 6Prevention and protection
  7. 7Assessment and uncertainty
  8. 8See also
  9. 9Notes
  10. 10References

An explosion is a rapid, potentially damaging rise in pressure, such as the one that follows when a combustible gas, vapour or dust mixed with air in a confined space is ignited. The flame travels through the mixture, and the confinement stops the hot combustion products from expanding freely.[1] NUREG-1805, a fire hazard analysis guide of the US Nuclear Regulatory Commission, gives a broader definition: a sudden and violent release of high-pressure gases into the surroundings.[2] This definition does not depend on what produced the pressure.[2] Explosions are often grouped into physical explosions, such as a heated tank that bursts, and chemical explosions, such as a burning cloud of gas or dust.[3]

Explosions matter in fire safety because ordinary buildings are weak against pressure. Glass windows typically fail at 2 to 7 kPa, whereas a premixed gas-air mixture ignited inside a building can produce up to about 8 bar, or 800 kPa.[4][5] Engineers distinguish a deflagration, in which the flame moves slower than sound, from a detonation, in which it moves faster, because the two need different protection.[6][7] Protection works by preventing a flammable mixture, by removing ignition sources, or by limiting the pressure through venting, suppression or an enclosure strong enough to resist it.[1][8]

What counts as an explosion

In its most widely accepted sense, an explosion is a bursting with a loud, sharp noise and an expanding pressure front, ranging from a supersonic shock wave to a relatively mild wind.[2] The NUREG-1805 definition rests on two ideas. The gas must be at a higher pressure than its surroundings, and it must be released so fast that its energy escapes as a shock wave.[2] The stored energy can be nuclear, chemical, electrical or pressure energy. A release counts as an explosion only if it produces a pressure wave that can be heard, even if it causes no damage.[3]

The rate of release decides how violent an explosion is. The energy stored in a car tyre can cause an explosive burst, but the same energy can also be released gradually.[9] In a combustion explosion the confinement keeps the hot gases from expanding freely.[1] The normal openings of process equipment are usually too small to prevent destructive pressures.[10]

Explosions are often classified by how the high-pressure gas is produced.[3] In a physical explosion the pressure comes from mechanical means without any chemical change. Examples are a tank that fails after external heating has raised its internal pressure, and a superheated liquid that is suddenly released and flashes into vapour.[3] A pressure vessel that ruptures without its gas igniting also causes a physical explosion, driven by the expansion of the gas to atmospheric pressure.[11] In a chemical explosion the pressure comes from a chemical reaction. This group includes combustion explosions of dust, gas, mists and backdraft gases, thermal explosions in runaway reactions, and explosions of high explosives and propellants.[3]

How a gas or dust explosion develops

For a deflagration, NUREG-1805 names three things that must be present at the same time: a flammable mixture of fuel and oxygen, usually from air, a means of ignition, and an enclosure.[10] The mixture can only explode between its lower and upper explosion limits, which are also called flammability limits.[12] Outside this range a flame cannot spread through it.[12] All glowing sparks are hot enough to ignite a mixture, but some carry too little energy, while a hot process surface may be cooler yet hold a large amount of energy.[10]

Once ignited, the flame spreads through the mixture and heats the gas it passes.[1] For a hydrocarbon fuel mixed with exactly enough air for complete burning, the temperature rises from 300 K to about 2,200 K.[13] If no heat is lost, the pressure in a closed vessel can then climb to a little over 7 bar. A vessel that is not designed for such pressures will rupture.[13] Measured maximum pressures for ten of twelve flammable gases and vapours lie between 6.8 and 8.1 bar above atmospheric pressure.[14] Near the flammability limits the pressures are lower, because the flame burns only part of the enclosure and heat is lost to the walls.[15]

The pressure does not rise evenly. In a worked example by Zalosh for a spherical vessel 7.15 m in radius, the pressure had risen by only about 14 kPa when the flame was halfway to the wall. The final rise was about 640 kPa.[16] In that example most of the pressure rise happened at the end, when flame filled almost the whole vessel.[16] Reaching the maximum pressure can take up to a few hundred milliseconds, depending on the size of the enclosure and the fuel.[10] With connected vessels, burning in one vessel raises the pressure in the next, and a flame that then enters it produces a correspondingly higher pressure.[13]

Explosions in the open air, called unconfined explosions, differ fundamentally from confined explosions in vessels, pipework or buildings, and need different countermeasures.[17] When a gas leaks into the open, mixes with air and is ignited, the flame front spreads through the cloud as a sphere wherever it can.[18] Inside a building, gases rushing out through a vent, door or window can create turbulence that produces a second pressure peak.[17]

Deflagration and detonation

A deflagration is an explosion in which the reaction spreads through the mixture by heat, at a speed below the speed of sound. It is slower than a detonation but can still cause damage.[8] In a detonation the combustion zone travels faster than sound through the unburned mixture.[19] A shock wave runs ahead of the reaction and compresses the gas, which speeds up the reaction.[10] A deflagration flame moves well below the speed of sound, about 340 m/s in room-temperature air, so its pressure rises almost evenly throughout an enclosure.[6] In a detonation the pressure rise is very uneven and almost instantaneous as the shock passes each point.[6]

Deflagration fills a vessel over time; detonation produces a sharp local shock front.
Deflagration and detonation create different pressure conditions. Based on Zalosh (2002), pp. 3-410–3-411 and Iqbal and Salley (2004), pp. F-20–F-21.

Typical detonation speeds are about 1,500 to 3,000 m/s, and NUREG-1805 gives a value of the order of 2,000 m/s.[13][10] In a gas mixture that starts at atmospheric pressure, the pressure just behind a detonation can be up to 20 bar.[13] Zalosh notes that this is roughly twice the pressure of the same mixture burning in a closed vessel.[7] Where a detonation is reflected from the closed end of a duct, the pressure can be up to 100 bar. Detonations are therefore much more destructive than an ordinary moving flame.[13]

The difference matters for protection. The sudden, uneven pressure rise of a detonation rules out explosion venting and explosion suppression, and its high, short pressure peaks need a special structural assessment.[7] Most accidental explosions are deflagrations, and NUREG-1805 describes detonations as very rare.[20][5] The minimum energy needed to start a detonation directly ranges from about 5 kJ for acetylene or hydrogen in air to an estimated 93,000 kJ for methane. These energies are far greater than those of accidental ignition sources.[21]

A deflagration can, however, accelerate and turn into a detonation, especially in long pipes, elongated enclosures or very turbulent conditions.[21] This deflagration-to-detonation transition comes sooner with a more reactive mixture, rough walls or obstacles, a larger diameter, a higher starting pressure or temperature, and more turbulence.[21][20] In tests with petrol vapour and air, a smooth pipe 15 cm wide and 16 m long gave no transition. A rough liner at the pipe entrance, however, produced a detonation after 74 pipe diameters.[20] Because the protection differs so much, Zalosh stresses that the possibility of a detonation has to be assessed, although there are no exact criteria for doing so.[21]

Explosion scenarios in fire safety

A dust explosion occurs in a cloud of combustible powder suspended in air. Dusts pose this hazard when their typical particle size is below about 100 to 400 micrometres, depending on the material.[22] The concentration must also lie between the lower and upper explosive limits, and for dusts finer than about 100 micrometres the lower limit is typically 30 to 60 g/m³.[22] A particular danger is the secondary explosion, in which a first explosion bursts a piece of equipment and its blast lifts dust that has settled on walls and floors.[23] Flame escaping from the equipment often ignites this new cloud, which in an occupied building can cause many casualties and structural damage.[23]

A vapour cloud explosion occurs when a very large cloud of flammable gas or vapour is ignited in a congested area, such as a chemical plant crowded with process equipment and storage vessels.[24] In risk studies of process plants, it usually involves more than about a tonne of flammable vapour, which forms a large cloud and explodes after delayed ignition.[25] The strength of its blast depends on the flame speed that develops and on how strongly the cloud is confined.[24] In a flash fire the flame is slower: it spreads quickly through the flammable part of the cloud but causes no blast damage.[25]

A BLEVE, short for boiling liquid expanding vapour explosion, occurs when a pressurised vessel of liquefied gas ruptures and the liquid suddenly expands into gas, producing blast waves.[26] It typically happens when the metal of a container has been overheated above about 538 °C. Usually only the wall around the vapour space is at risk, because the liquid absorbs heat from the wetted wall.[26] Barry describes it as a delayed event that generally follows 8 to 15 minutes of exposure to a fire.[25] The bursting vessel throws fragments and sends out a blast wave, and most BLEVEs also produce a rising fireball.[11] Fireballs in accidents with liquefied petroleum gas have usually been of the order of 100 m across and have lasted a few seconds.[27]

Heated vessel fails, releasing expanding vapour, fragments, blast and a fireball.
A BLEVE combines physical expansion with several possible hazards. Based on Iqbal and Salley (2004), p. 5-15 and Zalosh (2002), p. 3-419.

Fires inside buildings can produce explosions of their own. When a fire is short of air, unburned pyrolysis gases can collect and form, once mixed with air, a flammable mixture that needs only ignition energy.[28] A smoke explosion is the very rapid burning of such gases from decomposed or incompletely burned fuel.[29] It typically requires the gases and air to be reasonably premixed within the explosion limits, and it can be ignited by a spark, by glowing particles or by autoignition.[29] A backdraft is different: hot fire gases that lack only oxygen ignite explosively when fresh air suddenly enters, for example when a door is opened.[30] NUREG-1805 adds that a backdraft behaves like a premixed fuel-air deflagration, and that smoke explosions have been observed in smouldering polyurethane foam fires.[12]

Electric vehicles add a newer example. Battery cells in thermal runaway release flammable gases. If these do not ignite at once, they can build up into an explosive mixture in a confined or poorly ventilated space.[31] When the mixture reaches its lower explosion limit and is ignited, a fire gas explosion can occur.[31] Fires in electric vehicles can also involve repeated minor explosions, flying parts and flash fires.[32]

Effects on people and buildings

An explosion sends out a blast wave, a thin shell of strongly compressed air and hot gases that spreads in all directions. It can travel faster than sound, which is why a large explosion is heard as a loud bang.[33] Its strength is expressed as the peak overpressure, the highest pressure in the wave above normal atmospheric pressure.[33] This overpressure is highest near the source and falls quickly with distance, and nearby reflecting surfaces strongly influence the damage.[33] An explosion can also cause heat damage, damage from flying objects, ground shock, craters and injuries, although not every explosion does all of these.[34]

Buildings fail at low overpressures. Typical failure pressures are 2 to 3 kPa for room doors, 2 to 5 kPa for light partition walls and 7 to 15 kPa for unrestrained brick walls.[4] A gas explosion inside a building, at up to about 800 kPa, can therefore exceed the strength of such elements many times over.[5] For process plants, Barry lists about 3.4 to 6.9 kPa (0.5 to 1 psi) as causing glass damage and injuries from flying glass. He gives 17 to 34 kPa (2.5 to 5 psi) for heavy damage to buildings and process equipment.[35]

People can withstand fairly high blast pressures, and when blast waves kill people it is usually through falling objects rather than through the pressure itself.[5] Data quoted in NUREG-1805 give about 35 kPa as the threshold for eardrum rupture and about 70 kPa for lung damage.[36] The blast can also hurl fragments and debris, and it can throw people against the ground, walls or other fixed objects.[37] NUREG-1805 notes that the consequences for people depend strongly on whether anything nearby can strike them.[5]

A common way to estimate blast effects is to express an explosion as an equivalent mass of TNT (trinitrotoluene), whose blast effects have been studied in detail.[38] Such methods treat the explosion as an ideal blast released at a single point.[11][39] They cannot be valid inside or close to a flammable vapour cloud.[39] For vapour cloud explosions the real blast also weakens far less quickly with distance than an ideal one.[24] Current methods instead estimate the flame speed for the type of application, or use computer models that calculate how the flame spreads and accelerates.[24]

Prevention and protection

Explosion protection interrupts the chain of events that leads to an explosion. It can prevent a flammable fuel-air mixture from forming, for example by ventilation or inerting. It can remove ignition sources, for example by earthing equipment so that static electricity cannot discharge as a spark. It can also limit the pressure rise by explosion venting or explosion suppression.[1] Deflagration pressure containment instead designs a vessel and its fittings to withstand the maximum pressure of an internal deflagration.[8] In the United States, Factory Mutual Data Sheet 7-32 (1993 edition) recommends at least one of four protections for process equipment that could explode: venting, enough strength to resist the overpressure, inerting or suppression.[40]

Inerting adds a gas such as nitrogen or carbon dioxide to reduce the oxygen concentration to a safe level, the highest of which is called the limiting oxidant concentration.[1][22] For the flammable gases and vapours listed by Zalosh, apart from hydrogen, this limit is 9.5 to 12 per cent oxygen by volume with nitrogen. With carbon dioxide it is 11.5 to 14.5 per cent, while hydrogen needs about 5 per cent with either gas.[22] In the United States, the National Fire Protection Association standard NFPA 69 (1997 edition) states that oxygen should be kept at least 2 percentage points below this limit under continuous monitoring. Without such monitoring a larger margin is needed.[22] Uneven mixtures and the accuracy and response time of oxygen monitors also have to be considered.[22]

Explosion venting releases the combustion gases through openings during a deflagration, so that the pressure stays below the level that would damage the enclosure.[41] The openings are usually covered by blow-off panels, rupture discs or other vent devices, which usually open only after ignition. Venting works best when the vents open early, are as large as possible and let the gases escape freely. In the outside walls and roofs of buildings, the lowest practical opening pressure is usually just above the highest expected wind pressure, typically 0.96 to 1.44 kPa.[41] Explosion suppression instead detects an explosion in its first moments and discharges an extinguishing agent that quenches the flame.[42] It releases no flame or fuel, so it suits indoor equipment and toxic materials, but it is expensive to install and to reset.[42]

Neither venting nor suppression can protect against a detonation. Vents open only after the explosion has started, while a detonation reaches its maximum pressure the instant its shock front arrives.[41][7] Venting guidelines are also limited because they are based on tests and do not account for the turbulence and equipment obstructions of a particular application.[43] Most of them also ignore how much of the enclosure is filled with the flammable mixture.[43]

Prevention, venting and containment are mapped to deflagration; venting is excluded for detonation.
Protection must match the explosion mechanism. Based on Zalosh (2002), pp. 3-411–3-416.

Assessment and uncertainty

An explosion hazard evaluation usually has three steps. It recognises where a flammable fuel-air mixture could form, identifies ignition sources that could then be present, and works out the resulting pressures and possible structural damage.[1] In the chemical process industries, fire and explosion risk can also be assessed quantitatively, as a function of how often an event occurs and how severe its consequences are.[44] Such studies help to compare design options and to map explosion overpressure zones for emergency planning.[44] The amount of fuel in an accident depends on the release rate and duration, and so on how quickly a leak is detected and isolated.[45][25]

Evaluations have to allow for uneven or layered mixtures of gas and air.[22] Zalosh gives the example of about 200 g of butane leaking from an aerosol can into a filling room of 4 by 4 by 3 m.[6] Spread evenly through the 48 m³ room, this gas would be too little to form a flammable mixture. If it mixes in the right proportion with only part of the air and is ignited, the pressure in the unvented room still rises by about 0.46 bar.[6] That is more than rooms in industrial buildings can withstand, so some form of explosion venting and suppression would be needed.[6]

Several limits make explosion predictions uncertain, and NUREG-1805 notes that the pressures produced are usually difficult to predict.[5] Dust explosions are harder to model than gas explosions.[41] Detailed computer models of gas explosions are too complex for routine vent design, but some are used for special cases such as congested offshore platform modules.[41] Laboratory results depend on the ignition source and on the turbulence in the test vessel at the moment of ignition.[46] A standard method for dust tests was therefore developed so that different laboratories obtain reproducible results.[23] Barry also reports that human factors, such as management controls and operator errors, account for about 70 to 85 per cent of the primary causes of fires and explosions in the process industries.[47]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 Zalosh (2002), p. 3-402.
  2. ↑ 2.0 2.1 2.2 2.3 Iqbal and Salley (2004), p. 15-1.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 Iqbal and Salley (2004), p. 15-2.
  4. ↑ 4.0 4.1 Harris (1983), as cited in Iqbal and Salley (2004), p. 15-13.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 5.5 Iqbal and Salley (2004), p. 15-12.
  6. ↑ 6.0 6.1 6.2 6.3 6.4 6.5 Zalosh (2002), p. 3-410.
  7. ↑ 7.0 7.1 7.2 7.3 Zalosh (2002), p. 3-411.
  8. ↑ 8.0 8.1 8.2 Iqbal and Salley (2004), p. F-20.
  9. ↑ Iqbal and Salley (2004), p. 15-16.
  10. ↑ 10.0 10.1 10.2 10.3 10.4 10.5 Iqbal and Salley (2004), p. 15-3.
  11. ↑ 11.0 11.1 11.2 Zalosh (2002), p. 3-419.
  12. ↑ 12.0 12.1 12.2 Iqbal and Salley (2004), p. 15-4.
  13. ↑ 13.0 13.1 13.2 13.3 13.4 13.5 Simmons (2002), p. 1-147.
  14. ↑ Zalosh (2002), p. 3-407.
  15. ↑ Zalosh (2002), pp. 3-406–3-407.
  16. ↑ 16.0 16.1 Zalosh (2002), pp. 3-408–3-409.
  17. ↑ 17.0 17.1 Iqbal and Salley (2004), p. 15-5.
  18. ↑ Iqbal and Salley (2004), p. 15-6.
  19. ↑ Iqbal and Salley (2004), p. F-21.
  20. ↑ 20.0 20.1 20.2 Zalosh (2002), p. 3-413.
  21. ↑ 21.0 21.1 21.2 21.3 Zalosh (2002), p. 3-412.
  22. ↑ 22.0 22.1 22.2 22.3 22.4 22.5 22.6 Zalosh (2002), p. 3-403.
  23. ↑ 23.0 23.1 23.2 Zalosh (2002), pp. 3-409–3-410.
  24. ↑ 24.0 24.1 24.2 24.3 Zalosh (2002), p. 3-420.
  25. ↑ 25.0 25.1 25.2 25.3 Barry (2002), p. 5-184.
  26. ↑ 26.0 26.1 Iqbal and Salley (2004), p. 5-15.
  27. ↑ Beyler (2002a), p. 3-303.
  28. ↑ Van Rossum (2027), §4.8.
  29. ↑ 29.0 29.1 Van Rossum (2027), §4.10.
  30. ↑ Van Rossum (2027), §4.9.
  31. ↑ 31.0 31.1 Van der Graaf et al. (2021/2023), p. 27.
  32. ↑ Van der Graaf et al. (2021/2023), p. 29.
  33. ↑ 33.0 33.1 33.2 Iqbal and Salley (2004), p. 15-7.
  34. ↑ Iqbal and Salley (2004), p. 15-8.
  35. ↑ Barry (2002), p. 5-186.
  36. ↑ Fischer et al. (1995), as cited in Iqbal and Salley (2004), p. 15-12.
  37. ↑ Iqbal and Salley (2004), p. 15-9.
  38. ↑ Iqbal and Salley (2004), p. 15-10.
  39. ↑ 39.0 39.1 Iqbal and Salley (2004), p. 15-15.
  40. ↑ Iqbal and Salley (2004), pp. B-27–B-28.
  41. ↑ 41.0 41.1 41.2 41.3 41.4 Zalosh (2002), p. 3-414.
  42. ↑ 42.0 42.1 Zalosh (2002), p. 3-417.
  43. ↑ 43.0 43.1 Zalosh (2002), p. 3-416.
  44. ↑ 44.0 44.1 Barry (2002), p. 5-176.
  45. ↑ Barry (2002), p. 5-182.
  46. ↑ Zalosh (2002), p. 3-404.
  47. ↑ Barry (2002), p. 5-188.

References

  • 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.
  • 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.
  • 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 der Graaf, J., Hessels, T., & Rosmuller, N. (2023). Fire safety of indoor car parks accommodating electrically powered vehicles. Nederlands Instituut Publieke Veiligheid. (Original work published 2021)
  • Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.
  • Zalosh, R. (2002). Explosion protection. 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. 3-402–3-421). National Fire Protection Association; Society of Fire Protection Engineers.
Retrieved from "https://fire-safety.academy/w/index.php?title=Explosion&oldid=2853"