Skip to content
Upcoming courses

Oxygen depletion

Tools
From Fire Safety Academy Wiki
Revision as of 23:28, 1 November 2026 by Daan (talk | contribs) (New FSE wiki article: Oxygen depletion (written by Claude Opus 5.5))
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Contents
  1. 1Oxygen in air and in a fire
  2. 2Effect on the fire
  3. 3Effects on people
  4. 4Time to incapacitation
  5. 5Use in tenability analysis
  6. 6Lowering oxygen on purpose
  7. 7Modelling oxygen depletion
  8. 8Limits and common misconceptions
  9. 9See also
  10. 10Notes
  11. 11References

Oxygen depletion is the fall of the oxygen concentration in the air of a room or building below its normal value of about 21 per cent by volume. It is caused by a fire that consumes oxygen or by gases that displace it.[1][2] It affects both the fire and the people exposed to it. A fire that runs short of oxygen burns less efficiently, produces more carbon monoxide and smoke, and may die down or go out.[3][4] People who breathe air with too little oxygen suffer hypoxia, a lack of oxygen in the body that can lead to loss of consciousness and death.[5]

Low oxygen is one of the asphyxiant effects in fire toxicity, together with carbon monoxide and hydrogen cyanide.[6] Its effect on people is usually small down to about 15 per cent oxygen and becomes severe at about 10 per cent.[5] The SFPE Handbook gives a simple equation for the time to loss of consciousness at a given oxygen concentration, which can be combined with the doses of other gases in a fractional effective dose calculation.[2][7] The same physics is used on purpose by extinguishing systems that lower the oxygen concentration with inert gas, which raises the question of how long people may stay in such an atmosphere.[8][9]

Oxygen in air and in a fire

Air consists of about 78 per cent nitrogen, 21 per cent oxygen and 1 per cent other gases, all by volume.[1] The SFPE Handbook uses 20.9 per cent as the normal oxygen concentration at sea level.[2] Oxygen is one side of the fire triangle. When the oxygen level falls, a fire stops spreading and may eventually go out, depending on the exact concentration reached.[1]

The amount of oxygen a fire needs follows from stoichiometry, the fixed ratio between fuel and oxygen in a complete reaction. Introduction to Fire Safety Engineering expresses this as a stoichiometric constant. About 1.25 kg of oxygen is needed per kilogram of cellulose fuel such as wood, and about 3.46 kg per kilogram of hydrocarbon fuel such as petrol or diesel.[10] With this constant, an engineer can estimate how much oxygen a fire consumes and when it may smother for lack of oxygen. Some zone models use the constant as the starting point of this calculation.[10]

The oxygen in a closed room runs out quickly. According to the SFPE Handbook, a typical domestic room with the doors and windows shut will not support the complete combustion of more than about 1 kg of fuel. After that, the oxygen concentration has fallen to about 10 per cent and the fire goes out or dies down.[4] In a small bedroom, the decomposition of about 0.5 kg of material was calculated to be enough to lower the oxygen concentration to 10 per cent.[3] In a full-scale test in a furnished lounge, the oxygen concentration started to fall after about 5 minutes and the fire went out by itself at around 9 minutes.[11]

Effect on the fire

In the growth phase, a fire in a room is usually small compared with the room, and enough oxygen is present. Such a fire is fuel-controlled: its development depends on the rate at which the fuel releases combustible gases.[12] When the fire is large and the openings are small, the supply of air becomes the limit and the fire becomes a ventilation-controlled fire. The heat release rate then depends on the available oxygen rather than on the fuel. Unburnt pyrolysis gases can then leave the fire and form a flammable mixture where they meet fresh air.[12]

A fully developed fire is usually ventilation-controlled, and part of its gases burn as flames outside the openings.[12][13] The hot gases of a ventilation-controlled fire typically contain almost no oxygen.[14] Measurements in flashover experiments, quoted in the technical guide of the Fire Dynamics Simulator, show that the oxygen concentration in the upper layer drops to zero once its temperature exceeds about 600 °C.[15]

Oxygen depletion also changes what the fire produces. In closed-room fires, the combustion becomes inefficient once the oxygen concentration falls after a minute or so of burning, and the result is a dense smoke rich in carbon monoxide and other toxic products.[3] In early flaming fires, the ratio of carbon dioxide to carbon monoxide is high at first and falls to about 50 to 100 as the oxygen concentration drops towards 15 per cent.[16] In hot post-flashover fires above 800 °C with low oxygen, the pyrolysis products break down into small molecules with high concentrations of carbon monoxide and hydrogen cyanide.[17] For most hydrocarbons, the oxygen concentration varies with the equivalence ratio, the ratio of fuel to air relative to the stoichiometric ratio, in much the same way.[18]

A flame that burns in air diluted with combustion products goes out before all the oxygen has been used. The classic example is a candle under an inverted jar.[19] This is a form of flame extinction. Experiments by Morehart, Zukoski and Kubota found that flames went out at oxygen concentrations of 12.4 to 14.3 per cent. The lower value applied to a 50 cm pool burner and the higher value to a 9 cm burner.[20]

In diesel pool fire tests in a compartment, oxygen concentrations below 14 per cent could not be reached while the pool was burning. In water mist tests, the average oxygen concentration at extinction was 14.5 per cent for heptane spray fires and 13.5 per cent for pool fires.[20] These values hold for fairly calm conditions; at higher strain rates in the flow, the oxygen concentration at extinction rises.[20]

Oxygen depletion does not always end the hazard. In a closed space, a fire can consume oxygen while flammable gases build up at a high temperature. If a door is then opened and fresh air flows in, the gases can ignite suddenly, which is called a backdraft.[21] The Australian Fire Safety Verification Method data sheets note that a fire without an adequate air supply will put itself out. For design, however, they consider it generally conservative to assume that the air supply does not limit fire growth or the progress to flashover.[22]

Effects on people

The body can compensate for a moderate drop in the oxygen concentration of the inhaled air. When a person is exposed to low oxygen, the blood flow to the brain increases by reflex, and the blood releases its oxygen more efficiently.[5] In tests with monkeys, 15 per cent oxygen caused no harmful effects beyond a slight rise in heart rate, while 10 per cent caused a marked depression of brain function. In humans, too, there is little effect down to 15 per cent oxygen apart from a slightly reduced tolerance for exercise, but at about 10 per cent the effects suddenly become severe.[5] Much of the human data come from studies at high altitude, where the low air pressure has similar effects to a low oxygen concentration at sea level.[2] On this basis the SFPE Handbook describes four phases of hypoxia, each with an equivalent oxygen concentration at sea level:[5]

Phase Oxygen concentration (per cent) Main effects
Indifferent phase 20.9 to 14.4 minor effects on adjustment of the eyes to darkness; first effects on exercise tolerance towards 15 per cent
Compensated phase 14.4 to 11.8 slightly faster breathing and heart rate; some loss of performance in complex tasks, short-term memory and judgement
Manifest hypoxia 11.8 to 9.6 loss of critical judgement, dulled senses, poorer muscle control; behaviour may vary from indifference to euphoria
Critical hypoxia 9.6 to 7.8 rapid loss of judgement, then unconsciousness, followed by arrest of breathing and death

The manifest phase is particularly dangerous in fires. The handbook calls the passage into the critical phase, at about 10 per cent oxygen, the catastrophe point.[5] Up to this point the compensating mechanisms hide most of the effect, so performance hardly declines. Beyond it, the condition of the victim deteriorates very quickly, usually to unconsciousness, and escape is no longer possible.[5] Loss of consciousness occurs when the partial pressure of oxygen in the venous blood of the brain falls below 20 mmHg.[5]

Time to incapacitation

When a person suddenly enters air with less oxygen, it takes some time before the blood reaches equilibrium with the new air. During this time a kind of dose of hypoxia is taken up. Once equilibrium is reached, usually within a few minutes, the effect depends on the oxygen concentration and hardly changes with time.[23] The time to loss of consciousness depends on how far the oxygen concentration falls, on the activity of the person and on the volume of air breathed per minute.[2]

From altitude data for people at rest, the SFPE Handbook derives an equation for the time to loss of consciousness in a fire atmosphere with low oxygen.[2] The time is an exponential function of the oxygen vitiation, which is the drop of the oxygen concentration below the normal 20.9 per cent. Each extra percentage point of vitiation shortens the time by a fixed factor:[2] tIO=exp⁡[8.13−0.54(20.9−%O2)]

Symbol Description Unit
tIO time to incapacitation (loss of consciousness) by low oxygen min
%O2 oxygen concentration in the inhaled air per cent by volume
20.9−%O2 oxygen vitiation, the drop below normal air percentage points

The equation is based on the time of useful consciousness of resting people after a sudden drop in pressure, with data spanning sea-level equivalent oxygen concentrations of about 3.9 to 9.6 per cent.[2] Values calculated for higher oxygen concentrations lie outside the range of these data. The time also depends on the activity of the person, which raises the oxygen demand, and on the volume of air breathed per minute.[2]

For a varying oxygen concentration, the handbook divides the exposure into periods of one minute and adds the fractions 1/tIO of each minute. Loss of consciousness is predicted when the sum reaches 1.[2] In the handbook's example, a person breathes air with 10 per cent oxygen for 5 minutes, followed by 7.8 per cent for 1.5 minutes. At 10 per cent, tIO is exp(8.13 − 0.54 × 10.9), or about 9.43 minutes, so each minute adds 0.106. At 7.8 per cent, tIO is 2.87 minutes and each minute adds 0.348.[2] The sum is 0.106 × 5 + 0.348 × 1.5 = 1.05, so loss of consciousness is predicted at 6.5 minutes.[2]

The equation also shows that oxygen depletion does not follow Haber's rule, which assumes a constant product of concentration and time. At 17 percentage points of vitiation, the dose to incapacitation is about 5.6 per cent·min, but at 11.3 percentage points it is about 87 per cent·min.[2] A short exposure to severe hypoxia therefore incapacitates very quickly, while a long exposure to moderate hypoxia has little effect.[2] Applied to moderate values, the equation gives about 140 minutes at 15 per cent oxygen and about 28 minutes at 12 per cent (calculated with the handbook's equation).[2]

Use in tenability analysis

In an analysis of the time available for escape, low oxygen is one of several effects of smoke on people. The International Fire Engineering Guidelines (2005) note that life safety criteria often consider carbon monoxide, carbon dioxide and hydrogen cyanide, and also low oxygen.[24] In the incapacitation model of the SFPE Handbook, the fraction of an incapacitating dose of low oxygen is added to the fractions for carbon monoxide and hydrogen cyanide.[7] The fractions for these gases are first multiplied by a factor for the faster breathing caused by carbon dioxide. Incapacitation is predicted when the total reaches 1, and a total of 0.1 should allow nearly all exposed people, including sensitive groups, to escape safely.[7]

The gases interact. Carbon monoxide and low oxygen both reduce the oxygen carried by the blood, so some addition of their effects is expected.[25] Carbon dioxide, by contrast, has been shown to increase the resistance to incapacitation by low oxygen. It makes people breathe faster, which increases the uptake of oxygen, and it helps the blood release oxygen to the tissues.[25] In most practical fires, carbon monoxide is the most important asphyxiant, while hydrogen cyanide and low oxygen add to its effect and may shorten the time to incapacitation in some situations.[7]

Regulations take this into account in different ways. In New Zealand, Verification Method C/VM2 (second edition, 2025) requires the fractional effective dose for carbon monoxide to be calculated with ISO 13571 and to include the contributions of carbon monoxide, carbon dioxide and oxygen.[26] The commentary on the first edition explains that other species were left out because the data for them were considered unreliable.[27] Introduction to Fire Safety Engineering notes that people usually spend less than 30 to 60 seconds in smoke during escape, so the dose of toxic gases is often not decisive.[28]

Lowering oxygen on purpose

Some extinguishing systems use oxygen depletion as their working principle. In the Fire Dynamics Tools report of the U.S. Nuclear Regulatory Commission (NUREG-1805), inert gas agents, such as nitrogen and mixtures containing argon, are described as acting mainly by depleting oxygen.[29] According to the SFPE Handbook, an inert gas lowers the flame temperature by reducing the oxygen concentration and by raising the heat capacity of the atmosphere. Adding enough inert gas to bring the oxygen concentration below 12 per cent extinguishes flaming fires.[8]

For inert gases, the main health concern is the reduced oxygen concentration.[30] The SFPE Handbook gives the exposure limits that applied at the time of writing. Up to 43 per cent agent, which leaves 12 per cent oxygen, exposure is limited to 5 minutes. Between 43 and 52 per cent agent, or between 12 and 10 per cent oxygen, the limit is 3 minutes, and above 52 per cent agent it is 30 seconds.[9] Small amounts of carbon dioxide added to some inert gas blends appear to reduce the effects of hypoxia.[9]

Water mist can also deplete oxygen. When droplets turn to steam in a hot enclosure, the steam dilutes the oxygen, while the fire continues to consume it. If the combined effect brings the oxygen concentration below the limiting value, the fire goes out; for most hydrocarbon fuels this value is about 13 per cent.[31] The oxygen available depends on the size of the fire, the volume of the compartment and its ventilation. Enclosure effects increase the benefit of oxygen depletion for a water mist system, because the spray drives the hot, oxygen-poor gases of the upper layer down to the fire.[31]

Deliberate oxygen reduction can affect people outside the protected room. In 1992, at the Duane Arnold nuclear power plant in the United States, a test discharge of carbon dioxide into the cable spreading room leaked into the control room above it. Within a few minutes, oxygen levels of 17 per cent at chest height and 15 per cent at floor level were recorded. Both were below the plant's acceptance criterion of 19.5 per cent, and the operators put on breathing apparatus.[32]

Modelling oxygen depletion

Fire models handle oxygen depletion in different ways. Introduction to Fire Safety Engineering notes that most models deal poorly with a ventilation-controlled fire, so the user often has to specify what happens when oxygen runs short. The options are that all gases still burn in the fire room, that the surplus burns outside as external flames, or that combustion stops.[33] The zone model OZone, for example, offers a choice between ignoring the shortage, external flaming and an extended fire duration.[34] Zone models and CFD models can report the oxygen consumption in the fire room as a result.[35][36]

The zone model CFAST limits burning with a lower oxygen limit. By default, combustion stops when the oxygen mass fraction in the layer containing the fire falls to 0.15, a value that does not depend on temperature.[37] The Fire Dynamics Simulator uses a limiting oxygen volume fraction of 0.135 at an ambient temperature of 20 °C, consistent with the measurements of Morehart and co-workers.[15] In coarse simulations, the model also lets the fire burn at any oxygen level once the gas temperature exceeds a default of 600 °C, in line with the experiments described above.[15]

Limits and common misconceptions

A common misconception is that a fire in a closed room is harmless because it will soon go out. Closed-room fires can in fact produce a rapidly lethal atmosphere, because the low oxygen concentration comes together with dense smoke rich in carbon monoxide.[3] For the people in the room, carbon monoxide usually matters more than the lack of oxygen itself.[7]

A second misconception is that the oxygen concentration alone tells whether a flame can survive. The measured extinction limits range from about 12 to 15 per cent and depend on the size of the fire, the flow conditions and the gas that dilutes the air.[20] CFAST and the Fire Dynamics Simulator also express their limits differently, as a mass fraction and as a volume fraction.[37][15]

The equation for the time to incapacitation is based on healthy people at rest exposed to low air pressure at altitude, not on fire victims.[2] It does not include the effects of heat, irritant gases or carbon monoxide, which must be added separately.[7] The exposure limits for inert gases were set for extinguishing systems and cannot simply be transferred to an oxygen reduction system, for which the sources cited in this article give no data.[9]

See also

Notes

  1. ↑ 1.0 1.1 1.2 Van Rossum (2027), §4.2.
  2. ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10 2.11 2.12 2.13 2.14 2.15 Purser (2002), p. 2-107.
  3. ↑ 3.0 3.1 3.2 3.3 Purser (2002), p. 2-136.
  4. ↑ 4.0 4.1 Purser (2002), p. 2-155.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 5.5 5.6 5.7 Purser (2002), p. 2-106.
  6. ↑ Purser (2002), p. 2-91.
  7. ↑ 7.0 7.1 7.2 7.3 7.4 7.5 Purser (2002), p. 2-110.
  8. ↑ 8.0 8.1 DiNenno (2002), p. 4-174.
  9. ↑ 9.0 9.1 9.2 9.3 DiNenno (2002), p. 4-184.
  10. ↑ 10.0 10.1 Van Rossum (2027), §3.3.
  11. ↑ Purser (2002), p. 2-137.
  12. ↑ 12.0 12.1 12.2 Van Rossum (2027), §4.8.
  13. ↑ Van Rossum (2027), §4.11.
  14. ↑ Iqbal and Salley (2004), p. 2-4.
  15. ↑ 15.0 15.1 15.2 15.3 McGrattan et al. (2026), p. 63.
  16. ↑ Purser (2002), p. 2-135.
  17. ↑ Purser (2002), p. 2-123.
  18. ↑ Gottuk and Lattimer (2002), p. 2-79.
  19. ↑ McGrattan et al. (2026), p. 62.
  20. ↑ 20.0 20.1 20.2 20.3 Beyler (2002b), p. 2-184.
  21. ↑ Van Rossum (2027), §4.9.
  22. ↑ Australian Building Codes Board (2020a), Data Sheet B1, p. 20.
  23. ↑ Purser (2002), p. 2-88.
  24. ↑ National Research Council of Canada et al. (2005), p. 1.5-4.
  25. ↑ 25.0 25.1 Purser (2002), p. 2-109.
  26. ↑ Ministry of Business, Innovation and Employment (2025b), p. 17.
  27. ↑ Ministry of Business, Innovation and Employment (2013), p. 43.
  28. ↑ Van Rossum (2027), §7.7.
  29. ↑ Iqbal and Salley (2004), p. B-41.
  30. ↑ DiNenno (2002), p. 4-182.
  31. ↑ 31.0 31.1 Mawhinney and Back (2002), p. 4-314.
  32. ↑ Iqbal and Salley (2004), p. B-52.
  33. ↑ Van Rossum (2027), §8.4.
  34. ↑ Van Rossum (2027), §9.3 (How a zone model works).
  35. ↑ Van Rossum (2027), §9.3 (Results from zone models).
  36. ↑ Van Rossum (2027), §11.2.
  37. ↑ 37.0 37.1 Peacock et al. (2026), p. 13.

References

  • Australian Building Codes Board. (2020a). Data sheets: Fire Safety Verification Method (Handbook annex, Print version 1.1).
  • Beyler, C. L. (2002b). Flammability limits of premixed and diffusion flames. 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. 2-172–2-187). National Fire Protection Association; Society of Fire Protection Engineers.
  • DiNenno, P. J. (2002). Halon replacement clean agent total flooding systems. 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. 4-173–4-200). National Fire Protection Association; Society of Fire Protection Engineers.
  • Gottuk, D. T., & Lattimer, B. Y. (2002). Effect of combustion conditions on species production. 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. 2-54–2-82). 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.
  • Mawhinney, J. R., & Back, G. G., III. (2002). Water mist fire suppression systems. 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. 4-311–4-337). National Fire Protection Association; Society of Fire Protection Engineers.
  • McGrattan, K., Hostikka, S., Floyd, J., McDermott, R., Vanella, M., Mueller, E., & Paul, C. (2026). Fire Dynamics Simulator technical reference guide: Vol. 1. Mathematical model (6th ed., NIST Special Publication 1018-1). National Institute of Standards and Technology. https://doi.org/10.6028/NIST.SP.1018
  • Ministry of Business, Innovation and Employment. (2013). Commentary for Building Code clauses C1–C6 and Verification Method C/VM2.
  • Ministry of Business, Innovation and Employment. (2025b). Verification Method C/VM2: Framework for fire safety design (2nd ed.).
  • National Research Council of Canada, International Code Council, Department of Building and Housing, & Australian Building Codes Board. (2005). International fire engineering guidelines (2005 ed.). Australian Building Codes Board.
  • Peacock, R. D., McGrattan, K. B., Forney, G. P., & Reneke, P. A. (2026). CFAST – Consolidated Fire and Smoke Transport: Vol. 1. Technical reference guide (Version 7, NIST Technical Note 1889v1). National Institute of Standards and Technology. https://doi.org/10.6028/NIST.TN.1889v1
  • Purser, D. A. (2002). Toxicity assessment of combustion products. 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. 2-83–2-171). 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=Oxygen_depletion&oldid=206"