Skip to content
Upcoming courses

Incomplete combustion

Tools
From Fire Safety Academy Wiki
Contents
  1. 1Complete and incomplete combustion
  2. 2Causes
  3. 3Ventilation and the equivalence ratio
  4. 4Products and yields
  5. 5Effect of temperature
  6. 6Consequences in fire safety engineering
  7. 7See also
  8. 8Notes
  9. 9References

Incomplete combustion is burning in which part of the fuel is not fully oxidised to carbon dioxide and water, so that products such as carbon monoxide, soot, hydrogen and unburnt hydrocarbons are released.[1][2] Complete combustion is never reached in natural fires. Even with unrestricted ventilation, the smoke contains partly oxidised products, which shows that not all of the available energy has been released.[1] The degree of incompleteness rises sharply when a fire runs short of oxygen.[3]

Incomplete combustion is central to the hazards of fire. Carbon monoxide is the most common fire toxicant, and over half of all fire deaths have been attributed to its inhalation.[3] Soot makes smoke dark and blocks vision.[4] Unburnt gases in the smoke can ignite when they mix with air, which can cause a smoke gas explosion or flames outside the room of fire origin.[5][6] Engineers describe the degree of incompleteness with the combustion efficiency and with species yields. They link these to the ventilation of the fire through the equivalence ratio.[7][2]

Complete and incomplete combustion

Combustion is a set of mainly exothermic chemical reactions between a fuel and an oxidiser, usually oxygen, in which heat and light are produced.[8] In complete combustion, the fuel is burnt entirely to carbon dioxide and water.[2] For methane, one molecule reacts with exactly two molecules of oxygen to form one molecule of carbon dioxide and two of water.[9] The heat of combustion of a fuel is defined for this case, in which the fuel is oxidised completely to stable end products.[10]

In incomplete combustion, some carbon leaves the fire as carbon monoxide, as solid carbon in soot, or as unburnt or partly burnt hydrocarbons.[11] Some hydrogen leaves as hydrogen gas instead of water.[2] Drysdale lists carbon monoxide, aldehydes, ketones and particulate matter in the form of soot or smoke as typical partly oxidised products.[1] Introduction to Fire Safety Engineering describes the smoke of a fire as a mixture of decomposition gases from the fuel, combustion products and entrained air. Black smoke usually contains many soot particles.[12]

Because some fuel is not fully oxidised, incomplete combustion also releases less heat per kilogram of fuel. The ratio between the heat actually released and the heat of complete combustion is called the combustion efficiency.[13] The heat actually released per kilogram is the chemical heat of combustion, often called the effective heat of combustion.[7][14] Combustion efficiencies range from about 0.3 to 0.4 for heavily fire-retarded materials to 0.9 or more for oxygen-containing products.[1]

Tewarson illustrates the effect for polymethylmethacrylate (PMMA). In complete combustion it releases 24.9 kJ/g. When more and more carbon monoxide, solid carbon and ethylene are formed instead of carbon dioxide and water, this falls to 22.1, 18.2 and 11.5 kJ/g.[7] The corresponding combustion efficiencies are 0.89, 0.73 and 0.46. This behaviour is typical of ventilation-controlled fires, and below an efficiency of about 0.40 the flame becomes unstable and goes out.[7]

Causes

The main cause is a shortage of oxygen. Introduction to Fire Safety Engineering distinguishes fuel-controlled fires, in which enough oxygen is present, from ventilation-controlled fires, in which the supply of oxygen governs the burning.[5] A fire in a room starts as a fuel-controlled fire, because it is small compared with the room. A fire in the open air is always fuel-controlled.[5]

A ventilation-controlled fire arises when a large fire burns in a compartment with only limited openings. There may then be too little oxygen for complete combustion, and unburnt pyrolysis gases enter the smoke.[5] A fully developed fire is usually ventilation-controlled.[5]

Two room sections: left, small fire below a thin smoke layer; right, large fire, deep smoke layer, flames burning outside the door opening.
Over-ventilated fire (left) with near-complete combustion, and under-ventilated fire (right) with fuel-rich smoke that burns where it meets fresh air outside the room. Based on Gottuk and Lattimer (2002), p. 2-55, and Van Rossum (2027), §4.8.

Low temperatures are a second cause. The oxidation of a fuel first produces carbon monoxide, which then reacts further to carbon dioxide.[15] In a diffusion flame, the final burnout of carbon monoxide on the air side effectively stops when the temperature falls below about 1,250 K.[16] In the smoke layer of a room, temperatures between 800 and 900 K mark the transition from a very slow to a fast conversion. Below about 800 K, carbon monoxide hardly reacts further, so it accumulates.[15] The same quenching affects soot: a flame releases soot when the soot in its burnout zone cools below about 1,300 K, through radiation losses and mixing with cold air.[17]

The mode of burning also matters. Gottuk and Lattimer divide fires into smouldering fires, free-burning fires with an excess of air, and ventilation-limited flaming fires.[3] Smouldering is slow combustion at low gas temperatures without flames, and it can produce high levels of carbon monoxide. Free-burning fires are generally of little concern for toxic species.[3] Fire suppression has a similar effect. A fire that is held back by sprinklers burns less completely and produces more soot per kilogram of fuel, although it burns less fuel in total.[4]

Finally, the chemistry of the fuel plays a role. Soot forms in the fuel-rich regions of every flame and is normally partly burnt out again before it leaves the flame.[17] Aromatic fuels such as toluene and polystyrene form much more soot than other fuels.[18][17] Fuels that contain oxygen in their molecules, such as wood, can form carbon monoxide even without oxygen from the air. This makes them especially productive of carbon monoxide in a hot, oxygen-depleted smoke layer.[15]

Ventilation and the equivalence ratio

The overall ventilation of a fire is expressed with the equivalence ratio. It compares the actual ratio of fuel to air with the ratio needed for exactly complete combustion.[2] For a fire in a single compartment, it is calculated from the mass loss rate of the fuel and the air flow into the compartment:[2]

ϕ=m˙f/m˙ar

Symbols in the equivalence ratio
Symbol Description Unit
ϕ equivalence ratio –
m˙f mass loss rate of the fuel kg/s
m˙a mass flow rate of air into the fire or compartment kg/s
r stoichiometric fuel-to-air mass ratio –

A value below one describes an over-ventilated, fuel-lean fire, and a value above one an under-ventilated, fuel-rich fire.[2] At a value of one, an ideal process would burn all fuel to carbon dioxide and water with no oxygen left over. Above one there is not enough oxygen, so the products also contain excess fuel, carbon monoxide and hydrogen.[2] Gottuk and Lattimer warn that some tables give the air-to-fuel ratio instead of the fuel-to-air ratio, so tabulated values must be checked.[2]

The value depends on the chosen control volume, such as the fire plume, the hot upper layer or the whole compartment. A unique global equivalence ratio does not exist.[19] The upper layer collects gases over time, so its equivalence ratio can lag behind that of the plume. In their example, the plume of a growing fire changes from 0.5 to 1.5 within about 5 seconds, while the layer gases need about 20 seconds to be replaced. The layer then still has a value between 0.5 and 1.5.[19]

Gottuk and Lattimer give a worked example for upholstered furniture, taken as flexible polyurethane foam. Its stoichiometric fuel-to-air ratio is 0.1139, so each kilogram of foam needs 8.78 kg of air.[2] In the example, the foam burns at 9 g/s while the plume entrains 56 g/s of air.[20] The equivalence ratio is then (9/56)/0.1139 = 1.41, so the fire is under-ventilated.

The measured smoke layer composition confirms the incomplete burning. The carbon monoxide yield is 0.23 g per gram of foam, 16 % of the maximum possible. The carbon dioxide yield is only 63 % of the value for complete combustion.[20] The oxygen depletion is 69 % of the stoichiometric requirement, close to the 71 % that the ideal model, one divided by the equivalence ratio, would predict.[2][20] Almost all oxygen that reached the fire was therefore used, but it was not enough to burn the fuel completely.

Products and yields

The amounts of the products are expressed as yields, the mass of a species produced per mass of fuel burned. Normalised yields compare this with the theoretical maximum and range from 0 to 1. Under complete combustion, the normalised yields of carbon dioxide and water and the normalised oxygen depletion are all one.[11] As a fire burns less efficiently, these values decrease. The carbon balance requires that the carbon in carbon monoxide, carbon dioxide, hydrocarbons and residual carbon such as soot adds up to the carbon in the fuel.[11]

Hood experiments by Beyler showed that species yields correlate well with the equivalence ratio. Below an equivalence ratio of 0.6, little carbon monoxide was produced. Above 0.6, the yield rose and for most fuels levelled out above about 1.2.[18] The carbon monoxide was formed mainly at the expense of carbon dioxide, while oxygen and water yields stayed close to the ideal values.[18] Typical peak carbon monoxide concentrations in hood and compartment tests were 2 to 4 % by volume.[15] With burning wood linings in the hot upper layer, average concentrations reached about 10 % and peaks 14 %.[15][21]

Line chart of carbon monoxide yield against equivalence ratio: a step from 0 to 0.2 at 1, and a ramp from 0 at 0.5 to 0.2 at 1.5.
Simple engineering correlations for the carbon monoxide yield as a function of the equivalence ratio: a step at φ = 1 and a linear transition between φ = 0.5 and 1.5. Based on Gottuk and Lattimer (2002), p. 2-77.

Soot follows a similar pattern. Tewarson gives calculated yields of carbon monoxide and smoke for several materials at increasing equivalence ratios. For all three materials in the table below, both yields grow as the fire becomes more fuel-rich.[17]

Two line charts: carbon monoxide yield and smoke yield of polystyrene, polyethylene and wood rising from well-ventilated to equivalence ratio 3.
Calculated yields of carbon monoxide (left) and smoke (right) for polystyrene, polyethylene and wood at increasing equivalence ratio. Based on Tewarson (2002), p. 3-131.
Calculated yields of carbon monoxide (CO) and smoke in g/g at different equivalence ratios[17]
Material CO, well ventilated Smoke, well ventilated CO, φ = 1 Smoke, φ = 1 CO, φ = 2 Smoke, φ = 2 CO, φ = 3 Smoke, φ = 3
Polystyrene 0.060 0.164 0.070 0.202 0.137 0.331 0.162 0.417
Polyethylene 0.024 0.060 0.043 0.071 0.191 0.098 0.238 0.117
Wood 0.004 0.015 0.018 0.018 0.145 0.028 0.171 0.034

For wood, the carbon monoxide yield rises from 0.004 g/g in a well-ventilated fire to 0.171 g/g at an equivalence ratio of three, more than forty times higher.[17] For well-ventilated, pre-flashover fires, Introduction to Fire Safety Engineering lists soot yields from 0.015 kg/kg for wood to 0.23 kg/kg for polyurethane.[4] The book adds that most compartment fires are ventilation-controlled, so that these values then no longer apply directly.[4]

Effect of temperature

The temperature of the smoke layer decides whether incomplete products formed in the flame burn out afterwards. Below about 800 K the layer is chemically unreactive, and the combustion in the fire plume sets the final carbon monoxide level.[22] At such low temperatures, significant carbon monoxide can be produced even in some over-ventilated fires, with equivalence ratios between 0.5 and 1.[22] Above about 900 K the layer gases are reactive. For over-ventilated fires, nearly all carbon monoxide is then oxidised to carbon dioxide.[22]

This explains a difference between laboratory set-ups. In compartment fires of hexane and PMMA, with layer temperatures above 920 K, negligible carbon monoxide was produced until the fire became under-ventilated. In hood tests and in spruce and polyurethane compartment fires below 850 K, high yields occurred even in over-ventilated fires.[23][15] For under-ventilated fires, higher temperatures do not help. The available oxygen reacts preferentially with hydrocarbons, so carbon monoxide levels can even rise slightly.[22]

Consequences in fire safety engineering

Toxicity is the first concern. Carbon monoxide concentrations as low as 4,000 ppm (0.4 % by volume) can be fatal in less than an hour, and levels of several percent have been measured in full-scale compartment fires.[3] Nearly 75 % of deaths from smoke inhalation occur in places away from the room where the fire started.[3] Introduction to Fire Safety Engineering notes that, far from the fire, carbon monoxide concentrations can be very high while visibility still exceeds 20 m. Soot particles settle and spread less far than the gases.[12]

Incomplete combustion also creates flammable smoke. In a ventilation-controlled fire, unburnt pyrolysis gases can form a flammable mixture with oxygen, which then only needs an ignition source.[5] The book attributes smoke gas explosions to the very rapid burning of such gases, formed by decomposition of fuel or by incomplete combustion.[24] Hot, fuel-rich gases flowing out of a room can burn where they meet fresh air, a process called external burning.[6] In tests, external burning was first seen at a compartment equivalence ratio of 1.6, and it lowered the carbon monoxide levels further downstream.[25]

Design fires reflect these effects. In New Zealand, C/VM2 (second edition, 2025) prescribes a carbon monoxide yield of 0.04 kg/kg and a soot yield of 0.07 kg/kg for pre-flashover design fires.[26] After flashover, or once the fire becomes ventilation-limited, the yields rise to 0.40 kg/kg for carbon monoxide and 0.14 kg/kg for soot.[27] The carbon monoxide yield thus increases tenfold. The Australian FSVM data sheets (2020) likewise note that soot yields increase substantially and the effective heat of combustion falls in ventilation-controlled fires.[28]

Bar chart: carbon monoxide yield 0.04 before and 0.40 kg/kg after flashover; soot yield 0.07 before and 0.14 kg/kg after.
Design yields of carbon monoxide and soot in C/VM2 (second edition, 2025) before flashover and after flashover or ventilation limit. Based on Ministry of Business, Innovation and Employment (2025b), pp. 20 to 21.

For hand calculations, Gottuk and Lattimer give a simple engineering rule. Over-ventilated fires produce no carbon monoxide, and under-ventilated fires produce about 0.2 g of carbon monoxide per gram of fuel. This rule fits best for layer temperatures above 900 K.[29] Their correlations represent a lower bound and may not give the highest levels that a fire can produce.[29]

If external burning occurs and the smoke layer outside the room is shallow, the carbon monoxide can be partly oxidised in the adjacent space. A deep smoke layer outside the room, by contrast, inhibits this oxidation, so the incomplete products are carried on unchanged.[29] Computer models need special care as well. By default, the Fire Dynamics Simulator assumes that fuel and oxygen always react, which may not hold in an under-ventilated compartment.[30]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 Drysdale (2002), p. 1-95.
  2. ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10 Gottuk and Lattimer (2002), p. 2-58.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Gottuk and Lattimer (2002), p. 2-54.
  4. ↑ 4.0 4.1 4.2 4.3 Van Rossum (2027), §6.2.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 5.5 Van Rossum (2027), §4.8.
  6. ↑ 6.0 6.1 Gottuk and Lattimer (2002), p. 2-55.
  7. ↑ 7.0 7.1 7.2 7.3 Tewarson (2002), p. 3-103.
  8. ↑ Van Rossum (2027), §4.1.
  9. ↑ Van Rossum (2027), §3.3.
  10. ↑ Drysdale (2002), p. 1-92.
  11. ↑ 11.0 11.1 11.2 Gottuk and Lattimer (2002), p. 2-56.
  12. ↑ 12.0 12.1 Van Rossum (2027), §6.1.
  13. ↑ Tewarson (2002), p. 3-102.
  14. ↑ Australian Building Codes Board (2020a), Data Sheet B2, p. 19.
  15. ↑ 15.0 15.1 15.2 15.3 15.4 15.5 Gottuk and Lattimer (2002), p. 2-65.
  16. ↑ Simmons (2002), p. 1-152.
  17. ↑ 17.0 17.1 17.2 17.3 17.4 17.5 Tewarson (2002), p. 3-131.
  18. ↑ 18.0 18.1 18.2 Gottuk and Lattimer (2002), p. 2-60.
  19. ↑ 19.0 19.1 Gottuk and Lattimer (2002), p. 2-57.
  20. ↑ 20.0 20.1 20.2 Gottuk and Lattimer (2002), p. 2-59.
  21. ↑ Gottuk and Lattimer (2002), p. 2-66.
  22. ↑ 22.0 22.1 22.2 22.3 Gottuk and Lattimer (2002), p. 2-70.
  23. ↑ Gottuk and Lattimer (2002), p. 2-64.
  24. ↑ Van Rossum (2027), §4.10.
  25. ↑ Gottuk and Lattimer (2002), p. 2-72.
  26. ↑ Ministry of Business, Innovation and Employment (2025b), p. 20.
  27. ↑ Ministry of Business, Innovation and Employment (2025b), p. 21.
  28. ↑ Australian Building Codes Board (2020a), Data Sheet B2, p. 21.
  29. ↑ 29.0 29.1 29.2 Gottuk and Lattimer (2002), p. 2-77.
  30. ↑ McGrattan et al. (2026), p. 62.

References

  • Australian Building Codes Board. (2020a). Data sheets: Fire Safety Verification Method (Handbook annex, Print version 1.1).
  • Drysdale, D. D. (2002). Thermochemistry. 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. 1-90–1-98). 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.
  • 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. (2025b). Verification Method C/VM2: Framework for fire safety design (2nd ed.).
  • 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.
  • Tewarson, A. (2002). Generation of heat and chemical compounds in 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. 3-82–3-161). 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=Incomplete_combustion&oldid=2583"