Skip to content
From Fire Safety Academy Wiki
Revision as of 23:47, 7 November 2026 by Peter (talk | contribs) (FSE wiki: Fact-check: date source-bound uses, qualify textile and fluoropolymer conclusions, remove unsupported cause of value differences (written by GPT-6))
Contents
  1. 1Definition
  2. 2Test method
  3. 3Equations
  4. 4Worked example
  5. 5Typical values
  6. 6Uses
  7. 7Limitations
  8. 8See also
  9. 9Notes
  10. 10References

The oxygen index (OI), also called the limiting oxygen index (LOI), is the minimum concentration of oxygen, in a flowing mixture of oxygen and nitrogen, that just supports candle-like downward flaming combustion of a material.[1] It is measured with ASTM D2863, Test Method for Measuring the Minimum Oxygen Concentration to Support Candle-Like Combustion of Plastics (Oxygen Index), and is expressed as a volume percentage.[1][2] Normal air contains about 20.9 % oxygen. In principle, therefore, a material with an oxygen index above about 25 % should burn only under extreme conditions.[1] Tabulated values for common solids range from about 16 % for a kitchen candle or cotton to 95 % for polytetrafluoroethylene.[3]

Hirschler (2008a) described the test as repeatable and widely quoted in plastics specifications, quality control and research on flame retardants.[1] Rosales and Stoltzfus (2008) also described its use to evaluate materials for an oxygen-enriched atmosphere.[4] However, the 2008 Fire Protection Handbook describes it only as a measure of how easily a burning material can be extinguished. Because of its low heat input and artificial oxygen levels, it does not predict real-scale fire performance and is unsuitable for fire safety engineering calculations.[1]

Definition

Hirschler (2008a) places the oxygen index under ease of extinction, a fire property that matters once a material has ignited. The easier a burning material is to extinguish, the lower the fire hazard of products made from it.[1] For textiles, Stull and Chines (2008) define it as the minimum volume concentration of oxygen in an oxygen–nitrogen mixture that will just support sustained combustion. The material is ignited at its uppermost edge in a vertical position.[2] A fabric with a higher LOI is more likely to stop burning once the ignition source is removed.[2] The NUREG-1805 glossary describes the LOI as a characteristic of solid combustibles that gives a measure of relative flammability.[5]

The term has a second, related meaning in combustion science. For gas and liquid fuels, the oxygen index is the oxygen mole fraction of the oxidant stream at which a diffusion flame can no longer be stabilised.[6] Simmons and Wolfhard measured these limits by diluting the oxidant stream until the flame could not be held, as reported by Beyler (2002b).[6] Beyler stresses that the two kinds of value are not interchangeable. In the candle-type test for solids, extinction can be caused by both gas-phase and solid-phase effects, whereas the values for gases and liquids are governed by gas-phase effects alone.[7]

The oxygen index is linked to the fire triangle. Air contains about 78 % nitrogen, 21 % oxygen and 1 % other gases by volume. When the oxygen level falls, a fire stops spreading at some concentration that depends on the fuel, and may finally go out.[8] The oxygen index fixes that concentration for one material under one set of standard test conditions.

Test method

In the ASTM D2863 test a small specimen is placed vertically inside a glass column and ignited at its top with a small gas flame.[1] A controlled mixture of oxygen and nitrogen flows through the column, and the oxygen concentration is varied until the minimum that just sustains candle-like downward burning is found.[1][5] According to Hirschler (2008a), the method has excellent repeatability and reproducibility and produces numbers over a very broad range of fire performance.[1]

The test is part of the family of fire testing methods for plastics. In the comparison of standards in the 2008 handbook, ISO 4589 is listed as the international method similar to ASTM D2863.[9] ISO 4589 Part 2 is the ambient-temperature test and Part 3 an elevated-temperature version.[10] The values in the handbook's general table of oxygen indices were all determined with the ASTM test on materials without added fire retardants.[3] Greases, sealants and powders have also been tested with a cup method instead of as self-supporting specimens.[11]

The original oxygen index method for gases and liquids works differently. A counterflow diffusion flame forms at the stagnation region of a porous cylinder or sphere through which fuel vapour is fed, while a slow oxidant stream passes over it.[7] This arrangement gives the most favourable aerodynamic conditions for keeping a flame alive. Fuel and oxidant streams that burn in this apparatus may therefore not burn at higher velocities and shear.[7] For metals, flammability limits in the usual sense do not exist for most structural alloys, because they burn mainly in the liquid phase rather than as vapour.[12] Two practical measures are used instead: the threshold pressure, which is the minimum oxygen pressure that supports combustion of a standard sample, and the oxygen index at a given pressure.[12][13]

Equations

The oxygen index is simply the oxygen share of the oxygen–nitrogen mixture at the limiting condition, written as a percentage by volume.[2][1]

OI=100[O2][O2]+[N2]

Combustion science uses the same idea to explain flame extinction. A diffusion flame in a vitiated atmosphere goes out before it has used all the oxygen, as a candle does under an inverted jar.[14] The energy released by burning the available oxygen must heat the gas to a critical flame temperature. Below that temperature, the reactions cannot sustain the flame.[14] The Fire Dynamics Simulator (FDS) bases this critical flame temperature on the oxygen index. The oxygen mole fraction at the limit is first converted to a mass fraction.[14]

YOI=XOIWO2XOIWO2+(1−XOI)WN2

The adiabatic flame temperature at the limit then follows from an energy balance. The heat released per kilogram of oxygen consumed is shared by the gas mixture.[14]

TOI=T∞+YOIΔH/rO2cp

Symbol Description Unit
OI Oxygen index of a solid in the ASTM D2863 test % by volume
[O2], [N2] Volume concentrations of oxygen and nitrogen in the test gas at the limit –
XOI Oxygen mole fraction of the oxidant stream at the limit (gas-phase oxygen index) –
YOI Oxygen mass fraction at the limit –
WO2, WN2 Molar masses of oxygen (0.032 kg/mol) and nitrogen (0.028 kg/mol) kg/mol
TOI Adiabatic flame temperature at the oxygen index K
T∞ Ambient temperature K
ΔH/rO2 Heat released per unit mass of oxygen consumed, about 13,100 kJ/kg for most fuels kJ/kg
cp Average specific heat of the gases kJ/(kg·K)

The molar masses are from Drysdale (2002) and the heat release per kilogram of oxygen from McGrattan et al. (2026).[15][14] Simmons and Wolfhard found that the adiabatic flame temperature at the oxygen index equals that at the stoichiometric limit of a premixed system with the same fuel and diluent.[6] The equations are valid for gas and liquid fuels. Beyler warns that the oxygen index of a solid from the candle test is not directly relevant to gas-phase limits and should not be used to calculate adiabatic flame temperatures at the limit.[7]

Worked example

Beyler (2002b) tabulates an oxygen index of 0.139 (mole fraction) for methane, with an adiabatic flame temperature at that limit of 1,780 K.[6] The oxygen mass fraction at the limit follows from the second equation.[14][15]

YOI=0.139×0.0320.139×0.032+0.861×0.028=0.156

The energy balance needs a specific heat. Taking 1.1 kJ/(kg·K), the typical value Beyler (2002b) uses for hot gas layer estimates, and an ambient temperature of 293 K gives the following estimate.[7][14]

TOI=293+0.156×13,1001.1≈2,150 K

This simple estimate is about 370 K higher than the tabulated 1,780 K. Matching the table would need an average specific heat of about 1.37 kJ/(kg·K). The result is therefore very sensitive to the specific heat, which FDS averages over the composition and temperature range of the combustion products.[14] The same oxygen index also shows how much nitrogen would have to be added to air before even the favourable counterflow flame of the test could no longer be stabilised. Diluting air with about 0.50 m³ of nitrogen per cubic metre lowers its oxygen fraction from 0.209 to 0.139. These values were calculated for this article.

Typical values

The handbook tables give oxygen indices measured with ASTM D2863 on materials without added fire retardants.[3] Many common organic materials, such as cotton, polyethylene and polyurethane foam, have values below 20.9 % and therefore burn in normal air under the test conditions. Many commonly used materials ignite and burn in as little as 16 % oxygen.[3][4]

Oxygen index of selected materials (ASTM D2863, without added fire retardants unless stated)
Material Oxygen index (%) Source
Kitchen candle 16.0 [3]
Cotton 16.0–18.5 (19 in the textile chapter; 19.9 in the SFPE Handbook) [3][16][17]
Polyurethane foam, typical 16.5 [3]
Polymethylmethacrylate 17.3 [3]
Polyethylene, polypropylene 17.4 [3]
Red oak 23.0 [3]
Wool 23.8 (28 in the textile chapter; 25.2 in the SFPE Handbook) [3][16][17]
Nomex fibres 26.7–28.5 (30 in the textile chapter) [3][16]
Polybenzimidazole (PBI) 40.6–41.5 (38–49 in the textile chapter) [3][16]
Polyvinyl chloride 37.1–49 [3]
Polytetrafluoroethylene 95.0 [3]
Flame-retardant cotton (treated) 28–32 [16]

The spread between sources is large. For cotton the handbook gives 16.0 to 18.5 % in one table and 19 % in the chapter on textiles, while the SFPE Handbook gives 19.9 %.[3][2][17] For wool the values are 23.8, 28 and 25.2 % respectively.[3][16][17] The cited tables do not establish why these numbers differ. A single published value should therefore not be treated as an exact material constant.

Composition and additives change the value strongly. Flame-retardant treatments can raise cotton from 19 to 28 or 32 % and wool from 28 to 32 or 34 %, and adding fire retardants to polymers can increase the measured index substantially.[16][3] Fillers do not always help. In the oxygen-enriched atmospheres chapter, nylon 66 is listed at 30.5 % but a glass-filled grade at 23.5 %, while one silicone rubber rose from 29 to 36 % with 50 % glass.[11] The fluorocarbon plastics discussed in the source did not propagate flames even in 100 % oxygen.[4] For gases and liquids, the gas-phase oxygen indices in Beyler's table, expressed as mole fractions of oxygen, include 0.139 for methane, 0.127 for propane, 0.111 for methanol, 0.076 for carbon monoxide and 0.054 for hydrogen.[6]

Uses

Despite its limits, Hirschler (2008a) reported that the test was widely required in specifications for plastic materials and often quoted in data sheets.[1] The chapter described it as a quantitative quality control tool during manufacturing and a semiqualitative indicator of additive effectiveness in research and development. The results then apply mainly to low incident heat fluxes, not exceeding 20–25 kW/m².[1] Benedetti and Colonna (2008) wrote that small-scale tests such as the oxygen index test and the UL 94 flammability test were then used mostly for product development and production control.[18]

In textile flammability, the LOI has been used to compare fibres. Stull and Chines (2008) expected fabrics with high ignition temperatures and high LOIs to resist ignition and continued burning under the small-flame conditions discussed in their chapter.[2] Stull and Chines (2008) reported that LOI testing for textiles had abated by the time of their chapter because full-scale tests of the items were considered more reliable.[2]

Rosales and Stoltzfus (2008) describe the oxygen index test as a way to evaluate the ignition and propagation properties of materials for oxygen-enriched atmospheres.[4] Rosales and Stoltzfus (2008) define such an atmosphere as one in which the oxygen concentration exceeds 21 % by volume, the oxygen partial pressure exceeds 21.3 kPa (160 torr), or both.[19] Raising the oxygen partial pressure can move a material from the nonflammable to the flammable category.[20] The resistance of flame-retardant textiles in such atmospheres can likewise be inferred from their LOIs.[21]

The oxygen index has also served as a link between polymer chemistry and flammability. Van Krevelen showed that for many polymers the LOI is linearly related to the char yield measured by thermogravimetric analysis, as reported by Beyler and Hirschler (2002).[22] Since char yield can be estimated from molecular structure, the LOI of pure polymers with a substantial char yield is fairly computable.[22] Tewarson's data show the pattern: polybenzobisoxazole leaves 75 % char and has an LOI of 56 %, whereas polyethylene leaves no char and has 18 %. Halogenated polymers break the pattern: polyvinyl chloride leaves 11 % char but has an LOI of 50 %, and polytetrafluoroethylene leaves none but has 95 %.[23] In the 2026 FDS technical guide, the gas-phase oxygen index sets the critical flame temperature for local extinction.[14]

Limitations

The 2008 handbook is clear that the test is inappropriate as a predictor of real-scale fire performance, mainly because of its low heat input and the artificial high oxygen environments it uses.[1] For the same reason it is unsuitable for fire safety engineering calculations.[1] Tests with a small, low-intensity ignition source cannot evaluate the fire behaviour of materials under room or other full-scale fire conditions.[18] The claim that a material with an index above 25 % should burn only under extreme conditions holds in principle only. In reality the test measures how easy it is to extinguish a material that is already burning.[1]

The index says little about how much heat a material releases once a real fire has developed. In the data compiled by Beyler and Hirschler (2002), nylon-6 has a much higher LOI than polyacetal (25.6 against 15.7 %). In the cone calorimeter at 40 kW/m², however, its peak heat release rate was more than three times higher (1,300 against 360 kW/m²).[17] Polypropylene, with an LOI of 17.4 %, reached 1,500 kW/m² under the same conditions.[17] Beyler and Hirschler also report that attempts to relate thermal decomposition data to the LOI have fallen into disrepute, particularly because of the lack of confidence remaining in the LOI technique.[17]

The candle-type value for a solid also cannot be used as a gas-phase flammability limit, because solid-phase effects contribute to extinction in the test.[7] Even the gas-phase values describe the most favourable aerodynamic conditions, so a flame that survives in the apparatus may not survive at higher flow velocities.[7] Finally, the large scatter of published values for the same material means that comparisons are meaningful only between results from the same method and specimen type.[3][16][17]

See also

Notes

  1. ↑ 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.10 1.11 1.12 1.13 1.14 Hirschler (2008a), p. 6-42.
  2. ↑ 2.0 2.1 2.2 2.3 2.4 2.5 2.6 Stull and Chines (2008), p. 6-84.
  3. ↑ 3.00 3.01 3.02 3.03 3.04 3.05 3.06 3.07 3.08 3.09 3.10 3.11 3.12 3.13 3.14 3.15 3.16 3.17 3.18 Babrauskas (2008a), p. 6-282.
  4. ↑ 4.0 4.1 4.2 4.3 Rosales and Stoltzfus (2008), p. 9-233.
  5. ↑ 5.0 5.1 Iqbal and Salley (2004), p. F-41.
  6. ↑ 6.0 6.1 6.2 6.3 6.4 Beyler (2002b), p. 2-183.
  7. ↑ 7.0 7.1 7.2 7.3 7.4 7.5 7.6 Beyler (2002b), p. 2-186.
  8. ↑ Van Rossum (2027), §4.2.
  9. ↑ Hirschler (2008a), p. 6-38.
  10. ↑ Hirschler (2008a), p. 6-60.
  11. ↑ 11.0 11.1 Rosales and Stoltzfus (2008), p. 9-234.
  12. ↑ 12.0 12.1 Rosales and Stoltzfus (2008), p. 9-236.
  13. ↑ Rosales and Stoltzfus (2008), p. 9-237.
  14. ↑ 14.0 14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 McGrattan et al. (2026), p. 62.
  15. ↑ 15.0 15.1 Drysdale (2002), p. 1-91.
  16. ↑ 16.0 16.1 16.2 16.3 16.4 16.5 16.6 16.7 Stull and Chines (2008), p. 6-85.
  17. ↑ 17.0 17.1 17.2 17.3 17.4 17.5 17.6 17.7 Beyler and Hirschler (2002), p. 1-124.
  18. ↑ 18.0 18.1 Benedetti and Colonna (2008), p. 6-227.
  19. ↑ Rosales and Stoltzfus (2008), p. 9-225.
  20. ↑ Rosales and Stoltzfus (2008), p. 9-230.
  21. ↑ Stull and Chines (2008), p. 6-99.
  22. ↑ 22.0 22.1 Beyler and Hirschler (2002), p. 1-123.
  23. ↑ Tewarson (2002), p. 3-96.

References

  • Babrauskas, V. (2008a). Tables and charts. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-269–6-294). National Fire Protection Association.
  • Benedetti, R. P., & Colonna, G. R. (2008). Polymeric materials. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-209–6-231). National Fire Protection Association.
  • 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.
  • Beyler, C. L., & Hirschler, M. M. (2002). Thermal decomposition of polymers. 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. 1-110–1-131). National Fire Protection Association; Society of Fire Protection Engineers.
  • 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.
  • Hirschler, M. M. (2008a). Concepts and protocols of fire testing. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-35–6-60). National Fire Protection Association.
  • 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.
  • 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
  • Rosales, K., & Stoltzfus, J. M. (2008). Oxygen-Enriched atmospheres. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 9-225–9-243). National Fire Protection Association.
  • Stull, J. O., & Chines, S. A. (2008). Fibers and textiles. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-75–6-102). National Fire Protection Association.
  • 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.