Fire safety can be achieved by means of active or passive fire protection. Passive fire protection involves using materials or products with superior fire performance so as to either minimize the probability of ignition or, if ignition does occur, minimize the damaging effects of the resulting fire. Flame retardants offer one way of providing passive fire protection.

When a combustible material (often a polymer) is used in applications where fire safety is important, the lack of intrinsic fire safety must be addressed to provide passive fire protection. There are four possible approaches, the first two involving flame retardants.

  • Adding flame retardants (i.e., using additive flame retardants)
  • Creating new materials with better fire performance though syntheses of variations of the material (i.e., using reactive flame retardants)
  • Blending or otherwise compounding the material with other materials with better fire performance (i.e., creating blends or mixtures)
  • Encapsulating the material or separating it from potential exposure to the heat insult (e.g., using barriers).

Typical applications where fire safety is critical include consumer products (such as upholstered furniture or mattresses), electrical and electronics (such as wire and cable, circuit boards, computer or appliance housings), and building products (such as interior finish, insulation or roofing materials).

It is always possible to choose non-combustible materials for any or all of these applications, but such a choice would typically limit either the esthetics or the comfort of the product in question. For example, combustible foams are typically used as paddings for upholstered furniture because they provide comfort and resilience. Foams could be replaced by steel or concrete, but that would provide much lower comfort. Interior finish (such as wall linings) can be wood or decorative wall coverings, or it can be gypsum board or concrete, but the latter option involves a clear loss of appeal.

Flame retardants are materials intended for incorporation into combustible materials to improve their fire performance or to meet fire test requirements. Many studies have shown that flame retardants can improve ignitability and/or reduce flame spread. Clearly, fire will not occur if no ignition happens: thus, a delay in ignition will improve fire safety. However, fire hazard assumes that ignition has occurred and studies have shown that flame retardants also have a beneficial effect on heat release, which is the most important fire property.


Flame retardants are not a new invention: since the 18th Century, it has been found that the addition of some specific chemicals would help improve one or more fire properties. The earliest published example is a 1735 Wyld patent1 to improve the flammability of textiles and of paper by using "a mixture of alum, vitriol and borax.” In the same century, flame retardant coatings were used in "lighter than air balloons” launched by the Montgolfier brothers in France in 1783.2,3 Much earlier, the ancient Egyptians and Greeks used solutions of "alum” (an aluminum-based salt) as flame retardant coatings for wood construction during times of war.

The seven chemical elements that, when contained within materials incorporated into polymers as additive or reactive flame retardants, have the greatest effect in improving fire performance are: bromine, chlorine, phosphorus, nitrogen, aluminum, antimony, and boron.3 Flame retardant chemicals (more commonly known as flame retardants) often include one or more of these chemical elements, and their effectiveness is usually a direct function of the fraction of the active element present in the flame retardant.

Synergy (i.e., improvement in effectiveness beyond what would have been expected from the added effects of the individual components) often occurs when combining additives containing more than one of these chemical elements. For example, materials containing antimony as the only one of the above elements are almost invariably of little utility in terms of flame retardance unless used in combination with ones containing chlorine or bromine, or both.

Brominated flame retardants are probably the most efficient ones: a small proportion of a brominated additive or of a reaction that partially brominates the substrate polymer is often enough to cause a very significant effect on fire performance. The most common mechanism of action of brominated flame retardants is thermal breakdown to form free radicals that react, in the gas phase, and inhibit the chain reactions whereby the decomposition products of combustible materials propagate combustion. The mechanism involves converting very reactive free radicals into ones that are much less reactive.

Chlorinated flame retardants are also very efficient and can act in a similar way as brominated flame retardants, typically by decomposing to give off hydrogen chloride, with the same gas phase inhibition effect discussed above. However, chlorinated flame retardants also act in the condensed (or solid) phase by changing the decomposition mechanism and the burning rate.

One of the clearest indications of the beneficial effect of chlorine content on fire performance is the fact that polymers containing chlorine in their basic formula outperform polymers with the same structure and no chlorine. The typical example is PVC [polyvinyl chloride], which has much lower heat release and much better ignitability than polyethylene [PE]; the only difference in chemical structure between PE and PVC is that PVC has chlorine.

Flame retardants containing phosphorus may exert different modes of action, depending on the material used and the polymer (substrate) used. They may act in the condensed phase or in the gas phase or as a combination of both. For phosphorus-containing flame retardants, condensed phase action typically results in increased char formation, in some cases via a mechanism called intumescence.

The efficacy of such materials as flame retardants is often enhanced by materials containing halogens (chlorine or bromine) or materials containing nitrogen. This is sometimes done by developing materials that combine more than one such element within the structure or by using a series of additives as a system. Elemental phosphorus, as red phosphorus, is probably the only example of a flame retardant that is an elementary material. Its action is primarily in the gas phase and is strongly enhanced by synergists.

Intumescence requires three components (although the same material can have more than one role): a "carbonific” (supplying the raw material for char), a "spumific” (supplying evolving gases) and a "catalyst” (accelerating the process, often an ammonium phosphate). Certain compounds, either initially incorporated into or simply coated onto an organic polymer, either decompose or react with other materials in the condensed phase at high temperatures to give a protective barrier in which the gaseous products of polymer decomposition are trapped as they are formed.

An intumescent coating is then said to have been formed on the polymer surface. This non-flammable protective coating covers the polymer surface and helps insulate the flammable polymer from the source of heat and thus prevent the formation (or at least the escape into the gas phase) of combustible breakdown products; it may also insulate the gaseous oxidant (normally air or oxygen) from the surface of the polymer. Alternatively, direct application of a non-flammable layer on the surface of the polymer yields a non-intumescent coating.

Nitrogen-containing materials are rarely used as flame retardants on their own, but some polymers containing nitrogen (such as aromatic polyamides or natural materials such as wool or silk) have some inherent improved fire performance. The most common combinations are of materials containing nitrogen with those containing phosphorus or sulfur, often in the same molecule. One mechanism of action of these materials is the release of gases containing nitrogen, which dilute or cool the vapor phase and thus slow the combustion reactions. Flame retardant systems containing nitrogen can also be part of intumescent systems.

Alumina is the flame retardant that is most widely used, typically as a hydrated material (or aluminum hydroxide). It thermally decomposes to emit water, which then both cools and dilutes the vapor phase and makes combustion more difficult. It also acts as a filler, which decreases the amount of combustible material. Its action is primarily physical and usually requires very large amounts to be effective. Similar activity to hydrated alumina is obtained with magnesium hydroxide, but it decomposes at higher temperatures than hydrated alumina.

Antimony is primarily used as antimony oxide and it is an efficient synergist for halogenated materials (chlorine or bromine). Activity occurs principally in the gas phase, and it enhances the scavenging activity of the free radicals generated by the halogenated additives.

Boron is primarily used as either a substitute for antimony oxide (zinc borate) or with cellulosic materials, such as textiles or loose fill insulation (either borax or a borax/boric acid mixture). The mechanism of action is often a combination of the formation of glassy residues above the condensed phase, the enhancement of char production, and gas phase activity (including the release of water vapor). Boron has been used since the 19th Century.3

Many other elements can impart improved fire performance, on a limited basis. Typical examples are magnesium, sulfur (particularly as ammonium salts), and tin (as hydroxystannate or zinc hydroxystannate).

However, each flame retardant system (and modern ones may have multiple components) will be useful for a specific polymer system and a specific application. Thus, if a flame retardant system is very effective for a particular polymer when intended for a particular use, it may well not be useful for the same polymer for a different use or for a different polymer for the same use. Flame retardant systems must be tailored to fit both the material and the end use intended and the fire safety needs.


Flame retardants are a tool that can have a favorable effect on every key fire property, including ignitability, flame spread, heat release, and ease of extinction; albeit they do not necessarily have all of these effects simultaneously in every case. Some examples follow associated with well-known products in actual use where the effectiveness of the appropriate flame retardant system has been well demonstrated or even incorporated into codes or regulations. It is not possible to assign a particular flame retardant to a specific application, since many different flame retardants can be used for a variety of applications and vice versa.

Wood behaves very differently if it has been treated with flame retardants to obtain fire-retardant-treated wood (FRTW). Standard wood panels tend to exhibit a flame spread index (FSI) of between 75 and 200 (in the ASTM E844 test), while FRTW panels exhibit an FSI of under 25 and are accepted in many more applications than standard wood panels.

Recent heat release work has shown that the peak heat release rate [in the cone calorimeter] for both low density and medium density particleboards decreases when flame is retarded.5 Similar results exist for larch and thermowood pine.6 In all the cases reported, the treated wood materials were also less easily ignitable.

Cellulose loose-fill is a product used for insulating attics, although it has relatively poor fire performance. Regulations and codes require the product to meet a critical radiant flux. This fire performance can only be achieved if the cellulose loose-fill is treated with flame retardants, often boron materials.

Cables installed in plenums have long been required by the National Electrical Code (NEC)7 to be encased in non-combustible raceways. In the 1970s, fire hazard assessments showed that cables could be used safely in plenums without raceways if they met flame spread and smoke release requirements in a specific fire test designed for the application.8

After this was implemented in the NEC, it was found that suitable fire performance for plenum cable insulation and jacketing materials could be obtained with materials treated with flame retardants (often multi-additive systems). Small scale9,10 and intermediate scale10,11 heat release tests were conducted on the materials in these cables and on the cables themselves; the tests showed a significant decrease in flame spread and heat release associated with the flame retarded materials as compared to the ones that were not flame retarded. These are the materials typically used to make the cables used in plenums, meeting NEC requirements.

In the U.K., polyurethane foam used for upholstered furniture and mattresses has been required since the 1980s to meet a fire test.12 The test requires that the foam (when covered by a standard fabric) not spread flame to the extremities of the test cushion. The typical way in which furniture and mattresses using polyurethane foam (the vast majority of them) have been built in the U.K. is by incorporating flame retardants into the foam.

Cone calorimeter heat release tests have shown that systems with a fabric and a flame retarded foam that meet BS 5852 crib 5 requirements exhibit much lower heat release rates than those that are not flame retarded.13 Moreover, there is usually no ignition in those systems; as soon as the crib flame burns out, the fire ceases.

Full scale tests comparing an actual sofa containing BS 5852 crib 5 foam (purchased in the U.K.) with a sofa using a non-flame retarded (Non FR) foam (purchased in the U.S.) showed similar effects: the U.K. sofa did not ignite while the U.S. sofa ignited with a match equivalent ignition source and the sofa generated enough heat release to cause the compartment in which it was placed (with no other items) to go to flashover (See Figures 1 & 2).14

Figure 1: Sectional sofa with Non FR polyurethane foam ignited by BS 5852 ignition source 1 (match equivalent, 20 s) and same sofa after 4 min (close to flashover)14

Figure 2: U.K. sofa, with FR polyurethane foam complying with BS 5852 crib 5 before ignition and after exposure to BS 5852 ignition source 2 (lighter equivalent, 40 s)14

In this connection, it is worth discussing the test method used in California for assessing the flammability of upholstered furniture components, namely TB 117.15 The test, in effect since 1975, is a mild open flame test on bare foam; polyurethane foam requires flame retardants to pass the test.

Tests conducted on two identical sofas (one using Non FR foam and one using CA TB 117 foam) showed that the sofa with TB 117 foam did not ignite with the ignition source that caused the Non FR foam to ignite quickly; a significantly more severe ignition source was required and a much longer time elapsed before ignition.

Once ignited, both sofas caused the compartment in which they were placed to go to flashover.14 Recent work showed that when comparing two upholstered furniture mockups (See Figure 3) exposed to a severe ignition source (19 kW for 80 s, as per TB 133), a chair with a fire retarded cotton fabric and TB 117 foam ceased burning when the ignition source was removed while an identical chair with Non FR components was completely consumed.16 It also was shown that CA TB 117 foam exhibits lower heat release than Non FR foam.17

The ultimate proof of the effectiveness of a system for fire safety is whether fire hazard is reduced, as assessed by parameters such as heat release rate, tenability, or time available for escape. A study in 198818 demonstrated that flame retardants lower fire hazard.

In that study, five products were constructed containing flame retardants (FR) and not containing flame retardants (NFR) but otherwise substantially identical products. The flame retarded formulations were chosen to represent ones that were, at the time, commercially available and in common use, but which were anticipated to represent high quality performance.

Figure 3: Mockup of single sofa used in comparative fire tests16

In order to analyze the data for all products together, the full set of products were set, separately in an array in a room-corridor arrangement and exposed to a 50 kW burner. The FR products required the use of an auxiliary burner (120 kW) to avoid finding no flame propagation at all. The study showed that proper selection of flame retardants can improve fire and life safety by lowering total heat release (from 750 MJ to 200 MJ), toxic product release (by a factor of three), and mass loss (by more than half), while increasing time available for escape or rescue (from 113 s to 1789 s).

In summary, the FR products were associated with a much lower fire hazard. The authors stated that flame retarded products will not always be effective in lowering fire hazard, normally because the systems chosen are ineffective or the flame retardants are added at insufficient levels.

Note, for example, that the electrical cables were made with polyolefins (known for relatively poor fire performance) and that the polyurethane foam used for the upholstered furniture products was intended to perform better than foam intended for TB 117 use, but probably not as well as a BS 5852 crib 5 foam, and that the same fabric (a typical Non FR nylon) was used for both sets.

Marcelo M. Hirschler is with GBH International.


  1. Wyld, Obadiah, British Patent 551, March 17, 1735.
  2. Cullis, C. and Hirschler, M., The Combustion of Organic Polymers, Oxford University Press, Oxford, UK, 1981.
  3. Hirschler, M., "Recent Developments In Flame- Retardant Mechanisms,” in Developments in Polymer Stabilisation, Vol. 5, Applied Science Publ., London, 1982.
  4. ASTM E84, Standard Test Method for Surface Burning Characteristics of Building Materials, ASTM International, West Conshohocken, PA, 2014.
  5. Sumathipala, K. and White, R., "Cone Calorimeter Tests Of Wood Composites,” in Fire and Materials Conference 2013, Interscience Communications, London, UK, 2013.
  6. Tsantarides, L. and Östman, B., "Reaction To Fire Properties Of Surface Charred Wood” Interflam2013, Interscience Communications, London, UK, 2013.
  7. NFPA 70, National; Electrical Code, National Fire Protection Association, Quincy, MA 2014.
  8. NFPA 262, Standard Method of Test for Flame Travel and Smoke of Wires and Cables for Use in Air-Handling Spaces, National Fire Protection Association, Quincy, MA, 2011.
  9. Hirschler, M., "Heat release from plastic materials,” in Heat Release in Fires, Elsevier, London, UK, 1992.
  10. Barnes, M., Briggs, P., Hirschler, M., Matheson, A., and O’Neill, T., "A Comparative Study of the Fire Performance of Materials for Cable Applications. Part I. Tests on Materials and Insulated Wires,” Fire and Materials 20, 1-16, 1996.
  11. Barnes, M., Briggs, P., Hirschler, M., Matheson, A., and O’Neill, T., "A Comparative Study of the Fire Performance of Halogenated and Non-Halogenated Materials for Cable Applications. Part II. Tests on Cables”, Fire and Materials 20, 17-37, 1996.
  12. BS 5852, Methods Of Test For Assessment Of the Ignitability Of Upholstered Seating By Smouldering And Flaming Ignition Sources, British Standards Institution, London, 2006.
  13. Hirschler, M. and Smith, G., "Flammability Of Sets Of Fabric/Foam Combinations For Use In Upholstered Furniture,” Fire Safety J. 16, 13-31, 1990.
  14. Hirschler, M., "Residential Upholstered Furniture in the United States and Fire Hazard,” Business Communications Company Fifteenth Ann. Conference on Recent Advances in Flame Retardancy of Polymeric Materials, Norwalk, CT, 2004.
  15. Technical Bulletin 117, Requirements, Test Procedure and Apparatus for Testing the Flame Retardance of Resilient Filling Materials Used in Upholstered Furniture, State Of California, Department Of Consumer Affairs Bureau Of Home Furnishings and Thermal Insulation, North Highlands, CA, 2000.
  16. Janssens, M., Ewan, D., Gomez, C., Hirschler, M., Huczek, J., Mason, R., Overholt, K., and Sharp, J., "Reducing Uncertainty of Quantifying the Burning Rate of Upholstered Furniture,” SwRI Project No. 0.1.15998, Award No. 2010-DN-BX-K221, Southwest Research Institute, San Antonio, TX, 2012.
  17. Blais, M., Hirschler, M. and Janssens, M., "Fire Performance of Polyurethane Foam: California Technical Bulletin CA TB 117 and British Standard BS 5852,” in Fire and Materials Conference 2013, Interscience Communications, London, UK, 2013.
  18. Babrauskas, V., Harris, R., Gann, R., Levin, B., Lee, B., Peacock, R., Paabo, M., Twilley, W., Yoklavich, M. and Clark, H., "Fire Hazard Comparison of Fire-Retarded and Non-Fire-Retarded Products,” NBS Special Publ. 749, National Bureau of Standards, Gaithersburg, MD, 1988.