Plastics in fire

Plastics in fire describes how synthetic polymer materials, such as polyethylene, polystyrene, polyurethane foam and PVC, behave when they are heated and burn. In the classification of the SFPE Handbook, plastics are polymers that can be deformed only partly reversibly. Thermoplastics soften and may melt when heated, while thermosets change irreversibly and do not melt. Like wood, a plastic burns only after heat has broken it down into combustible gases.

Plastics matter in fire safety because they are everywhere in buildings and vehicles and because many of them burn intensely. Polyethylene and polypropylene release more than 43 kJ of heat per gram when burnt completely, about two and a half times as much as wood. Many plastics also produce much more soot than wood and so reduce visibility faster. Plastics that contain nitrogen release hydrogen cyanide, and PVC releases hydrogen chloride. Their behaviour varies widely, however: some plastics burn fiercely, while others such as PVC and PTFE have a very low peak heat release rate in standard tests.

Types of plastics

Polymers are large molecules built from repeating units. The SFPE Handbook classifies them by their physical properties into elastomers, plastics and fibres. Elastomers, or rubbers, can be stretched to several times their length and recover almost completely. Plastics can be stretched by 100 to 200 per cent before they break, and their deformation is only partly reversible. Plastics are further divided into thermoplastics, which soften on heating without permanent change, and thermosets, which undergo irreversible changes when heated.

The chemical composition gives a better indication of how a plastic behaves in fire. The main polymers containing only carbon and hydrogen are the polyolefins polyethylene (PE) and polypropylene (PP), and the styrenics such as polystyrene (PS). Polystyrene is widely used as a foam and for injection-moulded products.

Oxygen-containing plastics include poly(methyl methacrylate) (PMMA), used for its transparency, and polyesters such as polyethylene terephthalate (PET) and polycarbonate. Nitrogen-containing polymers include nylons, polyurethanes (PU) and polyacrylonitrile; polyurethanes are mainly used as flexible and rigid foams and as thermal insulation. Poly(vinyl chloride) (PVC) contains chlorine. Together with polyethylene and polypropylene it is one of the three most widely used synthetic polymers, both as a rigid and as a flexible, plasticised material. Fluorine-containing polymers such as polytetrafluoroethylene (PTFE) are very inert to heat and chemicals.

How plastics burn

Solid plastics do not burn directly. When heated, they undergo pyrolysis: the large molecules break down into smaller ones, which escape as combustible gases. Pyrolysis needs no oxygen, but the burning of the gases does. Burning becomes self-sustaining when the flame feeds enough heat back to the surface to keep producing these gases, a loop of heat, gas release and flame. The SFPE Handbook sums this up in one phrase: polymers cannot burn if they do not break down.

Burning of a plastic is a feedback loop: heat from the flame decomposes the solid into combustible gases, which burn and heat the surface again. Based on Beyler and Hirschler (2002), p. 1-110, and Van Rossum (2027), §3.1.

The products of decomposition depend on the polymer. Polyethylene and polypropylene form very little of their original building block, the monomer, and instead release up to about 70 different small molecules, mostly hydrocarbons. PMMA decomposes almost entirely back to its monomer and burns at a very steady rate. For this reason it is a favourite material in fire research. PVC first loses its chlorine as hydrogen chloride gas, nearly completely, between about 500 and 550 K by a process called chain stripping. In commercial PVC compounds the stabilisers delay this, so that hydrogen chloride is not released until above 520 K and the main stage starts at about 600 K.

Thermoplastics may melt and flow before they decompose. This can improve or worsen fire safety. A plastic that flows away from the heat source may avoid ignition or further fire growth, but plastic that flows towards the heat source can make the fire worse. Polyethylene melts and flows readily, while high-quality cast PMMA flows only slightly. Many standard fire tests let a melting material flow away from the heat source and are therefore unsuitable for assessing the hazard of materials that drip or flow. Thermosets cannot melt, so they undergo no notable physical change before they decompose.

Some plastics form a char, a carbon-rich solid residue. A char layer reduces the amount of combustible gas and protects the material below from heat, so it slows down burning. Char is not always an advantage, however: the char itself can keep burning as sustained smouldering combustion.

Heat release

The energy content of plastics varies widely. It depends mainly on the elements in the polymer: plastics made of carbon and hydrogen only release the most heat, while oxygen, chlorine and fluorine lower the value. The SFPE Handbook gives the following net heats of complete combustion, measured in an oxygen bomb calorimeter:

Material	Net heat of complete combustion (kJ/g)
Polyethylene	43.6
Polypropylene	43.4
Polystyrene	39.2
Expanded polystyrene	35.6 to 38.1
Nylon	30.8
Rigid polyurethane foams	25.0 to 28.0
Flexible polyurethane foams	23.2 to 27.2
Poly(methyl methacrylate)	25.2
Poly(vinyl chloride)	16.4
Polytetrafluoroethylene	6.2
Red oak (for comparison)	17.1

The heat of combustion of the hydrocarbon plastics is thus about two and a half times that of wood, while chlorine and fluorine lower it. PVC releases about as much heat per kilogram as oak, and PTFE much less. Introduction to Fire Safety Engineering lists a net heat of combustion of 18 MJ/kg for PVC roofing, the same as for beech wood. In a real fire, combustion is not complete, so the heat actually released is lower than these values.

The rate at which a plastic releases heat matters even more than its energy content. The heat release rate is the main variable that determines how much a material adds to the hazard of a fire in a room. In the cone calorimeter, at an irradiance of 40 kW/m² and a thickness of 6 mm, polypropylene reached a peak heat release rate of 1500 kW/m², high-density polyethylene 1400 kW/m² and polystyrene 1100 kW/m². PMMA reached 670 kW/m², PVC 180 kW/m² and PTFE only 13 kW/m². For comparison, values from other sources quoted in the same table are 450 kW/m² for cotton and 310 kW/m² for wool. The handbook warns that the temperature at which a polymer starts to decompose does not on its own predict its fire performance.

Net heat of complete combustion (left) and peak heat release rate in the cone calorimeter at 40 kW/m² (right) for common plastics. Based on Tewarson (2002), p. 3-108, and Beyler and Hirschler (2002), p. 1-124.

In large fires, the heat released per square metre of burning surface can be estimated from two material properties. The first is the mass loss rate per unit area. The second is the chemical heat of combustion, the heat actually released per gram of material lost in a real fire. The chemical heat release rate is the product of the two: $${\dot{Q}}^{\prime }{\text{'}}_{\mathrm{c}\mathrm{h}}={\dot{m}}^{″}\mathrm{\Delta }{H}_{\mathrm{c}\mathrm{h}}$$

Symbol	Description	Unit
${\dot{Q}}^{\prime }{\text{'}}_{\mathrm{c}\mathrm{h}}$	chemical heat release rate per unit area of burning surface	kW/m²
${\dot{m}}^{″}$	mass loss rate per unit area	g/(m²·s)
$\mathrm{\Delta }{H}_{\mathrm{c}\mathrm{h}}$	chemical heat of combustion	kJ/g

The SFPE Handbook applies this relation to the steady mass loss rates expected in large, well-ventilated fires. For polyethylene, the expected mass loss rate is 26 g/(m²·s) and the chemical heat of combustion 38.4 kJ/g, so the heat release rate is 26 × 38.4 = 998 kW/m². Polystyrene gives 36 × 27.0 = 972 kW/m². PVC, by contrast, gives only 16 × 5.7 = 91 kW/m², and FEP Teflon 7 × 4.1 = 28 kW/m². For PVC, the chemical heat of combustion is only about a third of its heat of complete combustion, while for polyethylene it is close to 90 per cent (calculated). The relation describes a fire that is not limited by its air supply; in a ventilation-controlled fire, combustion is less complete.

Plastics have changed the fire load of everyday objects. A study for the Netherlands Institute for Public Safety notes that modern cars, with their high plastic content, have a greater fire load and a higher heat release rate than older steel models. In indoor car parks this leads to very high temperatures and dense black smoke. Introduction to Fire Safety Engineering notes for the same reason that the design fire curves for cars in Dutch car park standards may be less conservative in some situations.

Smoke and soot

Plastics are major producers of smoke. The production of soot depends on the fuel and on the combustion conditions: plastics produce relatively much soot, wood much less, and ethanol very little. The soot yield is the mass of soot produced per mass of fuel burnt. Two sources give the following values for well-ventilated flaming combustion:

Material	Soot yield, Introduction to Fire Safety Engineering (kg/kg)	Particulate yield, NUREG-1805 (kg/kg)
Wood	0.015	0.015 (red oak)
Poly(methyl methacrylate)	–	0.022
Polypropylene	–	0.059
Acrylonitrile-butadiene-styrene (ABS)	–	0.105
Poly(vinyl chloride)	0.14	–
Polystyrene	0.16	0.164
Expanded polystyrene	0.195	–
Polyurethane	0.23	0.188 (flexible foam), 0.118 (rigid foam)

The values in Introduction to Fire Safety Engineering are pre-flashover values, for burning with excess oxygen. The NUREG-1805 values come from small-scale tests of turbulent flaming combustion compiled by Klote and Milke. In both sources, polystyrene and polyurethane produce roughly 8 to 15 times as much soot per kilogram as wood (calculated). A fire that is smothered, for example by a sprinkler, burns less completely and produces even more soot per kilogram of fuel.

Soot (particulate) yield of wood and plastics in well-ventilated flaming combustion. Based on Klote and Milke (2002), as cited in Iqbal and Salley (2004), p. 18-4.

Toxic gases

All fires produce carbon monoxide, but some plastics add specific toxic gases. Hydrogen cyanide is always present to some extent when nitrogen-containing materials burn, such as acrylics, polyurethane foams, melamine, nylon and wool. High concentrations are likely in large post-flashover fires. PVC releases hydrogen chloride as it decomposes, and acid gases such as hydrogen chloride are among the most important irritants in smoke. PTFE produces carbonyl fluoride, a highly toxic lung irritant. In vehicle fires, the burning plastic parts already create severely toxic smoke, which seriously limits the possibilities for escape.

The SFPE Handbook compares the toxicity of the smoke of material classes with a potency factor, where 1 corresponds to a typical lethal exposure dose of about 500 g·m⁻³·min. In non-flaming decomposition, most materials have a factor close to 1. Nitrogen-containing materials that release much hydrogen cyanide at low temperatures, such as polyacrylonitrile and rigid polyurethane foam, reach 3 to 8.

In early, well-ventilated flaming fires, wood and cotton are the least toxic, with a factor of 0.2. Plastics containing only carbon, hydrogen and oxygen have a factor of about 0.4, and PVC and fire-retarded materials about 1. After flashover, the potency of all materials rises. For nitrogen-containing materials it ranges from about 2.5 for flexible polyurethane foam to about 11 for polyacrylonitrile, and PVC is suggested to reach about 2.5.

Flame retardants

A flame retardant is added to a plastic to change the way it decomposes and burns, and its effect depends on chemistry and timing. In thermal analysis, the overlap between the decomposition stages of the polymer and of the additive indicates how effective an additive is. The mechanisms and likely effectiveness of flame retardants can therefore be predicted well from thermal decomposition data. In one example from the SFPE Handbook, a bromine compound in ABS worked together with antimony oxide, because the antimony bromide formed could evaporate. The same compound was ineffective with iron oxide, whose bromide does not evaporate, and with zinc oxide, whose bromide evaporates before the polymer starts to break down.

Flame retardants involve a trade-off. Many treatments increase the yield of toxic gases per gram of material but greatly reduce the rate at which the material burns, so that the overall fire hazard falls. In full-scale tests with upholstered chairs, fire-retarded covers increased the time before conditions became untenable through asphyxiant gases by about 1.5 minutes. The handbook concludes that some modern materials and fire retardant systems may present toxic or environmental problems, but give considerable overall benefits if they reduce the number of fires and slow fire development.

Plastics in buildings and regulation

Building regulations can limit the use of combustible linings through reaction-to-fire classes. In New Zealand, lining materials are classified by the time to flashover in a full-scale room test. Exposed polyurethane foams reach flashover within two minutes and fall into the worst class, Group Number 4, which is not acceptable in occupied spaces. Some plastic-containing products cannot be judged with small-scale tests. The New Zealand commentary names metal-faced sandwich panels and foil-faced combustible insulation as products for which the small-scale correlation does not work.

Limits and common misconceptions

A common misconception is that all plastics behave alike in a fire. The data above show that peak heat release rates in the cone calorimeter range from 13 to 1500 kW/m², and heats of combustion from 6.2 to 43.6 kJ/g. PVC is an example of a plastic with a low heat release rate but an important release of hydrogen chloride.

A second misconception is that a single property, such as the limiting oxygen index or the decomposition temperature, describes fire performance. The SFPE Handbook notes that the oxygen index technique has fallen into disrepute and that thermal decomposition data alone cannot predict fire performance. Toxicity data are also limited: the handbook calls the database of toxic potency tests very poor, especially for post-flashover conditions. Values for soot yield and toxicity depend strongly on the fire conditions, and the same material gives different results in well-ventilated and in smothered fires.

See also

• Euroclass system
• Fire resistance test
• Combustibility
• Cone calorimeter

References

• 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.
• 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.
• Janssens, M. (2002). Calorimetry. 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. 3-38–3-62). National Fire Protection Association; Society of Fire Protection Engineers.
• Ministry of Business, Innovation and Employment. (2013). Commentary for Building Code clauses C1–C6 and Verification Method C/VM2.
• 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.
• 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 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.