Contents
Thermoplastics and thermosets are the two groups into which plastics are divided according to their response to heat. A thermoplastic can be softened by heating and hardened by cooling again and again, and in the softened state it can be shaped by flow, for example by moulding or extrusion. A thermoset has been cured by heat or by other means and is then substantially infusible and insoluble.[1] The difference lies in the structure of the molecules. Rigid, cross-linked molecules give a thermosetting plastic, while more flexible chains that are not rigidly linked to each other give a thermoplastic.[1]
In a fire the distinction shapes how a plastic burns. Many thermoplastics melt; the melt may form a burning pool with a large surface area. Thermosets stay solid, tend to form a char and generally burn more slowly.[2][3] Melting thermoplastics can also produce flaming drips that start secondary fires.[4] Both groups are combustible, and the form of a product, its additives and the fire exposure often matter as much as the group to which its polymer belongs.[5][6] In the United States, the plastics grouping used for sprinkler protection, as described in the 2008 edition of the Fire Protection Handbook, places common thermoplastics such as polyethylene and polystyrene in the most hazardous group.[7]
Definitions and classification
The Fire Protection Handbook uses the ASTM definitions. A plastic is a material that contains one or more organic polymeric substances of large molecular weight as an essential ingredient. It is solid in its finished state, and at some stage of its manufacture or processing it can be shaped by flow.[1] The definitions of thermoplastic and thermoset given in the lead come from the same list of terms. The handbook chapter on the plastics industry repeats them and states that, whatever their properties or form, all plastics fall into one of these two groups.[8][9]
In the SFPE Handbook, Beyler and Hirschler classify polymers by their physical properties, in particular the elastic modulus and the elongation at fracture. This gives three classes: elastomers, plastics and fibres. Elastomers have elastic moduli of 105 to 106 N/m² and can be stretched by roughly 500 to 1,000 %. Plastics have moduli of 107 to 108 N/m² and elongations of 100 to 200 %, and fibres have moduli of 109 to 1010 N/m² and break after 10 to 30 %.[10] Within the plastics they separate thermoplastics, whose deformation at elevated temperatures is reversible, from thermosets, which undergo irreversible changes when heated.[10] The two definitions agree in substance: the ASTM wording stresses repeated softening and the infusible end state, while the SFPE wording stresses whether the change on heating can be reversed.
The NFPA handbook treats polymers as one division of the non-metals and divides them into elastomers, thermosets and thermoplastics.[1] Its examples of thermosets are alkyds, allyls, epoxies, melamine, phenolics, polyesters, polyurethane, silicones and urea. Its examples of thermoplastics are ABS (acrylonitrile-butadiene-styrene), acetals, acrylics, cellulosics, fluoroplastics, ionomers, nylons, olefins, polycarbonate, polyester, polyimide, polystyrene and the vinyls, which include polyvinyl chloride (PVC).[1] Elastomers, the synthetic rubbers, form a separate class.[5]
Most finished plastics are not pure polymers. They contain plasticisers, colourants, fillers, stabilisers, reinforcing agents, lubricants or other additives, and the number of product formulations runs into the thousands. The final form of a product, whether solid sections, films, sheets, foams, mouldings, fibres, pellets or powders, also has a significant effect on its fire properties.[5]
Molecular structure
Polymers are very large molecules formed in polymerisation reactions, which link small molecules into larger ones that contain repeating structural units.[1] Ouellette illustrates this with polyethylene. Its starting molecule, ethylene, is a gas with a molecular weight of 28. A chain of 6 of these units is a liquid, 35 units give a grease and 140 units a wax. At 430 units the molecular weight reaches 12,000 and the product is a hard resin.[11]
The step that separates the two groups is cross-linking. Cross-linking is the formation of chemical bonds between adjacent molecules, which creates a network that reduces solubility or resists flow on heating. When the plastic becomes highly resistant to flow at all temperatures up to thermal decomposition, it is called a thermoset.[1] Beyler and Hirschler add that cross-linking raises the melting temperature and, like stiffening of the polymer chain, can make a material infusible. Many aromatic polymers have melting temperatures above their decomposition temperatures, and this makes them thermosetting as well.[3]
The boundary between the groups is not sharp. Thermoplastics can be cross-linked after extrusion, chemically or by irradiation, to give a thermoset product.[5] Polyethylene treated in this way becomes a thermoset with high heat resistance.[12] Polyurethanes can be made either thermoplastic or thermosetting by varying the number of branchings in the molecule.[13] The same principle applies to rubber. Vulcanisation is a curing process that cross-links rubber, and vulcanised rubber cannot be reshaped or remoulded when it is reheated.[14]
Manufacture and processing
Thermosets are hardened into a permanent shape during manufacture. They usually cannot be softened by heating, and those that can be softened cannot be remelted or returned to the state they had before curing. Their components are sold as liquid monomer and polymer mixtures or as partially polymerised moulding compounds. They are usually formed with heat, sometimes under pressure, although some cure at room temperature.[5] Thermoplastics do not cure or set when heated, because heat softens them. They can be softened and rehardened repeatedly, but thermal ageing eventually degrades the material and limits the number of reheating cycles.[5]
The two groups therefore behave differently in plastics processing. Conversion into finished articles by moulding, extrusion, foaming or casting usually involves heating the plastic until it flows into a shape that it keeps on cooling. With thermoplastics no chemical reaction normally takes place. With thermosetting resins, the reaction that cross-links the material into its final, unchangeable form is a significant part of the operation.[15] For a thermoplastic, heat melts the compound and cooling hardens it in the mould; for a thermoset, heat triggers the chemical reaction that cures the resin in the mould under pressure.[16]
Some processes belong mainly to one group. Blow moulding of hollow products such as bottles and fuel tanks is generally used only with thermoplastics. Compression moulding is the most common way of forming thermosets, although it can also be used for thermoplastics. Extrusion forms thermoplastics into sheet, film, tubes, rods and profiles and is used to coat wire and cable.[16][17] Moulding and extrusion normally run at 149 to 343 °C (300 to 650 °F), depending on the plastic, so overheated heater bands and stuck temperature controllers are a known source of ignition in plastics plants.[18]
Behaviour on heating
In Introduction to Fire Safety Engineering, pyrolysis is described as the decomposition of a material by heating, in which large molecules are broken down into smaller ones and pyrolysis gases are formed. Pyrolysis needs no oxygen, but the combustion of the gases does.[19] All solid combustibles must undergo pyrolysis to produce the gaseous fuel for flaming combustion.[20] Gann and Bryner note that polymers exposed to enough heat typically first undergo a phase change, such as melting in the case of thermoplastics, and then decompose chemically. Both steps absorb heat and produce volatile products of low molecular weight.[6]
A thermoplastic can be softened without irreversible change as long as it stays below its minimum thermal decomposition temperature. Its behaviour depends on its crystallinity. A crystalline polymer has a well-defined melting temperature, whereas an amorphous polymer has a glass transition temperature above which it becomes soft and rubbery. The glass transition temperature is the upper limit for the practical use of a plastic and the lower limit for an elastomer. Most polymers are only partly crystalline, so melting usually extends over a range of 10 °C or more. Many polymers start to decompose before they melt.[21]
Thermoplastics such as polypropylene follow a two-step path: they melt into a liquid, and the melt then vaporises into gaseous fuel.[20] Thermoplastics generally soften and melt without forming char. Polymethyl methacrylate (PMMA) pyrolyses with very little melt and leaves no residue, but rigid PVC does char when burned.[22]
Thermosets are infusible and insoluble once formed, so simple phase changes on heating are not possible. Like cellulosic materials, they have no fluid state and show no notable physical transformation before they decompose.[21] Thermoset plastics form a char layer as they pyrolyse. The char insulates the unpyrolysed fuel beneath it from the heat source and slows pyrolysis unless the external heat flux increases.[20][22] Phenolic resin, for example, starts to decompose at about 575 K (302 °C), and in thermogravimetric tests at 3.3 °C/min it leaves 50 to 60 % of its mass as char. Epoxy resins are less stable than phenolic resins.[23]
The chemistry of decomposition follows four general mechanisms: random-chain scission, end-chain scission, chain stripping and cross-linking.[24] In simple thermoplastics the most common reaction breaks bonds in the main polymer chain. End-chain scission releases the monomer, a process known as unzipping, while random-chain scission produces monomers, short chains called oligomers and other species. Cross-linking builds larger structures that are less easily volatilised and is therefore important in char formation.[25] Charring is not restricted to thermosets, because PVC and polyacrylonitrile cross-link while they decompose. Char also has a drawback: it can sustain smouldering combustion after the flames have been reduced.[3]
Fire behaviour
These differences in melting and charring are a central theme of plastics in fire. Custer explains that thermoplastics produce a liquid melt when heated. The melt may form a pool with a large surface area and release energy more rapidly than polymers that stay solid. Thermosetting materials remain solid, generally tend to form a char layer and burn more slowly.[2] Gann and Bryner state the same tendency: highly cross-linked thermoset polymers normally burn less readily than thermoplastics.[6]
Melting can help or harm. A thermoplastic article may melt away from the flame front, which inhibits further burning, or it may produce flaming, tar-like drips that are difficult to extinguish and can start secondary fires.[4] The rate of melting compared with the burning rate, and the viscosity of the melt, are important for the hazard. A large amount of low-viscosity, possibly flaming melt can spread fire to new surfaces, which is especially dangerous when the melting material is part of a wall or ceiling lining.[22] Beyler and Hirschler give two contrasting examples: polyethylene melts and flows readily, while high-quality cast PMMA shows only a slight tendency to flow under fire conditions.[3]
The physical form of a thermoplastic changes its fire behaviour. Moulding pellets in bags, drums or cartons give the air little access to the surface. The same plastic moulded into hollow objects such as torch cases burns vigorously until the heat melts the shapes and reduces the exposed surface. Melting is hazardous when burning drips carry flame to a lower floor or spread fuel for later ignition.[26] Foamed polymers generally burn more rapidly than the same formulation in rigid, unexpanded form.[2] The surface flame spread of some plastics can reach about 0.6 m/s (2 ft/s), ten times the rate across most wood surfaces.[27]
In tests on office workstations for the World Trade Center investigation, the thermoplastic fabric on the privacy panels melted and rolled downward into a mass with much less surface area before it ignited. Its contribution to the peak heat release rate was therefore negligible. The office chair consisted largely of thermoplastic materials and approached its peak burning rate only when its support shell began to melt and flow to the floor.[28] Because such behaviour could not be predicted from cone calorimeter results alone, the computer simulations used cone calorimeter data directly only for the carpet, the desk and the privacy panels. The carpet and the panels were modelled as thermoplastics, with a burning rate proportional to the heat flux from the surroundings. The desk was modelled as a charring solid, in which a pyrolysis front leaves an insulating layer of char behind.[29]
In storage the hazard depends on chemical composition, physical form and storage arrangement. Ouellette lists polystyrene, ABS, polyurethane, polyethylene, polypropylene, plasticised PVC and polyester thermosets among plastics that can present a storage fire hazard. In a fire these materials may melt or break down into monomers or compounds of lower molecular weight, which can burn like flammable liquids. Fluorocarbons, unplasticised PVC and phenolics, by contrast, can be protected in the same way as ordinary Class III commodities such as wood and paper.[30]
Typical property values
Table 6.17.2 of the Fire Protection Handbook lists heats of combustion, per kilogram of fuel, and transition temperatures for a wide range of plastics. For thermosets it gives the entry "thermoset" in place of a melting temperature.[31][32] The table below shows selected values.
| Polymer | Group | Net heat of combustion (MJ/kg) | Glass transition temperature (°C) | Melting temperature (°C) |
|---|---|---|---|---|
| Polyethylene | Thermoplastic | 44.60 | −125 (high-density grade) | 120–137 (high-density grade) |
| Polypropylene | Thermoplastic | 43.10 | −13 | 160–176 |
| Polystyrene | Thermoplastic | 40.51 | 100 | 110–115 |
| Polymethyl methacrylate | Thermoplastic | 25.10 | 114 | 160–200 |
| Polyvinyl chloride, rigid | Thermoplastic | 16.89 | 81 | 210–230 |
| Epoxy, hardened | Thermoset | 28.90 | 120 | none (thermoset) |
| Phenol formaldehyde | Thermoset | 28.60 | 170 | none (thermoset) |
| Melamine formaldehyde | Thermoset | 18.13 | not listed | none (thermoset) |
| Urea formaldehyde | Thermoset | 14.61 | not listed | none (thermoset) |
The values show that the heat of combustion is not a property of the group. Among the thermoplastics listed, the net heat of combustion ranges from 16.89 MJ/kg for rigid PVC to 44.60 MJ/kg for polyethylene, a factor of about 2.6. The thermosets listed range from 14.61 to 28.90 MJ/kg.[31] Chemical composition matters more: polymers that contain halogen atoms, such as PVC, burn with difficulty, although plasticisers increase their propensity to burn.[6] Compiled transition temperatures also vary between sources. The SFPE Handbook gives PMMA a glass transition temperature of 50 °C and a melting range of 90 to 105 °C, against 114 °C and 160 to 200 °C in the NFPA table.[33][31]
Ignition temperatures show a similar spread. The International Fire Engineering Guidelines (2005), citing Babrauskas (2003), give piloted and autoignition temperatures by category of solid, with ± values whose statistical meaning is not specified in the table. For thermoplastics these are 369 ± 73 °C and 457 ± 63 °C, and for thermosetting plastics 441 ± 100 °C and 514 ± 92 °C. Halogenated plastics are listed at 382 ± 70 °C and 469 ± 79 °C; elastomers at 318 ± 42 °C and 353 ± 56 °C. Wood, paper and cotton are listed as ranges: 250 to 365 °C and 250 to 400 °C.[34] The reported central values are higher for thermosets than for thermoplastics, but their stated ± intervals overlap.
Fire testing and classification
Melting thermoplastics are a known problem in fire testing. When plastics were first used in construction, they were tested in the Steiner tunnel, even where the tunnel was not appropriate. Samples that cannot be held in place above the tunnel floor, or that melt and continue burning on the floor, which is typical of thermoplastics, still give results that are often misleading.[35] In the chapter on interior finish, Hirschler advises suspicion of tunnel results for thermoplastic materials that drip to the floor of the tunnel during testing.[36] Beyler and Hirschler make the same point more generally: many standard fire tests that allow materials to flow away from the heat source are unsuitable for assessing the hazard of flowing or dripping materials.[3]
Small-scale tests with a small, low-intensity ignition source, such as the oxygen index test (ASTM D2863) and UL 94, are used mostly for product development and production control. They are unsuitable for judging fire behaviour under room fire conditions.[37] Because formulation and product form affect fire properties so strongly, the handbook stresses testing products in their end-use configuration.[5] The International Fire Engineering Guidelines likewise note that the type of fuel likely to be present, thermoplastic or thermosetting, can affect fire dynamics and should be recorded when the fuel in a compartment is surveyed.[38]
In the United States, NFPA standards, as described by Golinveaux (2008), place plastics into three groups for the design of sprinkler protection in storage. Group A plastics present the most severe hazard, Group B plastics behave like Class IV commodities and Group C plastics like Class III commodities. Plastics can produce about 1.5 to 3 times as much heat per unit mass as wood or paper.[7] Group A contains many thermoplastics, including ABS, acrylic, acetal, polycarbonate, polyethylene, polypropylene, polystyrene and highly plasticised PVC, but also polyurethane and fibreglass-reinforced polyester. Group C contains the thermosets melamine, phenolic and urea formaldehyde together with rigid PVC and fluoroplastics such as polytetrafluoroethylene.[7][39] The grouping thus follows fire severity rather than the thermoplastic or thermoset label.
Flame retardants and limits of the distinction
Flame retardants are added to both groups during manufacture. They may be blended into the polymer mix or react with it and change its molecular structure. Halogen compounds tend to inhibit burning in the gas phase, while phosphorus compounds promote char formation and can form a glassy insulating layer. A retardant that works in one polymer does not necessarily work in another, so a system for polystyrene will not necessarily work for polyethylene or nylon.[40] Retardants are effective against small, short exposures, but in a severe exposure such as a fully developed room fire, treated and untreated plastics typically behave alike.[40]
A common misconception is that thermosets do not burn. All plastics are combustible to varying degrees, and plastics other than cellulose nitrate are classed as ordinary combustibles for fire fighting.[26] Inherently fire-resistant polymers and polymers with fire-retardant additives still burn under severe enough thermal conditions.[6] In the 2008 handbook, Benedetti and Colonna stated that lack of stability at high temperature and inherent combustibility had generally ruled out plastics where fire-resistance ratings were required. They identified special silicone foams for sealing openings in fire walls as an exception.[5]
The opposite misconception, that thermoplastics are always easy to ignite, is also wrong. Thermoplastic textile fibres such as nylon, olefin and polyester shrink away from a heat source and are generally more difficult to ignite than fabrics containing cellulose.[41] Medium and heavy thermoplastic upholstery fabrics generally resist ignition by cigarettes, although they do not resist flame ignition.[42] The handbook lists polyurethane among the thermosets, yet polyurethanes can also be made thermoplastic.[1][13] In a fire, flexible polyurethane foam can form a pool of burning liquid, similar to a fire in a combustible liquid with a high flash point.[26]
See also
- Group A plastics
- Rubber and elastomers
- Foam plastic insulation
- UL 94 flammability test
- Cellulose nitrate film
Notes
- ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Benedetti and Colonna (2008), p. 6-210.
- ↑ 2.0 2.1 2.2 Custer (2008), p. 2-50.
- ↑ 3.0 3.1 3.2 3.3 3.4 Beyler and Hirschler (2002), p. 1-123.
- ↑ 4.0 4.1 Benedetti and Colonna (2008), p. 6-225.
- ↑ 5.0 5.1 5.2 5.3 5.4 5.5 5.6 5.7 Benedetti and Colonna (2008), p. 6-211.
- ↑ 6.0 6.1 6.2 6.3 6.4 Gann and Bryner (2008), p. 6-12.
- ↑ 7.0 7.1 7.2 Golinveaux (2008), p. 16-73.
- ↑ Ouellette (2008), p. 9-53.
- ↑ Ouellette (2008), pp. 9-56–9-57.
- ↑ 10.0 10.1 Beyler and Hirschler (2002), p. 1-111.
- ↑ Ouellette (2008), p. 9-54.
- ↑ Benedetti and Colonna (2008), p. 6-221.
- ↑ 13.0 13.1 Benedetti and Colonna (2008), p. 6-222.
- ↑ Benedetti and Colonna (2008), p. 6-223.
- ↑ Ouellette (2008), p. 9-55.
- ↑ 16.0 16.1 Ouellette (2008), p. 9-57.
- ↑ Ouellette (2008), p. 9-58.
- ↑ Ouellette (2008), p. 9-62.
- ↑ Van Rossum (2027), §3.1.
- ↑ 20.0 20.1 20.2 Madrzykowski and Stroup (2008), p. 2-32.
- ↑ 21.0 21.1 Beyler and Hirschler (2002), p. 1-112.
- ↑ 22.0 22.1 22.2 Madrzykowski and Stroup (2008), p. 2-33.
- ↑ Beyler and Hirschler (2002), p. 1-127.
- ↑ Beyler and Hirschler (2002), p. 1-114.
- ↑ Beyler and Hirschler (2002), p. 1-118.
- ↑ 26.0 26.1 26.2 Benedetti and Colonna (2008), p. 6-230.
- ↑ Benedetti and Colonna (2008), p. 6-224.
- ↑ Ohlemiller et al. (2005), p. 41.
- ↑ Ohlemiller et al. (2005), p. 61.
- ↑ Ouellette (2008), p. 9-63.
- ↑ 31.0 31.1 31.2 31.3 Babrauskas (2008a), pp. 6-273–6-279.
- ↑ Babrauskas (2008a), p. 6-281.
- ↑ Beyler and Hirschler (2002), p. 1-113.
- ↑ Babrauskas (2003), as cited in National Research Council of Canada et al. (2005), p. 3.3-4.
- ↑ Hirschler (2008a), p. 6-43.
- ↑ Hirschler (2008b), p. 18-27.
- ↑ Benedetti and Colonna (2008), p. 6-227.
- ↑ Drysdale (1999), as cited in National Research Council of Canada et al. (2005), p. 2.4-3.
- ↑ Golinveaux (2008), p. 16-74.
- ↑ 40.0 40.1 Benedetti and Colonna (2008), p. 6-229.
- ↑ Stull and Chines (2008), p. 6-79.
- ↑ Babrauskas (2008b), p. 6-103.
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.
- Babrauskas, V. (2008b). Upholstered furniture and mattresses. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-103–6-128). 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., & 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.
- Custer, R. L. P. (2008). Dynamics of compartment fire growth. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 2-49–2-58). National Fire Protection Association.
- Gann, R. G., & Bryner, N. P. (2008). Combustion products and their effects on life safety. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-11–6-34). National Fire Protection Association.
- Golinveaux, J. E. (2008). Sprinkler systems for storage facilities. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 16-71–16-89). National Fire Protection Association.
- 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.
- Hirschler, M. M. (2008b). Interior finish. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 18-23–18-42). National Fire Protection Association.
- Madrzykowski, D., & Stroup, D. W. (2008). Flammability hazard of materials. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 2-31–2-48). National Fire Protection Association.
- 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.
- Ohlemiller, T. J., Mulholland, G. W., Maranghides, A., Filliben, J. J., & Gann, R. G. (2005). Fire tests of single office workstations (NIST NCSTAR 1-5C). National Institute of Standards and Technology.
- Ouellette, G. (2008). Plastics industry and related process hazards. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 9-53–9-65). 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.
- Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.