New FSE wiki article: Unstable reactive material (written by Claude Opus 5.5) |
m Remove source references from image captions |
||
| (3 intermediate revisions by 3 users not shown) | |||
| Line 1: | Line 1: | ||
An '''unstable reactive material''' | An '''unstable reactive material''' can undergo violent chemical change when exposed to heat, pressure, shock or contamination.<ref name="Lem-6-166">Lemoff (2008), p. 6-166.</ref><ref name="Dav-7-49">Davenport (2008), p. 7-49.</ref> Examples include rapid polymerisation or decomposition, and NFPA 55 also names vigorous condensation in its definition for gases.<ref name="Lem-6-166" /> NFPA 704 describes instability as a property of the material in itself: some such materials can detonate at normal temperatures and pressures, while others only become unstable when heated.<ref name="Wr-13-123">C. J. Wright (2008a), p. 13-123.</ref> Typical examples are [[acetylene]], [[ethylene oxide]], organic peroxides, monomers such as styrene that can polymerise, nitromethane and concentrated hydroxylamine.<ref name="Dav-7-50">Davenport (2008), p. 7-50.</ref><ref name="Zal-2-102">Zalosh (2008), p. 2-102.</ref><ref name="Lem-6-177" /> In the editions presented by the 2008 handbook, [[NFPA 704]] rates instability from 0 to 4 and NFPA 55 lists gas Classes 1 to 4.<ref name="Wr-13-123" /><ref name="Lem-6-166" /><ref name="LR-7-39">Lemoff and Rivkin (2008), p. 7-39.</ref> Davenport (2008) advises detached storage where possible, with pressure relief and explosion venting.<ref name="Dav-7-49" /> He recommends water and sprinklers for fires involving unstable chemicals and manual attack from a protected distance; water can warm refrigerated peroxides.<ref name="Dav-7-50" /><ref name="YN-2-91">Yu and Newman (2008), p. 2-91.</ref> | ||
== Definition and classes == | == Definition and classes == | ||
The handbook definition comes from NFPA 55 and is written for gases. Lemoff (2008) describes unstable reactive gases as gases that will vigorously polymerise, decompose or condense, become self-reactive, or otherwise undergo a violent chemical change under conditions of shock, pressure or temperature.<ref name="Lem-6-166" /> A Class 2 gas readily undergoes violent chemical change at elevated temperatures and pressures.<ref name="Lem-6-166" /> A Class 3 gas is itself capable of detonation, explosive decomposition or explosive reaction, but needs a strong initiating source or heating under confinement.<ref name="Lem-6-166" /> A Class 4 gas is readily capable of detonation, explosive decomposition or explosive reaction at normal temperatures and pressures.<ref name="Lem-6-166" /> NFPA 55 | The handbook definition comes from NFPA 55 and is written for gases. Lemoff (2008) describes unstable reactive gases as gases that will vigorously polymerise, decompose or condense, become self-reactive, or otherwise undergo a violent chemical change under conditions of shock, pressure or temperature.<ref name="Lem-6-166" /> A Class 2 gas readily undergoes violent chemical change at elevated temperatures and pressures.<ref name="Lem-6-166" /> A Class 3 gas is itself capable of detonation, explosive decomposition or explosive reaction, but needs a strong initiating source or heating under confinement.<ref name="Lem-6-166" /> A Class 4 gas is readily capable of detonation, explosive decomposition or explosive reaction at normal temperatures and pressures.<ref name="Lem-6-166" /> NFPA 55 lists detonable Classes 3 and 4 separately from nondetonable Classes 2 and 3; Class 3 can thus appear in either category, with different storage limits.<ref name="Lem-6-166" /><ref name="LR-7-39" /> | ||
The NFPA 704 instability scale | The NFPA 704 instability scale applies beyond gases. In the table reproduced by C. J. Wright (2008a) from the 2001 edition of NFPA 704, degree 4 covers materials readily capable of detonation or explosive decomposition at normal temperatures and pressures, including materials sensitive to localised thermal or mechanical shock.<ref name="Wr-13-123" /> The descriptions of degrees 3 and 2 parallel those of NFPA 55 gas Classes 3 and 2, but the two rating systems have different purposes.<ref name="Wr-13-123" /><ref name="Lem-6-166" /> Degree 1 covers materials that are normally stable but can become unstable at elevated temperatures and pressures, and degree 0 those that remain stable even under fire conditions.<ref name="Wr-13-123" /> The handbook uses several names for this quadrant: "reactivity hazard" in the text, "instability hazard" in the table and "self-reactivity (instability)" in the chapter on medical gases.<ref name="Wr-13-123" /><ref name="All-6-194">Allen (2008), p. 6-194.</ref> | ||
[[File:fse-unstable-reactive-material-v1.svg|thumb|upright=1.5|alt=Five NFPA 704 instability degrees next to separate NFPA 55 gas categories, with Class 3 shown in both detonable and nondetonable groups.|NFPA 704 instability degrees (2001 edition) and NFPA 55 gas categories (2005 edition) use similar words but serve different purposes; Class 3 appears under both detonable and nondetonable gas limits.]] | |||
Other United States rules use related terms. | Other United States rules use related terms. In the workplace regulation (29 CFR 1910) summarised by Hildebrand and Noll (2008), OSHA names "unstables (reactives)" among examples of a physical hazard.<ref name="HN-13-82">Hildebrand and Noll (2008), p. 13-82.</ref> In transport, Division 4.1 includes self-reactive materials, which are liable to undergo a strongly exothermal decomposition at normal or elevated temperatures because of excessively high transport temperatures or contamination.<ref name="Wr-13-120">C. J. Wright (2008a), p. 13-120.</ref> Under the Environmental Protection Agency rules summarised by Severson (2008), a waste is reactive if it is normally unstable and readily undergoes violent change without detonating.<ref name="Sev-7-60">Severson (2008), pp. 7-60–7-61.</ref> A waste that can detonate, readily or under a strong initiating source, is reactive as well.<ref name="Sev-7-60" /> | ||
== How unstable materials release energy == | == How unstable materials release energy == | ||
| Line 16: | Line 17: | ||
== Thermal runaway and critical conditions == | == Thermal runaway and critical conditions == | ||
Some unstable materials react slowly even at ordinary temperatures, as styrene does, and the hazard arises when the heat released is not removed.<ref name="Dav-7-50" /> Gray (2002) explains this with the balance between heat release and heat loss. Most chemical reaction rates rise almost exponentially with temperature, whereas heat loss by conduction rises only linearly.<ref name="Gra-2-212">Gray (2002), p. 2-212.</ref> Above a | Some unstable materials react slowly even at ordinary temperatures, as styrene does, and the hazard arises when the heat released is not removed.<ref name="Dav-7-50" /> Gray (2002) explains this with the balance between heat release and heat loss. Most chemical reaction rates rise almost exponentially with temperature, whereas heat loss by conduction rises only linearly.<ref name="Gra-2-212">Gray (2002), p. 2-212.</ref> Above the critical ambient temperature for a given body, heat release exceeds loss and the temperature rises; in a chemical reactor this is called a runaway reaction or [[thermal runaway]].<ref name="Gra-2-212" /><ref name="Gra-2-214">Gray (2002), p. 2-214.</ref> Below it, a stable balance point exists, usually 5 to 20 °C above the ambient temperature, at which the material self-heats without igniting but may degrade.<ref name="Gra-2-214" /> | ||
[[File:fse-unstable-reactive-material-v2.svg|thumb|upright=1.5|alt=Curved heat release line with three straight heat loss lines: one crosses it twice, one touches it, one stays below it.|Heat release (curve) and heat loss (straight lines) of a reacting mass at three ambient temperatures: subcritical with a stable balance point, critical where the lines touch, and supercritical runaway.]] | |||
The simplest theory, due to Semenov, assumes a uniform temperature in the reacting mass, as in a well-stirred liquid, and a single | The simplest theory, due to Semenov, assumes a uniform temperature in the reacting mass, as in a well-stirred liquid, and represents heat generation with a single effective reaction whose rate follows the Arrhenius form <math>Z\,e^{-E/RT}</math>.<ref name="Gra-2-213">Gray (2002), p. 2-213.</ref><ref name="Gra-2-215">Gray (2002), p. 2-215.</ref><ref name="Gra-2-216">Gray (2002), p. 2-216.</ref> That effective rate need not correspond to only one chemical step.<ref name="Gra-2-216" /> At the critical condition the heat release and heat loss curves just touch, and the mass is warmer than its surroundings by a critical amount. Gray gives that critical temperature rise as:<ref name="Gra-2-217">Gray (2002), p. 2-217.</ref> | ||
<math display="block">\Delta T_\mathrm{cr} = T_\mathrm{s,cr} - T_\mathrm{a,cr} \cong \frac{R\,T_\mathrm{a,cr}^2}{E}</math> | <math display="block">\Delta T_\mathrm{cr} = T_\mathrm{s,cr} - T_\mathrm{a,cr} \cong \frac{R\,T_\mathrm{a,cr}^2}{E}</math> | ||
| Line 40: | Line 42: | ||
The result holds only at criticality and is independent of the shape of the body.<ref name="Gra-2-217" /> Criticality exists only if the activation energy is large enough, <math>E \ge 4RT_\mathrm{a,cr}</math>; otherwise the temperature rises smoothly without an abrupt change.<ref name="Gra-2-217" /> Gray notes that R/E is about 0.0001 for most combustion reactions, and that the critical rise would typically be 20 to 30 °C for ambient temperatures around 30 to 40 °C.<ref name="Gra-2-217" /> | The result holds only at criticality and is independent of the shape of the body.<ref name="Gra-2-217" /> Criticality exists only if the activation energy is large enough, <math>E \ge 4RT_\mathrm{a,cr}</math>; otherwise the temperature rises smoothly without an abrupt change.<ref name="Gra-2-217" /> Gray notes that R/E is about 0.0001 for most combustion reactions, and that the critical rise would typically be 20 to 30 °C for ambient temperatures around 30 to 40 °C.<ref name="Gra-2-217" /> | ||
As a worked example, take a material stored at 30 °C (303 K) | As a worked example, take a material stored at a hypothetical critical ambient temperature of 30 °C (303 K). Assume its decomposition has E/R = 10,000 K, corresponding to R/E = 0.0001; Gray (2002) gives that ratio for most combustion reactions, but it is not measured here for a named chemical.<ref name="Gra-2-217" /> The model's critical rise is 303² / 10,000 = 9.2 K, provided the body size, heat transfer and reaction rate also place the system at the critical condition.<ref name="Gra-2-217" /> The body temperature at that point would be about 39 °C; a 9 K rise by itself is not a general runaway threshold. With the same activation energy, a rise of 10 K above 303 K multiplies the reaction rate by about 2.9, and a rise of 20 K by about 7.7. The typical 20 to 30 °C quoted by Gray corresponds to smaller activation energies, with E/R of roughly 3,100 to 4,900 K. This sensitivity is why hazard reduction for less-than-explosive instability concentrates on keeping the material from being heated to an unsafe level.<ref name="Sch-8-24">Schwab (2008a), p. 8-24.</ref> | ||
[[File:fse-unstable-reactive-material-v3.svg|thumb|upright=1.5|alt=Rising curve of reaction rate ratio from 1 to about 20 over a 30 K temperature rise, with the conditional critical rise of 9.2 K marked.|Relative reaction rate above 30 °C (303 K) for the assumed E/R = 10,000 K; the 9.2 K rise is critical only when the other Semenov conditions are met.]] | |||
Ouellette (2008) describes the same chain in stored, unstabilised polymer powders: decomposition liberates heat, the bulk temperature rises, the decomposition speeds up, and the process continues until the material reaches its self-accelerating decomposition temperature.<ref name="Oue-9-63">Ouellette (2008), p. 9-63.</ref> Monitoring the temperature of the containers is therefore important during storage.<ref name="Oue-9-63" /> Hydrogen peroxide shows the temperature effect in numbers: its decomposition rate increases about 1.5 times for each 10 °C (18 °F) rise, and near the boiling point it becomes very rapid.<ref name="Dav-7-46">Davenport (2008), p. 7-46.</ref> | Ouellette (2008) describes the same chain in stored, unstabilised polymer powders: decomposition liberates heat, the bulk temperature rises, the decomposition speeds up, and the process continues until the material reaches its self-accelerating decomposition temperature.<ref name="Oue-9-63">Ouellette (2008), p. 9-63.</ref> Monitoring the temperature of the containers is therefore important during storage.<ref name="Oue-9-63" /> Hydrogen peroxide shows the temperature effect in numbers: its decomposition rate increases about 1.5 times for each 10 °C (18 °F) rise, and near the boiling point it becomes very rapid.<ref name="Dav-7-46">Davenport (2008), p. 7-46.</ref> | ||
== Typical unstable reactive materials == | == Typical unstable reactive materials == | ||
Acetylene is classified as reactive, flammable and compressed, and the handbook shows it with 0 for health, 4 for flammability and 3 for reactivity on the NFPA 704 diamond.<ref name="Lem-6-177" /> It can react with copper and some copper alloys to form metallic acetylides, extremely shock-sensitive explosives that can initiate its decomposition | Acetylene is classified as reactive, flammable and compressed, and the handbook shows it with 0 for health, 4 for flammability and 3 for reactivity on the NFPA 704 diamond.<ref name="Lem-6-177" /> It can react with copper and some copper alloys to form metallic acetylides, extremely shock-sensitive explosives that can initiate its decomposition. Lemoff (2008) identifies brass with less than 65% copper as usable for acetylene piping, while allowing some other copper-containing components under specific conditions.<ref name="Lem-6-177" /> Methylacetylene and propadiene are unstable, reactive materials with hazards similar to acetylene; their commercial mixture is made stable enough by its content of propane and butanes.<ref name="Lem-6-184">Lemoff (2008), p. 6-184.</ref> | ||
Ethylene oxide is both flammable and unstable. Heat from its exothermic reactions with alcohols, acids, ammonia and many other compounds can start the polymerisation of the unreacted ethylene oxide, and its vapours can detonate if an initiating heat source is present.<ref name="Dav-7-49" /> Its flammable range in air is 3 to 80%; reported explosions of mixtures above 80% are the result of chemical decomposition.<ref name="Dav-7-49" /> Hydrogen cyanide tends to polymerise, a reaction catalysed by alkaline materials; because one product, ammonia, is itself alkaline, an explosive reaction eventually follows unless an acid is added to neutralise it.<ref name="Dav-7-49" /> | Ethylene oxide is both flammable and unstable. Heat from its exothermic reactions with alcohols, acids, ammonia and many other compounds can start the polymerisation of the unreacted ethylene oxide, and its vapours can detonate if an initiating heat source is present.<ref name="Dav-7-49" /> Its flammable range in air is 3 to 80%; reported explosions of mixtures above 80% are the result of chemical decomposition.<ref name="Dav-7-49" /> Hydrogen cyanide tends to polymerise, a reaction catalysed by alkaline materials; because one product, ammonia, is itself alkaline, an explosive reaction eventually follows unless an acid is added to neutralise it.<ref name="Dav-7-49" /> | ||
| Line 53: | Line 56: | ||
[[Organic peroxide|Organic peroxides]] are a widely used group, serving as polymerisation initiators, flour bleaches and catalysts.<ref name="Dav-7-49" /> All are combustible and increase the intensity of a fire, and some, such as methyl ethyl ketone peroxide, are detonatable.<ref name="Dav-7-49" /> Undiluted benzoyl peroxide burns with great rapidity, similar to black powder, and decomposes explosively if confined when heated.<ref name="Dav-7-50" /> Ethers form peroxides during storage, and pure, dry ether in a clear bottle develops detectable peroxides within one month; such peroxides can detonate when heated or concentrated.<ref name="Dav-7-50" /> | [[Organic peroxide|Organic peroxides]] are a widely used group, serving as polymerisation initiators, flour bleaches and catalysts.<ref name="Dav-7-49" /> All are combustible and increase the intensity of a fire, and some, such as methyl ethyl ketone peroxide, are detonatable.<ref name="Dav-7-49" /> Undiluted benzoyl peroxide burns with great rapidity, similar to black powder, and decomposes explosively if confined when heated.<ref name="Dav-7-50" /> Ethers form peroxides during storage, and pure, dry ether in a clear bottle develops detectable peroxides within one month; such peroxides can detonate when heated or concentrated.<ref name="Dav-7-50" /> | ||
Nitromethane is a combustible liquid that decomposes explosively at 315 °C (599 °F) and | Nitromethane is a combustible liquid that decomposes explosively at 315 °C (599 °F) and about 6.31 MPa gauge (915 psig).<ref name="Dav-7-49" /> Railroad tank cars of nitromethane detonated at Niagara Falls, New York, and at Mt. Pulaski, Illinois, in 1958.<ref name="Dav-7-49" /> The causes were not definitely determined; contamination by material previously carried in the tanks is a possible explanation.<ref name="Dav-7-49" /> | ||
Concentrated hydroxylamine caused two of the incidents listed by Zalosh (2008).<ref name="Zal-2-102" /> At Hanover Township, Pennsylvania, on 19 February 1999, 5 people died and 14 were injured; at Gunma, Japan, on 10 June 2000, 4 died and 58 were injured.<ref name="Zal-2-102" /> In both, the aqueous concentration was significantly higher than the recommended practice of less than 50%, and tests show that such concentrations with trace contaminants can decompose very violently, often as a detonation.<ref name="Zal-2-102" /> | Concentrated hydroxylamine caused two of the incidents listed by Zalosh (2008).<ref name="Zal-2-102" /> At Hanover Township, Pennsylvania, on 19 February 1999, 5 people died and 14 were injured; at Gunma, Japan, on 10 June 2000, 4 died and 58 were injured.<ref name="Zal-2-102" /> In both, the aqueous concentration was significantly higher than the recommended practice of less than 50%, and tests show that such concentrations with trace contaminants can decompose very violently, often as a detonation.<ref name="Zal-2-102" /> | ||
| Line 64: | Line 67: | ||
For gases, NFPA 55 (2005 edition), as described by Lemoff and Rivkin (2008), keeps unstable reactive gases of Classes 2, 3 and 4 at 6.10 m (20 ft) from other gases.<ref name="LR-7-38">Lemoff and Rivkin (2008), p. 7-38.</ref> Gases classed simply as "other gas" need no separation.<ref name="LR-7-38" /> The distance can be reduced without limit by a noncombustible barrier at least 1.52 m (5 ft) high with a fire resistance rating of at least half an hour.<ref name="LR-7-38" /> The same standard sets maximum allowable quantities per control area, a space separated by a fire barrier.<ref name="LR-7-38" /> For detonable Class 3 or 4 gases the quantity is zero in unsprinklered areas.<ref name="LR-7-39" /> In sprinklered areas it is 0.454 kg (1 lb) of liquefied or 0.283 m³ (10 ft³) of nonliquefied gas, doubled when a gas cabinet is used, while Class 1 gases have no limit.<ref name="LR-7-39" /> | For gases, NFPA 55 (2005 edition), as described by Lemoff and Rivkin (2008), keeps unstable reactive gases of Classes 2, 3 and 4 at 6.10 m (20 ft) from other gases.<ref name="LR-7-38">Lemoff and Rivkin (2008), p. 7-38.</ref> Gases classed simply as "other gas" need no separation.<ref name="LR-7-38" /> The distance can be reduced without limit by a noncombustible barrier at least 1.52 m (5 ft) high with a fire resistance rating of at least half an hour.<ref name="LR-7-38" /> The same standard sets maximum allowable quantities per control area, a space separated by a fire barrier.<ref name="LR-7-38" /> For detonable Class 3 or 4 gases the quantity is zero in unsprinklered areas.<ref name="LR-7-39" /> In sprinklered areas it is 0.454 kg (1 lb) of liquefied or 0.283 m³ (10 ft³) of nonliquefied gas, doubled when a gas cabinet is used, while Class 1 gases have no limit.<ref name="LR-7-39" /> | ||
The acetylene cylinder shows how a [[compressed gas cylinder]] can be designed around an unstable gas. It is filled with a porous mass saturated with acetone or dimethylformamide, so that the gas is stored in solution and any free gas volume is small, which limits the decomposition energy available.<ref name="Lem-6-177" /> | The acetylene cylinder shows how a [[compressed gas cylinder]] can be designed around an unstable gas. It is filled with a porous mass saturated with acetone or dimethylformamide, so that the gas is stored in solution and any free gas volume is small, which limits the decomposition energy available.<ref name="Lem-6-177" /> Lemoff (2008) reports a maximum charging pressure of 1724 kPa gauge (250 psig) at 21 °C (70 °F) under the United States and Canadian transport rules he describes.<ref name="Lem-6-177" /> He gives a general limit of 103 kPa gauge (15 psig) for most acetylene piping, with specially designed higher-pressure applications.<ref name="Lem-6-177" /> His chapter says the conventional codes of the time prohibit liquid or solid acetylene use.<ref name="Lem-6-178">Lemoff (2008), p. 6-178.</ref> | ||
[[File:fse-unstable-reactive-material-v4.svg|thumb|upright=1.5|alt=Cut-away of a United States or Canadian acetylene cylinder described in 2008, with porous filler, solvent and fusible plugs.|Cut-away of the United States and Canadian acetylene-cylinder design: acetylene is dissolved in acetone or DMF held in a porous mass, and fusible plugs open at about 100 °C.]] | |||
In process plants, the AIChE Center for Chemical Process Safety guidelines summarised by Zalosh (2008) combine four elements.<ref name="Zal-2-102" /> These are lists of unstable materials, software that estimates reaction energy, laboratory screening of reaction onset temperatures and sensitivities, and expert review.<ref name="Zal-2-102" /> Schwab (2008a) adds that most chemicals are reasonably stable, that data for the others come from suppliers, NFPA 704 and reactive-chemical handbooks, and that thermodynamic calculations and laboratory tests are needed where data are missing.<ref name="Sch-8-24" /> ASTM | In process plants, the AIChE Center for Chemical Process Safety guidelines summarised by Zalosh (2008) combine four elements.<ref name="Zal-2-102" /> These are lists of unstable materials, software that estimates reaction energy, laboratory screening of reaction onset temperatures and sensitivities, and expert review.<ref name="Zal-2-102" /> Schwab (2008a) adds that most chemicals are reasonably stable, that data for the others come from suppliers, NFPA 704 and reactive-chemical handbooks, and that thermodynamic calculations and laboratory tests are needed where data are missing.<ref name="Sch-8-24" /> Ural and Garzia's (2008) test list includes ASTM E476 for confined-system thermal instability, E537 for thermal stability by thermal analysis and E698 for Arrhenius kinetic constants.<ref name="UG-17-146">Ural and Garzia (2008), p. 17-146.</ref> Where the hazard is less than explosive, it is reduced by keeping the material from overheating, for example by lowering a boiling point under vacuum or by vacuum or freeze drying.<ref name="Sch-8-24" /> | ||
Hazard indexes build on the same ratings. In the Dow Fire and Explosion Index described by Watts (2002), the material factor is a number from 1 to 40.<ref name="Wat-5-129">Watts (2002), p. 5-129.</ref> It is determined from the flammability and reactivity of the material as described in NFPA 704 (1990 edition) and related NFPA data.<ref name="Wat-5-129" /> | Hazard indexes build on the same ratings. In the Dow Fire and Explosion Index described by Watts (2002), the material factor is a number from 1 to 40.<ref name="Wat-5-129">Watts (2002), p. 5-129.</ref> It is determined from the flammability and reactivity of the material as described in NFPA 704 (1990 edition) and related NFPA data.<ref name="Wat-5-129" /> | ||
== Fire fighting == | == Fire fighting == | ||
Davenport (2008) recommends water for fires involving unstable chemicals, including organic peroxides.<ref name="Dav-7-50" /> He calls automatic sprinklers the best protection, because such fires can lead to explosions, and calls for manual fire fighting from a protected distance.<ref name="Dav-7-50" /> A tank of ethylene oxide exposed to fire can heat its vapours rapidly to their ignition temperature unless it is kept wet with water spray.<ref name="Dav-7-49" /> Water at room temperature can, however, warm organic peroxides that are stored under refrigeration and promote their decomposition.<ref name="YN-2-91">Yu and Newman (2008), p. 2-91.</ref> | |||
Acetylene cylinders need special care in fire. | Acetylene cylinders need special care in fire. The fusible-metal relief devices described for United States and Canadian cylinders operate at about 100 °C (212 °F). Above about 300 °C (570 °F), acetylene begins to decompose rapidly into carbon and hydrogen, releasing much energy.<ref name="Lem-6-178" /> The decomposition can be so fast that the relief plugs cannot always prevent rupture, which can occur after a few minutes or up to several hours.<ref name="Lem-6-178" /> Lemoff (2008) advises spraying fire-exposed cylinders from a protected location and keeping people away; he warns that they can still explode 24 hours after exposure.<ref name="Lem-6-178" /> He reports that most European cylinders at that time lacked pressure-relief devices and were more susceptible to rupture.<ref name="Lem-6-178" /> | ||
== Limits and common misconceptions == | == Limits and common misconceptions == | ||
Instability is not the same as flammability. Nitromethane is only a combustible liquid, yet undiluted nitromethane can detonate under certain conditions of heat, pressure, shock and contamination, and NFPA 704 rates the two hazards in separate quadrants.<ref name="Dav-7-49" /><ref name="Wr-13-123" /> The upper flammable limit of 100% sometimes quoted for acetylene, instead of 81%, reflects the heat-producing decomposition that can occur without air; Lemoff (2008) calls this a misapplication of the flammable range concept.<ref name="Lem-6-177" /> In the same way, the explosions of ethylene oxide mixtures above 80% are decompositions, not combustion.<ref name="Dav-7-49" /> | Instability is not the same as flammability. Nitromethane is only a combustible liquid, yet undiluted nitromethane can detonate under certain conditions of heat, pressure, shock and contamination, and NFPA 704 rates the two hazards in separate quadrants.<ref name="Dav-7-49" /><ref name="Wr-13-123" /> The upper flammable limit of 100% sometimes quoted for acetylene, instead of 81%, reflects the heat-producing decomposition that can occur without air; Lemoff (2008) calls this a misapplication of the flammable range concept.<ref name="Lem-6-177" /> In the same way, the explosions of ethylene oxide mixtures above 80% are decompositions, not combustion.<ref name="Dav-7-49" /> | ||
Dilution and inhibitors help but do not remove the hazard. | Dilution and inhibitors help but do not remove the hazard. Davenport (2008) reports that many organic-peroxide formulations are diluted enough to prevent hazardous reactions, while others still need special precautions.<ref name="Dav-7-49" /> Inhibited monomers can still polymerise at fire temperatures.<ref name="Dav-7-49" /> Davenport also calls dilute hydrogen peroxide less chemically stable than concentrated material; that stability comparison alone does not establish which solution has the greater fire severity.<ref name="Dav-7-46" /> Small changes can matter: in the Tokai, Japan, incident of 11 March 1997, an excessive concentration of sodium bicarbonate lowered the reaction onset temperature of a mixture of asphalt and sodium salts.<ref name="Zal-2-102" /> | ||
The quantitative picture has limits too. The Semenov result is a first approximation that assumes a uniform temperature and a single reaction, and real values of the activation energy are often unknown or uncertain.<ref name="Gra-2-212" /><ref name="Gra-2-215" /> The classes and quantities given here are United States rules as presented in the 2008 edition of the ''Fire Protection Handbook''. The NFPA 704 scale is taken from its 2001 edition and the NFPA 55 limits from its 2005 edition.<ref name="Wr-13-123" /><ref name="LR-7-39" /> Relief venting also depends on correct installation, and Zalosh (2008) reports a poor record of many facilities in installing and maintaining pressure relief devices properly.<ref name="Zal-2-102" /> | The quantitative picture has limits too. The Semenov result is a first approximation that assumes a uniform temperature and a single reaction, and real values of the activation energy are often unknown or uncertain.<ref name="Gra-2-212" /><ref name="Gra-2-215" /> The classes and quantities given here are United States rules as presented in the 2008 edition of the ''Fire Protection Handbook''. The NFPA 704 scale is taken from its 2001 edition and the NFPA 55 limits from its 2005 edition.<ref name="Wr-13-123" /><ref name="LR-7-39" /> Relief venting also depends on correct installation, and Zalosh (2008) reports a poor record of many facilities in installing and maintaining pressure relief devices properly.<ref name="Zal-2-102" /> | ||
Latest revision as of 08:37, 13 November 2026
An unstable reactive material can undergo violent chemical change when exposed to heat, pressure, shock or contamination.[1][2] Examples include rapid polymerisation or decomposition, and NFPA 55 also names vigorous condensation in its definition for gases.[1] NFPA 704 describes instability as a property of the material in itself: some such materials can detonate at normal temperatures and pressures, while others only become unstable when heated.[3] Typical examples are acetylene, ethylene oxide, organic peroxides, monomers such as styrene that can polymerise, nitromethane and concentrated hydroxylamine.[4][5][6] In the editions presented by the 2008 handbook, NFPA 704 rates instability from 0 to 4 and NFPA 55 lists gas Classes 1 to 4.[3][1][7] Davenport (2008) advises detached storage where possible, with pressure relief and explosion venting.[2] He recommends water and sprinklers for fires involving unstable chemicals and manual attack from a protected distance; water can warm refrigerated peroxides.[4][8]
Definition and classes
The handbook definition comes from NFPA 55 and is written for gases. Lemoff (2008) describes unstable reactive gases as gases that will vigorously polymerise, decompose or condense, become self-reactive, or otherwise undergo a violent chemical change under conditions of shock, pressure or temperature.[1] A Class 2 gas readily undergoes violent chemical change at elevated temperatures and pressures.[1] A Class 3 gas is itself capable of detonation, explosive decomposition or explosive reaction, but needs a strong initiating source or heating under confinement.[1] A Class 4 gas is readily capable of detonation, explosive decomposition or explosive reaction at normal temperatures and pressures.[1] NFPA 55 lists detonable Classes 3 and 4 separately from nondetonable Classes 2 and 3; Class 3 can thus appear in either category, with different storage limits.[1][7]
The NFPA 704 instability scale applies beyond gases. In the table reproduced by C. J. Wright (2008a) from the 2001 edition of NFPA 704, degree 4 covers materials readily capable of detonation or explosive decomposition at normal temperatures and pressures, including materials sensitive to localised thermal or mechanical shock.[3] The descriptions of degrees 3 and 2 parallel those of NFPA 55 gas Classes 3 and 2, but the two rating systems have different purposes.[3][1] Degree 1 covers materials that are normally stable but can become unstable at elevated temperatures and pressures, and degree 0 those that remain stable even under fire conditions.[3] The handbook uses several names for this quadrant: "reactivity hazard" in the text, "instability hazard" in the table and "self-reactivity (instability)" in the chapter on medical gases.[3][9]
Other United States rules use related terms. In the workplace regulation (29 CFR 1910) summarised by Hildebrand and Noll (2008), OSHA names "unstables (reactives)" among examples of a physical hazard.[10] In transport, Division 4.1 includes self-reactive materials, which are liable to undergo a strongly exothermal decomposition at normal or elevated temperatures because of excessively high transport temperatures or contamination.[11] Under the Environmental Protection Agency rules summarised by Severson (2008), a waste is reactive if it is normally unstable and readily undergoes violent change without detonating.[12] A waste that can detonate, readily or under a strong initiating source, is reactive as well.[12]
How unstable materials release energy
Ordinary combustion is a set of exothermic reactions between a fuel and an oxidiser, usually oxygen from the air, and it needs fuel, oxygen and sufficient heat.[13][14] The decomposition of an unstable material does not need oxygen from the air; for acetylene, a hazardous, heat-producing reaction can occur in its absence.[6] When reactive groups such as the nitrate group are built chemically into a fuel, as in cellulose nitrate or trinitrotoluene (TNT), the resulting compound can be highly unstable and decompose violently.[15] Zalosh (2008) names the decomposition and polymerisation of inherently unstable compounds among the reactions that can proceed as fast as or faster than hydrocarbon combustion at elevated temperatures and produce several gaseous products.[5] Several of these reactions are autocatalytic: their products speed up the reaction of the remaining material.[5] Such reactions cause explosions in two ways: by producing gas faster than a vessel can vent it, or by a runaway exothermic reaction that rapidly raises the pressure of the gases and vapours already present.[5]
The reaction can be started in several ways. Acetylene owes its reactivity to an easily broken triple bond, and it can decompose rapidly into carbon and hydrogen with the release of heat.[6] The decomposition can be initiated by heat, and in the liquid or solid state also by mechanical impact.[6] Many organic peroxides decompose under heat, shock or friction, at a rate that depends on the formulation and the temperature.[2] Contamination acts as a trigger too: ethylene oxide polymerises violently when catalysed by anhydrous chlorides of iron, tin or aluminium, by iron rust, aluminium oxide or alkali metal hydroxides.[2] Light appears to be a more important factor than heat in the formation of peroxides in stored ethers, and freezing a peroxide solution can form sensitive crystals.[2][4]
Pressure changes the picture as well. Liquids such as propargyl bromide and nitromethane simply boil away when heated at atmospheric pressure, but under pressure their boiling point can rise until the liquid decomposes violently and the container explodes.[16] Pure acetylene, ethylene and nitrous oxide also decompose violently under pressure.[16] For acetylene, the higher the pressure, the easier decomposition is to initiate and the more violent its effects.[6] The reaction can end in detonation, as in the hydroxylamine incidents described below.[5]
Thermal runaway and critical conditions
Some unstable materials react slowly even at ordinary temperatures, as styrene does, and the hazard arises when the heat released is not removed.[4] Gray (2002) explains this with the balance between heat release and heat loss. Most chemical reaction rates rise almost exponentially with temperature, whereas heat loss by conduction rises only linearly.[17] Above the critical ambient temperature for a given body, heat release exceeds loss and the temperature rises; in a chemical reactor this is called a runaway reaction or thermal runaway.[17][18] Below it, a stable balance point exists, usually 5 to 20 °C above the ambient temperature, at which the material self-heats without igniting but may degrade.[18]
The simplest theory, due to Semenov, assumes a uniform temperature in the reacting mass, as in a well-stirred liquid, and represents heat generation with a single effective reaction whose rate follows the Arrhenius form .[19][20][21] That effective rate need not correspond to only one chemical step.[21] At the critical condition the heat release and heat loss curves just touch, and the mass is warmer than its surroundings by a critical amount. Gray gives that critical temperature rise as:[22]
| Symbol | Description | Unit |
|---|---|---|
| Critical temperature rise of the material above its surroundings | K | |
| Temperature of the material at the critical condition | K | |
| Critical ambient temperature | K | |
| Activation energy of the reaction | J/mol | |
| Universal gas constant | J/(mol·K) | |
| Pre-exponential factor of the reaction rate | depends on the reaction |
The result holds only at criticality and is independent of the shape of the body.[22] Criticality exists only if the activation energy is large enough, ; otherwise the temperature rises smoothly without an abrupt change.[22] Gray notes that R/E is about 0.0001 for most combustion reactions, and that the critical rise would typically be 20 to 30 °C for ambient temperatures around 30 to 40 °C.[22]
As a worked example, take a material stored at a hypothetical critical ambient temperature of 30 °C (303 K). Assume its decomposition has E/R = 10,000 K, corresponding to R/E = 0.0001; Gray (2002) gives that ratio for most combustion reactions, but it is not measured here for a named chemical.[22] The model's critical rise is 303² / 10,000 = 9.2 K, provided the body size, heat transfer and reaction rate also place the system at the critical condition.[22] The body temperature at that point would be about 39 °C; a 9 K rise by itself is not a general runaway threshold. With the same activation energy, a rise of 10 K above 303 K multiplies the reaction rate by about 2.9, and a rise of 20 K by about 7.7. The typical 20 to 30 °C quoted by Gray corresponds to smaller activation energies, with E/R of roughly 3,100 to 4,900 K. This sensitivity is why hazard reduction for less-than-explosive instability concentrates on keeping the material from being heated to an unsafe level.[23]
Ouellette (2008) describes the same chain in stored, unstabilised polymer powders: decomposition liberates heat, the bulk temperature rises, the decomposition speeds up, and the process continues until the material reaches its self-accelerating decomposition temperature.[24] Monitoring the temperature of the containers is therefore important during storage.[24] Hydrogen peroxide shows the temperature effect in numbers: its decomposition rate increases about 1.5 times for each 10 °C (18 °F) rise, and near the boiling point it becomes very rapid.[25]
Typical unstable reactive materials
Acetylene is classified as reactive, flammable and compressed, and the handbook shows it with 0 for health, 4 for flammability and 3 for reactivity on the NFPA 704 diamond.[6] It can react with copper and some copper alloys to form metallic acetylides, extremely shock-sensitive explosives that can initiate its decomposition. Lemoff (2008) identifies brass with less than 65% copper as usable for acetylene piping, while allowing some other copper-containing components under specific conditions.[6] Methylacetylene and propadiene are unstable, reactive materials with hazards similar to acetylene; their commercial mixture is made stable enough by its content of propane and butanes.[26]
Ethylene oxide is both flammable and unstable. Heat from its exothermic reactions with alcohols, acids, ammonia and many other compounds can start the polymerisation of the unreacted ethylene oxide, and its vapours can detonate if an initiating heat source is present.[2] Its flammable range in air is 3 to 80%; reported explosions of mixtures above 80% are the result of chemical decomposition.[2] Hydrogen cyanide tends to polymerise, a reaction catalysed by alkaline materials; because one product, ammonia, is itself alkaline, an explosive reaction eventually follows unless an acid is added to neutralise it.[2]
Several monomers polymerise with heat. Ethyl acrylate, methyl acrylate, methyl methacrylate and vinylidene chloride can polymerise at the elevated temperatures of a fire and violently rupture a closed container, and they usually contain an inhibitor.[2] Styrene polymerises slowly at ordinary temperatures; because the reaction is exothermic and accelerated by its own heat, it eventually becomes violent unless inhibitors are added.[4] Vinyl chloride, a toxic and flammable gas, behaves in a similar way in fire.[4]
Organic peroxides are a widely used group, serving as polymerisation initiators, flour bleaches and catalysts.[2] All are combustible and increase the intensity of a fire, and some, such as methyl ethyl ketone peroxide, are detonatable.[2] Undiluted benzoyl peroxide burns with great rapidity, similar to black powder, and decomposes explosively if confined when heated.[4] Ethers form peroxides during storage, and pure, dry ether in a clear bottle develops detectable peroxides within one month; such peroxides can detonate when heated or concentrated.[4]
Nitromethane is a combustible liquid that decomposes explosively at 315 °C (599 °F) and about 6.31 MPa gauge (915 psig).[2] Railroad tank cars of nitromethane detonated at Niagara Falls, New York, and at Mt. Pulaski, Illinois, in 1958.[2] The causes were not definitely determined; contamination by material previously carried in the tanks is a possible explanation.[2]
Concentrated hydroxylamine caused two of the incidents listed by Zalosh (2008).[5] At Hanover Township, Pennsylvania, on 19 February 1999, 5 people died and 14 were injured; at Gunma, Japan, on 10 June 2000, 4 died and 58 were injured.[5] In both, the aqueous concentration was significantly higher than the recommended practice of less than 50%, and tests show that such concentrations with trace contaminants can decompose very violently, often as a detonation.[5]
Storage, handling and process safety
Davenport (2008) lists four points that are specific to unstable chemicals in storage.[2] These are the catalytic effect of containers, other materials in the same area that could start a dangerous reaction, the presence of inhibitors, and the effect of direct sunlight or temperature changes.[2] Containers need pressure-relief vents and the storage area needs explosion venting, in addition to sprinkler or water spray protection and the removal of all combustible material.[2] Whenever possible, unstable chemicals should be stored in a detached outside location, and fire fighters should be briefed in advance on the procedures for a fire involving the store.[2]
For unstable flammable liquids, Slye (2008) notes that greater distances to property lines and between tanks, or extra fire protection, may be needed.[27] Where explosions of flammable vapours are possible, explosion venting is recommended for rooms holding at least Class IA liquids and unstable liquids.[28] Materials of construction must not combine with the liquid or contaminate it. Aluminium, for example, is avoided with a chlorinated compound that slowly releases hydrochloric acid, because the aluminium chloride formed may act as a catalyst.[29]
For gases, NFPA 55 (2005 edition), as described by Lemoff and Rivkin (2008), keeps unstable reactive gases of Classes 2, 3 and 4 at 6.10 m (20 ft) from other gases.[30] Gases classed simply as "other gas" need no separation.[30] The distance can be reduced without limit by a noncombustible barrier at least 1.52 m (5 ft) high with a fire resistance rating of at least half an hour.[30] The same standard sets maximum allowable quantities per control area, a space separated by a fire barrier.[30] For detonable Class 3 or 4 gases the quantity is zero in unsprinklered areas.[7] In sprinklered areas it is 0.454 kg (1 lb) of liquefied or 0.283 m³ (10 ft³) of nonliquefied gas, doubled when a gas cabinet is used, while Class 1 gases have no limit.[7]
The acetylene cylinder shows how a compressed gas cylinder can be designed around an unstable gas. It is filled with a porous mass saturated with acetone or dimethylformamide, so that the gas is stored in solution and any free gas volume is small, which limits the decomposition energy available.[6] Lemoff (2008) reports a maximum charging pressure of 1724 kPa gauge (250 psig) at 21 °C (70 °F) under the United States and Canadian transport rules he describes.[6] He gives a general limit of 103 kPa gauge (15 psig) for most acetylene piping, with specially designed higher-pressure applications.[6] His chapter says the conventional codes of the time prohibit liquid or solid acetylene use.[31]
In process plants, the AIChE Center for Chemical Process Safety guidelines summarised by Zalosh (2008) combine four elements.[5] These are lists of unstable materials, software that estimates reaction energy, laboratory screening of reaction onset temperatures and sensitivities, and expert review.[5] Schwab (2008a) adds that most chemicals are reasonably stable, that data for the others come from suppliers, NFPA 704 and reactive-chemical handbooks, and that thermodynamic calculations and laboratory tests are needed where data are missing.[23] Ural and Garzia's (2008) test list includes ASTM E476 for confined-system thermal instability, E537 for thermal stability by thermal analysis and E698 for Arrhenius kinetic constants.[32] Where the hazard is less than explosive, it is reduced by keeping the material from overheating, for example by lowering a boiling point under vacuum or by vacuum or freeze drying.[23]
Hazard indexes build on the same ratings. In the Dow Fire and Explosion Index described by Watts (2002), the material factor is a number from 1 to 40.[33] It is determined from the flammability and reactivity of the material as described in NFPA 704 (1990 edition) and related NFPA data.[33]
Fire fighting
Davenport (2008) recommends water for fires involving unstable chemicals, including organic peroxides.[4] He calls automatic sprinklers the best protection, because such fires can lead to explosions, and calls for manual fire fighting from a protected distance.[4] A tank of ethylene oxide exposed to fire can heat its vapours rapidly to their ignition temperature unless it is kept wet with water spray.[2] Water at room temperature can, however, warm organic peroxides that are stored under refrigeration and promote their decomposition.[8]
Acetylene cylinders need special care in fire. The fusible-metal relief devices described for United States and Canadian cylinders operate at about 100 °C (212 °F). Above about 300 °C (570 °F), acetylene begins to decompose rapidly into carbon and hydrogen, releasing much energy.[31] The decomposition can be so fast that the relief plugs cannot always prevent rupture, which can occur after a few minutes or up to several hours.[31] Lemoff (2008) advises spraying fire-exposed cylinders from a protected location and keeping people away; he warns that they can still explode 24 hours after exposure.[31] He reports that most European cylinders at that time lacked pressure-relief devices and were more susceptible to rupture.[31]
Limits and common misconceptions
Instability is not the same as flammability. Nitromethane is only a combustible liquid, yet undiluted nitromethane can detonate under certain conditions of heat, pressure, shock and contamination, and NFPA 704 rates the two hazards in separate quadrants.[2][3] The upper flammable limit of 100% sometimes quoted for acetylene, instead of 81%, reflects the heat-producing decomposition that can occur without air; Lemoff (2008) calls this a misapplication of the flammable range concept.[6] In the same way, the explosions of ethylene oxide mixtures above 80% are decompositions, not combustion.[2]
Dilution and inhibitors help but do not remove the hazard. Davenport (2008) reports that many organic-peroxide formulations are diluted enough to prevent hazardous reactions, while others still need special precautions.[2] Inhibited monomers can still polymerise at fire temperatures.[2] Davenport also calls dilute hydrogen peroxide less chemically stable than concentrated material; that stability comparison alone does not establish which solution has the greater fire severity.[25] Small changes can matter: in the Tokai, Japan, incident of 11 March 1997, an excessive concentration of sodium bicarbonate lowered the reaction onset temperature of a mixture of asphalt and sodium salts.[5]
The quantitative picture has limits too. The Semenov result is a first approximation that assumes a uniform temperature and a single reaction, and real values of the activation energy are often unknown or uncertain.[17][20] The classes and quantities given here are United States rules as presented in the 2008 edition of the Fire Protection Handbook. The NFPA 704 scale is taken from its 2001 edition and the NFPA 55 limits from its 2005 edition.[3][7] Relief venting also depends on correct installation, and Zalosh (2008) reports a poor record of many facilities in installing and maintaining pressure relief devices properly.[5]
See also
- Pyrophoric material
- Water-reactive material
- Chemical incompatibility
- Explosive material
- Ammonium nitrate
Notes
- ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Lemoff (2008), p. 6-166.
- ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10 2.11 2.12 2.13 2.14 2.15 2.16 2.17 2.18 2.19 2.20 2.21 2.22 Davenport (2008), p. 7-49.
- ↑ 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 C. J. Wright (2008a), p. 13-123.
- ↑ 4.00 4.01 4.02 4.03 4.04 4.05 4.06 4.07 4.08 4.09 Davenport (2008), p. 7-50.
- ↑ 5.00 5.01 5.02 5.03 5.04 5.05 5.06 5.07 5.08 5.09 5.10 5.11 Zalosh (2008), p. 2-102.
- ↑ 6.00 6.01 6.02 6.03 6.04 6.05 6.06 6.07 6.08 6.09 6.10 Lemoff (2008), p. 6-177.
- ↑ 7.0 7.1 7.2 7.3 7.4 Lemoff and Rivkin (2008), p. 7-39.
- ↑ 8.0 8.1 Yu and Newman (2008), p. 2-91.
- ↑ Allen (2008), p. 6-194.
- ↑ Hildebrand and Noll (2008), p. 13-82.
- ↑ C. J. Wright (2008a), p. 13-120.
- ↑ 12.0 12.1 Severson (2008), pp. 7-60–7-61.
- ↑ Van Rossum (2027), §4.1.
- ↑ Van Rossum (2027), §4.2.
- ↑ Drysdale (2008), p. 2-17.
- ↑ 16.0 16.1 Schwab (2008a), p. 8-25.
- ↑ 17.0 17.1 17.2 Gray (2002), p. 2-212.
- ↑ 18.0 18.1 Gray (2002), p. 2-214.
- ↑ Gray (2002), p. 2-213.
- ↑ 20.0 20.1 Gray (2002), p. 2-215.
- ↑ 21.0 21.1 Gray (2002), p. 2-216.
- ↑ 22.0 22.1 22.2 22.3 22.4 22.5 Gray (2002), p. 2-217.
- ↑ 23.0 23.1 23.2 Schwab (2008a), p. 8-24.
- ↑ 24.0 24.1 Ouellette (2008), p. 9-63.
- ↑ 25.0 25.1 Davenport (2008), p. 7-46.
- ↑ Lemoff (2008), p. 6-184.
- ↑ Slye (2008), p. 6-198.
- ↑ Slye (2008), p. 6-207.
- ↑ Slye (2008), p. 6-205.
- ↑ 30.0 30.1 30.2 30.3 Lemoff and Rivkin (2008), p. 7-38.
- ↑ 31.0 31.1 31.2 31.3 31.4 Lemoff (2008), p. 6-178.
- ↑ Ural and Garzia (2008), p. 17-146.
- ↑ 33.0 33.1 Watts (2002), p. 5-129.
References
- Allen, M. (2008). Medical gases. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-187–6-196). National Fire Protection Association.
- Davenport, J. A. (2008). Storage and handling of chemicals. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 7-43–7-55). National Fire Protection Association.
- Drysdale, D. (2008). Physics and chemistry of fire. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 2-3–2-20). National Fire Protection Association.
- Gray, B. (2002). Spontaneous combustion and self-heating. 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. 2-211–2-228). National Fire Protection Association; Society of Fire Protection Engineers.
- Hildebrand, M. S., & Noll, G. G. (2008). Public fire protection and hazmat management. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 13-81–13-98). National Fire Protection Association.
- Lemoff, T. C. (2008). Gases. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-165–6-186). National Fire Protection Association.
- Lemoff, T. C., & Rivkin, C. H. (2008). Storage of gases. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 7-35–7-42). National Fire Protection Association.
- 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.
- Schwab, R. F. (2008a). Chemical processing equipment. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 8-17–8-29). National Fire Protection Association.
- Severson, P. R. (2008). Hazardous waste control. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 7-57–7-69). National Fire Protection Association.
- Slye, O. M., Jr. (2008). Flammable and combustible liquids. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 6-197–6-208). National Fire Protection Association.
- Ural, E. A., & Garzia, H. W. (2008). Explosion prevention and protection. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 17-141–17-165). National Fire Protection Association.
- Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.
- Watts, J. M., Jr. (2002). Fire risk indexing. 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. 5-125–5-142). National Fire Protection Association; Society of Fire Protection Engineers.
- Wright, C. J. (2008a). Managing the response to hazardous material incidents. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 13-117–13-144). National Fire Protection Association.
- Yu, H.-Z., & Newman, J. S. (2008). Theory of fire extinguishment. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 2-79–2-92). National Fire Protection Association.
- Zalosh, R. (2008). Explosions. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 2-93–2-106). National Fire Protection Association.