Fire Chemistry: Combustion, Oxidation and the Fire Tetrahedron

The fire triangle, taught in elementary school science curricula worldwide, identifies three requirements for combustion: heat, fuel, and oxygen. The triangle served fire prevention education for decades. It was superseded as a professional model when suppression chemistry demonstrated that certain agents could extinguish a fire without significantly reducing temperature or oxygen concentration. The fourth face, the uninhibited chemical chain reaction, captures what the triangle misses.

Fuel is any material that can be oxidised to release energy. In forensic fire investigation, fuel mapping is a core task at every scene. Structural fuels (timber framing, engineered wood products, insulation, roofing felt) are distributed by the construction design. Contents fuels (furniture, electronics, clothing, stored liquids) are distributed by the occupant. The burn pattern that survives the fire reflects where fuel was densest, where it was arranged to create pathways, and whether an unusually intense burn at a geometrically improbable location suggests an introduced fuel (the classic accelerant pour pattern scenario).

Oxidiser in most structural fires is atmospheric oxygen at roughly 21 per cent by volume. Oxygen concentration governs whether combustion is fuel-controlled or ventilation-controlled, a distinction examined in the next topic in this module. At standard atmospheric pressure, combustion slows markedly below about 15 per cent oxygen and extinguishes below approximately 12 to 14 per cent in flaming combustion. Smouldering combustion, which involves surface oxidation rather than gas-phase flame, can persist at lower oxygen concentrations, which is why a deeply buried smouldering fire can survive beneath debris apparently starved of air.

Heat is the energy input that raises fuel temperature to its ignition point, sustaining the reaction against heat losses to the environment. Once a fire is established, most of its heat is self-generated by the exothermic reaction, but the initial heat source is critical to origin determination. Competent fire investigation requires characterising both the ignition source (an open flame, an electrical arc, a hot surface, a chemical reaction) and the first fuel ignited.

Uninhibited chain reaction refers to the free-radical propagation mechanism that sustains flaming combustion. The gas-phase combustion of a hydrocarbon fuel proceeds through a cascade of intermediate steps, each producing free radicals (hydroxyl radical OH, hydrogen atom H, oxygen atom O) that attack adjacent fuel molecules, liberating more radicals and more energy. Halogenated suppression agents (historically Halon 1301 and Halon 1211, now largely replaced by cleaner alternatives such as FK-5-1-12 and IG-541 in new installations) work by scavenging these radicals, breaking the chain without cooling the environment below ignition temperature. CO2 and water mist act primarily by the other three mechanisms (oxygen displacement, fuel surface cooling, and heat absorption).

Combustion is a chemical reaction, and chemical reaction rates obey the Arrhenius equation. In its simplest form, the rate constant k for a reaction is:

k = A × exp(-Ea / RT)

where A is the pre-exponential frequency factor, Ea is the activation energy of the reaction (in joules per mole), R is the universal gas constant (8.314 J mol⁻¹ K⁻¹), and T is the absolute temperature in kelvin. The critical implication for fire science is the exponential dependence on temperature: as T rises, the exp term grows rapidly, meaning reaction rate increases dramatically with small temperature increments. A rule of thumb used in fire engineering is that for many organic oxidation reactions, a 10-kelvin temperature rise approximately doubles the reaction rate.

This temperature dependence explains why fire growth is not linear but acceleratory. A small ignition source raises local temperature, the elevated temperature accelerates oxidation in adjacent fuel, the accelerated oxidation generates more heat, which raises temperature further. This positive feedback loop is what transforms a localised ignition into a room fire in tens of seconds, and into a fully developed compartment fire in minutes. The Arrhenius relationship also explains why cooling is an effective suppression strategy: reducing temperature by even a modest amount reduces the reaction rate significantly, breaking the feedback loop.

For forensic fire investigators, Arrhenius kinetics underlie the concept of ignition temperature. Every fuel has a characteristic minimum temperature at which the gas-phase oxidation reaction becomes self-sustaining. Below the ignition temperature, combustion reactions occur but release heat more slowly than it is lost to the environment; above it, the heat release rate exceeds heat loss and the fire is self-sustaining. This is distinct from the auto-ignition temperature (the spontaneous ignition temperature in the absence of a pilot flame) and the flash point (the minimum temperature at which a liquid produces sufficient vapour for ignition by a pilot flame), which are important in accelerant investigation.

Activation energy varies between fuel types. Cellulosic fuels (wood, paper, cotton) have complex multi-step decomposition pathways with different activation energies at different temperature ranges. Synthetic polymers (polyurethane foam, polyethylene, polypropylene) typically pyrolyse to simpler monomeric or oligomeric fragments before burning, and their combustion kinetics differ markedly from cellulose. Understanding these differences helps the investigator interpret why different fuels leave different char patterns, different char depths per unit time of burning, and different residues.

Combustion is not a single process. Two distinct modes are encountered in forensic fire investigation, and they leave different patterns, produce different hazardous gases, and respond differently to suppression.

Flaming combustion (also called gas-phase combustion) occurs when the volatile pyrolysis products of a fuel reach the gas phase and react with oxygen above the solid or liquid surface, producing a visible flame. The flame is the zone of intense chemical reaction: temperatures in a diffusion flame (the type produced by most solid fuels in air, where fuel and oxidiser are not pre-mixed) range from approximately 800°C at the outer cool edge to 1,200°C or more in the reaction zone. Light is emitted because carbon particles heated to incandescence in the flame produce the characteristic yellow-orange luminosity of an open fire. Flame temperature is high enough to cause soot particles to oxidise on exit, but in fuel-rich or ventilation-limited conditions soot survives and is deposited as black smoke.

Glowing combustion (also called smouldering or surface combustion) occurs at the solid surface without a gas-phase flame. Oxygen from the environment diffuses to the surface of the fuel, where it reacts directly with the solid char or carbonised material. This reaction is slower and lower-temperature than flaming combustion, typically 400 to 700°C at the surface. Glowing combustion generates large amounts of CO and other partially oxidised products because the lower temperature and restricted oxygen access prevent complete oxidation to CO2 and water. This is the mechanistic reason why smouldering fires are disproportionately responsible for fire fatalities: a smouldering upholstered chair in a closed room can fill the space with lethal CO concentrations without producing a visible flame that would trigger a smoke alarm or alert occupants.

The transition between the two modes is significant in fire investigation. A smouldering fire can transition to flaming combustion when: additional air reaches the smouldering surface (a door opening, a window breaking), the temperature at the smouldering front accumulates to a threshold at which volatilisation and gas-phase ignition occur, or a flame from an adjacent source contacts the smouldering fuel. This transition can appear as a sudden, intense fire to witnesses, sometimes misinterpreted as an explosion or as evidence of an introduced accelerant. The investigator must consider the scenario of a slow-building smoulder transitioning to flaming combustion before concluding that a witness account of a sudden intense fire is inconsistent with accidental ignition.

In the United Kingdom, the work of the Building Research Establishment (BRE) and the Fire Research Station has extensively documented smouldering transitions in residential upholstered furniture, a key driver of the UK Furniture and Furnishings (Fire) (Safety) Regulations 1988 and subsequent amendments. In the United States, NIST and Underwriters Laboratories have published parallel research characterising the smoulder-to-flame transition in upholstery and mattress fires.

Smoke is not a single substance. It is a heterogeneous aerosol of solid particles, liquid droplets, and gases produced by incomplete combustion and pyrolysis of the fuel load. Its composition varies with the fuel type, the combustion mode (flaming or smouldering), the temperature regime, and the oxygen availability. For forensic investigators, smoke chemistry has two direct applications: understanding what hazardous gases were present (relevant to victim toxicology and cause of death determination) and understanding what residues the smoke deposits on surfaces (relevant to fire pattern analysis and accelerant screening).

Carbon monoxide (CO) is the most important fire gas from a toxicological perspective. It is produced whenever incomplete combustion occurs: insufficient oxygen, reduced temperature, or fuel-rich conditions all increase CO yield. CO binds to haemoglobin with an affinity approximately 200 times that of oxygen, forming carboxyhaemoglobin (COHb) and impairing oxygen delivery to tissues. COHb levels above 30 to 35 per cent are typically incapacitating; levels above 50 to 60 per cent are typically lethal, though individual tolerance varies. Forensic pathologists in the US, UK, Australia, India, and across the EU routinely measure blood COHb in fire fatalities; elevated COHb in a fire victim confirms alive-in-fire exposure and is inconsistent with post-mortem body placement at the scene.

Hydrogen cyanide (HCN) is produced by the combustion of nitrogen-containing polymers, particularly polyurethane (PU) foam, which is ubiquitous in modern residential upholstered furniture, nylon fabrics, and certain carpet backings. HCN inhibits cytochrome c oxidase, the final enzyme of the mitochondrial electron transport chain, producing cellular asphyxia. At the concentrations encountered in compartment fires involving modern furnishings, HCN can produce incapacitation faster than CO, though the two agents act synergistically and most fire fatality investigations measure both. Postmortem blood cyanide is less stable than COHb and must be collected promptly, a consideration for forensic pathologists and medical examiners in all jurisdictions.

Soot in fire debris is both a marker of combustion completeness and a substrate for chemical analysis. Soot particles consist largely of polycyclic aromatic hydrocarbons (PAHs) adsorbed onto elemental carbon cores. The PAH profile of soot can, in some cases, provide information about the fuel type. More directly relevant to arson investigation, soot deposits on ceilings, walls, and surfaces form the smoke horizon that fire investigators use to reconstruct the hot gas layer height at the time of smoke deposition, an input into fire modelling.

Acrolein and other irritants are produced by the combustion of specific polymers. Acrolein (propenal, CH2=CH-CHO) is a potent upper respiratory irritant generated by the thermal decomposition of polyurethane and polyethylene. PVC combustion produces hydrogen chloride (HCl) gas, which causes immediate upper airway irritation and delayed-onset pulmonary injury. These gases are forensically relevant because they can cause rapid incapacitation at sub-lethal concentrations, explaining why fire victims are sometimes found in locations where escape appeared physically possible.

Heat moves from hotter to cooler regions by three mechanisms. In a developing compartment fire, all three operate simultaneously and interactively. Understanding each mode is essential for interpreting the geometry of burn patterns: why ceiling damage appears before floor damage, why the area directly above an ignition point is typically the most severely burned, and why remote ignition (apparent fire origin in a location that never had a direct flame connection to any fuel) indicates either radiation-driven ignition or a separate ignition event.

Conduction is the transfer of heat energy through a solid material by molecular vibration. The rate of conductive heat transfer depends on the thermal conductivity of the material, the temperature gradient across it, and the area and thickness of the conducting path. Metals are excellent conductors; structural steel can transmit heat rapidly over significant distances. Wood, concrete, and brick are poor conductors compared to metal, but over extended fire durations, the temperature at the back face of a thick concrete wall will rise significantly due to conduction. In forensic fire investigation, conductive heat transfer through structural elements can explain fire spread between compartments via metal pipes, conduit, or steel beams, even when visible burn paths suggest an improbable spread route.

Convection is the dominant heat transfer mode in compartment fires during the growth phase. Hot combustion gases, being less dense than cooler air, rise by buoyancy. This buoyant plume carries thermal energy vertically, creating the characteristic pattern of greater ceiling damage directly above the ignition source before lateral spread. The hot gas layer that accumulates at ceiling level in a compartment fire transfers heat by convection to the ceiling and upper wall surfaces, and by radiation (see below) to the contents below. The convective component of fire spread is strongly influenced by ventilation geometry: openings at ceiling level allow hot gases to exit and cool air to enter at low level; sealed compartments accumulate the hot gas layer more rapidly, driving flashover conditions.

Convective flow also governs smoke transport through a building, which is a critical input into the investigator's reconstruction. By mapping where smoke was deposited on vertical surfaces (the smoke horizon), and by considering the geometry of vents, doors, and windows open during the fire, the investigator can often establish the sequence of events leading from ignition to full room involvement.

Radiation is the transfer of heat by electromagnetic waves, primarily in the infrared range, without requiring a material medium. All objects at temperatures above absolute zero emit thermal radiation; at fire temperatures (500 to 1,200°C), radiant heat flux from flames and hot surfaces can be intense enough to ignite remote fuels. The Stefan-Boltzmann law predicts that radiated power scales as the fourth power of absolute temperature, meaning a doubling of temperature (in kelvin) produces a 16-fold increase in radiated power. In practical compartment fire terms, this means that once a room reaches near-flashover temperatures (above 500 to 600°C in the hot gas layer), every exposed fuel surface receives radiant heat flux sufficient to drive pyrolysis and ignition simultaneously, regardless of proximity to the original flame.

Radiant ignition of remote targets is a recognised mechanism in forest fire spread (spotting) and in forensic investigation of fires where burn patterns suggest multiple simultaneous ignition points. The investigator must distinguish radiation-driven near-simultaneous ignition from multiple separate ignition events (a pattern associated with deliberate fire-setting with separated pour patterns) by reference to the spatial geometry, the hot gas layer height at the time of ignition, and the fuel arrangement.

Extinguishment by design targets one or more faces of the fire tetrahedron. In practice, most suppression agents act on more than one face simultaneously, and the forensic investigator must consider not only the pre-suppression fire behaviour but also how suppression activities modified the scene.

Fuel removal involves physically removing the combustible material (or cutting off its supply, as in closing a gas valve). Fuel starvation is rarely the primary suppression mechanism in structural fire-fighting but is central to wildfire management (fire lines, controlled burns ahead of the fire front) and to some industrial fire scenarios (flammable liquid fires where the supply can be isolated).

Oxidiser reduction (oxygen exclusion) is the operating principle of CO2, nitrogen, and inert gas flooding systems used in equipment rooms, server facilities, and vaults. These systems displace atmospheric oxygen to below the 12 to 14 per cent level needed for flaming combustion. In forensic investigation, finding evidence of a total-flooding system discharge is significant: it tells the investigator that the fire had reached a detection threshold before suppression, and it establishes a timeline relative to the fire alarm.

Heat removal is the primary mechanism of water application. Water has exceptional heat absorption capacity: its specific heat capacity (4.18 kJ kg⁻¹ K⁻¹) is higher than most materials, and its latent heat of vaporisation (2,260 kJ kg⁻¹) means that converting one kilogram of water to steam at 100°C absorbs more than 2.5 MJ of energy. Water mist systems exploit the fine droplet size to maximise the surface area available for heat exchange, accelerating cooling while reducing total water volume (critical for protecting water-sensitive equipment). The investigator considers the dilution and redistribution effects of fire-fighting water when interpreting chemical analysis of fire debris: accelerant residues may be diluted, displaced, or volatilised by the application of large water volumes during suppression.

Chain reaction interruption is the mechanism of halogenated agents. Halon 1301 (bromotrifluoromethane, CBrF3) was the dominant clean agent through the 1990s; under the Montreal Protocol, its production ceased in developed countries after 1994. Current clean agents in new installations include FK-5-1-12 (dodecafluoro-2-methylpentan-3-one, Novec 1230) and inert gas blends (IG-541: 52% N2, 40% Ar, 8% CO2). Halon 1301 systems installed before the phase-out remain in legacy installations in many countries, including in critical infrastructure sites in India, the United States, and the United Kingdom; fire investigators working in older buildings must be aware that a Halon discharge event may be part of the fire history.

1. Identify fuel type and load
   Map structural and contents fuels at the origin area. Document material composition where possible (cellulosic vs synthetic). Fuel type governs combustion chemistry and the toxicological gas mix produced.
2. Characterise the ignition source
   Identify the energy source (electrical, thermal, chemical, mechanical). Match ignition source temperature to the first fuel's ignition temperature. Inconsistency suggests an introduced ignition source or accelerant.
3. Assess oxidiser availability
   Reconstruct the ventilation geometry at the time of ignition. Window and door positions, HVAC state, and building envelope integrity all affect whether the fire was fuel-controlled or ventilation-controlled from the outset.
4. Interpret heat transfer patterns
   Read the burn pattern geometry for conductive, convective, and radiative signatures. Ceiling damage directly above a low-level origin is convective; remote surface char consistent with line-of-sight to a hot ceiling is radiative.
5. Account for suppression effects
   Document the timing and method of fire-fighting intervention. Water application affects accelerant distribution; agent flooding affects oxygen levels; timing affects what burn patterns could have developed before suppression.
6. Apply NFPA 921 / BS 4422 hypothesis testing
   Formulate one or more hypotheses for origin, cause, and spread. Test each against the physical, chemical, and witness evidence. A hypothesis that requires violation of basic combustion chemistry is excluded.