Soot

Soot is the carbonaceous particulate material produced mainly by incomplete combustion. It is a component of smoke rather than the whole smoke mixture, and contributes to dark colour, optical extinction and flame radiation. Its amount and properties depend on fuel chemistry, combustion mode, oxygen supply and transport.

Soot forms through growth processes in fuel-rich combustion regions and can be consumed later through oxidation. Flaming smoke often contains elemental graphitic carbon, while pyrolysis and smouldering also produce droplets with different properties. A complete description therefore distinguishes soot particles from other aerosol and gas components.

Soot matters because it can obscure escape routes, expose people and structures to radiation and provide a measurable indicator of incomplete combustion. Its effects depend on production per unit fuel, total fuel consumed and subsequent transport and deposition.

Formation in flames

Incomplete combustion leaves carbon in partially oxidised or elemental forms instead of converting it all to carbon dioxide. Soot grows in fuel-rich flame regions and can later burn out where oxygen and temperature are suitable. Emission reflects the balance between these formation and consumption processes.

Fuel chemistry affects this balance. Mulholland notes that aromatic polymers such as polystyrene produce more smoke than hydrocarbons with single carbon-carbon bonds in the cited comparison. Material structure therefore influences soot production alongside oxygen and heat.

The fuel-rich side of a hydrocarbon-air diffusion flame favours thermal decomposition and carbon formation. This region is associated with yellow luminosity from hot carbon particles. Other flame regions have different oxygen availability and can support further oxidation.

Soot forms and is consumed in different flame regions. Based on Mulholland (2002), p. 2-258 and Simmons (2002), pp. 1-151–1-152.

Soot production is not limited to large fires. Small diffusion flames can also emit particles, and weakly visible flames can indicate low rather than absent soot production. Appearance alone cannot quantify the source.

Soot yield

Soot yield is the mass of soot produced per unit mass of fuel burned, commonly expressed as kilograms of soot per kilogram of fuel or as a percentage. It is a source property under specified combustion conditions, not the concentration at a remote receiver. Concentration also depends on smoke volume and transport.

Van Rossum gives pre-flashover values of 1.5% for wood, 23% for polyurethane, 14% for PVC, 16% for polystyrene and 19.5% for expanded polystyrene. These values apply to oxygen excess in the cited source. They illustrate strong material dependence but should not be transferred unchanged to ventilation-limited fires.

Pre-flashover soot yields for selected materials. Based on Van Rossum (2027), §6.2.

Mulholland reports smoke conversion factors from 0.001 to 0.17 for flaming combustion and from 0.01 to 0.17 for pyrolysis and smouldering across the reviewed experiments. Most were measured on small samples under free-burning ambient conditions, and reduced ventilation can strongly affect them. The term smoke conversion factor includes a broader aerosol definition than soot alone.

A higher yield does not necessarily mean more total soot. Suppressed or smouldering fires can have higher yield per kilogram while consuming less fuel and producing less power. Total source mass therefore requires both yield and fuel-consumption history.

From fuel consumption to soot mass

Soot yield relates emitted soot mass to consumed fuel mass under specified combustion conditions. When released energy and an applicable effective heat of combustion are known, they also give the consumed fuel mass. A homogeneous-layer calculation then divides soot mass by layer volume to obtain concentration. $$m_s=y_s{m}_f=y_s\frac{E}{\mathrm{\Delta }{H}_{c,\mathrm{e}\mathrm{f}\mathrm{f}}},C_s=\frac{m_s}{V}$$

Symbol	Description	Unit
$m_s$	Soot mass entering the modelled layer	kg
$y_s$	Soot mass per mass of fuel consumed	kg/kg
$m_f$	Fuel mass consumed	kg
$E$	Energy released	MJ
$\mathrm{\Delta }{H}_{c,\mathrm{e}\mathrm{f}\mathrm{f}}$	Effective heat of combustion	MJ/kg
$C_s$	Soot mass concentration	kg/m³
$V$	Layer volume	m³

Van Rossum's shop example releases 120 MJ, uses 20 MJ/kg and distributes the products through 450 m³. Substituting the wood soot yield of 0.015 kg/kg consumes 6.0 kg of fuel and gives 0.090 kg of soot. The resulting soot concentration is 0.000200 kg/m³.

Using the book's decimal optical-density coefficient of 3,300 m²/kg gives 0.66 m⁻¹. Its non-luminous visibility approximation, 1 divided by that optical density, gives about 1.5 m. This is a source-based sensitivity example: changing the assumed yield changes visibility while the energy release and layer volume remain fixed.

The calculation assumes uniform mixing and no removal of soot from the selected layer. Deposition would remove airborne particles; coagulation alone changes their number and size without removing their combined mass. Optical properties can nevertheless change with particle size, shape and composition, so one mass coefficient is an approximation.

The wood value belongs to the pre-flashover conditions summarised by Van Rossum. His table compiles earlier material data through Robbins and Wade, so it does not establish a constant yield for every timber product or ventilation state. A different combustion regime requires evidence appropriate to that regime.

Optical density and visibility

Optical density describes light attenuation using a decimal logarithm. Van Rossum uses an average mass coefficient of 3,300 m²/kg with this decimal convention. The natural-logarithm extinction coefficient is larger by a factor of ln(10), so the same numerical coefficient cannot be used for both conventions.

Particle mass concentration is calculated from soot mass and the volume over which it is distributed. Uniform mixing is a modelling assumption, and real rooms can have strong spatial variation. Optical density along an actual sightline may therefore require integration rather than one room-average value.

Visibility in smoke also depends on object contrast, illumination and the observer. Soot particles block and scatter light, reducing the contrast needed to recognise an object or sign. More lighting can increase scattering and fail to improve recognition.

Van Rossum's worked shop-fire example uses 1 MW for 120 seconds, effective heat of combustion of 20 MJ/kg, soot yield of 0.1 kg/kg and a 450 m³ layer. It produces 0.6 kg of soot and an optical density of 4.4 per metre. The corresponding rough visibility estimates are 0.23 m for a non-luminous object and 0.57 m for a luminous one, showing how quickly a layer can become effectively opaque.

Soot yield strongly affects optical density and visibility. Based on Van Rossum (2027), §6.5.

The same example with wood soot yield of 0.015 kg/kg gives an optical density of 0.66 per metre and visibility of about 1.5 m. These calculations depend on the stated yield, volume and uniform-smoke assumptions. They should not be treated as expected values for another room.

Flame radiation

Flame radiation includes emission from hot soot particles, which radiate as approximate blackbody sources. This contributes to yellow and orange flame appearance and to heat transfer towards fuel and surrounding objects. Low soot production can make a flame difficult to see without eliminating its thermal hazard.

Soot radiation often contributes more than gaseous radiation in the situations discussed by Tien and colleagues. Water vapour, carbon dioxide and hydrocarbon gases also participate, so total emission depends on the complete medium. Soot is important but not the only radiating component.

Radiative feedback supports pyrolysis and fire spread by heating unburnt fuel. Hot soot in flames and upper layers can therefore influence objects beyond direct flame contact. The receiver's geometry and exposure time still determine its response.

Reduced-scale experiments can change the radiative share because flame size affects radiation relative to convection. Soot measurements and radiation data should consequently retain scale and configuration. A small test does not automatically represent a full-scale soot-radiation environment.

Particle size and transport

Particle size distribution influences optical properties, deposition and detection. Mulholland treats size distribution and smoke concentration as the principal determinants of many smoke properties. A distribution often spans several orders of magnitude and is conveniently shown on a logarithmic scale.

The geometric mean diameter and geometric standard deviation summarise many approximately log-normal distributions. These describe a distribution's central size and width rather than one identical particle size for the whole smoke. A monodisperse distribution has a geometric standard deviation of one.

During transport, particles can coagulate, partially evaporate or deposit through diffusion and sedimentation. Additional droplets can form through condensation. These processes change concentration and size distribution before smoke reaches a remote location.

Van Rossum notes that soot can settle while gases spread farther through a building. Visibility can then improve even though carbon monoxide remains high. Deposition reduces local particle concentration but does not demonstrate that the complete smoke hazard has disappeared.

Soot and other smoke components

Smoke includes soot, other particles, droplets, gases and entrained air. Soot contributes to colour and extinction but does not determine the entire toxic or thermal hazard. Carbon monoxide, hydrogen cyanide and other gases require separate assessment.

Smouldering aerosol can have much larger particles than sooty flaming smoke. In one cellulosic-insulation study described by Ohlemiller, the mass mean particle size was 2 to 3 micrometres, compared with much smaller cigarette and flaming-combustion particles. This affected detector sensitivity and transport through the fuel bed.

Soot is also not identical to ash. It is emitted particulate material associated with combustion, whereas residue can include char and inorganic material left at the fuel. Char can insulate underlying material or oxidise later, affecting further soot and gas production.

Measurement and modelling

Soot-yield measurement must match the combustion mode, oxygen supply and material form. Small free-burning tests provide useful data, but compartment ventilation and scale can change the result. A single material value should not be used without these conditions.

Mass released can be estimated from energy released and effective heat of combustion, then multiplied by soot yield. This calculation requires a consistent energy basis and a defensible effective heat of combustion. It does not directly measure concentration until the smoke volume is specified.

Particle-size instruments respond to different physical properties, including electrical mobility and optical behaviour. Their measurement basis can affect the resulting distribution, so a comparison should use compatible methods or an appropriate conversion. The full distribution is often more informative than one average diameter.

Smoke models then connect source yield with plume entrainment, layer volume and transport. Uniform-layer assumptions can obscure local gradients, while deposition and condensation can change the aerosol along its path. Optical and toxic results should be evaluated at the receiver.

Limits and common misunderstandings

Soot is not all smoke. Droplets and gases can remain after soot concentration falls, and toxic gases can persist when visibility improves. Assessing only dark particles would miss these components.

A high soot yield does not automatically produce a larger fire or higher total soot mass. Fuel consumption and heat release rate determine how much material is actually burned. Yield and mass must be evaluated together.

Soot is not only a visibility problem. It contributes to flame radiation and heat transfer, and its particles can influence detector response and deposition. These effects require optical, thermal and transport analysis rather than colour alone.

a generic mass-extinction coefficient is not exact for every aerosol. Particle size, composition and measurement method affect the appropriate value. Using it without its source and assumptions can create false precision in visibility estimates.

Source apportionment requires the same combustion-stage detail as yield measurement. Early well-ventilated burning, developed fuel-rich burning and smouldering can produce different particles and yields. Assigning one yield to an entire fire would hide those changes and could distort both optical and radiative calculations.

Soot mass should also be balanced against char and unburnt carbon remaining at the source. Not all fuel carbon becomes airborne soot, and residual char can oxidise later to carbon monoxide and carbon dioxide. This makes the emitted particle source only one part of the complete carbon pathway.

Measurement of emitted mass does not by itself establish human or structural exposure. The particles must be transported to a receiver, where concentration, size and incident radiation determine the effect. Deposition and coagulation can change all three along the path.

For detector assessment, particle size and concentration must match the detection principle. Ohlemiller's cellulosic-smoulder example had much larger particles than the flaming comparison and reduced ionisation-detector sensitivity. A detector response based only on a sooty flaming source cannot be applied to that different aerosol without evidence.

For radiation assessment, source temperature and emissivity should accompany soot concentration. Particles contribute strongly to emission in many fires, but gases and heated boundaries also participate. Adding these contributions requires consistent geometry and wavelength treatment rather than a universal soot-radiation fraction.

A reproducible calculation should show the fuel-consumption estimate, soot yield, layer volume, extinction coefficient and visibility rule. It should also state whether deposition and non-uniform distribution were neglected. These details reveal which assumptions most strongly affect the predicted visibility.

Where those assumptions are uncertain, alternative yields and volumes can be compared. The shop example demonstrates how changing soot yield alone changes optical density and visibility by an order of magnitude. Sensitivity analysis is therefore more useful than presenting the rough visibility rules as exact distances.

Soot's changing role through the fire should also be recognised. Particles can form in one region, oxidise in another and accumulate in a layer or on a surface. Their final location affects whether they contribute to flame radiation, remote obscuration or deposited evidence rather than simply to the initial emission estimate.

A final check is whether the evidence supports the chosen soot-specific conclusion. Source-yield data may not establish remote visibility, and a dark plume may not establish the local radiant flux. Linking each conclusion to the appropriate measurement and location prevents soot from becoming an imprecise proxy for the entire fire.

See also

• Flame
• Diffusion flame
• Emissivity
• Plastics in fire

References

• Beyler, C. L., & Hirschler, M. M. (2002). Thermal decomposition of polymers. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Drysdale, D. D. (2002). Thermochemistry. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Jin, T. (2002). Visibility and human behavior in fire smoke. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Mulholland, G. W. (2002). Smoke production and properties. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Ohlemiller, T. J. (2002). Smoldering combustion. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Purser, D. A. (2002). Toxicity assessment of combustion products. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Simmons, R. F. (2002). Premixed burning. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Tien, C. L., Lee, K. Y., & Stretton, A. J. (2002). Radiation heat transfer. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.
• Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.