Flame spread

Flame spread is the movement of a flame across or adjacent to a condensed fuel whose evaporation or pyrolysis supplies the flame with gaseous fuel. It involves two advancing regions: the visible gas-phase flame and the evaporation or decomposition front in the solid or liquid. The rate of the condensed-phase front defines surface flame-spread velocity.

The process differs from flame propagation through a premixed gas. In surface flame spread, fuel and oxygen are not already distributed uniformly throughout the gas. Heat from the advancing flame must first produce new vapour ahead of the reaction region.

Flame spread influences how rapidly a small ignition develops into a larger fire. It can accelerate as more fuel and stronger heat feedback become involved, or stop when forward heating, fuel continuity or combustion conditions become inadequate.

Flame and decomposition fronts

The visible flame occupies the gas phase, while the condensed fuel remains beneath or beside it. Heat transferred from the flame evaporates liquid or decomposes solid ahead of the current burning area. The new vapour mixes with air and allows the flame to advance.

Surface flame spread involves two advancing fronts. Based on Quintiere (2002), p. 2-246.

The evaporation or pyrolysis front is harder to observe than the flame, but it marks the region being converted into gaseous fuel. Its position and movement connect the visible flame with fuel consumption. A flame can appear substantial while the underlying front moves comparatively slowly.

This mechanism can be viewed as successive Ignition of new material. Each newly heated region must reach a condition supporting piloted reaction and continued vapour supply. Flame spread therefore depends on the target's initial temperature and ignition response as well as on the existing flame.

A premixed gas flame follows a different mechanism. Fuel and oxidiser are already present in the gas, so propagation can be much faster and does not require an advancing solid-decomposition front. The two processes should not share an unqualified spread-rate value.

Heat transfer ahead of the flame

Forward heat transfer raises fuel from its initial temperature towards its ignition condition. Conduction through gas and solid, convection from hot gases and radiation from the flame can all contribute. Their relative importance depends on flow, orientation, scale and material properties.

Quintiere's conceptual energy balance compares heat delivered ahead of the front with the energy needed to prepare new fuel. Lower initial temperature requires more heating, while preheating by the wider fire reduces the additional energy needed. Spread can therefore accelerate as the surroundings become hotter.

Heat losses can prevent the front from sustaining itself. Energy entering a thick solid can drain into its interior, and convection can cool the reaction region. Stronger opposed flow may enhance some heat-transfer paths while reducing the time available for chemical reaction, so its effect is not universally positive or negative.

Flame temperature and heat flux are also coupled with oxygen and chemistry. Reduced oxygen can lower reaction and heat release, while excessive flow can quench the reaction region. Material data alone cannot predict spread without these fire conditions.

Direction and airflow

Upward flame spread benefits from buoyant hot gases and flame contact preheating fuel above the front. This can produce a large heated region and rapid acceleration on vertical surfaces. Van Rossum identifies this mechanism as an important reason for faster upward spread.

Flow direction changes flame-spread heat transfer. Based on Van Rossum (2027), §4.3 and Quintiere (2002), p. 2-247.

Opposed-flow flame spread occurs when the front advances against the surrounding gas flow. Forward heat must reach fuel despite that opposing transport. Quintiere distinguishes this from wind-aided spread, where the surrounding flow supports movement in the same direction.

Lateral and downward configurations have different heat-transfer paths and generally different rates from upward spread. The direction of gravity changes buoyancy, flame shape and which fuel is preheated. Material orientation must therefore accompany any reported rate.

External wind and fire-induced natural convection can both create the relevant flow. A value measured in still air cannot be transferred automatically to a ventilated shaft, open facade or compartment with strong fire-driven circulation.

The distance heated ahead of the front helps explain the difference between the flow arrangements. Quintiere gives illustrative heating lengths of about 1–3 mm for opposed spread and 0.1–10 m for wind-aided spread. The latter can extend over much of the adjacent flame and hot-gas region. These lengths belong to the simplified configurations and do not replace a heat-flux calculation.

Material thickness and thermal response

Thermally thin materials have relatively little temperature variation through their thickness in the simplified model. Paper, garments and draperies can sometimes approach this behaviour. Thermally thick materials retain important internal gradients, and heat conducted into them is unavailable at the surface.

Geometric thickness alone does not define the thermal class. Conductivity, density, specific heat, exposure time and backing conditions all influence the response. Quintiere gives approximate material-specific examples, but these should not be converted into universal thickness thresholds.

Low density can favour rapid response because less energy is required to prepare a given volume for ignition. Substrates can add further heat storage or loss. A thin combustible finish on a massive backing can consequently behave differently from the same finish without that backing.

Moisture, melting and char formation change the response during heating. They alter heat consumption, surface integrity and volatile release. Treating the material as homogeneous and unchanged is a modelling simplification whose limits must be stated.

Spread rate and fire growth

Quintiere illustrates growth with an upholstered-chair ignition region 6 cm in diameter and an unenhanced spread rate of about 1 mm/s. After one minute in the example, the pyrolysing region reaches 18 cm in diameter as the front advances 6 cm. The larger fuel area can then strengthen heat feedback.

For an ideal circular region, increasing the diameter from 6 cm to 18 cm multiplies the area by nine. The initial area is about 0.0028 m² and the later area about 0.025 m². This calculation assumes that the entire enclosed region remains involved. It illustrates how even a constant front speed can produce increasing burning area.

The example demonstrates area growth rather than a universal chair-fire rate. Spread rate can change as heat release increases, and fuel burnout can limit later development. Early spread and long-term fire growth therefore require a time-dependent description.

The handbook's order-of-magnitude table lists smouldering at 0.001 to 0.01 cm/s, lateral or downward spread on thick solids around 0.1 cm/s, and upward spread on thick solids from 1 to 100 cm/s. Premixed laminar deflagration is listed from 10 to 100 cm/s. These ranges show different mechanisms and scales, not values for direct substitution into another scenario.

Order-of-magnitude flame-spread rates from the SFPE Handbook. Based on Quintiere (2002), p. 2-247.

Fire spread beyond the first object adds further pathways, including radiation across gaps, direct flame contact and burning debris. Surface flame spread is one mechanism within this wider process. A room or building fire should not be modelled as one continuous sheet of fuel unless that geometry actually applies.

Materials and fire performance

Combustibility describes whether and how a material can burn under relevant conditions, but it does not specify a flame-spread rate by itself. Ignition response, heat capacity, decomposition and geometry all affect movement across the material. Two combustible products can therefore spread fire very differently.

Reaction to fire testing examines product behaviour under a specified fire exposure. For flame spread, relevant observations include ignition, front movement and heat release. These describe the product in its tested configuration, whereas a spread model additionally needs the heat-transfer conditions around the advancing front.

Surface orientation and installation affect measured performance. A product tested horizontally may not behave identically on a wall or ceiling, and joints or substrates can alter heat transfer. The tested specimen and the installed product must be compared explicitly.

A material can resist initial ignition yet still spread flame under a larger exposure. The threshold crossed in one test does not prove performance in every fire. Scenario severity and duration remain part of the assessment.

Testing and measurement

The Single Burning Item test is described in the 2002 handbook as a European regulatory heat-release-rate test for building products. Two panels form a corner without a ceiling. This geometry differs from a complete room, so its result does not directly represent every installed exposure.

Quintiere discusses cone-calorimeter and lateral ignition and flame-spread tests as sources of material data. These methods helped replace less meaningful flammability indices with measurements linked to ignition and spread models. They do not remove all uncertainty in heat-transfer and material-property inputs.

A spread-rate measurement needs an identifiable front, direction and exposure condition. Visible flame length alone does not establish how quickly the pyrolysis region moves. Measurement definitions can therefore change the reported result.

A room-corner test provides another configuration for interior finishes. Quintiere describes studies combining cone-calorimeter data with flame-spread models to reproduce its fire growth. The calculation still needs relationships for flame length and heat transfer in that particular geometry. Obtaining a material property in one apparatus therefore supplies part of the prediction, rather than the whole installed-product response.

Tests at reduced scale can underrepresent radiation and room feedback. As the fire grows, upper-layer radiation and heated boundaries can preheat new regions beyond the local test arrangement. Applying small-scale rates to a developed enclosure fire requires an appropriate model.

Calculating spread and recognising its limits

The basic energy balance equates forward heating with the energy needed to prepare fresh material for burning. In a simplified steady model, spread speed increases when more energy reaches the fuel ahead of the front. It decreases when that fuel needs a larger temperature rise. Density and heat capacity determine how much energy the heating requires.

$$\rho Vc(T_{ig}-T_s)={\dot{q}}^{″}$$

Symbol	Description	Unit
$\rho$	Fuel density	kg/m³
$V$	Spread speed	m/s
$c$	Specific heat capacity	kJ/(kg·K)
$T_{ig}$	Effective surface ignition temperature	°C
$T_s$	Fuel temperature ahead of the front	°C
${\dot{q}}^{″}$	Net heat flux preparing fresh fuel in the conceptual balance	kW/m²

The temperature difference has the same numerical value in kelvin and degrees Celsius. The heat flux in this balance is the effective forward heating, rather than the fire's total heat release rate. A detailed model must establish how much energy actually reaches the fresh fuel. It must also represent the distance over which that heat acts.

For opposed-flow spread over thermally thick materials, Quintiere gives a relation fitted using lateral-spread tests. The denominator combines the material's thermal inertia with the square of the required temperature rise. The fitted parameter represents the spread conditions and fuel response. It is not a universal constant for all flows or oxygen concentrations.

$$V=\frac{\mathrm{\Phi }}{k\rho c(T_{ig}-T_s{)}^2}$$

Symbol	Description	Unit
$k\rho c$	Effective thermal inertia, conductivity × density × specific heat	kW²·s/(m⁴·K²)
$\mathrm{\Phi }$	Empirical flame-spread parameter for the material and exposure	kW²/m³
$V,T_{ig},T_s$	Spread speed and temperatures as defined above	m/s, °C, °C

The test method preheats a vertical specimen by external radiation before lateral spread begins. The surface is allowed to approach a steady temperature for each exposure. The fitted properties describe that arrangement under natural convection in air. The spread relation does not calculate the preceding time needed to establish the preheated state. Downward-spread results were similar for the materials examined, except where excessive melting and dripping changed the behaviour.

For 1.27 cm Douglas fir particleboard, the handbook lists an effective ignition temperature of 382 °C. Its thermal inertia is 0.94 kW²·s/(m⁴·K²), and its spread parameter is 12.7 kW²/m³. The listed minimum surface temperature supporting continued spread is 210 °C. These are effective properties from the reported experiments, rather than universal values for all timber.

As an illustration, choose a preheated surface temperature of 250 °C, above that measured minimum. The resulting calculated spread speed is approximately 0.78 mm/s. Choosing 300 °C with the same fitted properties gives approximately 2.0 mm/s. These two chosen temperatures demonstrate the effect of preheating; they are not separate measured test results.

$$V_{250}=\frac{12.7}{0.94(382-250{)}^2}=0.000775\dots \mathrm{m}/\mathrm{s}\approx 0.78\mathrm{m}\mathrm{m}/\mathrm{s}$$ $$V_{300}=\frac{12.7}{0.94(382-300{)}^2}=0.002009\dots \mathrm{m}/\mathrm{s}\approx 2.0\mathrm{m}\mathrm{m}/\mathrm{s}$$

The relation cannot be continued without limit towards the ignition temperature. Its assumptions describe a particular spread process with fitted properties and established heating conditions. Widespread ignition, changing heat flux or a different airflow requires another treatment. Below the measured minimum temperature, continuing spread was not sustained, despite the algebraic formula still returning a positive number.

Burnout introduces a separate limitation. Fuel behind the front can finish burning while fresh material is still being heated. If spread is fast relative to burning time, the active burning area can grow; if it is slow, growth can decelerate or stop. A calculation of front speed alone therefore does not determine the complete fire's heat release.

Liquid fuels introduce an additional mechanism that the solid-material relation does not include. Surface-tension differences and circulation can transport heat within the liquid ahead of the flame. Quintiere distinguishes spread governed by this liquid motion from faster propagation through an existing flammable vapour layer. The liquid's initial temperature relative to its flash point helps distinguish these regimes.

See also

• Euroclass system
• Travelling fire
• Trench effect
• Wildfire

References

• Australian Building Codes Board, National Research Council of Canada, International Code Council, & Department of Building and Housing New Zealand. (2005). International fire engineering guidelines (2005 ed.). Australian Building Codes Board.
• Babrauskas, V. (2002). Heat release rates. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-1–3-37). National Fire Protection Association.
• Beyler, C. L., & Hirschler, M. M. (2002). Thermal decomposition of polymers. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 1-110–1-131). National Fire Protection Association.
• Drysdale, D. D. (2002). Thermochemistry. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 1-90–1-98). National Fire Protection Association.
• Kanury, A. M. (2002). Flaming ignition of solid fuels. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 2-229–2-245). National Fire Protection Association.
• Quintiere, J. G. (2002). Surface flame spread. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 2-246–2-257). 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., pp. 1-73–1-89). National Fire Protection Association.
• Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.