Contents
A Flame is the visible part of Combustion in which gaseous fuel reacts with an oxidiser, usually oxygen from the surrounding air. The reaction region releases heat, and light can come from hot particles and excited chemical species. Flaming is therefore a gas-phase form of combustion, distinct from flameless surface processes such as smouldering.[1][2]
Flames vary in shape, luminosity, temperature and stability. Fuel and oxidiser can be mixed before ignition or can come together within the flame. Flow can be smooth or turbulent, and the flame can exchange heat with fuel, nearby surfaces and the surrounding gas. These differences determine how a flame looks and how it affects its surroundings.[1][3]
A visible flame is not a complete measure of a fire. Some flames produce little light, while a hot or decomposing fuel can remain hazardous without visible flaming. Flame observation therefore needs to be combined with knowledge of fuel supply, oxygen, heat transfer and combustion products.[1][4]
Premixed and diffusion flames
In a Premixed flame, fuel and oxidiser are mixed before ignition. A Bunsen burner provides a familiar example. Simmons describes a luminous inner cone, where rapid oxidation produces a steep temperature rise, surrounded by hot combustion gases. With a fuel-rich initial mixture, the outer region can behave as a diffusion flame while remaining fuel burns with oxygen from the surrounding atmosphere.[5]
In a Diffusion flame, fuel and oxidiser are initially separated and mix within the flame. A candle and many fires involving solids or liquids follow this broad pattern. Their fuel vapour originates at a surface, while oxygen arrives from the surrounding air. The reaction zone develops where mixing and temperature support combustion.[1][6]
The distinction concerns mixing arrangement rather than two wholly different chemistries. Simmons notes that the basic chemistry in a hydrocarbon-air diffusion flame is related to that in a premixed flame, but local conditions vary across the reaction region. Fuel-rich and oxygen-rich zones support different reactions and products. A single overall fuel-to-air ratio cannot describe every location in the flame.[6]
A premixed flame has a burning velocity, defined as its speed normal to the flame surface through the adjacent unburnt mixture. In an ideal flat flame, stability requires this velocity to balance the incoming mixture velocity. Excess flow can blow the flame off, while insufficient flow can allow flashback towards the burner. A conical burner flame can adjust its area and remain stable over a wider flow range.[5]
Diffusion flames can also depend on local premixed regions. Near the base of a gas-jet flame, low gas velocity allows fuel and air to mix before burning. The balance between local flow and burning velocity helps determine whether the flame remains attached or blows off. Recirculation behind an obstruction can similarly create a stable combustion region.[6]
Laminar flow, turbulence and buoyancy
A laminar flame has an orderly flow structure. In a laminar diffusion flame, molecular diffusion is an important means of oxygen transport. Van Rossum uses a candle flame to illustrate this slow mixing process and the relatively hot outer reaction region. The description applies to the selected flame conditions and not to every candle-like fire at every scale.[1]
Turbulence introduces continually changing velocity components. These fluctuations deform the reaction region and increase mixing between fuel, oxidiser and hot products. In a premixed flame, a sharply defined laminar front can become a thicker, brush-like reaction region as turbulence increases. The flame is then not a smooth surface with uniform properties.[7]
Many real fires are turbulent diffusion flames whose flow is driven largely by buoyancy. Hot combustion products have lower density than the surrounding cooler gas and rise, drawing air into the fire region. Simmons explains that this turbulent motion can control mixing and hence the overall reaction rate, even when the underlying chemistry could proceed faster.[6][3]
This interaction produces a changing flame shape rather than a fixed outline. Turbulent motion can fold and stretch the reaction zone, while air supply and fuel vapour production vary with time. The instantaneous visible boundary does not identify all locations where reactions or heat release occur.[3]
Colour, soot and radiation
The yellow or orange luminosity of many diffusion flames is associated with hot Soot particles. These particles emit visible radiation because of their temperature. Fuels that produce little soot can have flames that are difficult to see, so low luminosity does not necessarily mean low temperature or low heat release.[1]
Local chemistry influences this appearance. On the fuel-rich side of a hydrocarbon-air diffusion flame, limited oxygen favours thermal decomposition and carbon formation. Simmons connects this region with the characteristic yellow carbon zone. The hot blue region on the air side has different composition and reaction conditions.[6]
Premixed flames also emit radiation from excited chemical species. Simmons describes radiation from electronically excited hydroxyl radicals contributing to the luminous inner cone of a Bunsen flame. Flame light consequently has more than one physical origin, and colour alone cannot identify temperature or completeness of combustion.[8]
Flame radiation transfers energy to surrounding surfaces without requiring a gas to carry heat between them. Together with convective heat transfer, it can heat unburnt fuel and other objects. Van Rossum gives a rough division of approximately 30% radiation and 70% convection for fire energy, while explicitly noting that the actual proportions differ between fires.[1]
Radiative feedback is particularly relevant to sustained burning of liquids and solids. Heat returned to the fuel supports evaporation or pyrolysis, which supplies more vapour to the flame. The flame and fuel therefore form a coupled system. Changing surface orientation or heat loss can alter growth even when the fuel chemistry remains unchanged.[9]
Temperature and heat release
Flame temperature is not uniform. In a premixed Bunsen flame, Simmons places the maximum temperature slightly downstream of the inner cone, while heat losses produce steep gradients towards the surroundings. Diffusion flames likewise have fuel-rich, reaction and air-side regions with different temperatures and compositions.[5][6]
The adiabatic flame temperature is a theoretical value calculated without heat exchange with the surroundings. It is reached only in restricted regions of sufficiently large flames under suitable assumptions. Small premixed flames and turbulent flames commonly fail to reach it because their heat losses are appreciable.[10]
High-temperature chemistry adds another limitation. Above approximately 1800 K, calculations may need to account for energy absorbed by dissociation of combustion products. Drysdale similarly explains that radiation and dissociation make real temperatures lower than a simple theoretical maximum. A temperature calculation therefore depends on both energy conservation and chemical composition.[10][11]
Heat release rate and flame size are related but not interchangeable. Heat release rate is the energy produced per unit time, while flame geometry describes where the reaction region extends. Van Rossum notes that greater fire power is associated with greater Flame height, but fuel, mixing and flow still influence the observed shape.[12][1]
A flame can also receive fuel produced in one place and burn it elsewhere. In an oxygen-limited enclosure, unburnt volatiles can leave through an opening and ignite after mixing with fresh air. The visible external flame then belongs to the same overall fire system, although part of its heat release occurs outside the original compartment.[4]
Flames and fire growth
A flame can sustain or enlarge a fire by heating new fuel. Upward spread over a vertical surface benefits from heat transfer to material above the burning region. This preheating can increase pyrolysis and accelerate spread. The process depends on orientation and heat feedback, not only on whether the material can burn in a small test.[1]
Enclosures modify this feedback. Hot gases and surfaces can radiate to other combustible objects, while limited ventilation changes oxygen supply and the movement of unburnt vapour. Flames can consequently influence remote ignition as well as the fuel immediately beneath them.[1][4]
Flame stability results from a balance of reaction, mixing and heat loss. At a flammability limit, heat losses can exceed heat production and prevent sustained propagation. Chemical inhibitors provide another route by promoting termination of chain reactions, reducing reaction until the flame cannot continue.[13][8]
Extinction of visible flaming does not prove that all combustion has stopped. Smouldering can continue within a porous solid, and hot fuel can later produce a flammable vapour mixture. A flame assessment therefore needs to consider the remaining fuel condition and the possibility of renewed ignition.[14][15]
Measuring flame behaviour
Burning velocity is measured by relating the flow of unburnt mixture to the area and position of the flame front. For propane-air mixtures, Simmons reports maximum laminar burning velocities mainly between 37 and 45 cm/s, with most values in the range 41 ± 2 cm/s. These values relate to room-temperature, atmospheric-pressure initial conditions and should not be transferred to turbulent fire plumes.[7]
The measurement method contributes to variation between reported values. Different definitions of the visible front can produce different areas, while the burner rim removes heat and changes the local reaction. Flow near the cone tip can also differ from the idealised flat-flame arrangement. A numerical burning velocity therefore needs its apparatus and flame definition.[7]
Temperature mapping addresses a different property. Simmons' example of a premixed natural-gas flame shows a changing temperature field above and around the inner cone rather than one flame temperature. The highest region is downstream of the primary reaction zone, and the outer gases cool towards the surroundings. A single point measurement cannot represent that whole field.[5][10]
Heat release rate requires another measurement again. Oxygen-consumption calorimetry estimates it from oxygen consumed during combustion, using approximately 13.1 kJ/g of oxygen for many fuels with a correction associated with carbon monoxide. The method can be applied from small specimens to larger furniture tests. It does not determine heat release from flame luminosity alone.[11]
Radiative and convective heat transfer connect these measurements with fire effects. A flame can have substantial heat release while transferring different proportions of energy to a fuel surface, depending on configuration. Measurement of total energy production therefore does not remove the need to assess exposure at the receiving object.[1]
Chemical models add further choices. In a turbulent methane-air example discussed by Simmons, assuming thermodynamic equilibrium predicted a peak carbon-monoxide mole fraction about 2.5 times the experimental value. A laminar-flamelet approach produced agreement within 10% for that case. This comparison is specific to the reported model and experiment, not a general error allowance.[3]
The difference arises because actual composition need not reach equilibrium throughout a turbulent diffusion flame. Flow can move gases away before slower reaction and burnout processes finish. Flame models consequently combine chemical assumptions with conservation of mass, energy and species, and their predictions require experimental support.[3]
Flame observation, temperature measurement, heat-release measurement and chemical modelling therefore answer separate questions. A complete description identifies which property was obtained, under what fuel and flow conditions, and with which spatial and temporal resolution. This prevents a visible flame dimension from being used as a substitute for all other flame properties.[7][11]
Interpretation and limits
Flame appearance provides qualitative evidence but not a complete engineering measurement. Luminosity depends on soot and excited species, while temperature and reaction rates vary through the flame. Two flames with different colours or shapes cannot be ranked reliably without information about fuel, scale and flow.[1][8]
Experimental burning velocities also depend on measurement definitions. Simmons reports that discrepancies can arise from how the flame-front position and area are determined, together with burner-rim quenching and local flow effects. A numerical flame property therefore needs its test method and conditions.[7]
Models of turbulent flames must make assumptions about chemistry and equilibrium. Simmons notes that local equilibrium can apply only in limited regions and that real fires add feedback between chemistry, heat transfer and fluid motion. A simple flame picture cannot capture all these interactions.[3]
For fire safety work, the useful description includes the fuel source, mixing mode, geometry, heat feedback and surroundings. These features connect the flame to fire growth, detection and extinguishment without treating its visible outline as the fire itself.[1][9]
A flame should also be distinguished from the burning object. The object supplies fuel, while the flame occupies a gas-phase reaction region that can change position and shape. This distinction remains valid when flames extend beyond the original fuel package.[16][4]
See also
Notes
- ↑ 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.10 1.11 1.12 Van Rossum (2027), §4.3.
- ↑ Ohlemiller (2002), p. 2-200.
- ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Simmons (2002), p. 1-152.
- ↑ 4.0 4.1 4.2 4.3 Drysdale (2002), p. 1-95.
- ↑ 5.0 5.1 5.2 5.3 Simmons (2002), p. 1-144.
- ↑ 6.0 6.1 6.2 6.3 6.4 6.5 Simmons (2002), pp. 1-151–1-152.
- ↑ 7.0 7.1 7.2 7.3 7.4 Simmons (2002), p. 1-146.
- ↑ 8.0 8.1 8.2 Simmons (2002), p. 1-150.
- ↑ 9.0 9.1 Beyler and Hirschler (2002), p. 1-110.
- ↑ 10.0 10.1 10.2 Simmons (2002), p. 1-145.
- ↑ 11.0 11.1 11.2 Drysdale (2002), p. 1-96.
- ↑ Van Rossum (2027), §4.5.
- ↑ Drysdale (2002), p. 1-97.
- ↑ Ohlemiller (2002), p. 2-206.
- ↑ Ohlemiller (2002), p. 2-208.
- ↑ Van Rossum (2027), §4.1.
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.
- Ohlemiller, T. J. (2002). Smoldering combustion. 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.
- Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.