Cool flame is a flame having maximal temperature below about 400 °C (752 °F). It is usually produced in a chemical reaction of a certain fuel-air mixture. Contrary to conventional flame, the reaction is not vigorous and releases very little heat, light and carbon dioxide. Cool flames are difficult to observe and are uncommon in everyday life, but they are responsible for engine knock – the undesirable, erratic, and noisy combustion of low-octane fuels in internal combustion engines.
Cool flames were accidentally discovered in the 1810s by Sir Humphry Davy, who noticed that certain types of flame did not burn his fingers or ignite a match. He also found that those unusual flames could change into conventional ones and that at certain compositions and temperatures, they did not require an external ignition source, such as a spark or hot material. Harry Julius Emeléus was the first to record their emission spectra, and in 1929 he coined the term "cold flame".
|Compound||CFT (°C)||AIT (°C)|
|Methyl ethyl ketone||265||515|
|Methyl isobutyl ketone||245||460|
Cool flame can occur in hydrocarbons, alcohols, aldehydes, oils, acids, waxes and even methane. The lowest temperature of a cool flame is poorly defined and is conventionally set as temperature at which the flame can be detected by eye in a dark room (cool flames are hardly visible in daylight). This temperature slightly depends on the fuel to oxygen ratio and strongly depends on gas pressure – there is a threshold below which cool flame is not formed. A specific example is 50% n-butane–50% oxygen (in volume percent) which has a cool flame temperature (CFT) of about 300 °C at 165 mmHg (22.0 kPa). One of the lowest CFTs (156 °C) was reported for a C2H5OC2H5 + O2 + N2 mixture at 300 mmHg (40 kPa). The CFT is significantly lower than the auto-ignition temperature (AIT) of conventional flame (see table).
The spectra of cool flames consist of several bands and are dominated by the blue and violet ones – thus the flame usually appears pale blue. The blue component originates from the excited state of formaldehyde (CH2O*) which is formed via chemical reactions in the flame:
- CH3O• + •OH → CH2O* + H2O
- CH3O• + CHnO• → CH2O* + CHnOH
A cool flame does not start instantaneously after the threshold pressure and temperature are applied, but has an induction time. The induction time shortens and the glow intensity increases with increasing pressure. With increasing temperature, the intensity may decrease because of the disappearance of peroxy radicals required for the above glow reactions.
Whereas in a usual flame molecules break down to small fragments and combine with oxygen producing carbon dioxide (i.e. burn), in a cool flame, the fragments are relatively large and easily recombine with each other. Therefore, much less heat, light and carbon dioxide is released; the combustion process is oscillatory and can sustain for a long time. A typical temperature increase upon ignition of a cool flame is a few tens of degrees Celsius whereas it is on the order of 1000 °C for a conventional flame. Most experimental data can be explained by the model which considers cool flame just as a slow chemical reaction where the rate of heat generation is higher than the heat loss. This model also explains the oscillatory character of the cool flame: the reaction accelerates as it produces more heat until the heat loss becomes appreciable and temporarily quenches the process.
Cool flames may contribute to engine knock – the undesirable, erratic, and noisy combustion of low-octane fuels in internal combustion engines. In a normal regime, the conventional flame front travels smoothly in the combustion chamber from the spark plug, compressing the fuel/air mixture ahead. However, the concomitant increase in pressure and temperature may produce a cool flame in the last unburned fuel-air mixture (the so-called end gasses) and participate in the autoignition of the end gasses. This sudden, localized heat release generates a shock wave which travels through the combustion chamber, with its sudden pressure rise causing an audible knocking sound. Worse, the shock wave disrupts the thermal boundary layer on the piston surface, causing overheating and eventual melting. The output power decreases and, unless the throttle (or load) is cut off quickly, the engine can be damaged as described in a few minutes. The sensitivity of a fuel to a cool-flame ignition strongly depends on the temperature, pressure and composition. The cool flame initiation of the knock process is likely only in highly throttled operating conditions, since cool flames are observed at low pressures. Under normal operating conditions, autoignition occurs without being triggered by a cool flame. Whereas the temperature and pressure of the combustion are largely determined by the engine, the composition can be controlled by various antiknock additives. The latter mainly aim at removing the radicals (such as CH2O* mentioned above) thereby suppressing the major source of the cool flame.
- Lindström, B.; Karlsson, J.A.J.; Ekdunge, P.; De Verdier, L.; Häggendal, B.; Dawody, J.; Nilsson, M.; Pettersson, L.J. (2009). "Diesel fuel reformer for automotive fuel cell applications" (PDF). International Journal of Hydrogen Energy. 34 (8): 3367. doi:10.1016/j.ijhydene.2009.02.013.
- Pearlman, Howard; Chapek, Richard M. (1999). Cool Flames and Autoignition: Thermal-Ingnition Theory of Combustion Experimentally Validated in Microgravity. NASA. p. 142. ISBN 978-1-4289-1823-8., Web version at NASA
- Peter Gray; Stephen K. Scott (1994). Chemical oscillations and instabilities: non-linear chemical kinetics. Oxford University Press. p. 437. ISBN 978-0-19-855864-4.
- Stephen K. Scott (1993). Chemical chaos. Oxford University Press. p. 339. ISBN 978-0-19-855658-9.
- H. Davy (1817) "Some new experiments and observations on the combustion of gaseous mixtures, with an account of a method of preserving a continued light in mixtures of inflammable gases and air without flame," Philosophical Transactions of the Royal Society of London, 107 : 77-86. On p. 79, Davy inserted a hot platinum wire into a mixture of air and diethyl ether vapor: "When the experiment in the slow combustion of ether is made in the dark, a pale phosphorescent light is perceived above the wire, which is most distinct when the wire ceases to be ignited. This appearance is connected with the formation of a peculiar acrid volatile substance possessed of acid properties."
- A number of other investigators subsequently also observed cold flames:
- H. B. Miller (1826) "On the production of acetic acid, in some original experiments with metallic and non-metallic substances over ether, alcohol, etc.," The Annals of Philosophy, new series, 12 : 17-20. From page 19: "The tip of the glass rod held over the ether emits the blue flame from the whole of its surface ; acetic acid formed in abundance."
- (Döbereiner) (1834) "Sauerstoffabsorption des Platins" (Oxygen absorption by platinum), Annalen der Physik und Chemie, 31 : 512. From page 512: "Eine andere nicht uninteressante Beobachtung von Döbereiner ist: dass Aether schon bei der Termperatur von 90° R. verbrennt, und zwar mit einer nur im Dunkeln wahrnehmbaren blassblauen Flamme, die nicht zündend wirkt, aber selbst so entzündbar ist, dass sie sich bei Annäherung einer brennenden Kerze augenblicklich in eine hochlodernde, hellleuchtende Flamme verwandelt." (Another not uninteresting observation of Döbereiner is that ether burns even at the temperature of 90° Réaumur with a pale blue flame that is perceptible only in the dark, which does not cause [things to] ignite, but itself is so flammable that on approach of a burning candle, it transforms instantly into a blazing, brightly glowing flame.)
- Boutigny (1840) "Phénomènes de la caléfaction", Comptes rendus … , 10 : 397-407. On page 400, Boutigny stated that when diethyl ether was added dropwise to a red-hot platinum crucible, an irritating, acidic vapor was produced. " … il est bien à présumer qu'il s'opère là une combustion lente, … " ( … it is well to presume that a slow combustion is taking place there … )
- Pierre Hippolyte Boutigny, Études sur les corps à l'état spheroidal: Nouvelle branche de physique [Studies on bodies in a spheroidal state: a new branch of physics], 3rd ed. (Paris, France: Victor Masson, 1857), pp. 165-166. On page 166, Boutigny noted that when he poured some diethyl ether into a hot crucible: "Dans une obscurité profonde, on aperçoit, à toutes les phases de l'expérience, une flamme d'un bleu clair peu apparent, qui ondule dans le creuset dont elle remplit toute la capacité. Cette flamme rare et transparente est le signe d'une métamorphose profonde qui subit l'éther ; elle est caractérisée par le dégagement d'une vapeur dont l'odeur vive et pénétrante irrite fortement la muquese nasale et les conjonctives." (In deep darkness, one perceives, at all stages of the experiment, a flame of an inconspicuous light blue, which ripples in the crucible which it fills completely. This rare and transparent flame is a sign of a profound metamorphosis that the ether undergoes ; it is characterized by the release of a vapor whose sharp and penetrating odor strongly irritates the nasal mucosa and conjunctiva [of the eyes].)
- W.H. Perkin (1882) "Some observations on the luminous incomplete combustion of ether and other organic bodies," Journal of the Chemical Society, 41 : 363-367.
- Harry Julius Emeléus (1929) "The light emission from the phosphorescent flames of ether, acetaldehyde, propaldehyde, and hexane," Journal of the Chemical Society (Resumed), pp. 1733-1739.
- H. J. Pasman; O. Fredholm; Anders Jacobsson (2001). Loss prevention and safety promotion in the process industries. Elsevier. pp. 923–930. ISBN 0-444-50699-3.
- Hazards XIX: process safety and environmental protection : what do we know? where are we going?. IChemE. 2006. p. 1059. ISBN 0-85295-492-1.
- Griffiths, John F.; Inomata, Tadaaki (1992). "Oscillatory cool flames in the combustion of diethyl ether". Journal of the Chemical Society, Faraday Transactions. 88 (21): 3153. doi:10.1039/FT9928803153.(this reference cites evidence of cool fire at 430 K, which is 156 C, not 80 C)
- Barnard, J (1969). "Cool-flame oxidation of ketones". Symposium (International) on Combustion. 12 (1): 365. doi:10.1016/S0082-0784(69)80419-4.
- Jones, John Clifford (September 2003). "Low temperature oxidation". Hydrocarbon process safety: a text for students and professionals. Tulsa, OK: PennWell. pp. 32–33. ISBN 978-1-59370-004-1.
- George E. Totten; Steven R. Westbrook; Rajesh J. Shah, eds. (2003). Fuels and lubricants handbook: technology, properties, performance, and testing. ASTM International. p. 73. ISBN 0-8031-2096-6.
14. ^Recently, self-sustained, stable cool flames has been established by adding ozone into oxidizer stream: Won, S. H., Jiang, B., Diévart, P., Sohn, C. H., & Ju, Y. (2014). Self-sustaining n-heptane cool diffusion flames activated by ozone. Proceedings of the Combustion Institute, Vol.35, 2014. doi:10.1016/j.proci.2014.05.021
- Peter Gray; Stephen K. Scott (1994). Chemical oscillations and instabilities: non-linear chemical kinetics. Oxford University Press. pp. 429 ff. ISBN 0-19-855864-3. - an explanation of the oscillatory nature of cool flame.