Cool Flame Formation Research Articles

Two-dimensional simulation of cool and double flame formation induced by the laser ignition under shock-tube conditions

The laser ignition-induced spherical cool and double flame’s initiation, propagation, and transition are studied using a two-dimensional simulation with a detailed mechanism using n-heptane/O2/Ar/He mixture. The current study’s primary goal is to reveal how the radicals and flow field produced by the laser ignition lead to cool, hot, or double flame formation and to determine the flame propagation modes. It is found that after the laser spark, the over-driven shock wave results in vortex pairs, which significantly distorts the temperature field and the spatial distribution of the radicals. A torus-like hot flame is ignited at the ignition kernel center instantaneously. The results show that the laser-ignited hot flame can transit to a cool or double flame depending on the laser pulse energy. On the one hand, the hot flame extinguishes under a small laser ignition energy condition, at which the radicals at the hot flame front are transported by the vortex pair and lead to a cool flame formation. On the other hand, when the laser spark energy is large, a transient premixed double flame structure is observed. The double flame can be localized, and the appearance location is primarily affected by the spatial distribution of laser-induced radicals. The cool flame can coexist with a trailing hot flame for milliseconds, dramatically accelerating the hot flame propagation and fundamentally impacting the flame speed interpolation. Computational singularity perturbation (CSP) analysis identifies key reactions that belong to typical low-temperature chemistry in the double flame zone. A comparative simulation case with an initial temperature above the low-temperature chemistry regime confirms that no double flame structure can be formed under high-temperature conditions. The present study provides essential insight and guidance for the flame speed measurement using laser ignition under engine-relevant conditions.

Analysis of ignition temperature range and surrogate fuel requirements for GCI engine

Gasoline compression ignition (GCI) technology aims to improve internal combustion engine (ICE) efficiency. However, much is still unknown about the detailed combustion behavior of fuels under this novel ICE technology, such as relevant ignition temperature regimes and surrogate requirements under various engine load and speed conditions as well as for different fuels. Engine and spray simulations require the use of carefully designed surrogate fuels that mimic the properties of real fuels. Generally, these surrogates range from simple primary reference fuels (PRFs) to complex multi-component mixtures. This study aims to identify the relevant temperature conditions and the surrogate requirements for GCI engines covering the entire engine speed-load map; engine loads between 1 and 22 bar indicated mean effective pressure (IMEP) and engine speeds between 800 and 2500 revolutions per minute (RPM). Simulations were performed using the constant volume zero-dimensional (0-D) and homogeneous charge compression ignition (HCCI) modules in Chemkin-Pro for two gasoline fuels with widely different sensitivities (11 and 0). Furthermore, homogeneous ignition delay times and Livengood-Wu integrals were used to analyze the fundamental characteristics of GCI combustion. It was found that, except at excessively high-load, the relevant ignition temperature in a GCI engine is in the 750–1000 K range. The low-temperature (less than700 K) reactivity exhibited negligible contribution to the ignition of the high-sensitivity fuel. Two stage ignition and cool flame formation were found to dominate mid-load kinetics of the low-sensitivity fuel. Simple PRF surrogate was found to be sufficient to mimic the ignition requirement of the low-sensitivity gasoline fuel at almost the entire GCI engine speed-load map. However, for the high-sensitivity gasoline fuel, TPRF and TPRFE surrogates were needed for a better representation of the auto-ignition behavior. The obtained knowledge of the surrogate requirement and the ignition temperature regimes inside GCI engine will aid in guiding GCI engine optimization and fuel design.

Cool flame characteristics of methane/oxygen mixtures

As one of the most important clean fuels, methane plays a significant role in the energy supply system, and its combustion characteristics directly affect the efficiency and emission. In order to avoid the production of a large number of nitrogen oxides, the low temperature combustion associated with cool flame is getting more and more attention. But until now the cool flame characteristics of methane is still unclear, even its existence is controversial. In this work, the property of premixed methane/oxygen cool flame has been experimentally studied in a cylindrical reactor and numerically analyzed using different mechanisms. The effects of pressure, temperature and equivalence ratio on methane cool flame have been investigated experimentally. The experimental results show that the cool flame can be obtained in the range of 0.2–4.8 equivalent ratio. The lower limit of pressure of the cool flame region decreases with the increasing initial temperature. The sensitivity and reaction path have been analyzed through the numerical simulation, which reveals that the bifurcation in CH3 oxidation process has remarkable effects on cool flame formation.

On the propagation limits and speeds of premixed cool flames at elevated pressures

The flame speeds and propagation limits of premixed cool flames at elevated pressures are numerically modeled using dimethyl ether mixtures. The primary focus is paid on the effects of pressure, mixture dilution, computation domain, and heat loss on cool flame propagation. The results showed that cool flames exist on both fuel lean and fuel rich sides and dramatically extend the lean and rich flammability limits of conventional hot flames. There exist three different flame regimes: the hot flames, lean and rich cool flames, and double flames. A new flame flammability diagram including both cool flames and hot flames at elevated pressures is obtained. The results show that pressure significantly changes cool flame propagation and burning limits. It is found that the increase of pressure affects the propagation speeds of lean and rich cool flames differently due to the negative temperature coefficient effect. On the lean side, the increase of pressure accelerates the cool flame chemistry and shifts the transition limit of cool flame to hot flame to a lower equivalence ratio. At lower pressure, there is an extinction transition from hot flame to cool flame. However, above a critical pressure, the hot flame directly transfers to a cool flame without hot flame extinction. Moreover, increases in dilution reduce the heat release of the hot flame and promote cool flame formation. Furthermore, the results show that a smaller downstream computation domain and higher heat loss also extend the cool flame transition limit and promote cool flame formation.

Numerical simulations of premixed cool flames of dimethyl ether/oxygen mixtures

The formation and dynamics of premixed cool flames are numerically investigated by using a detailed kinetic mechanism of dimethyl ether mixtures in both freely-propagating and stretched counterflow flames with and without ozone sensitization. The present study focuses on the dynamics and transitions between cool flames and high temperature flames. The impacts of mixture temperature, inert gas temperature, and ozone concentration on low temperature ignition, cool flame formation, and flammable regions of different flame regimes are investigated. For the freely-propagating flames, three different flame structures (high temperature flames, double flames, and cool flames) are found. The present study shows that the flammability limit of dimethyl ether is significantly extended by the appearance of cool flames and that the conventional concept of the flammability limit of a high temperature flame ought to be reconsidered. Furthermore, the results demonstrate that the cool flame propagation speed can be significantly higher than that of near-limit high temperature flames and that ozone addition dramatically accelerates the formation of cool flames at low temperatures and extends the flammability limit. A schematic of a modified flammability limit diagram including both high temperature flames and cool flames is proposed. For stretched counterflow flames, the results also show that multiple flame regimes exist with and without ozone addition. It is demonstrated that at the same mixture enthalpy, ozone addition kinetically extends the cool flame extinction limit to a higher stretch rate. Moreover, with ozone addition, two different cool flame transition regimes: a low temperature ignition transition and a direct cool flame transition without an ignition limit at higher temperature, are predicted. The present results suggest that cool flames can be an important combustion process in affecting flammability limits and flame regimes as the mixture temperature, turbulent mixing, and radical production/recirculation are increased. The results also provide guidance in observing self-sustaining premixed cool flames in experiments.

Numerical modeling of auto-ignition of isolated fuel droplets in microgravity

In this work we present and apply a mathematical model to simulate the auto-ignition of isolated fuel droplets burning in microgravity conditions. The aim is to demonstrate the fundamental role of the low-temperature mechanisms on the auto-ignition process and to show that several experimental observations cannot be explained without considering the formation of cool-flames around the burning droplet.Thus, in order to better clarify the importance of the low-temperature chemistry, a detailed kinetic scheme (with hundreds of species and thousands of reactions) was adopted to model the spontaneous ignition of isolated droplets of n-heptane, n-decane, and n-dodecane in air, in a wide range of operating conditions (with environment temperatures from 600K to 1100K and pressures from 1bar to 20bar).The model was able to correctly identify the typical auto-ignition regimes of n-alkane oxidation. The comparison with the experimental measurements available in the literature was satisfactory: both first-stage and total induction times were reasonably captured by the numerical simulations. The simulations confirmed that the low-temperature chemistry plays a role of paramount importance in the auto-ignition process. In particular, the competition between low- and high-temperature mechanisms was found to explain the different types of auto-ignition which can be experimentally observed.

Chemical kinetics of the oxidation of methyl tert-butyl ether (MTBE)

An experimental investigation in a conventional static apparatus of the oxidation of equimolecular methyl tert-butyl ether (MTBE)-oxygen mixtures has shown that MTBE oxidizes far less readily than alkanes like neopentane or 2,2-dimethylbutane. In the investigated pressure (up to 1000 Torr) and temperature ranges (300–500°C) we have not observed any cool flame formation; the mixtures either react slowly or ignite. Such behaviour is in line with the well-known antiknock effect of MTBE since this ether is already largely used as an octane booster in commercial gasolines, especially in Germany and in the United States. An evaluation of the rate constants of the possible elementary steps by the methods of thermochemical kinetics has allowed us to propose a reaction scheme which is in satisfactory agreement with our experimental results. It is shown that the reaction of MTBE-oxygen mixtures around 450°C does not really present the characteristics of a genuine oxidation reaction and appears more as a thermal degradation. This reaction chiefly gives formaldehyde and rather unreactive t-C 4H 9·radicals which react with oxygen to give i-C 4H 8 and HO 2·. The HO 2·radicals cannot be efficient propagating radicals in oxidation reactions. In the presence of 2% added HBr, the ignition limits of equimolecular mixtures of MTBE and O 2 are considerably lowered (∼50°C), and with 10% added HBr, the usual cool flame behavior of hydrocarbon oxidations are similarly observed for MTBE/O 2 mixtures.

Cool flame quench distances

The results of a brief experimental investigation are presented which confirm the expectation that cool flame quenching distances should be larger than hot flame quenching distances. It is also discovered that whereas quenching distances for hot flames reach their minimum values near stoichiometric conditions, cool flame quenching distances are least under rich conditions. Rich conditions are well known to favor cool flame formation.

The cool-flame oxidation of n-heptane. Part I. The kinetic features of the reaction

The oxidation of n-heptane has been investigated at subatmospheric pressures and temperatures from 220° to 360°C, and detailed kinetic data have been collected in the cool-flame region in an attempt to understand the phenomenon of cool-flame formation. A detailed ignition diagram has been constructed for a 1:1 fuel-oxygen mixture. At high pressures at temperatures above 355°C, ignition is a single stage process while at lower temperatures it is always initiated by a cool-flame, and appears in a two-stage phenomenon. Although there is a large temperature rise during this first stage it is insufficient to cause thermal ignition of the fuel, suggesting that initiation of the second stage is both chemical and thermal. During a cool-flame the temperature rise is sufficient to take the reaction mixture into the high temperature slow combustion region, when, presumably, the intermediate responsible for cool-flame formation is no longer formed. The flame is quenched, the temperature falls, and if sufficient intermediate is formed again, another flame propagates. Most of the work reported in this paper supports the “thermokinetic” theory of cool-flame formation involving one chain-branching intermediate. It is shown that in the cool-flame region there are complex relationships between cool-flame parameters (induction periods, pressure rise and maximum rate of pressure rise) and initial temperatures and pressures. These give rise to the observed complex shapes of the cool-flame and ignition limits. In turn all these phenomena reflect the complex temperature dependence of the propagation and branching reactions involved in the oxidation.

Cool flame phenomena in the oxidation of n-pentane

The temperature and pressure changes associated with the passage of a cool flame through mixtures of n-pentane and oxygen (mainly 1:1 and 1:2 mixtures) have been examined simultaneously in the temperature range 277–364°C. The maximum temperature rise can be up to ∼250°C: it reached a “ceiling” value at high initial pressures of reactants. As the initial pressure was reduced, however, the magnitude of the temperature pulse suddenly decreased sharply and at the lower pressures the maximum temperature rise was < 50°C. The passage of a cool flame gave a permanent increase in the pressure that was approximately half the magnitude of the pressure pulse, and this is consistent with analytical data, which show that a significant proportion of the n-pentane is consumed during the passage of the cool flame. The use of a fast-response recording system has shown that the pressure pulse associated with the cool flame possesses an abrupt, almost isothermal step that precedes the thermal component of the pressure pulse. Thus the present results do not indicate that a critical temperature rise is required for the appearance of a cool flame. It is suggested, therefore, that the chain-branching reactions are important in the mechanism leading to cool flame formation when n-pentane is the fuel and that the corresponding very fast reaction produces the thermal effects accompanying the passage of a cool flame.

The cool flame combustion of ethanol

The kinetics and products of the cool flame combustion of ethanol between about 280 and 330°C have been studied using a static system. The oxidation exhibited a long induction period, during which very little reaction occurred, followed by a period during which the rate, as measured by temperature and pressure increase and by consumption of reactants, accelerated exponentially. The passage of a cool flame was marked by a sharp increase in rate. During the acceleration period before a flame, the main products were hydrogen peroxide, water, carbon oxides, acetaldehyde, formaldehyde and methanol, the concentrations of which increased continuously except for acetaldehyde which rose to a constant value or even a maximum. Added acetaldehyde did not completely remove the induction period and did not reduce the acceleration period unless the amount added was greater than that normally present just before a cool flame passed. Coating the vessel with potassium chloride markedly inhibited cool flame formation. The temperature rise in a mixture up to the passage of a cool flame (∆ T c. f. ) was found to be constant at a constant ambient temperature ( T 0 ) on varying the mixture composition and total pressure and on addition of inert gases, but probably increased slightly with increasing T 0 . When only slow oxidation occurred the maximum temperature rise was less than Δ T c. f. . In contrast, the amount of alcohol consumed up to the cool flame was much less for a 1:2 than for a 1:1 ethanol/oxygen mixture at constant T 0 and varied with total pressure and on addition of inert gases. Addition of He, CO 2 and H 2 O increased the cool flame pressure limit at constant T 0 , whereas Ar and Xe reduced it. The results indicate that branching was probably due to the decomposition of an acetyldehyde-hydrogen peroxide compound and that thermal factors, particularly the thermal conductivity of the mixture, were of predominant importance in determining whether a cool flame occurred. The variation in the consumption of ethanol up to the cool flame with conditions has been explained qualitatively by considering the oxidation simply as a self-heating reaction. The sudden increase in rate when a cool flame occurred may have been due to spontaneous ignition of a peroxidic intermediate.

The low-temperature combustion of n-pentane

The reaction of n -pentane with oxygen has been investigated in the temperature range 257°–280°C at initial pressures which cover both the slow-oxidation and cool-flame regions. The oxidation has been followed by the direct analysis of reaction mixtures, together with the changes in pressure and temperature. The analytical results show that acetone is the major primary organic product of the slow reaction, and that it accounts for about 60% of the pentane consumed in the early stages of the reaction. In contrast, cool-flame formation is accompanied by the production of relatively large amounts of olefins, while the importance of acetone as a product is markedly decreased in comparison with the slow-oxidation region. These results are discussed in terms of a reaction mechanism that includes isomerization reactions of the alkylperoxy radical, and it is concluded that acetone must be formed through the formation of a hydroperoxyalkylperoxy radical rather than a decomposition reaction of a hydroperoxyalkyl radical. It is shown qualitatively that the relative importance of the various products obtained is in accord with the suggested mechanism for the over-all reaction. Simultaneous records of the pressure and temperature changes during the appearance of a cool flame show that the temperature rise is preceded by an abrupt, almost isothermal pressure rise. With n -pentane as fuel, therefore, it is suggested that branching reactions play an important role in cool-flame formation, and that the cool-flame limit is not governed solely by thermal factors. * Present address: Department of Chemical Engineering, University of Salford.

Negative temperature-coefficient of reaction rate during hydrocarbon oxidation

A new region of negative temperature-coefficient of rate is observed at the low-temperature limit for cool flame formation.

The non-isothermal oxidation of 2-methylpentane. II. The chemistry of cool flames

The products of all the modes of non-isothermal oxidation of 2-methylpentane by molecu­lar oxygen and of the attendant slow combustion reactions have been analysed by gas-liquid chromatographic and chemical methods. Oxidation in the cool-flame temperature range produces more than forty molecular species, including O -heterocycles, peroxides, alkenes and saturated and unsaturated aldehydes and ketones. A good qualitative description of the mode of formation of this complex mixture and of its variation with temperature is afforded by the alkylperoxy radical isomerization theory. This theory is developed semi-quantitatively and is in reasonable agreement with the quantitative experimental results. It is concluded that chain propagation in the cool-flame region occurs predominantly by attack on the fuel by hydroxyl radicals; the resulting oxidation is rapid and unselective. In contrast, at temperatures too low for cool-flame formation alkylperoxy radicals are the likely chain-propagating species, whereas at temperatures above the upper cool-flame limit hydroperoxy radicals probably propagate the chain. The mechanism of chain branch­ing is not clear but it is established that, in the cool-flame region, peroxidic compounds are involved.

The cool flame combustion of hydrocarbons II—Propane and n-heptane

The combustion of n-heptaneoxygen and propaneoxygen mixtures between 200° and 350°C has been studied using a static system and different reaction vessel surfaces, the oxidation being followed by measuring pressure and temperature changes and by analysis. Using an uncoated vessel cool flames were very readily obtained with n-heptane, the ‘lower temperature limit’ for a 1:1 mixture being about 230°C compared with 285° to 290°C with propane. With the former, packing the vessel or coating it with lead oxide inhibited cool flame formation to some degree, but a potassium chloride coating had little effect. The yield of peroxide, particularly of propylhydroperoxide, from the propane combustion was small even at 275°C. With n-heptane at lower temperatures the main peroxidic products were dihydroperoxyheptane and its aldehyde addition compounds, while at high temperatures or after the passage of a cool flame they were replaced by hydrogen peroxide. Possible elementary reactions occurring in the system, especially those involving the various peroxy radicals, and the role of the surface are discussed. The rate of internal hydrogen abstraction in alkylperoxy radicals probably governs the relative ease of cool flame formation with different hydrocarbons.

The cool flame combustion of hydrocarbons I—Cyclohexane

The combustion of cyclohexaneoxygen mixtures between 230° and 350°C has been studied using a static system and different reaction vessel surfaces. The oxidation was followed by measuring pressure and temperature changes and by analysis. The cool flame limits were determined both for a 1:1 mixture (temperature/total pressure limits) and at constant temperature (total pressure/composition limits). Cool flames were most easily obtained with an uncoated vessel. Packing this vessel or coating the unpacked vessel with potassium chloride or lead oxide inhibited cool flame formation. Outside the cool flame limits there was a zone of slow combustion, except with a potassium chloride coated vessel. At around 250°C the main organic products in the initial stages of the combustion were cyclohexylhydroperoxide, lower aldehydes and acetone, whereas at above 300°C, or after the passage of a cool flame, they were olefins (particularly cyclohexene). By preforming C 6 H 11 OOH in the uncoated vessel it was shown that its concentration governed the length of the induction period before a cool flame. Possible elementary processes occurring in the system, in particular reactions of cyclohexylperoxy radicals and the general mechanism of cool flame formation and quenching are discussed.

Cool flames in the combustion of toluene and ethylbenzene

The oxidation of toluene and ethylbenzene has been studied in a static system using a spherical reaction vessel (700 ml.) over the temperature range 300 to 500°C, and at total pressures up to 600 mm. Cool flames were observed in the oxidation of both hydrocarbons, but only the reaction of ethylbenzene gave rise to a ‘blue’ flame at higher temperatures. With neither hydrocarbon did periodicity in light intensity, or pressure pulses, occur. The ignition diagrams for 4 to 1 fuel + oxygen mixtures have been mapped out. With ethyl­benzene, the cool flame was maintained in a flow system, its spectrum was photographed and shown to be similar to that of fluorescent formaldehyde. The products of the reaction con­tained acetophenone, benzaldehyde and benzoic acid, phenol, quinol, hydrogen peroxide and methoxyhydroperoxide. The results have been compared with corresponding data for the oxidation of paraffin hydrocarbons, and it is concluded that, with both aromatic compounds, the processes allowing the possibility of cool-flame formation are themselves secondary in nature.

The oxidation of weak methane-air mixtures at high temperatures

The motored engine technique for studying hydrocarbon combustion processes has been used by Downs, Street, Walsh and Wheeler ~, 2 for investigating cool flame formation, and by Pastell et al2, 4 and Levedahl et al. ~ to study precombustion reactions and cool flame limits. Detailed analyses of exhaust products obtained by this method have been made by Pahnke, Cohen and Sturgis ~ and have been discussed in relation to the appearance of cool flames. In addition, Melby 7 has used the method to study the effect of fuel additives on the precombustion reactions of nheptane. The experimental method is to introduce a fuel air mixture to a motored engine and observe the reaction while the compression ratio is slowly increased until auto-ignition occurs. In the present work, however, an engine of fixed compression ratio was used and the induced charge passed through a controlled electric heating system. The initial mixture temperature was therefore the experimental variable, and could be closely controlled during observation of the reactions. Closely related work on the oxidation of methane has been carried out by Burgoyne and Hirsch s using a heated tube, the temperature ranges are similar but pressures are very much lower (atmospheric as compared with approximately 30 atm in the present work) and residence time somewhat shorter. The ignition of methane when injected into a high temperature moving air stream has been studied by Mullins 9.

Relation of Octane Number to Cool Flame Formation

Relation of Octane Number to Cool Flame Formation

A study of the reactions that lead to 'knock' in the spark-ignite engine

A study of the reactions that lead to 'knock' in the spark-ignite engine