The predictive capability of an automatically generated combustion chemistry mechanism: Chemical structures of premixed iso-butanol flames

The chemical compositions of four low-pressure premixed flames of iso-butanol are investigated with an emphasis on assessing the predictive capabilities of an automatically generated combustion chemistry model. This kinetic model had been extensively tested against earlier experimental data [S.S. Merchant, E.F. Zanoelo, R.L. Speth, M.R. Harper, K.M. Van Geem, W.H. Green, Combust. Flame (2013), http://dx.doi.org/10.1016/j.combustflame.2013.04.023.] and also shows impressive capabilities for predicting the new flame data presented here. The new set of data consists of isomer-resolved mole fraction profiles for more than 40 species in each of the four flames and provides a comprehensive benchmark for testing of any combustion chemistry model for iso-butanol. Isomer-specificity is achieved by analyzing flames, which are burner-stabilized at equivalence ratios of ? = 1.0–1.5 and at pressures between 15 and 30 Torr, with molecular-beam mass spectrometry and single-photon ionization by tunable vacuum-ultraviolet synchrotron radiation. Predictions of the C₂H₄O, C₃H₆O, and C₄H₈O enol–aldehyde–ketone isomers are improved compared to the earlier work by Hansen et al. [N. Hansen, M. R. Harper, W. H. Green, Phys. Chem. Chem. Phys. 13 (2011) 20262-20274] on similar n-butanol flames. A reaction path analysis identifies prominent fuel-consumption and oxidation sequences. Almost all of the species mole fraction data reported here are predicted within the measurement uncertainties of a factor of two to three. Some significant differences with previous published models are highlighted.     

Effects of buffer gas composition on autoignition

This work quantifies the chemical kinetic and thermal effects of buffer gas composition on autoignition of three fuels at conditions relevant to engines, combustors, and experimental facilities used to study ignition kinetics. Computational simulations of autoignition of iso-octane, n-heptane, and of n-butanol were used to characterize the effects of buffer gas composition on ignition delay time and heat release rate. Stoichiometric mixtures, ? = 1.0, and a temperature range of 600–1100 K were considered. Iso-octane and n-heptane were studied at initial pressures of 9.0 atm and 60.0 atm, and n-butanol was studied at initial pressures of 3.2 atm and 60.0 atm. Two dilution levels of buffer gas to O₂ of 3.76:1 (mole basis) and 5.64:1 were considered (∼21% and ∼15% O₂ respectively, mole basis). The fuels and simulation conditions were selected based on the relevance to engine operating conditions and previously published ignition studies. The buffer gases considered were argon, nitrogen, water, and carbon dioxide. Simulation results predicted changes of greater than a factor of 2 in ignition delay time and heat release rate as a function of buffer gas composition in the negative temperature coefficient (NTC) region for n-heptane and iso-octane. Outside the NTC region, the predicted effects of changes in buffer gas composition were small (<20%); however, experimental data for n-heptane indicate larger effects of buffer gas composition on ignition delay time at higher temperatures (>a factor of 2). The heat release rates were also sensitive to buffer gas composition, with carbon dioxide exhibiting relatively high levels of early and late heat release relative to the other buffer gases. Sensitivity analysis of the third-body collision efficiencies for the buffer gases showed the effects of uncertainties in the third body collision efficiencies on ignition delay time and heat release rate. The results highlight the significance of buffer gas composition on low-temperature combustion chemistry, particularly via H₂O₂ and HO₂ decomposition and recombination reactions.     

Experimental and kinetic modeling study of 2-butanol pyrolysis and combustion

2-Butanol (sC₄H₉OH) pyrolysis has been studied in a flow reactor with the synchrotron vacuum ultraviolet photoionization mass spectrometry combined with the molecular-beam sampling technique. The pyrolysis species were identified and their mole fractions were determined. Four pressures of 5, 30, 150 and 760 Torr were selected to study the pressure dependence of 2-butanol pyrolysis chemistry. The temperature- and pressure-dependent rate constants of unimolecular reactions of 2-butanol were calculated with the RRKM/Master Equation method. With the help of theoretical calculations, a detailed kinetic model consisting of 160 species and 1038 reactions was developed to simulate the 2-butanol pyrolysis. It is concluded that the mole fractions of pyrolysis species are very sensitive to the 2-butanol unimolecular reaction rates. To enhance the accuracy, the model is further validated by the species profiles in shock tube pyrolysis, a rich laminar premixed flame, oxidation data from jet-stirred reactor, ignition delay times, and laminar flame speed. Good agreements between the predicted and measured results were obtained.     

Laminar flame propagation of atmospheric iso-cetane/air and decalin/air mixtures

Laminar flame speeds of iso-cetane/air and decalin/air mixtures were measured in the counterflow configuration at atmospheric pressure and an elevated unburned mixture temperature of 443 K. Axial flow velocities were measured along the stagnation streamline using the digital particle image velocimetry. The laminar flame speeds were determined by determining the variation of a reference flame speed as a function of strain rate and computationally assisted non-linear extrapolations. The data are the first to be reported in the literature, and they were modeled using a recently developed kinetic model that includes 187 species and 6086 elementary reactions. In general, the computed results were found to be in close agreement with the data. In order to get insight into kinetic effects on flame propagation, detailed sensitivity and reaction path analyses were performed using the computed flame structures. The results revealed that at the same equivalence ratio, laminar flame speeds of iso-cetane/air mixtures are lower than those of n-hexadecane/air mixtures. Additionally, it was found that the laminar flame speeds of iso-cetane/air and decalin/air mixtures are sensitive largely to C₀–C₄ kinetic subset, and that the lower reactivity of iso-cetane compared to n-hexadecane could be attributed to the higher production of relatively stable intermediates.     

A modelling study of aromatic soot precursors formation in laminar methane and ethene flames

A relatively short kinetic mechanism (93 species and 729 reactions) was developed to predict the formation of poly-aromatic hydrocarbons (PAH) and their growth of up to five aromatic rings in methane and ethane-fueled flames. The model is based on the C₀-C₂ chemistry with recent well-established chemical kinetic data. Reaction paths for mostly stable and well studied PAH molecules were delineated and the reaction rate constants for PAH growth were collected. These were obtained by analysing the data reported in the literature during the last 30 years, or by using the estimates and optimisations of experimentally measured concentration profiles for small and PAH molecules. These profiles were collected by 12 independent work groups in laminar premixed CH₄ and C₂H₄ flames under atmospheric pressure or in shock tube experiments under elevated pressure. The simulated flame speeds, temporal profiles of small and large aromatics and also soot particles volume fraction data are in good agreement with the experimental data received for different temperatures, mixing ratios and diluents. The extensive analysis of PAH reaction steps showed that the main reaction routes can be conditionally divided into “low temperature” reaction routes, dominating at T < 1550 K and “high temperature” reaction routes, which contribute mostly to PAH formation at T > 1550 K. The presented mechanism can be used as the basis for further extensions or reductions applied in kinetic schemes involving PAH and soot production in practical fuel combustion.     

Sooting behaviors of n-butanol and n-dodecane blends

This work focuses on understanding the formation and oxidation of soot when adding n-butanol, an oxygenated fuel, to n-dodecane. A two-stage burner was used to characterize the oxidation of soot from different n-butanol blends, 10%, 30%, and 60 mol% in n-dodecane. The two-stage burner isolates the soot oxidation process from the formation process. Soot is formed in a first-stage premixed burner under fuel-rich conditions, while in a second stage, the soot is oxidized under slightly fuel-rich conditions. A scanning mobility particle sizer (SMPS) was used to measure the soot particle size distributions in the flame at different heights during oxidation. Results showed a decrease in particle mass concentration (g/cm³) as the fraction of n-butanol increased, which indicates the capability of n-butanol to reduce soot particle number and mass. On the other hand, the results demonstrated that the increasing n-butanol reduces the difference between initial mass of soot particles entering and the final mass of soot particles leaving the second burner. This result implies that increasing the n-butanol concentration decreases the rate of soot oxidation. Two different fuel quality indicators are used to quantify our observations. The first one, “sooting tendency”, is calculated to show how the amount of soot formed in the flame is affected by using different n-butanol percentages. The second one, “sooting stability”, is defined for quantifying the stability of soot particles against oxidation. The results demonstrated that by increasing the n-butanol percentage, soot formation was suppressed. However, sooting stability increased with higher concentrations of n-butanol. The soot nanostructure was quantified by high-resolution electron microscopy and digital image processing. Image analysis revealed layer arrangement is in correlation with sooting stability. The results of interlayer spacing showed a decrease by increasing n-butanol at the same sampling height.     

Detailed kinetic study of anisole pyrolysis and oxidation to understand tar formation during biomass combustion and gasification

Anisole was chosen as the simplest surrogate for primary tar from lignin pyrolysis to study the gas-phase chemistry of methoxyphenol conversion. Methoxyphenols are one of the main precursors of PAH and soot in biomass combustion and gasification. These reactions are of paramount importance for the atmospheric environment, to mitigate emissions from wood combustion, and for reducing tar formation during gasification. Anisole pyrolysis and stoichiometric oxidation were studied in a jet-stirred reactor (673–1173 K, residence time 2 s, 800 Torr (106.7 kPa), under dilute conditions) coupled with gas chromatography–flame ionization detector and mass spectrometry. Decomposition of anisole starts at 750 K and a conversion degree of 50% is obtained at about 850 K under both studied conditions. The main products of reaction vary with temperature and are phenol, methane, carbon monoxide, benzene, and hydrogen. A detailed kinetic model (303 species, 1922 reactions) based on a combustion model for light aromatic compounds has been extended to anisole. The model predicts the conversion of anisole and the formation of the main products well. The reaction flux analyses show that anisole decomposes mainly to phenoxy and methyl radicals in both pyrolysis and oxidation conditions. The decomposition of phenoxy radicals is the main source of cyclopentadienyl radicals, which are the main precursor of naphthalene and heavier PAH in these conditions.     

Effects of ultra-high injection pressure and micro-hole nozzle on flame structure and soot formation of impinging diesel spray

The effects of ultra-high injection pressure (P_inj = 300 MPa) and micro-hole nozzle (d = 0.08 mm) on flame structure and soot formation of impinging diesel spray were studied with a high speed video camera in a constant volume combustion vessel. Two-color pyrometry was used to measure the line-of-sight soot temperature and concentration with two wavelengths of 650 and 800 nm. A flat wall vertical to the injector axis is located 30 mm away from the injector nozzle tip to generate impinging spray flame. Three injection pressures of 100, 200 and 300 MPa and two injector nozzles with diameters of 0.16 and 0.08 mm were used. With the conventional injector nozzle (0.16 mm), ultra-high injection pressure generates appreciably lower soot formation. With the micro-hole nozzle (0.08 mm), impinging spray flame shows much smaller size and lower soot formation at the injection pressure of 100 MPa. The soot formation is too weak to be detected with the micro-hole nozzle at injection pressures of 200 and 300 MPa. With eliminating the impact of injection rate on soot level, both ultra-high injection pressure and micro-hole nozzle have an obvious effect on soot reduction. Soot formation characteristics of impinging spray flame were compared with those of free spray flame using both the conventional and micro-hole nozzles. With the conventional nozzle, flat wall impingement deteriorates soot formation significantly. While soot formation characteristics of free spray flame with the micro-hole nozzle are not altered obviously by flat wall. Liquid length of the 0.16 mm nozzle is longer than the impingement distance and liquid length of the 0.08 mm nozzle is shorter than the impingement distance. Liquid impingement upon the wall is responsible for the deteriorated soot level of impinging flame compared to that of free flame with the conventional nozzle.     

Experimental and kinetic modeling study of i-butanol pyrolysis and combustion

i-Butanol (iC₄H₉OH) pyrolysis has been studied in a flow reactor with synchrotron vacuum ultraviolet photoionization mass spectrometry combined with molecular-beam sampling technique. The pyrolysis species were identified and their mole fractions were determined. Three pressures of 30, 150 and 760 Torr were selected to study the pressure effect of i-butanol chemistry. A detailed kinetic model consisting of 186 species and 1294 reactions was developed to simulate i-butanol high temperature chemistry. To enhance the accuracy, the model was further validated by the species profiles in shock tube pyrolysis, laminar premixed flames, oxidation data from jet-stirred reactor, ignition delay times, and flame speeds. Good agreement between the predicted and measured results was obtained.     

Experimental and kinetic modeling study of pyrolysis and oxidation of n-decane

The pyrolysis of n-decane was investigated in a flow reactor at 5, 30, 150 and 760 Torr, and the oxidation of n-decane at equivalence ratios of 0.7, 1.0 and 1.8 was studied in laminar premixed flames at 30 Torr. In both experiments, synchrotron vacuum ultraviolet photoionization mass spectrometry (SVUV-PIMS) was used to identify combustion species and measure their mole fraction profiles. A new detailed kinetic model of n-decane with 234 species and 1452 reactions was developed for applications in intermediate and high temperature regions, and was validated against the experimental results in the present work. The model was also validated against previous experimental data on n-decane combustion, including species profiles in pyrolysis and oxidation in high pressure shock tube and atmospheric pressure flow reactor, jet stirred reactor oxidation, atmospheric pressure laminar premixed flame, counterflow diffusion flame and global combustion parameters such as laminar flame speeds and ignition delay times. In general, the performance of the present model in reproducing these experimental data is reasonably good. Sensitivity analysis and rate of production analysis were conducted to understand the decomposition processes of n-decane.     

Modeling fuel NOx formation from combustion of biomass-derived producer gas in a large-scale burner

This study investigates the characteristics of fuel NO_x formation resulting from the combustion of producer gas derived from biomass gasification using different feedstocks. Common industrial burners are optimized for using natural gas or coal-derived syngas. With the increasing demand in using biomass for power generation, it is important to develop burners that can mitigate fuel NO_x emissions due to the combustion of ammonia, which is the major nitrogen-containing species in biomass-derived gas. In this study, the combustion process inside the burner is modeled using computational fluid dynamics (CFD) with detailed chemistry. A reduced mechanism (36 species and 198 reactions) is developed from GRI 3.0 in order to reduce the computation time. Combustion simulations are performed for producer gas arising from different feedstocks such as wood gas, wood + 13% DDGS (dried distiller grain soluble) gas and wood + 40% DDGS gas and also at different air equivalence ratios ranging from 1.2 to 2.5. The predicted NO_x emissions are compared with the experimental data and good levels of agreement are obtained. It is found out that NO_x is very sensitive to the ammonia content in the producer gas. Results show that although NO–NO₂ interchanges are the most prominent reactions involving NO, the major NO producing reactions are the oxidation of NH and N at slightly fuel rich conditions and high temperature. Further analysis of results is conducted to determine the conditions favorable for NO_x reduction. The results indicate that NO_x can be reduced by designing combustion conditions which have fuel rich zones in most of the regions. The results of this study can be used to design low NO_x burners for combustion of gas mixtures derived from gasification of biomass. One suggestion to reduce NO_x is to produce a diverging flame using a bluff body in the flame region such that NO generated upstream will pass through the fuel rich flame and be reduced.     

Experiments and modeling of propane combustion with vitiation

The chemical species composition of a vitiated oxidizer stream can significantly affect the combustion processes that occur in many propulsion and power generation systems. Experiments were performed to investigate the chemical kinetic effects of vitiation on ignition and flame propagation of hydrocarbon fuels using propane. Atmospheric-pressure flow reactor experiments were performed to investigate the effect of NO_x on propane ignition delay time at varying O₂ levels (14–21 mol%) and varying equivalence ratios (0.5–1.5) with reactor temperatures of 875 K and 917 K. Laminar flame speed measurements were obtained using a Bunsen burner facility to investigate the effect of CO₂ dilution on flame propagation at an inlet temperature of 650 K. Experimental and modeling results show that small amounts of NO can significantly reduce the ignition delay time of propane in the low- and intermediate-temperature regimes. For example, 755 ppmv NOx in the vitiated stream reduced the ignition delay time of a stoichiometric propane/air mixture by 75% at 875 K. Chemical kinetic modeling shows that H-atom abstraction reaction of the fuel molecule by NO₂ plays a critical role in promoting ignition in conjunction with reactions between NO and less reactive radicals such as HO₂ and CH₃O₂ at low and intermediate temperatures. Experimental results show that the presence of 10 mol% CO₂ in the vitiated air reduces the peak laminar flame speed by up to a factor of two. Chemical kinetic effects of CO₂ contribute to the reduction in flame speed by suppressing the formation of OH radicals in addition to the lower flame temperature caused by dilution. Overall, the detailed chemical kinetic mechanism developed in the current work predicts the chemical kinetic effects of vitiated species, namely NO_x and CO₂, on propane combustion reasonably well. Moreover, the reaction kinetic scheme also predicts the negative temperature coefficient (NTC) behavior of propane during low-temperature oxidation.     

Effects of water vapor addition to the air stream on soot formation and flame properties in a laminar coflow ethylene/air diffusion flame

The effects of adding water vapor to the air stream on flame properties and soot volume fraction were investigated numerically in a laminar coflow ethylene/air diffusion flame at atmospheric pressure by solving the fully elliptic conservation equations and using a detailed C₂ reaction mechanism including PAH up to pyrene and detailed thermal and transport properties. Thermal radiation was calculated using the discrete-ordinates method and a statistical narrow-band correlated-k based wide band model for the absorption coefficients of CO₂ and H₂O. Soot formation was modeled using a PAH based inception model and the HACA mechanism for surface growth and oxidation. Addition of water vapor significantly reduces radiation heat loss from the flame primarily through reduced soot loading and flame temperature. The added water vapor affects soot formation and flame properties through not only dilution and thermal effects, but also through chemical effect. The chemical effect is as significant as the dilution and thermal effects. The primary pathway for the chemical effect of water vapor is the reverse reaction of OH + H₂ ↔ H + H₂O. Our numerical results confirm that the reduced H radical concentration leads to lower PAH concentrations and consequently lower soot inception rates. In contrast, the radiation effect due to the added water vapor was found to have a minor influence on both flame structure and soot formation in the laminar diffusion flame investigated.     

An experimental and kinetic modeling study of n-butanol combustion

n-Butanol is a fuel that has been proposed as an alternative to conventional gasoline and diesel fuels. In order to better understand the combustion characteristics of n-butanol, this study presents new experimental data for n-butanol in three experimental configurations. Species concentration profiles are presented in jet stirred reactor (JSR) at atmospheric conditions and a range of equivalence ratios. The laminar flame speed obtained in an n-butanol premixed laminar flame is also provided. In addition, species concentration profiles for n-butanol and n-butane in an opposed-flow diffusion flame are presented. The oxidation of n-butanol in the aforementioned experimental configurations has been modeled using an improved detailed chemical kinetic mechanism (878 reactions involving 118 species) derived from a previously proposed scheme in the literature. The proposed mechanism shows good qualitative agreement with the various experimental data. Sensitivity analyses and reaction path analyses have been conducted to interpret the results from the JSR and opposed-flow diffusion flame. It is shown that the main reaction pathway in both configurations is via H-atom abstraction from the fuel followed by β-scission of the resulting fuel radicals. Several unimolecular decomposition reactions are important as well. This study gives a better understanding of n-butanol combustion and the product species distribution.     

Experimental study of ethylene counterflow diffusion flames perturbed by trace amounts of jet fuel and jet fuel surrogates under incipiently sooting conditions

The structure of an ethylene counterflow diffusion flame doped with 2000 ppm on a molar basis of either jet fuel or two jet fuel surrogates is studied under incipient sooting conditions. The doped flames have identical stoichiometric mixture fractions (z_f = 0.18) and strain rates (a = 92 s⁻¹), resulting in a well-defined and fixed temperature/time history for all of the flames. Gas samples are extracted from the flame with quartz microprobes for subsequent GC/MS analysis. Profiles of critical fuel decomposition products and soot precursors, such as benzene and toluene, are compared.The data for C7-C12 alkanes are consistent with typical decomposition of large alkanes with both surrogates showing good qualitative agreement with jet fuel in their pyrolysis trends. Olefins are formed as the fuel alkanes decompose, with agreement between the surrogates and jet fuel that improves for small alkenes, probably because of an increase in kinetic pathways which makes the specifics of the alkane structure less important.Good agreement between jet fuel and the surrogates is found with respect to critical soot precursors such as benzene and toluene. Although the six-component Utah/Yale surrogate performs better than the Aachen surrogate, the latter performs adequately and retains the advantage of simplicity, since it consists of only two components.The acetylene profiles present a unique multimodal behavior that can be attributed to acetylene’s participation in early stages of formation of soot precursors, such as benzene and other large pyrolysis products, as well as in the surface growth of soot particles.     

Oxidation of 1-butanol and a mixture of n-heptane/1-butanol in a motored engine

The oxidation of neat 1-butanol and a mixture of n-heptane and 1-butanol was studied in a modified CFR engine at an equivalence ratio of 0.25 and an intake temperature of 120 °C. The engine compression ratio was gradually increased from the lowest point to the point where significant high temperature heat release was observed. Heat release analyses showed that no noticeable low temperature heat release behavior was observed from the oxidation of neat 1-butanol while the n-heptane/1-butanol mixture exhibited pronounced cool flame behavior. Species concentration profiles were obtained via GC-MS and GC-FID/TCD. Quantitative analyses of the reaction products from the oxidation of neat 1-butanol indicate that 1-butanol is consumed mainly through H-atom abstraction. Among the H-atom abstraction reactions, it is observed that the H-atom abstraction from the α-carbon of 1-butanol is particularly favored. The investigation on the oxidation of the mixture of n-heptane/1-butanol showed that the oxidation of 1-butanol is facilitated at low temperatures through the radical pool generated from the oxidation of n-heptane.     

Measurements of sooting tendency in laminar diffusion flames of n-heptane at elevated pressure

This research focuses on the effects of an increasing pressure on the soot formation during combustion of vaporized liquid fuel. Therefore soot formation is measured in a laminar diffusion flame, with n-heptane as fuel, over a range of pressures from 1.0 to 3.0 bar. The soot volume fraction in the diffusion flames has been measured using Laser-Induced Incandescence (LII) calibrated by means of the Line Of Sight Attenuation (LOSA) technique. The values of the calibration factors between LII intensities and soot volume fraction from LOSA are slightly varied for different pressure. The integral soot volume fractions show power law dependence on pressures, being proportional to pⁿ, with n being 3.4 ± 0.3 in the pressure range of 1.0–3.0 bar.     

Study on combustion and ignition characteristics of natural gas components in a micro flow reactor with a controlled temperature profile

Combustion and ignition characteristics of natural gas components such as methane, ethane, propane and n-butane were investigated experimentally and computationally using a micro flow reactor with a controlled temperature profile. Special attention was paid to weak flames which were observed in a low flow velocity region. The observed weak flame responses for the above fuels were successfully simulated by one-dimensional computations with a detailed kinetic model for natural gas. Since the position of the weak flame indicates the ignition characteristics as well as the reactivity of each fuel, the experimental and computational results were compared with research octane number (RON) which is a general index for ignition characteristics of ordinary fuels. At 1 atm, ethane showed the highest reactivity among these fuels, although RON of ethane (115) is between those of methane (120) and propane (112). Since the pressure conditions are different between the present experiment and the general RON test, weak flame responses to the pressure were investigated computationally for these fuels. The order of the fuel reactivity by the reactor agreed with that by RON test when the pressure was higher than 4 atm. Reaction path analysis was carried out to clarify the reasons of the highest reactivity of ethane at 1 atm among the employed fuels in this study. The analysis revealed that C₂H₅ + O₂ ⇔ C₂H₄ + HO₂ is a key reaction and promotes ethane oxidation at 1 atm. The effect of the pressure on the fuel oxidation process in the present reactor was also clarified by the analysis. In addition, weak flame responses to various mixing ratios of methane/n-butane blends were investigated experimentally and computationally. The results indicated a significant effect of n-butane addition in the blends on combustion and ignition characteristics of the blended fuels.     

Experimental study and detailed kinetic modeling of the effect of exhaust gas on fuel combustion: mutual sensitization of the oxidation of nitric oxide and methane over extended temperature and pressure ranges

New experimental results were obtained for the mutual sensitization of the oxidation of NO and methane in a fused silica jet-stirred reactor operating at 1-10 atm, over the temperature range 800-1150 K. Probe sampling followed by on-line FTIR analyses and off-line GC-TCD/FID analyses allowed the measurement of concentration profiles for the reactants, stable intermediates, and final products. Detailed chemical kinetic modeling of the experiments was performed. An overall reasonable agreement between the present data and modeling was obtained, whereas previously published models failed to properly represent these new data. According to the proposed model, the mutual sensitization of the oxidation of methane and NO proceeds through the NO to NO₂ conversion by HO₂ and CH₃O₂. The modeling showed that at 1-10 atm, the conversion of NO to NO₂ by CH₃O₂, is more important at low temperatures (800 K) than at higher temperatures (850-900 K), where the reaction of NO with HO₂ dominates the production of NO₂. The NO to NO₂ conversion is enhanced by the production of HO₂ and CH₃O₂ radicals from the oxidation of the fuel. The production of OH resulting from the oxidation of NO promotes the oxidation of the fuel: NO + HO₂ ? OH + NO₂ is followed by OH + CH₄ ? CH₃. At low temperature, the reaction further proceeds via CH₃ + O₂ ? CH₃O₂, CH₃O₂ + NO ? CH₃O + NO₂. At higher temperatures, the production of CH₃O involves NO₂: CH₃ + NO₂ ? CH₃O. The sequence is followed by CH₃O ? CH₂O + H, CH₂O + OH ? HCO, HCO + O₂ ? HO₂, and H + O₂ ? HO₂. ? CH₂O + H, CH₂O + OH ? HCO, HCO + O₂ ? HO₂, and H + O₂ ? HO₂.