Thompson M. Sloane

Combutational studies of end-wall flame quenching at low pressure: The effects of heterogeneous radical recombination and crevices

Abstract A computational study of the effect of heterogeneous radical destruction and crevices on end-wall flame quenching in a lean CH4O2Ar mixture at low pressure has been performed. The unsteady conservation equations for a chemically reacting multicomponent gas have been solved with the commercially available partial differential equation solver package PDEPACK. Heterogeneous radical destruction at a cooled surface was found to have a minor effect on flame quenching compared to homogeneous destruction in the cooled flame gases near the surface. These results are in qualitative agreement with experiments performed in a side-wall geometry. Cooled crevices, however, had a much greater effect on the postquench oxidation of the fuel. Whereas the fuel layer left after quenching near a flat wall burns up more rapidly as the pressure is increased, unburned fuel left in a crevice after quencing burns up more slowly as the pressure is increased. This reaffirms the potential importance of crevices as sources of unburned hydrocarbons in an engine.

Energy Requirements for Spherical Ignitions in Methane-Air Mixtures at Different Equivalence Ratios

Abstract Computational studies of ignition are reported where the ignition energy is added as an initial condition in a form which serves to raise the gas temperature (heat) or as a combination of dissociation of the fuel and oxygen with heat. The effect of these different energy addition methods on the time history of the ignition kernel and the minimum ignition energy has been investigated in atmospheric-pressure methane-air mixtures. The ignition energy in all of these calculations is added as an initial condition. At low ignition energy densities where the maximum initial temperature in the center of the ignition kernel region is aboul 2000 K. depositing 10% of the ignition energy in dissociation of methane and oxygen leads to a reduced ignition delay and a reduced minimum ignition energy compared to addition of the ignition energy in the form of heat only. These comparisons were made at the same ignition energy and ignition energy density. However, the observed reductions disappear if the ignition en...

Reaction product identification from O(3P)+benzene, toluene, and 1,3,5‐trimethylbenzene collisions in crossed molecular beams

Although absolute rate coefficients have been measured for a number of reactions of aromatic molecules with O(3P) atoms, very little is known about the mechanisms of these reactions. In order to identify the products of O atom–aromatic molecule reactions, the products of single reactive collisions between oxygen atoms and benzene, toluene, and 1,3,5,‐trimethylbenzene have been observed in crossed molecular beam experiments. The products were detected with a quadrupole mass spectrometer. Two product paths were observed for O+benzene: (1) the O atom–benzene adduct, which is most likely phenol, and (2) carbon monoxide and a C5H6 hydrocarbon which is probably 3‐penten‐l‐yne. An additional path involving the formation of (methyl‐substituted) benzaldehyde and H2 was observed for the reaction with methyl‐substituted benzenes. Only the adducts, the benzaldehydes, and the olefins were observed directly. The identities of the corresponding products were inferred from the difference between the mass‐to‐charge ratio ...

Numerical Simulation of Electric Spark Ignition in Atmospheric Pressure Methane-Air Mixtures

Abstract A model has been developed to describe the growth of an ignition kernel produced by an electric spark in quiescent methane-air mixtures at atmospheric pressure. The model involves the solution of the conservation equations of mass, momentum, and energy in a chemically reacting fluid in one-dimensional spherical coordinates. Detailed chemical kinetics for methane oxidation are included in the equations, along with an accurate method of calculating the transport coefficients. The electric spark energy is added in the form of internal energy; ionization in the discharge is neglected. The temporal and spatial dependence of this internal energy addition can be taken from the current, voltage, and duration of the discharge which produces the ignition kernel being modeled. The agreement with experimental results for arc and glow discharges is quite good.

The Effect of Selective Energy Deposition on the Homogeneous Ignition of Methane and its Implication for Flame Initiation and Combustion Enhancement

Abstract This paper describes a computational study of the ignition of methane mixtures by selective ignition energy deposition. In this work the ignition energy was added as an initial condition to the combustible mixture in a specific manner, either in the form of heat or a combination of heat and selected, highly reactive atoms or molecules. The resulting induction time for the ignition of homogeneous combustion was found to decrease with increasing fraction of ignition energy deposited into dissociation for the same total ignition energy at a given equivalence ratio. In addition to these calculations performed at constant energy, other calculations were performed where equal amounts of the radicals H, O, and OH were added to assess the effect of direct addition of these radicals on the induction time. Oxygen atoms yielded the shortest induction time in the stoichiometric case and hydrogen atoms yielded the shortest induction time in the lean case. Hydroxyl radicals yielded by far the longest induction...

A Comparison of Ignition Phenomena Modelled with Detailed and Simplified Kinetics

Recent numerical calculations have predicted minimum ignition energies for stoichiometric methane-air mixtures which are a factor of 70 lower than experimental values. We show that most of this discrepancy is due to the use of a one-step kinetic model whose numerical values were chosen to reproduce burning velocities and heat-release profiles of steady one-dimensional flames. When a detailed chemical mechanism is employed, the predicted minimum ignition energies are close to the experimental values,especially when losses present in the experiments are considered. The reasons why these one-step models do not provide good quantitative predictions of ignition phenomena are discussed and the implications for the modelling of transient name phenomena with simplified chemical models are given.

A molecular beam mass spectrometer study of side-wall flame quenching at low pressure by cooled noncatalytic and catalytic surfaces

Abstract Molecular beam mass spectrometer sampling has been used to probe the chemistry of a flat lean CH 2 O 2 Ar flame at 4.0 kPa which is cooled by a gold and a platinum surface whose temperature is held at 373K. The gold surface is chemically inert, and its effect on the flame should be limited to cooling of the flame gases. The platinum surface should in addition recombine H, O, and OH radicals which collide with the surface. This approximates a situation which may occur in the engine if the wall deposits are capable of recombining or otherwise destroying radicals which collide with the cylinder wall. Our experimental arrangement then constitutes a study of side-wall flame quenching by noncatalytic and catalytic cooled surfaces. Previous studies of side-wall quenching by an inert cooled surface have included measurements of only the stable flame components. This work reports measurements of several stable and radical component mole fractions, affording a detailed picture of the flame chemistry near the surfaces.

Numerical Simulation of Electric Spark Ignition in Methane-Air Mixtures at Pressures Above One Atmosphere

Abstract A model previously developed for describing ignition kernel growth in spark ignition experiments conducted in atmospheric-pressure methane-air mixtures has been tested at pressures of 200 and 400 kPa. A comparison of published results of experiments performed at 200 kPa showed that the measured size of the ignition kernel as a function of time was approximately twice that obtained from the calculations. The measured and calculated curves were roughly parallel, however, indicating that the calculations gave reasonable agreement with the measured time-dependent laminar burning velocity. The calculated ignition kernel growth at 400 kPa was found to agree reasonably well with experimental measurements at that pressure. The calculated time-dependent temperature profile was also in moderately good agreement with measured temperatures in the ignition kernels in the 400 kPa experiments, except at the center of the ignition region where heat loss to the electrodes, which is not included in the model, shou...

A Computational Study of Ignition by Oxygen Dissociation

Abstract Numerical calculations have been performed on the ignition of methane-oxygen-argon mixtures due to dissociation of molecular oxygen into ground state oxygen atoms. The computations are caried out using a time and spatially dependent fluid mechanics model in one-dimension coupled with a detailed oxidation mechanism for methane. A calculation of the minimum ignition energy as a function of equivalence ratio shows that ignition by oxygen dissociation is clearly different from laboratory experimental results of spark ignition. Whereas the minimum spark ignition energy occurs at Math for methane-oxygen-argon mixtures, the computations show a steadily declining minimum ignition energy from Math. Increasing the amount of oxygen dissociation decreases the induction time and increases the flame propagation rate during tht early part of the burning. This enhanced flame propagation rate lasts only until the oxygen atom concentration relaxes to a level which is typical of a flame moving at a constant speed. ...

Time-Resolved Mass Spectrometry of a Propagating Methane-Oxygen-Argon Flame

Abstract An apparatus has been constructed for molecular beam sampling of transient combustion phenomena using time-resolved mass spectrometry. This new method of flame diagnostics has significant advantages over other currently available time-resolved techniques. A large number of flame components can be detected with excellent spatial and time resolution at the expense of a small flame disturbance. The method has been demonstrated by time-resolved measurements of the partial pressures of several chemical components in a stoichiometric methane-oxygen-argon mixture burning at an initial pressure of 32 kPa (240 torr). In addition to the major stable flame components, the important OH radical can also be detected. This new method will be applied to the study of the chemical initiation and modification of combustion.