Symposium (International) on Combustion最新文献

The effect of a flammable spray on thermal explosion in a preheated combustible gas mixture is investigated using a simplified model that contains the essentials of the basic physical processes at work. The study represents a re-examination of the question of the ignition of a spray of droplets from the viewpoint of an explosion problem, in which the droplets are taken to be a source of endothermicity. Use is made of various methods for the qualitative analysis of systems of differential equations in order to examine the dynamics of the system. Possible types of dynamical behavior of the system are looked into and parametric regions of their existence are determined analytically. Peculiarities, of these dynamical regimes are investigated, and their dependence on the physical system parameters are analyzed. In particular, analytical formulas are developed for ignition delay times by exploiting the sensitivity of the process to the chemical activation energy. A qualitative comparison of predicted ignition times with independent experimental measurements from the literature yields good order of magnitude agreement.

Simulations of laminar combustion and other reactive flow processes (like chemical vapor deposition, plasma etching, etc.) are presently carried out in most cases using the transport code TRANFIT attached to the CHEMKIN package. The approach used is based on experimental data from 1975 and is now outdated, especially in view of recent work presented in the literature.

The new approach described here seeks to remove the deficiencies of former transport models by using the following features: (1) representation of transport data of light species at high temperature by switching to an exponential repulsive potential, (2) use of effective potential parameters to handle the intermolecular forces in an easy and elegant way, if polar molecules are considered, and (3) use of a simplified formula for binary thermal diffusion factors, based on an expansion for large values of the mass ratio of the species included.

This paper presents the new transport model in terms of a complete set of equations. The molecular parameters provided allow a complete treatment of the oxidation of H₂ and H₂/CO mixtures (data for species taking place in the oxidation of hydrocarbons and in other reaction systems are not yet available). To demonstrate the consequences of the new transport model for combustion processes, results have been generated by implementing the model in a code for the simulation of premixed laminar flames.

Results of two-dimensional numerical computations of turbulent methane flames using detailed and reduced chemistry are analyzed in the context of a new theory for premixed turbulent combustion. This theory defines the thin reaction zones regine, where the Kolmogorov scale is smaller than the preheat zone thickness but larger than the reaction zone thickness. The two numerical computations considered in this paper fall clearly within this regime. A lean and a stoichiometric flame are considered. The former is characterized by a large ratio of the turbulence intensity to the laminar burning velocity and the latter by a smaller value of that ratio.

The displacement speed of the reaction zone relative to the flow is defined as the displacement speed of the isoscalar line at a fuel mass fraction corresponding to 10% of the upstream value. The three different mechanisms that are contributing to the displacement of the reaction zone, namely, normal and tangential diffusion and reaction, are analyzed and their probability density functions are evaluated. Although these contributions fluctuate considerably, the mean value of the overall displacement speed is found to be only around 40% different from the burning velocity of a plane premixed flame at the same equivalence ratio. Furthermore, the contribution of tangential diffusion, which can be expressed as a curvature term, cancels as far as the mean overall displacement speed is concerned, while the contributions of normal diffusion and reaction are large but have opposite signs. These contributions depend implicitly on curvature. This dependence is small for the lean flame but considerable for the stoichiometric flame where it leads to an enhanced diffusivity. This diffusivity is compared to the Markstein diffusivity that describes the equivalent curvanture effect in the corrugated flamelet regime.

The rate constant for the H+NO+N₂ reaction (R1,N₂) has been determined in the temperature range 1000–1170K from flow-reactor experiments on the CO/O₂/H₂O/N₂ system perturbed with different amounts of NO. The initiation temperature of this system is highly sensitive to reaction R1, which is the rate-controlling step in the nitric oxide catalyzed removal of hydrogen atoms. Based on the flow-reactor results and the limited amount of data reported in literature, a rate constant for the H+NO+N₂ reaction of 4.0×10²⁰T^−1.75 cm⁶/(mol²s) was determined. This value is in good agreement with the recent result of Allen and Dryer at 1000K but significantly lower at high temperatures than the recommendation of Tsang and Herron. With the recently determined value of ΔH_f,298 (HNO) of 26.0 kcal/mol, which is 2 kcal/mol higher than previous estimates, our results correspond to a rate constant of 1.7×10¹⁹T^−1.5 exp(−23,400/T) cm³/(mol s) for the HNO+N₂ dissociation reaction in the 1000–2500 K range. The sharp drop-off in the rate constant for H+NO+M at high temperatures suggested by the flow-reactor results are supported by reinterpretation of data reported in literature on H₂/O₂/N₂ flames doped with NO. Theoretical considerations suggest that the effect can be attributed to weak-collision effects.

A fuel-rich, near-sooting, low-pressure, premixed haphthalene/oxygen/argon flame has been analyzed for condensible raction intermediates, mainly mono- and polycyclic aromatic hydrocarbons (PAHs), and for free radicals. Samples were taken through a nozzle beam that was expanded and condensed together with the radical scavenger dimethyldisulfide in a cold trap. Product separation was done with gas chromatography (GC) equipped with a mass spectrometer and a special detector for radical scavenging products. More than 310 different nonradicalic compounds and 23 mainly aromatic radicals could be detected. Among them were partially oxidized aromatics such as 1- and 2-naphthol phenol, benzofuran, inden-1-one, and benzaldehyde. Aromatic radicals with an even number of C and σ-radicals, for example, phenyl and two naphthyls: those with an odd number are π-radicals, such as indenyl and cycloheptatrienyl, the latter being detected for the first time in a flame. An important initiation of growth reactions toward larger PAHs is the formation of various biaryls by the reaction of aromatic radicals with the fuel molecule and aromatic degradation products from it. Biaryls are themselves highly reactive intermediates that first undergo intramolecular cyclodehydrogenations, if they posses four-or five-sided carbon bays, and then grow to highly condensed PAHs by addition of acetylene.

Acetylene is readily converted to perchlorinated gas-phase intermediates including hexachlorobenzene, hexachlorobutadiene, and tetrachloroethylene and heavier perchlorinated species via heterogeneous gas-solid reactions with HCl and cupric oxide on borosilicate under postcombustion conditions. Experiments were conducted using an integrated gas-solid flow-reactor and analytical system at temperatures ranging from 150 to 500°C for gas-phase residence times of 2.0 s and total reaction times of 60 min. Chlorine addition and chlorine net substitution mechanisms mediated by the conversion of Cu(II)Cl₂ to Cu(I)Cl are proposed to account for the observed or inferred C₂ reaction products including tetrachloroethylene, trichloroethylene, and dichloroacetylene. The formation of condensation products including tetrachlorovinylacetylene, hexachlorobutadiene, and hexachlorobenzene are proposed to be catalyzed by copper chloride species and involve the following steps: (1) chemisorption of a chlorinated ethylene or acetylene by HCl elimination or 1,2-Cu−Cl addition, respectively: (2) physisorption of additional chlorimated ethylenes or acetylenes followed by cis-insertions: and (3) carbon-to-copper chlorine transfer followed by desorption of the molecular growth product. The mechanism accounts for product isomer distributions and branching desorption of the higher molecular weight products, and regeneration of the copper chloride catalyst.

Self-propagating reactions in multilayer foils are analyzed using an unsteady computational model. The reactions are described in terms of the energy conservation equation and the evolution equation for a conserved scalar. The model is applied to analyze combustion waves in reacting foils that consist of alternating layers of Ni and Al. The individual layers have thicknesses, 2δ, in the range 20 to 200 nm, and the foils are 1 to 100 μm thick. The interfaces between the layers are assumed to be diffuse, with a characteristic mixed-zone thickness of 4Ω. The propagation of the flame is analyzed in terms of δ and Ω. Consistent with experimental observations and steady-state calculations, computed results show that the flame speed increases with decreasing δ, until a critical value, δ_c, is reached. Below δ_c, the trend is reversed—that is, the flame speed decreases with δ. Meanwhile, the flame speed increases monotonically with decreasing Ω. However, the calculations show that propagation of the reaction occurs in an unsteady fashion. Periodic and quasi-periodic, large-amplitude oscillations in the burning rate and the flame width are observed. As the flame speed increases, the amplitude of the oscillations increases and their characteristic period decreases. The occurrence of superadiabatic temperatures within the flame suggests that the oscillations result in an average propagation speed that is larger than the steady-state prediction.

An extensive investigation has been carried out on the thermal behavior of biomass pyrolysis oil derived from different feedstocks (mixed hardwood, pine, and poplar). Experiments were performed on streams of monodispersed droplets (50–100 μm) and on large suspended droplets (300–600 μm).

The small droplets were generated by a single-droplet generator and injected in a drop-tube furnace, the temperature of which ranged from 300 to 850°C. Light-scattering methods have been used to follow the process undergone by the pyrolysis oil droplets. The solid material, residual of the droplet heating and vaporization processes, has been examined by means of optical and scanning electron microscopy. Two different morphologies of residual particles have been observed at the exit of the furnace: (a) more compact, mechanically resistant spheres, with typical diameter of 10–40 μm, and (b) fragile, glasslike cenospheres with thin walls and menisci, with diameter of 100–200 μm.

The large droplets were suspended at the exit of the furnace, using both a quartz fiber and an exposed junction thermocouple. The furnace temperature was varied between 400 and 1200°C. The phenomena, occurring during the vaporization and burning of each droplet, were followed by coupling the thermocouple measurement to high-speed visualization. The imaging was performed by means of fast digital video recording (50–250 frames/s) and high-speed cinematography (400–1000 frames/s). Temperature-time curves of the burning droplets show a stepped behavior, with two zones at constant temperature, at ∼100 and ∼450°C. The first step is due to the vaporization of water. The second plateau corresponds to the heating of heavy compounds due to selective vaporization and liquid-phase pyrolysis. Swelling, shrinking, and microexplosions with ejection of matter characterize this phase. The combustion of the droplets starts with an enveloping blue flame. The flame then develops a yellow tail of increasing size. After the flame extinction, the remaining char particle burns without residual.

In this study, we extend the results of previous combined numerical and experimental investigations of an axisymmetric laminar diffusion flame in which difference Raman spectroscopy, laser-induced fluorescence (LIF), and a multidimensional flame model were used to generate profiles of the temperature and major and minor species. A procedure is outlined by which the number densities of ground-state CH (X²II), excited-state CH (A²δ, denoted CH^*), and excited-state OH (A²Σ, denoted OH^*) are measured and modeled. CH^* and OH^* number densities are deconvoluted from line-of-sight flame-emission measurements. Ground-state CH is measured using linear LIF. The computations are done with GRI Mech 2.11 as wel as an alternate hydrocarbon mechanism. In both cases, additional reactions for the production and consumption of CH^* and OH^* are added from recent kinetic studies. Collisional quenching and spontaneous emission are responsible for the de-excitation of the excited-state radicals.

As with our previous investigations, GRI Mech 2.11 continues to produce very good agreement with the overall flame length observed in the experiments, while significantly under predicting the flame liftoff height. The alternate kinetic scheme is much more accurate in predicting lift-off height but overpredicts the overall flame length. Ground-state CH profiles predicted with GRI Mech 2.11 are in excellent agreement with the corresponding measurements, regarding both spatial distribution and absolute concentration (measured at 4 ppm) of the CH radical. Calculations of the excited-state species show reasonable agreement with the measurements as far as spatial distribution and overall characteristics are concerned. For OH^*, the measured peak mole fraction, 1.3×10⁻⁸, compared well with computed peaks, while the measured peak level for CH^*, 2×10⁻⁹, was severely underpredicted by both kinetic schemes, indicating that the formation and destruction kinetics associated with excited-state species in flames require further research.

Quantitative laser-induced fluorescence (LIF) measurements of the concentration of HCO are made in three 25-torr methane-oxygen-nitrogen flames: fuel lean (φ=0.81), near stoichiometric (φ=1.07), and fuel rich (φ=1.28). LIF is excited in the (000)-(000) band of the B-X system near 258 nm. The LIF signal from the flame is calibrated against nonflame measurements of a known HCO concentration produced by laser photolysis of acetaldehyde. The LIF signal is adjusted for the variation in the fraction of the population of the laser-excited level as the measured temperature changew with position in the flame and for the measured variation in quenching. The resulting concentration measurements agree well with model predictions for the fuel-lean and near-stoichiometric flame. The measurements in the fuel-rich flame are significantly larger than the model predictions: however, these measurements are subject to increased uncertainty due to the large, broadband background in the rich flame.