James C. Keck

Burning velocities of mixtures of air with methanol, isooctane, and indolene at high pressure and temperature

Burning velocities of mixtures of air with methonol, isooctane, and indolene (RMFD303) have been measured using the constant volume bomb method for fuel-air equivalence ratios φ = 0.8-1.5 over the pressure and temperature ranges p = 0.4–50 atm and T = 298–700K. The effect of adding simulated combustion products to stoichiometric isooctane-air mixtures was also studied for diluent mass fractions f = 0−0.2. Over the range studied, the results can be fit within ± 10% by the functional form Su = Su0 (TuT0) α(pp0)β (1–2.1f), where Su0 depends on fuel type and equivalence ratio and α and β depend only on equivalence ratio. In overlapping ranges, the results agree well with those previously reported.

Experimental and Theoretical Study of Nitric Oxide Formation in Internal Combustion Engines

Abstract The nonequilibrium formation of nitric oxide within the internal combustion engine cylinder is examined. A thermodynamic model which predicts the properties of the burnt and unburnt gases during the combustion process is developed. A set of reactions which govern the formation of nitric oxide is proposed, and rate equations for nitric oxide concentrations as a function of time in the post-flame gases are derived. The results of time-resolved measurements carried out on a CFR engine are described, where emitted light intensities at wavelengths selected to record radiation from the CO + O and NO + O continua were used to determine the nitric oxide concentration. The comparison between theoretical and experimental results for fuel-lean mixtures confirms that the important features of the model presented are correct.

Two-stage ignition in HCCI combustion and HCCI control by fuels and additives

A Rapid Compression Machine (RCM) has been used to study the effects of fuel structure and additives on the Homogeneous Charge Compression Ignition (HCCI) of pure hydrocarbon fuels and mixtures under well-determined conditions. Such information is needed for understanding ignition delays and burning rates in HCCI engines, and “knock” in spark-ignition engines. It is also valuable for validating basic chemical kinetic models of hydrocarbon oxidation. The pure fuels used in the study include: paraffins (n-heptane, iso-octane), cyclic paraffins (cyclohexane, methylcyclohexane), olefins (1-heptene, 2-heptene, 3-heptene), cyclic olefins (cyclohexene, 1,3-cyclohexadiene), and an aromatic hydrocarbon (toluene). The additives were 2-ethyl-hexyl-nitrate and di-tertiary-butyl-peroxide. It was found that fuels which contained the structure -CH2-CH2-CH2- showed two-stage ignition with relatively short ignition delays and that the ignition delay depended strongly on the energy released during the first-stage. For primary reference fuel mixtures (n-heptane + iso-octane), the ignition delay depended only on the molar ratio of n-heptane to oxygen and was independent of the octane number (percent iso-octane). On the other hand, the burn rate depended on both these parameters, which uniquely determine the equivalence ratio. When additives were included in the air/fuel mixtures, the ignition delay was reduced but the burn rate was not affected. These results indicate that for HCCI combustion, the ignition delay and the burn rate can be independently controlled using various fuel mixtures and additives.

Laminar Burning Velocity of Propane-Air Mixtures at High Temperature and Pressure

The laminar burning velocity of propane-air mixtures has been measured in the pressure range 0.4 to 40 atm and temperature range 298 to 750 K for equivalence ratios from 0.8 to 1.5. The measurements were made in a constant-: volume spherical combustion bomb which could be heated to 500 K. A thermodynamic analysis was used to calculate the laminar burning velocity from a pressure time history of the combustion process. The measured values were correlated using both power law and exponential expressions.

A reduced chemical kinetic model for HCCI combustion of primary reference fuels in a rapid compression machine

A model for the Homogeneous Charge Compression Ignition (HCCI) of Primary Reference Fuels (PRFs) in a Rapid Compression Machine (RCM) has been developed. A reduced chemical kinetic model that included 32 species and 55 reactions was used and the affect of wall heat transfer on the temperature of the adiabatic core gas was taken into account by adding the displacement volume of the laminar boundary layer to the cylinder volume. A simple interaction between n-heptane and iso-octane was also included. The results showed the well-known two-stage ignition characteristics of heavy hydrocarbons, which involve low and high temperature cycles followed by a branched chain explosion. The first stage energy release decreases and the ignition delay increases nonlinearly with increasing octane number and decreasing the initial pressure. The energy release rate and total energy released were determined primarily by the rate of CO oxidation during the explosive phase following the ignition delay. The model reproduced the pressure curves obtained in the RCM experiments over a wide range of conditions remarkably well and was very sensitive to the fuel structure, the mixture composition and the initial temperature and pressure. Thus, the model can be easily adapted for predicting “knock” in spark-ignition engines and ignition-delays and burning rates in HCCI engines.

Diffusion Theory of Nonequilibrium Dissociation and Recombination

The coupled vibration—dissociation—recombination process for molecules and atoms has been examined. Techniques for solving the appropriate master equations for both quantum (discrete) and classical (continuous) models are given. It is shown that the process is most easily treated classically and that in this case the master equation can be reduced to an equivalent diffusion equation. It is assumed that, after an initial vibration transient, during which reactions are negligible, the process may be treated using the steady‐state approximation. During the steady‐state phase, the usual phenomenological rate equations are valid and the ratio of the forward and reverse rate constants is the equilibrium constant, though the individual rate constants are depressed below their equilibrium values.Comparison of the results with other theoretical work shows general agreement for similar models; comparison with shock‐tube experiments on molecular dissociation and stellarator experiments on ionic recombination is enco...

Rate-controlled partial-equilibrium method for treating reacting gas mixtures

A rate-controlled partial-equilibrium method for treating reacting gas mixtures has been developed that is very much simpler than the usual technique of integrating the full set of rate equations. The method is based on the observation that a rate-controlling reaction is equivalent to a “passive constraint” on the gas, so that the corresponding partial-equilibrium state can be calculated by minimizing the appropriate free energy function. An example illustrating the application of the method to the “freezing” of threebody reactions in an internal combustion engine is given.

Turbulent flame propagation and combustion in spark ignition engines

Abstract Pressure measurements synchronized with high-speed motion picture records of flame propagation have been made in a transparent piston engine. The data show that the initial expansion speed of the flame front is close to that of a laminar flame. As the flame expands, its speed rapidly accelerates to a quasi-steady value comparable with that of the turbulent velocity fluctuations in the unburned gas. During the quasi-steady propagation phase, a significant fraction of the gas behind the visible front is unburned. Final burnout of the charge may be approximated by an exponential decay in time. The data have been analyzed in a model independent way to obtain a set of empirical equations for calculating mass burning rates in spark ignition engines. The burning equations contain three parameters: the laminar burning speed sl, a characteristic speed uT, and a characteristic length lT. The laminar burning speed is known from laboratory measurements. Tentative correlations relating uT and lT to engine geometry and operating variables have been derived from the engine data.

Stand-off distances on a flat flame burner

Abstract For a given stand-off distance of a laminar flame on a porous metal burner, it has been shown both experimentally and theoretically that there exists two solutions, a low-speed flame and a high-speed flame. For small enough stand-off distance there is no solution. The minimum stand-off distance is identified as the quasi-steady approximation to the extinction length for flames quenching in flows perpendicular to a heat sink. Measurements of flame speed, maximum flame temperature, and stand-off distance have been correlated for hydrogen, ethylene, and methane flames by a Peclet number dependent only on the ratio of the heat of combustion to the heat loss. The correlation agrees quantitatively with solution of one-dimensional flame equations where a Dirac-delta function models the reaction rate.

High Reynolds number flow in a moving corner

The problem of a piston moving in a cylinder is studied experimentally using flow visualization techniques. A vortex motion is observed at the piston face and cylinder wall interface as the cylinder wall moves toward the piston. Non-dimensional scaling parameters for the vortex size and stability are determined and semi-empirical theories for the size of the vortex are presented.