[en] Ignition of hydrocarbon–oxygen mixtures by means of a nanosecond surface dielectric barrier discharge (NSDBD) was studied experimentally. The propagation velocity of the flame wave and the ignition delay time in mixtures of oxygen with methane, ethane, ethylene, and dimethyl ether were measured using a high-speed camera. The experiments were carried out at room temperature and gas mixture pressures in the range of 0.75–1.25 atm. It is shown that, for all hydrocarbons under study, the flame velocity decreases with reducing pressure and stoichiometric ratio, as well as when the mixture is diluted with molecular nitrogen. Theoretical analysis of the processes in the NSDBD plasma and measurements of the flame velocity in hydrocarbon-containing mixtures without plasma agree qualitatively with the measurement results, except for the increasing dependence of the flame velocity on the pressure, which is decreasing in experiments without a discharge plasma.

[en] The dynamics of radiation intensity at 337 nm emitted by the surface dielectric barrier discharge driven by 50 ns duration and 8 kV amplitude voltage pulse of both polarities has been analyzed experimentally and numerically. The calculations were performed in a 2D approach for experimental conditions to check the existing numerical models and understand what processes manage the discharge length and the spatial distribution of the discharge radiation intensity. Experimentally measured and numerically simulated profiles of radiation intensity along the discharge length notably differ from each other and the performed modification of the discharge physical model does not essentially change this difference. Radiation intensity from the discharge channel has been shown to be extremely sensitive primarily to the value of reduced electric field. We suppose that the discrepancy between experiment and numerical simulation could be due to 2D instead of 3D simulation and corresponding incorrect estimation of electric field distribution along the discharge channel. The numerical simulation of discharge length is in good agreement with experiment for the discharge stroke at the leading front of the voltage pulse and gives few times less values for the backward discharge at the trailing edge of the voltage pulse. (paper)

[en] Electron detachment from O^-₂ ions has been theoretically studied in oxygen (i) when the gas is vibrationally excited and (ii) when the ions are heated in a strong external electric field. Assuming that electron detachment proceeds via the formation of vibrationally excited temporary O^-₂ ions, the detachment rate was determined on the basis of the statistical approach for the vibrational transfer and relaxation in collisions between O^-₂ ions and O₂ molecules. The approach used in our previous work to study electron detachment in vibrationally excited oxygen was amended to take into account the conservation of total angular momentum in O^-₂-O₂ collisions. The calculated detachment rates agree well with available measurements in oxygen under equilibrium conditions. The method was used to calculate detachment rates under strongly non-equilibrium conditions in which the vibrational temperature of molecules or ion translational temperature is elevated. The obtained results were compared with available measurements in drift tubes.

[en] Three-body electron attachment to O₂ molecules and electron detachment from O₂^- ions have been theoretically studied in vibrationally excited oxygen and O₂-containing mixtures. Assuming that electron attachment and detachment proceed via the formation of vibrationally excited temporary O₂^- ions, the rates of these processes were determined on the basis of the statistical approach for the vibrational transfer and relaxation in collisions between O₂^- ions and O₂ molecules. The calculated attachment and detachment rate constants turned out to agree well with available measurements in unexcited oxygen. This method was extended to calculate attachment and detachment rates in vibrationally excited oxygen. It was shown that the effect of vibrational excitation on electron detachment is profound, whereas attachment of low-energy electrons to vibrationally excited O₂ is inefficient. The calculated vibrational distribution of stable O₂^- ions turned out to be non-equilibrium in an excited gas and the effective vibrational temperature of the ions was much lower than the vibrational temperature of molecules. An analytical method was suggested to determine this distribution and the effective vibrational temperature. The calculated rate constants were used to simulate the formation and decay of an electron-beam-generated plasma in N₂ : O₂ mixtures at elevated vibrational temperatures. The calculations showed that vibrational excitation of molecules leads to orders of magnitude increase in the plasma density and in the plasma lifetime, in agreement with available observations.

[en] We present the results of the experimental study of high-voltage nanosecond repetitively pulsed discharge and of its afterglow in ethane, dimethyl ether and their mixtures with oxygen. The measurements were made for room temperature and pressures from 2 to 6 Torr. The measured specific deposited energy and mean discharge current varied with the number of voltage pulses monotonously in ethane and dimethyl ether and nonmonotonously in the hydrocarbon:oxygen mixtures. A microwave interferometer was used for time-resolved electron density measurements in the discharge afterglow. The effective recombination coefficients were obtained from the analysis of the measured data. These coefficients varied with the number of voltage pulses monotonously in ethane and peaked in pure dimethyl ether and its mixture with oxygen. Possible mechanisms of the nonmonotonous behavior of the discharge characteristics were discussed. (paper)

[en] We study experimentally and numerically the kinetics of ignition in lean and stoichiometric C₂H₂ : O₂ : Ar mixtures after a high-voltage nanosecond discharge. The ignition delay time is measured behind a reflected shock wave with and without the discharge using detection of CH radiation. Generation of the discharge plasma is shown to lead to a decrease in ignition delay time. Discharge processes followed by chain chemical reactions with energy release are simulated during ignition in the C₂H₂ : O₂ : Ar mixtures. The generation of atoms, radicals and excited and charged particles in the discharge phase is numerically simulated. The calculations are based on the measured time-resolved discharge current and electric field. The calculated densities of the active particles produced in the discharge on a nanosecond time scale are employed as input data to simulate plasma-assisted ignition on a microsecond scale. The calculated ignition delay times are compared with the experimental data. It is shown that the effect of the discharge plasma on ignition of the acetylene-containing mixtures is associated with active species production in the discharge phase rather than with gas heating during the discharge and in its afterglow. A sensitivity analysis is made to determine limiting reactions in acetylene autoignition and ignition after the discharge under the conditions studied. (paper)

[en] Ignition of stoichiometric hydrocarbon : air mixtures by a nanosecond surface dielectric barrier discharge has been experimentally studied at room temperature and atmospheric and subatmospheric pressures. Observations were made for different voltage polarities and shapes of the high-voltage electrode. The ignition delay time and the velocity of the combustion wave were measured in a C_2H_2 : air mixture versus applied voltage by processing discharge gap images. It was concluded that the mixtures are ignited easier by the discharge for a negative voltage polarity and when it develops from a gear-like electrode. A 2D simulation of the discharge was performed to calculate the temporal and spatial distributions of generated active species and gas temperature during the discharge and in its afterglow for both electrode polarities. It was shown that the voltage threshold for ignition by a negative-polarity discharge is lower than that for a positive-polarity discharge, in qualitative agreement with observations. This is due to the formation of a region with efficient active species production and fast gas heating in the immediate vicinity of the high-voltage electrode when a voltage of negative polarity is applied to it. (paper)

[en] The results of the experimental and numerical study of high-voltage nanosecond discharge afterglow in pure methane, ethane and propane are presented for room temperature and pressures from 2 to 20 Torr. Time-resolved electron density during the plasma decay was measured with a microwave interferometer for initial electron densities in the range between 5  ×  10¹⁰ and 3  ×  10¹² cm⁻³ and the effective recombination coefficients were obtained. Measured effective recombination coefficients increased with gas pressure and were much higher than the recombination coefficients for simple molecular hydrocarbon ions. The properties of plasma in the discharge afterglow were numerically simulated by solving the balance equations for charged particles and electron temperature. Calculations showed that electrons had time to thermalize prior to the recombination. The measured data were interpreted under the assumption that cluster hydrocarbon ions are formed during the plasma decay that is controlled by the dissociative electron recombination with these ions at electron room temperature. Based on the analysis of the experimental data, the rates of three-body formation of cluster ions and recombination coefficients for these ions were estimated. (paper)

[en] The decay of air plasma produced by a high-voltage nanosecond discharge at room temperature and gas pressures in the range of 1–10 Torr was studied experimentally and theoretically. The time dependence of the electron density was measured with a microwave interferometer. The initial electron density was about 10¹² cm⁻³. The discharge homogeneity was monitored using optical methods. The dynamics of the charged particle densities in the discharge afterglow was simulated by numerically solving the balance equations for electron and ions and the equation for the electron temperature. It was shown that, under these experimental conditions, plasma electrons are mainly lost due to dissociative and three-body recombination with ions. Agreement between the measured and calculated electron densities was achieved only when the rate constant of the three-body electron-ion recombination was increased by one order of magnitude and the temperature dependence of this rate constant was modified. This indicates that the mechanism for three-body recombination of molecular ions differs from that of the well-studied mechanism of atomic ion recombination.

[en] This paper presents the measurements of discharge length for surface dielectric barrier discharge driven by single voltage pulses with different voltage slopes (0.05–3.4 kV ns⁻¹) and different amplitudes of the pulses (7–15 kV). The measurements were carried out for different barrier dielectric materials characterized by relative dielectric permittivity ε from 2 to 35 and different thickness (0.8–5 mm). For all the voltage pulses, the discharge has a channeled structure and the discharge length decreases with increased ε value. The discharge length dependence on dielectric thickness is not notable. The comparison of experimental results with predictions of the analytical model leads to the conclusion that for positive voltage pulses, the discharge development occurs as a set of separate 3D streamers rather than a 2D plasma sheet even in the case of maximal considered 3.4 kV ns⁻¹ voltage slope. The measured discharge lengths lay in between the predictions of the 2D and 3D models. (paper)