Inchul Choi

Plasma assisted ignition and high-speed flow control: non-thermal and thermal effects

The paper reviews recent progress in two rapidly developing engineering applications of plasmas, plasma assisted combustion and plasma assisted high-speed flow control. Experimental and kinetic modeling results demonstrate the key role of non-thermal plasma chemistry in hydrocarbon ignition by uniform, repetitively pulsed, nanosecond pulse duration, low-temperature plasmas. Ignition delay time in premixed ethylene‐air flows excited by the plasma has been measured in a wide range of pulse repetition rates and equivalence ratios and compared with kinetic modeling calculations, showing good agreement. Comparing ignition delay time predicted by the model for plasma assisted ignition and for ignition by equilibrium heating demonstrated that chain reactions of radicals generated by the plasma reduce ignition time by up to two orders of magnitude and ignition temperature by up to 300K. These results provide additional evidence of the non-thermal nature of low-temperature plasma assisted ignition. Experiments and flow modeling show that the dominant mechanism of high-speed plasma flow control is thermal, due to heating of the flow by the plasma. Development and characterization of pulsed dc and pulsed RF localized arc filament plasma actuator arrays for control of high-speed atmospheric pressure jet flows are discussed. Actuator power is quite low, ∼10W at 10% duty cycle. Plasma emission spectra show that a greater fraction of the pulsed RF discharge power goes to heat the flow (up to 2500 ◦ C), while a significant fraction of the pulsed dc discharge power is spent on electrode and wall heating, resulting in their erosion. Rapid localized heating of the flow by the pulsed arc filaments, at a rate of ∼1000K/10 µs, results in the formation of strong compression/shock waves, detected by schlieren imaging. Effect of flow forcing by repetitively pulsed RF actuators is demonstrated in a M = 1.3 axisymmetric jet. These two case studies provide illustrative examples of isolating non-thermal (non-equilibrium plasma chemistry) and thermal (Joule heating) effects in plasmas and adapting them to develop efficient large-volume plasma igniters and high-speed flow actuators. (Some figures in this article are in colour only in the electronic version)

Energy coupling to the plasma in repetitive nanosecond pulse discharges

A new analytic quasi-one-dimensional model of energy coupling to nanosecond pulse discharge plasmas in plane-to-plane geometry has been developed. The use of a one-dimensional approach is based on images of repetitively pulsed nanosecond discharge plasmas in dry air demonstrating that the plasma remains diffuse and uniform on a nanosecond time scale over a wide range of pressures. The model provides analytic expressions for the time-dependent electric field and electron density in the plasma, electric field in the sheath, sheath boundary location, and coupled pulse energy. The analytic model predictions are in very good agreement with numerical calculations. The model demonstrates that (i) the energy coupled to the plasma during an individual nanosecond discharge pulse is controlled primarily by the capacitance of the dielectric layers and by the breakdown voltage and (ii) the pulse energy coupled to the plasma during a burst of nanosecond pulses decreases as a function of the pulse number in the burst. T...

Pure rotational CARS thermometry studies of low-temperature oxidation kinetics in air and ethene–air nanosecond pulse discharge plasmas

Pure rotational CARS thermometry is used to study low-temperature plasma assisted fuel oxidation kinetics in a repetitive nanosecond pulse discharge in ethene–air at stoichiometric and fuel lean conditions at 40 Torr pressure. Air and fuel–air mixtures are excited by a burst of high-voltage nanosecond pulses (peak voltage, 20 kV; pulse duration, ~ 25 ns) at a 40 kHz pulse repetition rate and a burst repetition rate of 10 Hz. The number of pulses in the burst is varied from a few pulses to a few hundred pulses. The results are compared with the previously developed hydrocarbon–air plasma chemistry model, modified to incorporate non-empirical scaling of the nanosecond discharge pulse energy coupled to the plasma with number density, as well as one-dimensional conduction heat transfer. Experimental time-resolved temperature, determined as a function of the number of pulses in the burst, is found to agree well with the model predictions. The results demonstrate that the heating rate in fuel–air plasmas is much faster compared with air plasmas, primarily due to energy release in exothermic reactions of fuel with O atoms generated by the plasma. It is found that the initial heating rate in fuel–air plasmas is controlled by the rate of radical (primarily O atoms) generation and is nearly independent of the equivalence ratio. At long burst durations, the heating rate in lean fuel air–mixtures is significantly reduced when all fuel is oxidized.

Hydroxyl Radical Kinetics in Repetitively Pulsed Hydrogen–Air Nanosecond Plasmas

Absolute hydroxyl radical (OH) concentration is determined in stoichiometric hydrogen-air mixtures at P = 54-94 torr and initial temperature of T = 100°C-200°C, which are both functions of time, after the application of a single approximately 25-ns-duration approximately 20-kV discharge pulse and 60 μs after the final pulse of a variable-length burst of pulses, using single-photon laser-induced fluorescence (LIF). Relative LIF signal levels are put on an absolute number density scale by means of calibration with a standard atmospheric-pressure near-adiabatic Hencken flat-flame burner. By obtaining OH LIF data in both the plasma and the flame and correcting for differences in the collisional quenching and vibrational energy transfer rates, absolute OH number density has been determined. For a single discharge pulse, the absolute OH temporal profile is found to rise rapidly during the initial ~0.1 ms after discharge initiation and decay relatively slowly, with a characteristic time scale of ~1 ms. In repetitive burst mode, the absolute OH number density is observed to rise rapidly during the first approximately ten pulses (0.25 ms) and then level off to a near steady-state plateau. In all cases, a large secondary rise in OH number density is also observed, which is clearly indicative of ignition, with ignition time ranging from 5 to 10 ms, for initial temperatures of 100 °C and 200°C and pressures in the range of 54-94 torr. Plasma kinetic modeling predictions capture this trend quantitatively, using both a full 22-hydrogen-air-chemical-reaction set and a reduced 9-reaction set.

Spatially and Temporally Resolved Atomic Oxygen Measurements in Short Pulse Discharges by Two Photon Laser Induced Fluorescence

Two Photon Laser Induced Fluorescence (TALIF) is used to measure absolute oxygen atom concentrations as a function of time in O2/He, O2/N2, and methane/air plasmas produced with a 20 nanosecond duration, 20 kV pulsed discharge. While the pulser is capable of repetition rates as high as 50 kHz, the data reported here was purposefully obtained at much lower repetition rate, 10 Hz, in order to limit the number of pulses experienced by the flowing, room temperature gas sample during its resonance time within the plasma to one, or at most two. While not directly measured, the low repetition rate insures negligible temperature rise due to direct plasma heating. Relative atomic oxygen concentration data are put on an absolute scale by means of a calibration procedure in which the observed signal level is compared to that of reference TALIF spectra of atomic xenon, obtained under known conditions of pressure and temperature, and identical optical and spectroscopic conditions. Calibrated TALIF spectra show that a single discharge pulse creates initial atomic oxygen concentrations in the range (2.0 – 3.0) x 10 14 cm -3 for air, 10% O2 in helium and 10% O2 in nitrogen mixtures. Peak atomic oxygen concentration is a factor of approximately two lower in fuel lean (φ=0.5) methane/air mixtures. In pure helium buffer, the initially formed atomic oxygen decays monotonically, with decay time consistent with formation of ozone and oxygen. In all nitrogen containing mixtures, atomic oxygen concentrations are found to initially increase, for time scales on the order of 10-100 microseconds, due presumably to additional O2 dissociation caused by collisions with electronically excited nitrogen. The long time scale decay in O2/N2 mixtures occurs on a time scale, which is similar to that of O2/He, whereas in the methane/air mixture, the decay rate is greater by a factor of approximately five.

Kinetics of Low-Temperature Hydrogen Oxidation and Ignition by Repetitively Pulsed Nonequilibrium Plasmas

Two photon and single photon laser induced fluorescence of atomic oxygen and the OH radical, respectively, complemented with UV ICCD imaging, is used to study low-temperature plasma assisted fuel oxidation kinetics in repetitive nanosecond pulse discharges in hydrogen-air mixtures at 40 Torr pressure. Air and premixed fuel-air mixtures are excited by a burst of high-voltage nanosecond pulses at a 10 kHz or 40 kHz pulse repetition rate and burst repetition rate of 10 Hz. The number of pulses in the burst is varied from one pulse to a few hundred pulses. Time-resolved relative OH concentration measurements are in good agreement with predictions of a new hydrogen-air plasma chemistry model which incorporates non-equilibrium plasma discharge processes, low temperature H2 – air chemistry, nonempirical scaling of nanosecond discharge pulse energy coupled to the plasma, and quasi-onedimensional conduction heat transfer. Kinetic model prediction of a significant reduction in O atom concentration in the presence of hydrogen has been confirmed qualitatively by the experimental data. Kinetic sensitivity analysis shows that in room temperature hydrogen-air discharges, OH formation and decay, as well as the initial heating rate are controlled by the three process sequence: O + HO2 + M → HO2 + M, O + HO2 → OH + O2 and OH + H2 → H2O + H, essentially without radical chain branching. At intermediate temperatures, 500 – 600 K, significant chain branching, with associated additional energy release, occurs in reactions of O with HO2, as well as in O + H2 → OH + H reaction. Both chain branching and net exothermic heat release in plasma chemical reactions becomes more pronounced at higher temperatures, eventually resulting in ignition. Sensitivity analysis also shows that generation of radicals in the plasma is key to low-temperature plasma chemical fuel oxidation and associate heat release, while ignition is primarily controlled by the well known chain branching sequence O + H2 → OH + H and H + O2 → OH + O.

Pure Rotational CARS Measurements of Thermal Energy Release and Ignition in Nanosecond Pulse Burst Air and Hydrogen-Air Plasmas

Pure rotational Coherent Anti-Stokes Raman Spectroscopy (CARS), complemented by UV emission and ICCD imaging, is used to study low-temperature plasma assisted fuel oxidation kinetics and ignition in a repetitive nanosecond pulse discharge in hydrogen-air at stoichiometric and fuel lean conditions at 40 Torr pressure. Air and premixed fuel-air mixtures are excited by a burst of high-voltage nanosecond pulses (peak voltage 20 kV, pulse duration ~25 nanoseconds) at a 40 kHz pulse repetition rate and burst repetition rate of 10 Hz. The number of pulses in the burst is varied from a few pulses to a few hundred pulses. The results are compared to a new hydrogen-air plasma chemistry model which incorporates nonequilibrium plasma processes, low temperature H2 – air chemistry, non-empirical scaling of nanosecond discharge pulse energy coupled to the plasma with the pulse waveform and the number density, and quasi-one-dimensional conduction heat transfer. Experimental centerline time-resolved temperature and O2 mole fraction, determined as a function of number of pulses in a burst, are found to agree well with model predictions. The results demonstrate that the heating rate in low temperature hydrogen-air plasmas is much faster than in pure air plasmas, primarily due to energy release from the exothermic reactions of fuel with O and H atoms generated in non-equilibrium quantities in the plasma. Specifically, it is found that the initial heating rate at room temperature is controlled by the low temperature processes, O + HO2 → OH + O2 and OH + H2 → H2O + H, where HO2 is formed by three-body recombination of O and H2. At intermediate temperatures, 500 – 600 K, significant chain branching, with associated additional energy release, occurs in reactions of O with HO2, as well as in O + H2 → OH + H reaction. Both chain branching and net exothermic heat release in plasma chemical reactions becomes more pronounced at higher temperatures, eventually resulting in ignition. Sensitivity analysis also shows that generation of radicals in the plasma is key to low-temperature plasma chemical fuel oxidation and associated heat release, while ignition is primarily controlled by the well known chain branching sequence O + H2 → OH + H and H + O2 → OH + O. Rapid plasma chemical hydrogen oxidation, in ϕ = 0.5 and ϕ = 1.0 mixtures, leads to a distinct maximum in temperature which is both predicted, and observed, after approximately 700 discharge pulses (17.5 msec), concurrent with a predicted and observed rapid loss in O2. UV emission and ICCD imaging provides further evidence that plasma chemical reactions lead to volumetric ignition at pressures in the approximate range 40 - 100 Torr, and equivalence ratios in the approximate range φ = 0.3 to φ = 1. Experimental ignition delay times are found to be a strong function of pressure, but a weak function of equivalence ration, general trends which are consistent with modeling predictions.

Hydroxyl Radical Kinetics In Repetitively Pulsed Hydrogen-Air Nanosecond Plasmas

Absolute hydroxyl radical (OH) concentration is determined in stoichiometric hydrogen-air mixtures at P = 54-94 torr and initial temperature of T = 100°C-200°C, which are both functions of time, after the application of a single approximately 25-ns-duration approximately 20-kV discharge pulse and 60 μs after the final pulse of a variable-length burst of pulses, using single-photon laser-induced fluorescence (LIF). Relative LIF signal levels are put on an absolute number density scale by means of calibration with a standard atmospheric-pressure near-adiabatic Hencken flat-flame burner. By obtaining OH LIF data in both the plasma and the flame and correcting for differences in the collisional quenching and vibrational energy transfer rates, absolute OH number density has been determined. For a single discharge pulse, the absolute OH temporal profile is found to rise rapidly during the initial ~0.1 ms after discharge initiation and decay relatively slowly, with a characteristic time scale of ~1 ms. In repetitive burst mode, the absolute OH number density is observed to rise rapidly during the first approximately ten pulses (0.25 ms) and then level off to a near steady-state plateau. In all cases, a large secondary rise in OH number density is also observed, which is clearly indicative of ignition, with ignition time ranging from 5 to 10 ms, for initial temperatures of 100 °C and 200°C and pressures in the range of 54-94 torr. Plasma kinetic modeling predictions capture this trend quantitatively, using both a full 22-hydrogen-air-chemical-reaction set and a reduced 9-reaction set.

Nanosecond Pulse Burst Ignition of Ethylene and Acetylene by Uniform Low-Temperature Plasmas 1

Nanosecond pulse burst plasma ignition measurements and kinetic modeling calculations have been used to analyze kinetics of low-temperature plasma assisted ignition of hydrocarbon fuels. Uniform low-temperature plasmas have been generated by high voltage, nanosecond duration pulses at high pulse repetition rates. Pulse bursts of up to 1000 pulses have been used to ignite premixed ethylene-air and acetylene-air flows. Ignition delay time has been determined by measuring time-resolved OH, CH, and C2 Swan band emission from the flow, which produces a well pronounced overshoot during ignition. Ignition delay time has been measured in a wide range of pulse repetition rates and equivalence ratios. Kinetic modeling of a high-voltage, nanosecond duration pulse discharge demonstrate that charge accumulation on dielectric plates covering the electrodes results in strong shielding of the plasma, which significantly lowers gap voltage and limits the pulse energy coupled to the plasma. The effective reduced electric field value inferred from plasma emission spectra, E/N=330 ± 30 Td, is approximately a factor of 3 lower than the value estimated from the pulse peak voltage. Discharge pulse energy predicted by the model is in good agreement with our previous work, where the pulse energy was inferred from the O atom density measurements. Experimental ignition delay times have been compared with results of kinetic modeling of repetitively pulsed hydrocarbon-air plasma. Modeling calculations predict considerable chemical energy release from the fuel species due to exothermic fuel oxidation in reactions with radicals generated by electron impact. This effect results in significant additional heating of fuel-air mixtures in the lowtemperature plasma. Comparing calculated plasma assisted ignition delay time with ignition by equilibrium heating demonstrated that radicals generated by the plasma reduce ignition temperature by up to 300 0 C and reduce ignition delay time by up to 2 orders of magnitude. This demonstrates conclusively the non-thermal nature of low-temperature plasma assisted ignition. Comparison of experimental ignition delay time in ethylene-air with kinetic modeling calculations shows very good agreement. In methane-air, the model predicts no ignition at the present conditions, consistent with the experiments, in which no methane-air ignition has been detected. In acetylene-air, the model overpredicts ignition delay time by 50100% compared with the experimental data.

Hydroxyl Radical Kinetic Measurements in Low Temperature Nanosecond Pulsed Non-equilibrium Plasmas

Laser Induced Fluorescence (LIF) spectroscopy is used to study the temporal evolution of hydroxyl radical (OH) concentration in hydrogen-air repetitively pulsed nanosecond discharges at equivalence ratios of 0.5 and 1.0, and pressure and initial temperature of 40 Torr and 300 K, respectively, with the number of pulses in a 40 kHz burst varied between one and one thousand. Relative OH concentration data is put on an absolute scale by calibration with an atmospheric pressure near-adiabatic flat flame Hencken burner. Experimental time-resolved absolute OH radical concentrations as a function of time after initiation of a single discharge pulse, and as a function of number of pulses in the 40 kHz burst are found to agree well with predictions from a recently developed hydrogen-air plasma chemistry model which incorporates non-equilibrium plasma processes, low temperature H2 – air chemistry, non-empirical scaling of nanosecond discharge pulse energy coupled to the plasma, and quasi-one-dimensional conduction heat transfer.