[American Institute of Aeronautics and Astronautics 47th AIAA Aerospace Sciences Meeting including The New Horizons Forum and Aerospace Exposition - Orlando, Florida ()] 47th AIAA

American Institute of Aeronautics and Astronautics

1

Unstable Pulse Discharge in Mixing Layer of Gaseous Reactants

Sergey B. Leonov*, Yury I. Isaenkov†, and Dmitry A. Yarantsev‡ Joint Institute for High Temperature RAS, Moscow, 125412 Russia,

Igor V. Kochetov§, Anatoly P. Napartovich** TRINITI, Troitsk, Moscow region, Russia

and

Michail N. Shneider†† Princeton University, NJ, USA

The paper describes results of experimental and computational efforts on the filamentary transversal pulse discharge dynamics in ambient conditions, different gases (mixtures), and high-speed airflow. The effect of fast turbulent expansion of the post-discharge channel is studied experimentally in order to enhance mixing processes of fuel and oxidizer and reduce the time of ignition. The idea of extra benefit possessed with nonequilibrium, unsteady, and non-homogeneous discharge is verified numerically.

I. Introduction fundamental problem faced by designers of a hypersonic scramjet is the acceleration of mixing and ignition of a fuel-air mixture at a supersonic velocity, which can provide mixture ignition within reasonable distances from

the flow inlet. In order to solve those problems the extra method is required to achieve high mixing efficiency with a minimum total pressure loss [1-2].

The idea of using plasma-assisted methods of fuel ignition (Plasma-Assisted Ignition and Combustion) is based on non-equilibrium generation of chemically active species that speed up the combustion process. It is believed that gain in energy consumed for combustion acceleration by plasmas is due to the non-thermal nature of discharge plasma, which allows radicals to be produced in an above-equilibrium amount [3-5]. At the same time such an approach ignored the fact of limited rate of mixing, which has to be mostly completed the first, before an ignition.

The problem of plasma-assisted mixing was discussed earlier, particularly in frames of MHD approach [6]. Recently the effect of fast development of the post-discharge channel instability at stagnant conditions and in high-speed flow was observed experimentally [7-9]. This instability possesses form of lateral jets escaping. The mechanism of acceleration of post-discharge channel turbulent expansion was described earlier [10] for ambient conditions. But the experiment has shown a wider zone of disturbance and a higher speed of spreading out.

The second discussed idea is the filamentary discharge movement in medium at gradient concentration of different components. The discharge position and dynamics of mixing layer depend on the discharge parameters and physical properties of gases involved. It was considered that the result of interaction can be controlled by means of small additives in the gas [11]. The effects found are supposed to be applied for high-speed combustion enhancement due to non-equilibrium excitation of air/fuel composition and mixing acceleration of non-premixed multi-components flow. * DSc, Head of Laboratory, AIAA Associate Fellow, [email protected] † PhD, Senior Scientist ‡ PhD Student, Research Staff § PhD, Senior Scientist ** DSc, Professor, Head of Division, [email protected] †† DSc, Research Scientist, [email protected]

A

47th AIAA Aerospace Sciences Meeting Including The New Horizons Forum and Aerospace Exposition5 - 8 January 2009, Orlando, Florida

AIAA 2009-820

Copyright © 2009 by Sergey Leonov. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.


2

II. Test Facility Description The experimental approach includes the following key points: high-speed duct-driven airflow; direct injection

of gaseous fuel (co-flow jet of fuel or model gas: Ar or CO2, preliminary); transversal short-pulse repetitive electrical discharge, crossing air-fuel-air zone or along air-fuel boundary; plasma-induced enhanced turbulence; zone of improved mixing. Two facilities were used in experiments: (1) the first is to study discharge properties in different gases at ambient conditions and in mixing layer of neighboring gases; and (2) to study the discharge dynamics in high-speed transversal flow.

The scheme of the first facility is shown in Fig.1. The facility consists of test section, system of the pulse-repetitive feeding (pulse-repetitive discharger PD-100/2.5), and diagnostic arrangement. It is equipped with high-quality optical windows. The diagnostics consists of the schlieren device with spatial resolution not worse than 0.2 mm and temporal resolution 0.1 μs; set of transducers and probes for pressure, voltage, current, radiation measurements; spectroscopic system, etc. The pulse transversal discharge is realized by means of the specially designed power supply made on the base of Tesla coil with impact excitation. The typical parameters of the tests are the following: gas pressure P=0.1-2 Bar, discharge’s main pulse duration τ=50-150 ns, maximal voltage Umax=120 kV, maximal current Imax=2.4 kA, maximal power Wmax=90 MW, typical energy deposition in short-pulse mode E=(1.2-2.5) J. Two configurations of electrodes were applied in tests described below: in the first scheme the top electrode looks like a single needle on axis of gas-feeding pipe; in the second scheme two anodes were installed near the wedge of the pipe. This was done to locate the top electrode close to a boundary of gas jet flown out of the top pipe to bottom one.

Fig. 1. Experimental scheme of test in quiescent air. Two configurations of electrodes arrangement.

The high-speed experimental facility consists of blow-down wind tunnel PWT-50 [12], system of the pulse-repetitive discharge feeding (pulse-repetitive discharger PD-100/2.5), and the same diagnostic equipment. The gas dynamic (GD) duct includes rectangular channel Y*Z=60*72mm made of ceramics with flush-mounted electrodes. Two electrodes were installed transversally on opposite horizontal walls. The pulse transversal discharge was excited by means of the pulse-repetitive power supply of the discharge made on the base of Tesla transformer. The typical parameters of the tests are the following: air pressure P≈(0.1-1) Bar, flow velocity V=(0-500) m/s, discharge pulse duration in a range t=40-100 ns, maximal voltage, current and power were Umax=120 kV, Imax=2.5 kA, Wmax=90 MW.

III. Experimental Results At high level of pressure and fixed 2-electrode configuration the discharge appears in form of filamentary long

spark. The filamentary discharge penetrates cross flow between “hot” electrodes with the speed about V≈107m/s=1cm/ns. The plasma temperature was estimated in Tpl=15-17 kK on the base of spectroscopic measurements of continuum luminescence. Estimation on the base of deposited energy gives the same result. The initial volume of the discharge channel is about Vol=10-8 m3. Specific energy release (specific enthalpy) is about H=108J/kg (about 25% of the energy is lost due to radiation). Such enthalpy corresponds to air with the following parameters: temperature T0=16 kK, ionization degree α≈0.9, mean conductivity is about σ≈102 (Ohm*cm)-1, channel resistance is about R=25 Ohm (it depends on discharge mode and sort of gas – see below). The maximal radius of the heated zone expansion can be estimated on the base of Sedov’s hypothesis, that the expansion is defined only by


3

integral energy release. By that way the maximal radius is equal: ;5

2

0max pl

Er××

=π

where l is the length of the

channel, p0 is the external pressure. The estimation gives rmax≈5 mm that is close to experimental data. The typical value of parameter E/P0 is shown in Fig. 2 for three different pulse durations.

Fig.2. Measured value of E/P0 (energy release reduced to pressure), which defines the size of post-discharge thermal

cavern.

The visualization of the plasma filament in ambient conditions and high-speed flow was done by means of fast camera, schlieren method and line-scan (streak) camera. The thermal distortion, which is generated by the spark, was measured by schlieren method. The discharge channel diameter and dynamics were recorded by streak technique. As the result, we are evaluating that the diameter of high-temperature zone at the maximum of discharge current is about d=0.5 mm. At the same time, a visible diameter of the plasma channel is much larger that corresponds with expanded stage of heated zone.

On the first stage of expansion (t<50 μs) the post-discharge channel looks quite classical: cylindrical thermal cavity. At t≈80-150 μs (depending on conditions) the shape of the after-spark channel occurs unstable as shown in Fig.3. The physical mechanism of this instability was determined in [8-9]: cooling of the axial zone leads to the pressure decrease with sequential gas reverse movement. Such a movement occurs unstable due to the Rayleigh-Taylor mechanism. The graph in Fig.3 shows comparison between the experimental and computational data that proves the fact of the reverse flow.

Fig.3. Dynamics of the after-spark channel expansion and schlieren photo of the disturbed zone in early stage of instability development.

The later stage of expansion can be characterized by generation of intensive lateral jets followed by fast turbulization of significant volume of the gas (the diameter is inter-gap distance, approximately). The sample of schlieren image of the jets generation is shown in Fig.4. Thus the analysis of experimental data gives some not obvious results concerning the size of disturbed zone: its value occurs several times bigger than it could be in accordance with laminar or turbulent diffusion mechanisms. The result of measurements and of respective simulation is presented in Fig.5 [9]. It is well seen that the second phase of expansion is not predicted by the model. An observation of the experimental graph allows us to conclude two important points: the diameter of the plasma zone after the fast expansion is about d=8 mm, and the mean “velocity” of the plasma disturbances expansion on the


4

phase of jets generation is more than V=100 m/s under the conditions of this experiment. The data on the base of fast-response pressure transducers have verified the data of schlieren observations as a whole.

Fig.4. Typical schlieren image of post-discharge thermal distortion, t=300μs.

Fig.5. Plasma disturbance expansion as a function of time. Experiment (left) and simulation (right). Air, P0=1 Bar; initial flow velocity V=0.

Instability vs pressure. The pressure decrease leads to reduction of the disturbance evolution intensity in post discharge channel. Two experimental series were performed: (1) measurements of the short-pulse discharge parameters in air at variable pressure; (2) and study of pulse discharge dynamics in high-speed flow (M=0.5 and M=2).

The first series of measurements was undertaken because the static pressure in high-speed flow is lower than atmospheric value: Pst=100-250 Torr at M=2. The tests were made in test-section of facility PWT-50. The electrical parameters were measured in parallel with schlieren images acquisition to detect the appearance of instability. Power supply has been operated in three modes listed in table below. In some range of parameters a shorter pulse can be more efficient in energy releasing to low-density gas.

Capacitance of

secondary circuit, pF

Maximal voltage of secondary circuit, kV

Maximal energy saved, J

Nominal energy saved, J

Half-cycle duration at R=0, ns

Mode 1 940 110 5.7 4.7 94 Mode 2 470 110 2.85 2.4 65 Mode 3 254 105 1.4 1.25 45

Fig.6 summarizes the electrical parameters of the discharge as a function of gas pressure. The first graph gives the detailed data for the power deposition. The next figure presents data for the energy release in a single pulse depending on the operation mode and pressure.


5

Fig.6. Comparison of the discharge modes in terms of power and energy deposition.

It is seen that the current pulse in Mode 3 is too short in comparison with the discharge channel formation time

at P=1 Bar. However, at lower gas density the reduced energy density deposited into gas is higher. Despite the considerable difference in the energy saved in capacitance, the energy deposition at lower gas density is nearly the same as at 1 Bar. This fact is posed as critically important for further work. Typical schlieren images for Mode 2, described above, are presented in Fig.7 at different pressures. Time delay was 150 μs in these cases. The other Modes demonstrate similar behavior.

Fig. 7. Typical schlieren images vs air pressure.

Fig. 8. Typical schlieren image of pulse discharge in M=0.5 and M=2 flow.

Typical schlieren images for Mode 3 in M=0.5 and M=2 flows are shown in Fig.8 at static pressure

Pst=670 Torr and Pst=220 Torr, respectively. Time delay was 150 and 100 μs in these cases. The graphs in Fig.9 demonstrate the data on the thermal cavern movement and expansion at flow and ambient conditions. A specific


6

feature of the after-discharge channel dynamics is an amplification of instability in subsonic flow. This effect was observed previously [8]; it has no proper explanation yet.

Fig. 9. Thermal cavern movement (solid lines) and expansion (drop lines) under M=0, M=0.5, and M=2 flow. Following conclusions can be currently drawn from the experimental data obtained recently:

The “amplitude” of instability of after-discharge thermal distortion drops with gas density despite of increase of energy input per density number;

The relative velocity of the post-discharge thermal cavern in subsonic and supersonic flow is near zero, if the pulse duration is less than 1μs;

The shorter pulses are the higher is efficiency of energy deposition in explored range of gas parameters.

IV. Some Ideas and Experimental Demonstrations The next important issue is the discharge’s filament localization in inhomogeneous medium. It was found

experimentally that in the most cases the discharge voltage is increased significantly when the air-fuel-plasma interaction takes place. A sample of oscillograms is shown in Fig.10.

Fig.10. Voltage-radiation records of the discharge in airflow and under the hydrogen injection.

The explanation of this fact includes the idea that the discharge localization is managed by the rule of minimal reduced voltage drop along the line shown schematically in Fig.11a. A direction of movement is shown here conditionally as a sample (see below). It can be supposed that the discharge current “prefers” to run through the path in the fuel, oxidizer or between them depending on discharge evolution stage. Currently the criterion is under formalization. The complexity of the problem is aroused by necessity to make adequate account for the gas movement. Fig.11b shows results of numerical simulations of discharge initiation by electric field linearly growing


7

in time in air, ethylene and their stoichiometric mixture for experimental parameters of the channel defined above (note, the discharge’s current density is homogeneous here). The fuel provides the best conditions for discharge development on this phase. With further growth of the electric current the situation can be changed because of two reasons: chemical reactions modify the composition, and the discharge structure (current density distribution at filamentary shape) depends on medium. This statement is confirmed by measuring history of E/N and current in high-current phase of the discharge under study in different gases (see below).

0 1 2 3 4 5 6

0

20

40

60

80

10-910-810-710-610-510-410-310-210-1100101102103

E/N

, Td

Time, μs

Jdisch

J disc

h, A/c

m2

E/N

T = 300 K, P = 1 atm, C = 650 pF

Air Air+C2H4

C2H4

Fig.11. Explanation scheme of the plasma movement from one media to another (a - left). Calculated discharge

current waveforms at the begining of discharge in air, ethylene and their mixture (b - right).

Discharge and after spark channel in different gases. The tests were performed in Air, C2H4, and CO2 (oxidizer, fuel, and product). The pulse duration was t≈100 ns, rate of the voltage rise before breakdown dU/dt≈3×1010V/s, gas pressure p=1Bar, inter-electrodes gap d=55mm. The table below presents some factual data.

Gas Air C2H4 CO2

Typical breakdown voltage, kV 100 98 93

Energy deposition in main pulse, J 2.1 2.4 2.9

Typical plasma channel resistance at t/2, Ohm 18 30 65

Fig.12. Experimental data for 100ns discharge in Air, C2H4, and CO2. Volt-ampere characteristics (upper),

dynamics of reduced electrical field (below).


8

Fig.12 presents the result of reduced electrical field recalculation based on experimental data. The volt-ampere characteristics of the discharge in air, CO2, and ethylene are shown for the reference, as well. It is seen that in the first stage of the discharge the highest field is observed in CO2, but in the second stage the higher value is realized in air. Despite such different behaviors, the power release was almost the same! A small difference in energy release (see table) can be explained by the different durations. The physical reason of discrepancy in calculated and experimental data can be in different structure of the current channel at gas type variation. This difference is visible apparently on schlieren photos of the post discharge channel at small delays after the pulse that is illustrated in Fig.13. In some part of the discharge channel in CO2 the current “splits” to multiple or diffuse shape. It may be supposed that this part of the “filament” has higher resistance. These results can be posed as an experimental demonstration of the discharge location and structure controllability with electrical parameters. This mechanism can enhance mixing additionally.

Fig. 13. Typical schlieren images of pulse discharge in Air and CO2 at delay time 10 μs and 50 μs.

Fig. 14. Schlieren images of pulse discharge interaction with mixing layer of CO2 and Air.

On one side, the breakdown voltage is lower in a fuel (experimentally and by the calculations, as well), on the other side the discharge channel loses the resistance much faster in the air. A priori it is difficult to predict the discharge filament location, when the electrodes position is close to boundary of two different gases. Special tests were performed to clarify this point experimentally. It was found, that in the most cases the discharge was located inside a mixing layer between the gases. Fig.14 illustrates this statement: CO2 jet in air (the first frame) was applied as a model of mixing layer. This fact looks very positive and stimulates future efforts.

V. Filamentary Discharge for Ignition: Computation Results Such parameters of a discharge as E/N (E is the electric field strength, N gas number density) and electric current

density or electron number density are strictly correlated by discharge nature. In order to find rates of processes involving electrons it is necessary to address the electron Boltzmann equation where from the electron energy distribution function (EEDF) could be calculated. To account properly excitation and dissociation of molecules in the discharge one has to guarantee acceptable accuracy of electron scattering cross sections used in the model. The


9

key criterion for evaluation of their accuracy is a good agreement between calculated and available measured transport and kinetic coefficients for each component of the gas mixture. Traditionally, a set of cross sections for a given species, which satisfies this criterion, is called the ‘self-consistent’ set.

The next problem encountered in modeling plasma of inflammable gases is the necessity to combine approaches of high non-thermal plasma kinetics and of classic thermal combustion. Such unification was made by the authors [13] who showed numerically that for plasma ignition of ethylene-air mixture within a reasonable length of a supersonic flow rather high energy input per mass of gas flow is required (about 210 J/g).

Actually, development of a non-thermal discharge with required characteristics in high-speed, high-pressure supersonic gas flow is questionable. As was noted by authors [14], there are experimental data indicating that plasma chemical conversion of methane is more effective when discharges have filamentary form. We anticipate that usage of non-uniform (filamentary) plasma may accelerate essentially ignition of premixed fuel – air flows.

To examine this assumption the model was developed for burning initiation by a series of periodically positioned transverse filamentary-like discharges in approximation of distributed mixing of excited and non-excited gas streams. The model includes self-consistent simulations of the discharge of a small radius in supersonic flow of ethylene-dry air mixture [13] with followed gradual mixing of excited gas with main flow. Mixing process was characterized by time interval between beginning of mixing and its termination by exhaust of main stream, mixtΔ , and the ratio of the final volume to the initial plasma volume, V/V0.

Fig. 15 shows calculated dynamics of gas temperature and discharge power density. It is noteworthy that evolution of the gas temperature is non-monotonous. The temperature comes through the maximum and lowers to the end of mixing. Such a behavior is explained by competition between cooling associated with immixture with the relatively cold (700 K) gas and heating in chemical reactions. At the parameters selected this competition leads to the temperature maximum in the beginning of mixing zone. This effect is more evident in Fig. 16 where the linear time scale is used. Fig. 16 presents calculated temperature evolution for variable dilution degree (V/V0) and for two values of the mixing time.

10-7 10-6 10-5 10-4 10-3

1000

2000

3000

End of mixing

Ein = 320 J/g

V/V0=7

T, K

Time, μsBeginning of mixing

T

0.0

0.5

1.0

1.5

Pdisch

Pdisch, MW/cm3

0 100 200 300 400 500

1000

2000

3000

5

2

T, K

Time, μs

1 2 5 V/V0=7

1020

Ein = 320 J/g

Solid line - mixing time 100 μsDashed line - 50 μs

Fig. 15. Calculated static gas temperature and discharge power density evolution.

Fig. 16. Calculated static gas temperature evolution for variable dilution degree and two mixing time values.

Curve 1 is for the case of uniform discharge excitation with the reduced energy input 320 J/g. The dilution degree

equal to 7 at mixing time 100 μs provides induction time about 300 μs, while faster mixing mixtΔ = 50 μs results in much longer induction time. Modeling plasma ignition by non-uniform discharge, the mixing process starts just after discharge termination (about 1 μs). In this case, faster mixing of excited and non-excited streams leads to prevalence of cooling by mixing process and to lengthening of the induction period.

Fig. 17 shows how the predicted induction time for combustion of ethylene-air mixture depends on the energy input per mass of gas flow in the case of filamentary discharge for two values of mixing time. At mixtΔ =100 and

indtΔ =500 μs the required reduced energy input is about 40 J/g, that is remarkably lower than for uniform discharge, 210 J/g.


10

Fig. 17. Calculated induction time for ignition of С2H4:O2:N2=1:3:12 mixture flow with M = 2.6, Pst=1Bar, Tst=700K as a function of the reduced energy input.

Fig. 17 shows how the predicted induction time for combustion of ethylene-air mixture depends on the energy

input per mass of gas flow in the case of filamentary discharge for two values of mixing time. At mixtΔ =100 and

indtΔ =50 μs the required reduced energy input is about 40 J/g, that is remarkably lower than for uniform discharge.

VI. Concluding Remarks In this paper it is assumed that effect of fast turbulent expansion of the post-discharge channel is effective for an

acceleration of mixing and ignition in the non-premixed multi-component flow.

Now the plasma of electrical discharges in airflow is considered as the quite promising method for combustion enhancement. In the case of high-speed combustion a new technique has to solve several important problems: fuel ignition by direct heating/active radicals’ deposition/flow structure modification, mixing in non-premixed flow, flame-holding, etc. Preliminary analysis has shown that the short-pulse repetitive transversal electric discharge looks as the most prospective low-energy method of the mixing intensification and, probably, ignition. From the other side the properties of the filamentary plasma under so specific conditions are known not perfectly. The effect of fast turbulent expansion of the post-discharge channel in high-speed flow was experimentally observed and described in details. Now such an effect is supposed to be effective for a mixing acceleration of non-premixed multi-components flow. The plasma influence occurs in a gas portion and a negligible, in average, energy release doesn’t lead to a dramatic change of the flow bulk parameters in duct and to thermal chocking.

Experimental and theoretical results presented in this work demonstrate that turbulent and directed motion arising in the after-spark channel can essentially enhance the rate of fuel-gas mixing, which may control the mixing rate in engines with high-speed gas flow. It is predicted by numerical simulations that usage of filamentary discharge for ethylene-air mixture ignition in supersonic flow can lead to significant reduction of required energy input.

Acknowledgments The experimental work is funded through EOARD-ISTC project #3793p (special thanks to Dr. J. Tishkoff and

Dr. S. Surampudi). Some parts of this work were supported by the Russian Academy of Science (coordinator Academician Gorimir Chernyi) and RF President Grant for Leading Research Schools No NSh-1573.2008.2.

The authors express their gratitude to Dr. Valentin Bityurin for multiple discussions on the mixing nature, to Mr. Konstantin Savelkin and Mr. Paul Makeev of Joint Institute for High Temperatures, Russian Academy of Sciences for valuable assistance in experimental work.

References 1. M. G. Mungal & P. E. Dimotakis “Mixing and Combustion with Low Heat Release in a Turbulent Shear Layer,” J. Fluid Mech., (1984), 148, 349-382.


11

2. T. Rossmann, M. G. Mungal & R. K. Hanson “Mixing Efficiency Measurements Using a Modified Cold Chemistry Technique,” Expts. Fluids, 37, 566-576, (2004).

3. Leonov, S. B., Yarantsev, D. A., Napartovich, A. P., Kochetov, I. V. “Plasma-Assisted Combustion of Gaseous Fuel in Supersonic Duct”, Plasma Science, IEEE Transactions on, 2006, Volume: 34, Issue: 6, pp.2514-2525.

4. L. Jacobsen, C. Carter, R. Baurie, T. Jackson “Plasma-Assisted Ignition in Scramjet”, AIAA-2003-0871, 41st AIAA Aerospace Meeting and Exhibit, 6-9 January, Reno, NV, 2003.

5. Starikovskaia S.M. Plasma Assisted Ignition and Combustion, J.Phys. D: Appl. Phys. 2006. V. 39. R265. 6. Bocharov A., Bityurin V., Klement’eva I., Leonov S. Experimental and Theoretical Study of MHD Assisted Mixing and Ignition in Co- Flow Streams // Paper AIAA 2002- 2228, 40th AIAA Aerospace Sciences Meeting & Exhibit, 14-17 January 2002/ Reno, NV, P.8.

7. Sergey B. Leonov, Dmitry A. Yarantsev, Yury I. Isaenkov “Properties of Filamentary Electrical Discharge in High-Enthalpy Flow”, 43rd AIAA Aerospace Sciences Meeting & Exhibit, 10-13 January 2005/ Reno, NV, AIAA-2005-0159.

8. S. Leonov, M. N. Schneider, D. Yarantsev “Fast Mixing by Pulse Discharge in High-Speed Flow” // 14th AIAA/AHI Space Planes and Hypersonic Systems and Technologies Conference, Canberra, Australia, Nov. 6-9, 2006, Paper AIAA-2006-8129

9. S. B. Leonov, Y. I. Isaenkov, M. N. Shneider, Suppression of the Turbulent Decay of Afterspark Channel with Residual Current, Physics of Plasmas, v.15 (2007)

10. M. N. Shneider, Turbulent decay of after-spark channels, Physics of Plasmas, v.13, 073501-11, (2006) 11. S. Leonov, Yu. Isaenkov, D. Yarantsev “Pulse Discharge in Mixing Layer of Reacting Gases”, Book of Abstracts, 61th Annual Gaseous Electronics Conference 14- 17 October, 2008, Dallas-Addison Marriott Quorum, Dallas, TX, USA

12. Sergey B. Leonov, Campbell Carter, Konstantin V. Savelkin, Valery N. Sermanov, Dmitry A. Yarantsev, “Experiments on Plasma-Assisted Combustion in M=2 Hot Test-Bed PWT-50H”, AIAA-2008-1359, 46th AIAA Aerospace Meeting and Exhibit, 7-10 January, Reno, NV, 2008

13. Kochetov I. V., Leonov S. B., Napartovich А. P., High Energy Chemistry, 40, 94, 2006 14. Pushkarev A., Zhu Ai-Min, Li Xio-Song, Sazonov R., V International Symposium on Theoretical and applied Plasma Chemistry, l.1, 23, 2008

Documents

[American Institute of Aeronautics and Astronautics 47th AIAA Aerospace Sciences Meeting including The New Horizons Forum and Aerospace Exposition - Orlando, Florida ()] 47th AIAA