Kenichi Kubota

Comparison of Simulated Plasma Flow Field in a Two-Dimensional Magnetoplasmadynamic Thruster With Experimental Data

Comprehensive comparisons of the numerically simulated results of plasma flow fields in a 100-kW-class 2-D magnetoplasmadynamic thruster with the available experimental data are conducted. The propellant is argon of 1.25 g/s, and the discharge current is varied from 8 to 12 kA. The physical model includes a nonequilibrium single level of ionization and a collisional radiative model for argon ion to assess the reaction processes in detail. The data we mainly compared are the current path, electron number density, and electron temperature. There is qualitative agreement between the calculated and experimental results except for the electron temperature. In order to explain the disagreement of the electron temperature, we estimate the excitation temperature from the distributions of the excited ions in 4s and 4p states, the radiation of which was employed to determine the electron temperature in the experiment. As a result, it is found that the calculated excitation temperature becomes close to the measured result and that the plasma deviates from the partial local thermodynamic equilibrium near the anode surface. Regarding the thrust and thrust efficiency, their features against variation of the discharge current are well captured by the simulation, although they are slightly overestimated compared with the measured values.

Numerical Study of Plasma Behavior in a Magnetoplasmadynamic Thruster Around Critical Current

A self-field magnetoplasmadynamic thruster is numerically studied for an argon mass flow rate of 0.8 g/s. We focus on the plasma behavior around the critical current, at which full propellant ionization is theoretically expected. At a discharge current higher than the critical current, the Hall parameter is large at the anode surface, which leads to an obliquely skewed current profile. This skewed current profile, together with the voltage―current characteristics, show that the Hall effect plays an important role around the critical current. When operating the magnetoplasmadynamic thruster above the critical current, it is shown that the current concentration and the gas density depletion at the anode edge induce a shortage of current carriers. To circumvent the carrier shortage at high discharge currents, we employ a segmented anode. It is found that a segmented anode is effective for suppressing the current concentration and the plasma density depletion on the anode surface without degrading the performance of the thruster.

Study of electron transport in a Hall thruster by axial–radial fully kinetic particle simulation

Electron transport across a magnetic field in a magnetic-layer-type Hall thruster was numerically investigated for the future predictive modeling of Hall thrusters. The discharge of a 1-kW-class magnetic-layer-type Hall thruster designed for high-specific-impulse operation was modeled using an r-z two-dimensional fully kinetic particle code with and without artificial electron-diffusion models. The thruster performance results showed that both electron transport models captured the experimental result within discrepancies less than 20% in thrust and discharge current for all the simulated operation conditions. The electron cross-field transport mechanism of the so-called anomalous diffusion was self-consistently observed in the simulation without artificial diffusion models; the effective electron mobility was two orders of magnitude higher than the value obtained using the classical diffusion theory. To account for the self-consistently observed anomalous transport, the oscillation of plasma properties was speculated. It was suggested that the enhanced random-walk diffusion due to the velocity oscillation of low-frequency electron flow could explain the observed anomalous diffusion within an order of magnitude. The dominant oscillation mode of the electron flow velocity was found to be 20 kHz, which was coupled to electrostatic oscillation excited by global ionization instability.

Parametric Kinetic Simulation of an IHI High Specific Impulse SPT-Type Hall Thruster

A 1 kW class magnetic layer type Hall thruster designed for high specific impulse operation by IHI Corporation, Japan was modeled by a fully kinetic particle code. The measured maximum performance of the thruster was 64% in anode efficiency and 3,200s (3.1kW, 800V) in anode specific impulse. The thruster performance, wall heat loss and erosion, and the plasma property distributions were numerically investigated for the operation conditions ranged from 300V to 700V in discharge voltage, and 2mg/s to 4mg/s in xenon mass flow rate. Simulations with two numerical models: with and without the Bohm diffusion assumption were performed for each thruster operation conditions to characterize the uncertainty caused by the Bohm diffusion model. The simulation results were compared with the measured results, and exhibited excellent agreement with the maximum performance error of 20% for both models. It is suggested that as engineering tools, with and without Bohm simulations can be respectively used as the worst and best case analysis for the performance, heat load, and erosion.

Flowfield Analysis in an MPD Thruster with Applied Magnetic Field

Two-dimensional numerical analyses on the flow field in a several tens of kilowatts class Magnetoplasmadynamic thruster with an applied magnetic field are conducted. With a newly developed numerical code, the influences of the strength of the applied magnetic field and its divergence angle on the plasma flowfield are examined under the conditions of a constant argon mass flow rate of 100 mg, and a constant discharge current of 1 kA. It is shown that the discharge current expands downstream (to the plume region) when the applied magnetic field strength is increased. The thruster with a rapidly diverging magnetic field is superior to a slowly diverging case in terms of plasma acceleration. The Hall acceleration is found to occur mainly inside the discharge chamber, and it is increased with raising the applied magnetic field. The highest thrust of the Hall acceleration in this paper amounts to 52 % of the total thrust increment with a rapidly diverging applied magnetic field of 0.25 T. The energy flux analysis suggests that a considerable friction loss or a thermal conduction loss on the electrode exist in the discharge chamber of the thruster. On the other hand, in the plume region, effective energy conversion from the azimuthal kinetic energy or the specific enthalpy into the axial kinetic energy seems to occur.

Numerical simulation of microwave neutralizer including ion’ s kinetic effects

In order to analyze a microwave neutralizer in ion’s time/space scales, a threedimensional Hybrid-PIC (Particle-In-Cell) solver which treats ions and electrons as particles and fluid was developed. In this analysis, microwave power absorption distribution was estimated by means of another electromagnetic PIC solver. The results show that qualitative agreement on voltage-current characteristics was achieved, whereas a discontinuous current jump between a low current mode and a high current mode was still diffused. It is found that, in the high current mode, the electric potential is gradually increased toward the plume region inside the orifice, which promotes high ion production rate there. It is also shown that the sputtering rate of an antenna is comparable with the measured data, where doublyionized ions mainly produced inside the orifice considerably aggravate the sputtering rate.

Numerical Simulation of Plasma Flow in a Self-field MPD Thruster Coupled with Electrode Sheath

4Plasma flows in a 100-kWe-class, steady-state, self-field magnetoplasmadynamic (MPD) thruster were simulated by a plasma flow solver coupled with an electrode sheath model, which enables us to evaluate electrode fall voltages quantitatively. In this paper, influences of the coupling with the electrode sheath model on discharge pattern are discussed as well as dependences of thruster performances on the propellant mass flow rate and the discharge current. By the coupling, it is shown that a thrust is not significantly affected while a discharge voltage is increased attributed to a cathode fall voltage comparable with a potential fall just in the bulk plasma. The thrust and discharge voltage evaluated with the electrode sheath roughly agree with existing experimental results. For an argon mass flow rate of 2.0 g/s and a discharge current of 8 kA, the average cathode fall voltage was estimated to be 7.1 V, which is comparable with the average bulk fall voltage (7.1 V). Thus, it can be said that energy consumption within the cathode sheath is a significant loss factor of the MPD thruster.

Numerical Study of a Hydrogen Plasma Flow Field in a Self-Field Magnetoplasmadynamic Thruster

Numerical simulation was conducted for the MY-II magnetoplasmadynamic (MPD) thruster (flared type, H2 propellant) including the real gas effect (i.e., nonequilibrium dissociation and ionization, three temperature model). In the case of Jdis = 5 kA and m& = 0.4 g/s, the calculated discharge current patterns reproduced measured results well. On the other hand, difference current patterns between the calculated and the experimental result are found in the case of Jdis = 10 kA and m& = 0.4 g/s, which corresponds to critical current. Below the critical current, it is found that numerical simulation could accurately predict the measured thrust (Fcalc. = 15.8 – 26.9 N) for Jdis = 5-9 kA, m& = 1.37 g/s.

Numerical Simulation on Magnetoplasmadynamic Thruster with an Electrode Model

An electrode model was incorporated into the two-dimensional numerical simulation of plasma flow in a magnetoplasmadynamic thruster. Quantitative prediction of the sheath voltages on a cathode and an anode provides self-consistent evaluation of the discharge voltage and thrust efficiency. In this model, the anode sheath was treated as an ion-rich sheath. For argon propellant of 0.8 g/s and a discharge current of 5 kA, the cathode sheath voltage was about 20-25 V, and the anode sheath voltage was 1-2 V along the anode surface. The input power and thrust efficiency estimated with the electrode model was 200 kW and 14%, although the input power on the experiment was 246 kW. The discrepancy might be attributed to the cathode spot formation in the quasi-steady experiment. The evaluation of the heat losses indicates that the anode heat loss is dominant energy loss mechanism.