Search

Hollow cathodes serve as the electron source in ion and Hall thrusters. One of the life limiting factors of the propulsion system is cathode failure due to erosion from high energy ion bombardment. Despite the successful application of hollow cathodes on commercial and deep-space missions, the fundamental physical processes of erosion are not fully understood, particularly the source of the high energy ions. A recent experimental study of the near-plume in a high current hollow cathode confirmed the existence of ion acoustic turbulence (IAT), a phenomenon that was only previously suspected and modeled in numerical simulations. Theoretical analyses and turbulence measurements in the plume established that ion acceleration due to turbulence could explain the existence of high energy ions. This paper is a continuation of this work, focusing on detecting and quantifying instabilities in the cathode near-plume under conditions relevant for the thruster being developed for the proposed Asteroid Robotic Redirect Mission (ARRM). Ion acoustic turbulence and ion energy measurements were taken at the nominal discharge currents and cathode flow rates expected for ARRM, as well as off-nominal conditions to determine where it might be susceptible to erosion by high energy ions. Dual ion saturation probes were used to measure fluctuations in ion saturation current and assess the wave dispersion relation. A retarding potential energy analyzer was used to measure ion energy distribution functions for ions with velocities perpendicular to the plume flow. The effect of cathode orifice size on the onset of IAT has also been studied. Peaks in wave amplitude were found at extremely low flow rates and high discharge currents, where instabilities other than IAT dominate. In addition, the onset of IAT was found to occur at discharge currents as low as 35 A. Measurements also confirm that at high discharge currents above 30 A, a larger orifice size reduces the magnitude of turbulence. Howeve

Distributions of ion energy-per-charge were collected with an electrostatic energy analyzer directed along various lines-of-sight in the near-plume region of a hollow cathode operating in a laboratory discharge chamber. While the analyzer was stationary, the cathode and anode assembly were mounted on a rotary and a linear stage, allowing measurements along the cathode axis and at off-axis angles up to 90°; the linear stage enabled translational motion which allowed for measurements parallel to the cathode axis. At angles below 45°, three distinct ion populations were identified: two low-energy narrow peaks and a broad high-energy “tail” with energies-per-charge in excess of 5 times the discharge voltage. Detection of low-energy ions above 45° was not possible since they were confined by the anode voltage. At large angles, the intensity of the high-energy ion signal decreased to become indistinguishable from noise in the data system; however, based on previous results, the failure to detect high-energy ions at large angles has been attributed to the large distance between the ESA line-of-sight and the cathode axis increased from 0° to 16°, the location of the low-energy ion peaks shift but there is no apparent difference in the high-energy ion tail. Increasing the separation angle further yields a decrease in intensity at all energies. For low energies-per-charge, the shape and intensity of the unidirectional ion-energy distribution does not change when the cathode is translated up to 2.5cm; however, as the distance from the cathode centerline increases, the signal intensity of the high-energy ions decreases in a manner similar to the decrease in high-energy ions seen with increasing angles above 30°. These results indicate that magneto-gas dynamic forced did not accelerate the ions observed in this experiment; additional experiments will elucidate the mass and energy dependence of the mechanisms responsible for high-energy ions.

The effect of tungsten erosion, transport and redeposition on the operation of dispenser hollow cathodes was investigated in detailed examinations of the discharge cathode inserts from an 8200 hour and a 30,352 hour ion engine wear test. Erosion and subsequent redeposition of tungsten in the electron emission zone at the downstream end of the insert reduces the porosity of the tungsten matrix, preventing the flow of barium from the interior. This inhibits the interfacial reactions of the barium-calcium-aluminate impregnant with the tungsten in the pores. A numerical model of barium transport in the internal xenon discharge plasma shows that the barium required to reduce the work function in the emission zone can be supplied from upstream through the gas phase. Barium that flows out of the pores of the tungsten insert is rapidly ionized in the xenon discharge and pushed back to the emitter surface by the electric field and drag from the xenon ion flow. This barium ion flux is sufficient to maintain a barium surface coverage at the downstream end greater than 0.6, even if local barium production at that point is inhibited by tungsten deposits. The model also shows that the neutral barium pressure exceeds the equilibrium vapor pressure of the impregnant decomposition reaction over much of the insert length, so the reactions are suppressed. Only a small region upstream of the zone blocked by tungsten deposits is active and supplies the required barium. These results indicate that hollow cathode failure models based on barium depletion rates in vacuum dispenser cathodes are very conservative.

Ion energy measurements from a ¼ inch cathode were made with krypton and xenon expellants at a range of flow rates and discharge currents using an electrostatic energy analyzer mounted on axis. High energy ion content scaled strongly with discharge current and inversely with mass flow rate and molecular mass. At high current conditions for krypton and xenon peak ion energies exceeded 130eV (>17kms⁻¹) and 75eV (>10kms⁻¹) with high energy tails extending well beyond 140eV and 200eV respectively. No conclusive correlation was round with discharge noise in the form of high frequency voltage oscillations and high energy ion content.

Hollow cathodes that operate with impurities in the xenon propellant exhibit poisoning of the emitter surface, resulting in a higher surface work function and increased operating temperatures. The goal of this paper is to provide insight Into the physical processes responsible for cathode degradation during short-duration oxygen poisoning over a range of discharge currents and surface temperatures. The experimental results presented here show that the cathode is more susceptible to poisoning at higher oxygen concentration levels, lower discharge currents, and lower initial insert temperatures. The threshold current at which the emitter surface becomes susceptible to oxygen poisoning was determined to be between 8.24 A and 10.17 A for an oxygen concentration level of 10 ppm, and between 10.17 A and 12.35 A for oxygen concentration levels of 50 and 100 ppm. To determine the threshold surface temperature, further work must be conducted to isolate the effects of xenon ion sputtering and surface temperature on cathode operation during poisoning.

Magnetoplasmadynamic thrusters (MPDTs) are coaxial devices that utilize Lorentz forces to accelerate a plasma discharge. Such devices are capable of processing high power (from 100s or kW to multi-MW per device) and are or interest for future spacecraft propulsion applications. However, the performance of these thrusters is limited by the "onset" instability condition, in which radial pinch forces deplete the anode region plasma, resulting In high anode fall voltages and excessive erosion beyond a characteristic current level. A new MPDT was developed utilizing a multi-channel hollow cathode with an extended anode and multi-channel hollow cathode. The thruster was tested in the "quasi-steady" mode with argon propellant at 1.5, 1.0, and 0.6 g/s mass now rates. The thruster was operated initially in the self-generated magnetic field mode, and data are presented on the voltage-current response characteristics to identify early indicators or transition to the MPD "onset" instability in this geometry. This work serves to prepare for later experiments to compare results with various externally-applied magnetic fields to investigate the influence of the magnetic field on the characteristic current for the onset condition.