Investigation of Hall effect thruster channel wall erosion mechanisms
Schinder, Aaron M.
MetadataShow full item record
The primary life-limiting mechanism for Hall effect thrusters (HETs) is the plasma erosion of the discharge channel wall. Over the course of tens of thousands of hours, energetic ions sputter material from the annular discharge channel wall of the HET, wearing away the material in the 1-2 cm near the exit plane of the thruster. If the channel wall is completely worn away in these areas, the magnetic circuit is exposed, and continued operation of the thruster will lead to the ejection of ferrous material into the spacecraft environment and eventual failure of the magnetic circuit. Qualifying HETs for a minimum 1.5 times desired mission life is an expensive process requiring tens of thousands of hours of chamber time. Computational modeling of thruster lifetime can make predictions about the average erosion depth, but present models cannot explain certain features that appear during testing. One such feature is the anomalous erosion ridge phenomenon in HETs. In order to improve HET life modeling, a better understanding of the formation of features during plasma erosion is needed. In this work, an investigation into the details of the plasma erosion of materials is conducted. The way in which the material microstructure and the mechanical stress in materials modify the process of plasma erosion is studied with experiments and computational modeling. A 3D raytracing model of the development of surfaces in a complex heterogeneous material is created. The model reproduces the development of surface features observed in SEM microscopy of the eroded AFRL/UM P5 channel wall. SEM imaging of borosil reveals a complex heterogeneous microstructure composed of boron nitride grains in a silica matrix. The role of the microstructure in the development of observed erosion features is explored. The strain relief hypothesis, which proposes that the presence of mechanical stress in materials will lead to the existence of unstable surface modes under erosion, is investigated. The SRH predicts that surface features with wavelengths dependent on applied mechanical stress will grow during erosion. An experiment to test the dependence of the plasma erosion process on the presence of mechanical stress in materials is designed and conducted. Two materials, amorphous fused silica and M26 borosil, are placed under varying amounts of mechanical stress up to 25 MPa and exposed to argon plasma for 12 hours. Microscopy and detailed surface statistics are collected before and after each exposure. During each exposure, a pair of samples: one under a compressive mechanical load, and the other unloaded, are exposed. The results of these experiments reveal that different mechanisms for each material lead to the development of complex surface patterns. For fused silica, a complex cell pattern is generated from initial roughness present in the surface. The development of this cell pattern can be explained as resulting from the angle-dependence of the sputtering yield of silica. For M26 borosil, it is found that the difference in the sputtering yield between the boron nitride and silica components of the material is the dominant mechanism leading to the development of surface features. For both M26 and fused silica samples, for applied loads of up to 25 MPa, no dependence of the development of the surface features on the mechanical stress has been detected. This work has found that the ion impact angles, the initial surface structure in the case of fused silica, and the heterogeneous nature of borosil composites all play a role in the generation of microstructural surface features during plasma erosion. However, no evidence has been found for the sensitivity of the plasma erosion of M26 and fused silica to mechanical stresses of up to 25 MPa.