A differential sputter yield measurement technique is described, which consists of a quartz crystal monitor that is swept at constant radial distance from a small target region where a high current density xenon ion beam is aimed. This apparatus has been used to characterize the sputtering behavior of various forms of carbon including polycrystalline graphite, pyrolytic graphite, and PVD-infiltrated and pyrolized carbon-carbon composites. Sputter yield data are presented for pyrolytic graphite and carbon-carbon composite over a range of xenon ion energies from 200eV to 1keV and angles of incidence from 0° to 60°.
Sputtring is the physical removal of material from a surface through the means of energetic partical bombardment. In film deposition applications, sputtering can be mode by which thin film coatings of target material are applied to surfaces to enhance their mechanical, thermal, or optical properties. In the context of space propulsion using ion or plasma thrusters, sputtering is never desired. In essence, it is an erosion phenomenon that limits the lifetime of components used in these thrusters. Although sputtering has been studied for the past 150 years, the sputter yields for many materials have not been fully characterized, especially for heavy particles like xenon ions with moderate to low energies and non-zero incidence angles that are bombarding materials suitable for electric propulsion components such as pyrolytic graphite (PG) and carbon-carbon composites (C/C).
As the interest in electric propulsion (EP) builds, it is becoming important for analysts and designers to have sputter yield data for a wide range of incident ion angles and energies in order to accurately gauge thruster life time. Both total sputter yield and differential sputter yield are needed. These data are of interest for calculating erosion of spacecraft surfaces and thruster components that can be struck by high-energy ions produced in the beams of these thrusters. Of equal interest is sputtering caused by ions with lower energies. Sputter erosion of surfaces at or near cathode potential that face regions of dense palsma can also occur via bombardment from multi-charged ions. These multi-charged ions can obtain energies corresponding to multiples of the anode and cathode potential deifference, which can be above the sputtering threshold for most materials.
Sputter yield data are also of interest for calculating back sputter rates onto thruster and vacuum chamber surfaces from beam dumps in vacuum facilities where life tests and flight acceptance tests are performed. Sputter resistanct components fabricated from various forms carbon have been proposed for life critical components like accel and screen grids, keeper electrodes, discharge chamber liners, etc and for vacuum facility beam dumps and all liners. However, due to the incredibly long life times required of EP systems, even these components may become highly eroded. Long term life tests of new carbon-based components are prohibitively expensive, and their resistance to erosion will need to be evaluated through the use of models in combination with high quality sputter yield data.
The work described in this paper was initiated response to the need for sputter yield measurements of materials unique to thruster designs with long life time capability- namely pyrolytic graphite and specialized carbon-carbon composite materials. In this effort we utilized a technique to measure total and differential sputter yields of pyrolytic graphite and C/C composites subjected to xenon ion bombardment at energies ranging from 200 eV to 1000 eV and for angles of incidence from 0° to 60°. The heart of the technique was a quartz crystal monitor that was swept at constant radial diatance from a small target region where a high-current-density xenon ion beam was aimed. The QCM measurements allowed calculation of differential sputter yields versus the polar angel of the QCM measured from the target normal direction. Differential sputtering yields were measured over a 180° arc in a plane that included the ion beam centerline and the normal vector of the target surface. Total sputter yields can be calculated from differential sputter yield data using simple integration procedures and comparisons to total sputter yield calculations were made to sputter yields obtained from the literature wherever possible.