Experimental and modeling studies of low energy ion sputtering in ion thrusters

The successful demonstration of the NASA Solar electric propulsion Technology Application Readiness (NSTAR) 30 cm diameter ion thruster on the Deep Space 1 mission has paved the way for more ambitious electric propulsion missions. The DAWN mission, scheduled to launch in 2006, will use three NSTAR thrusters to flyby asteroids Vesta and Ceres.¹ NASA's Evolutionary Xenon Thruster (NEXT), a 40 cm diameter thruster currently under development, will enable outer planet and sample return missions using electric propulsion.² These new missions will require substantially longer life time. For instance, the DAWN mission requires a xenon propellant throughput of 288 kg for its NSTAR thrusters, which is 2.4 times more than the currently flight qualified throughput.² Most failure modes of an electrostatic ion thruster are associated with accumulated damage from ion bombardment sputtering.³ The cathode, cathode keeper, screen grid, and accelerator grid are the engine parts that are susceptible to failure due to sputtering damage. For instance, sputtering of the cathode keeper and the screen grid is thought to be mostly the result of doubly charged ion impingement. Chargeexchange ions produced in ion optics can impinge upon the accel grid leading to grid erosion. Sputtering in ion thrusters is caused by low energy impingement ions. Hence, knowledge of low energy sputtering yields and the sputtering threshold are critical for accurate prediction of a thruster's lifetime.² However, sputtering yields for xenon-molybdenum systems are not well known for low ion energies, especially for energies below 100 eV, the energy range of interest to ion thruster applications. Very few experimental measurements ion sputtering for energies below 100 eV exist, and there is a large discrepancy among the published data. Modeling results are also unreliable in the low energy range. As a result, current predictions of damage that engine parts would sustain from low energy sputtering are highly uncertain. This paper presents initial results from an experimental and modeling investigation of low energy ion sputtering. Experimental measurements involve sputtering molybdenum onto a foil substrate using a xenon ion beam. Rutherford Backscattering Spectrometry (RBS) is used to measure the amount of sputtered material deposited on the substrate in order to calculate the sputtering yield. Experimental results are compared with TRIM sputtering simulations, and the Yamamura sputtering model. Section II discusses experimental procedures. Section III characterizes experimental conditions and discusses their effects on results. Section IV discusses modeling using the TRIM code and the Yamamura model. Section V compares experimental results and modeling results. Section VI contains conclusions.