2122

IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 36, NO. 5, OCTOBER 2008

Decel Grid Effects on Ion Thruster Grid Erosion

Richard E. Wirz, John R. Anderson, Dan M. Goebel, Fellow, IEEE, and Ira Katz
AbstractJet Propulsion Laboratory (JPL) is currently assessing the applicability of the 25-cm Xenon Ion Propulsion System (XIPS) as part of an effort to infuse low-cost technically mature commercial ion thruster systems into NASA deep space missions. Since these mission require extremely long thruster lifetimes to attain the required mission V, this paper is focused on understanding the dominate wear mechanisms that effect the life of the XIPS three-grid system. Analysis of the XIPS three-grid conguration with JPLs CEX3D grid erosion model shows that the third decel grid effectively protects the accel grid from pits and grooves erosion that is commonly seen with two-grid ion thruster grid systems. For a three-grid system, many of the charge-exchange ions created downstream of the grid plane will impact the decel grid at relatively low energies (25 V), instead of impacting the accel grid at high energies (200 V), thus reducing overall erosion. JPLs CEX3D accurately predicts the erosion patterns for the accel grid, although it appears to overpredict the pits and grooves erosion rates due, mainly, to uncertainties in incident energies and angles for sputtering ions and since it does not account for local redeposition of sputtered material. Since the model accurately simulates the erosion pattern but tends to overpredict the erosion rates for both the two- and three-grid sets, this comparative analysis clearly shows how the decel grid signicantly suppresses the erosion of the downstream surface of the accel grid as observed in experimental tests. The results also show that the decel grid has a relatively small effect on barrel erosion (erosion of the aperture wall) and erosion of the upstream surface of the accel grid. Decreasing the accel grid voltage of the XIPS can reduce barrel (and total) erosion of the accel grid and should be considered for high-V missions. Index TermsDecel grid, grid erosion, ion thruster, Xenon Ion Propulsion System (XIPS).

Fig. 1. SEM images of the NSTAR accel grid after 30 352 h of operation. The accel grid experienced signicant erosion on the downstream surface in a pits & grooves pattern. In the center region of the grids, the pits were worn through the grid around many of the apertures. Grid apertures were widened by 25% while minimal erosion occurred on the upstream surface.

I. I NTRODUCTION ASA/JET Propulsion Laboratory (JPL) is considering the use of commercial ion thruster technology for deep space missions. L-3 Electron Technologies, Inc.s 25-cm Xenon Ion Propulsion System (L-3 ETI 25-cm XIPS) is an attractive candidate due to its ight heritage, reliability, and performance. Several XIPS thrusters have thousands of hours of ight and ground test data; however, to accommodate high-V deep space missions, we must validate the thruster for tens of thousands of hours of life. The Long-Duration Test (LDT), and then later the Extended Life Test (ELT), of the NASA Solar Electric Propulsion Technology Application Readiness (NSTAR) ion thruster identied the erosion of the molybdenum accelerator

Manuscript received November 1, 2007; revised March 2, 2008 and April 25, 2008. Current version published November 14, 2008. This work was carried out by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. The authors are with the Electric Propulsion at the Jet Propulsion Laboratory, Pasadena, CA 91109 USA (e-mail: Richard.E.Wirz@nasa.jpl.gov). Color versions of one or more of the gures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identier 10.1109/TPS.2008.2001041

ion extraction grid (or simply accel grid) as a primary lifelimiting mechanism for two-grid ion thruster life [1], [2]. During the ELT, the accel grid apertures were eroded to as much as 25% beyond their original diameter; and, as shown in Fig. 1, pits were eroded entirely through the downstream accel grid face between many of the apertures near the thruster axis. The erosion on the downstream face is generally referred to as pit & grooves erosion while the erosion on the inside of aperture walls is referred to as barrel erosion. Both of these erosion types are important to minimize since they can compromise the structural integrity of the grids, increase neutral propellant loss, and increase the voltage required to prevent electron backstreaming. The XIPS ion thruster uses a third molybdenum decelerator grid (decel grid), in addition to screen and accel molybdenum grids. The decel grid is used on XIPS to minimize pits and groove erosion of the accel grid, thus minimizing the amount of eroded molybdenum grid material that is ejected from the thruster. The primary disadvantage of a decel grid is the increase in complexity of the physical thruster assembly. To predict the long-term life and performance of the XIPS thruster, we must understand how the decel grid affects the erosion of the accel grid. Reference [3] discusses that the decel grid reduces ion current to the accel grid. Experimental tests by Brophy et al. [4] showed that the erosion rate for a 30-cm ion thruster found a greater than 100 times reduction in accel grid erosion when using a three-grid compared to a two-grid assembly. More recently, 2-D and 3-D grid erosion models CEX2D and CEX3D were developed at JPL to help understand these erosion mechanisms [2], [5]. These models reveal that the erosion of the ion extraction grids is due to charge-exchange (CEX) ions created between and downstream of the grids; this result was also found by grid erosion models created at other institutions [6]. The culprit CEX ions are created when fast ions that are accelerated through the grid apertures gain an electron from slow neutral atoms, thus creating a slow ion and a neutral atom

0093-3813/$25.00 2008 IEEE

WIRZ et al.: DECEL GRID EFFECTS ON ION THRUSTER GRID EROSION

2123

that quickly travels downstream. A signicant percentage of these slow ions (typically referred to as CEX ions) are born in potential elds that accelerate them toward the accel grid at velocities sufcient to erode the grid surface. In previous efforts, the CEX2D and CEX3D models were compared against LDT and ELT data for the NSTAR grid assembly and compare well with the experimentally observed erosion behavior; however, further code development is necessary to validate the exact erosion rates as discussed in [2] and [5]. For example, and most importantly to this effort, CEX3D accurately simulates the pits and grooves erosion patterns observed on the accel grid from the LDT; however, the erosion rates from [2] overestimate the initial wear rate by about 50%. This difference is due at least in part to the uncertainties in sputter yield as a function of incident energy and angle, determination of local incident angles, and the fact that the codes do not account for redeposition of sputtered grid material locally at the erosion sites. Therefore, in this effort, we use the model results to perform a relative comparison of erosion rates for a given geometry (using the beginning-of-life (BOL) geometry) and not to determine longterm erosion values. This approach is valid for comparing the erosion of two- versus three-grid geometries since the uncertainties previously mentioned predominately affect the local surface morphology over long durations; therefore the relative accuracy of the instantaneous erosion rates will not be affected by differences between the two- and three-grid geometries. A. Objective In this paper, we examine the erosion characteristics of the XIPS accel grid in the presence of the decel grid and discuss how these characteristics will affect thruster life. To facilitate this investigation, we rst compare the erosion of the XIPS three-grid accel grid with erosion of the NSTAR two-grid assembly and then compare the erosion of the XIPS accel grid with and without the decel grid. We also investigate methods for reducing overall grid wear. The results presented herein are for comparison of two- and three-grid systems and not intended to provide absolute erosion rates. II. A NALYSIS AND R ESULTS In this section, we compare the accel grid erosion of twoand three-grid geometries using JPLs CEX3D grid erosion models. CEX3D is used since a 3-D domain is necessary to accurately simulate the upstream erosion of the accel grid. All comparisons are for apertures near the thruster axis where erosion is expected to be greatest for both the XIPS and NSTAR thrusters due to the higher current density and intermediate neutral density near the axis [1], [7]. JPLs ion thruster discharge model was used to determine upstream ion density, neutral density, and electron temperature conditions [8]. A detailed description of how CEX3D propagates these conditions into the beamlet and computes erosion rates is given in [2]. The axial domain for all problems extends 5-cm downstream of the upstream surface of the screen grid to capture the erosion effects due to CEX ions created far downstream of the grid plane (note: Extending the axial domain to 10 cm produces a change

TABLE I TYPICAL THROTTLE POINTS FOR XIPS AND NSTAR

in the erosion rates predicted by the code of less than 0.5% since the majority of CEX that contribute to erosion are created within 5 cm of the grid). The grid geometries and thruster conditions used in the models are for BOL and include the hotgrip gap spacing observed during thruster testing [9]. Double ions are not included in the analysis; however, equivalent beam current densities are used. The propellant for all simulations is xenon, the propellant used by XIPS, NSTAR, and most noble gas electric thrusters. Throttle points used for model inputs are shown in Table I [1], [10]. A. Comparison of Accel Grid Erosion for XIPS and NSTAR Grid Assemblies In this section, we compare the BOL erosion rates of the accel grid for the XIPS three-grid and NSTAR two-grid systems. The XIPS and NSTAR grid systems have similar ion transparency. Ion transparency is a measure of the beamlet current extracted by a grid geometry for a given upstream plasma density and grid voltage, which is analogous to the ChildLangmuir law for parallel electrodes. The screen and accel grids for these thrusters have identical aperture diameters; however, the XIPS screen grid is thinner, and the screen-accel grid gap is larger than the NSTAR thruster grid set. The thinner XIPS screen grid tends to increase the ion transparency while the larger XIPS screen-accel spacing tends to decrease the ion transparency. Since these effects nearly offset, the performance of the grid sets is comparable, although NSTAR has a slightly higher ion transparency. We compare the XIPS and NSTAR at the TH15 maximum throttle condition for NSTAR operation as indicated in Table I. Since the diameter of the active grid area of XIPS is 4 cm less than NSTAR, we used an upstream ion density for XIPS that is 18% greater than that used for NSTAR in these calculations to simulate equivalent thrust levels. A XIPS thruster was successfully tested at the TH15 throttle condition [10]. The results from Fig. 2 show that the erosion on the downstream face is much greater for the two-grid geometry, while only a small amount of downstream erosion occurs at the periphery of the accel grid aperture for the three-grid geometry. The corresponding downstream erosion rates for a single NSTAR and XIPS accel grid aperture are 0.826 and 0.0173 mg/khr, respectively. To investigate the reason for this large difference in erosion rate between the two- and threegrid geometeries, we must determine where the CEX ions that

2124

IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 36, NO. 5, OCTOBER 2008

Fig. 2. Comparison of erosion of the downstream face of the accel grid for the NSTAR (two grid) and XIPS (three grid) geometries at TH15 as calculated by CEX3D. These data show that the decel grid of the three-grid geometry effectively shields the accel grid from erosion of the downstream face.

Fig. 3. Erosion rate (mg/khr/m) of downstream surface of accel grid due to CEX ions with respect to the position in the beamlet at which the CEX ions are generated; (above) XIPS three grid and (below) NSTAR two grid. The discharge sources are at the left-hand end of these gures. Axial domain used for CEX3D erosion calculations is approximately three times as long as shown in these images. This result shows that the decel grid effectively eliminates erosion of the downstream surface of the accel grid due to CEX ions created downstream of the grid plane.

contribute to downstream erosion are generated in the beam. To do this, we generated contour plots that show the magnitudes of downstream erosion caused by CEX ions with respect to the location from which they are generated. In Fig. 3, the downstream erosion rate due to CEX ions created in the beam is plotted

as mg/khr/m, where mg is milligrams of molybdenum and m is meter of axial distance. These units avoid overweighing the importance of downstream cells, since the cell sizes grow larger axially further downstream where CEX3D requires less resolution. From the erosion rates in Fig. 3, we see that the decel

WIRZ et al.: DECEL GRID EFFECTS ON ION THRUSTER GRID EROSION

2125

Fig. 4. Potential () proles for (above) XIPS three grid and (below) NSTAR two grid for throttle point TH15. Potentials for the two-grid geometry lead to high-energy impact (200 eV) of CEX ions on downstream accel grid surface, while the potentials created by the decel grid lead to relatively small impact energies (25 V) and, hence, minimal erosion. TABLE II COMPARISON OF ACCEL GRID EROSION RATES (mg/khr) AT BOL

grid effectively eliminates almost all of the erosion due to CEX ions created downstream of the grid plane. The CEX ions generated in the beamlet originally have very low velocity since they are made from the relatively slow moving neutrals. The potentials in Fig. 4 show that the CEX ions generated in the downstream region of the beamlet are inuenced by the local electric eld to move away from the beamlet axis. Therefore, for a two-grid geometry, a large fraction of the CEX ions created downstream are guided to the pits & grooves region between the apertures and accelerated through a large electric eld to energies (over 200 eV) that will cause relatively signicant erosion on the upstream face of the accel grid as shown in Figs. 1 and 2. For the three-grid geometry, most of these CEX ions are accelerated through smaller electric elds due to the much higher potential of the decel grid (0 V), resulting in relatively low impact energies (25 eV) incident to the decel grid. As discussed in [11], the sputter yield for molybdenum is three orders of magnitude smaller at these lower energies (in comparison to energies 200 eV) and is therefore insufcient to cause comparably noticeable erosion. This minimal erosion of the decel grid agrees with observations from XIPS long-duration tests [5]. The erosion rates for the TH15 condition (as well as other conditions, discussed in the next section) are summarized in Table II. From these values, we see that the barrel erosion is similar for both geometries, although slightly lower for XIPS. The erosion of the upstream face of the accel grid is much smaller for the NSTAR grid. The lower upstream erosion for the NSTAR geometry is due to the smaller spacing of the screen and accel grids, which provides a smaller region for barrel

erosion CEX ions, as discussed in [5]; also, the upstream ion density used for XIPS is 18% higher. B. Effect of Decel Grid on Accel Grid Erosion for the XIPS Three-Grid Conguration In this section, we employ results from JPLs CEX3D code to understand how the decel grid affects accel grid erosion by comparing results for identical accel and screen grids with and without a decel grid, assuming XIPS grid geometries and potentials. Fig. 5 shows the comparison of the pits & grooves and bridge erosion for the XIPS three-grid with a hypothetical XIPS two-grid conguration (assuming the decel grid is removed) for the XIPS high-power throttle point. As discussed in the previous section, these results show that the decel grid effectively shields the downstream face of the accel grid from erosion except for a small amount of erosion just around the periphery of the accel grid aperture. Referring to Table II, the decel grid reduces the downstream accel grid erosion over an order of magnitude for both the high- and low-power operating conditions. Referring to Figs. 6 and 7, we see that the electric elds are such that the majority of the CEX ions generated downstream are guided away from the center of the beamlet and back toward the region between the apertures. As discussed in the previous section, for the two-grid geometries the potential elds result in high ion impact energy on the pits and grooves region of the accel grid, while the three-grid geometries create a potential eld that causes these same CEX ions to impact the decel grid at much lower energies, resulting in signicantly less sputtering.

2126

IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 36, NO. 5, OCTOBER 2008

eld (Fig. 7) is such that a larger fraction of the downstream CEX ions are generated in a potential eld that will guide them back to the accel grid hole wall. Upstream erosion of the accel grid is slightly higher for the two-grid geometry, 0.225 mg/khr compared to 0.213 mg/khr for the three-grid geometry. At the low-power point, we see (from Table II) that the accel barrel erosion is 11% higher, and the upstream accel erosion is 27% higher for the three-grid conguration; however, the total erosion rate for the accel grid two-grid system is several times higher due to the much larger downstream erosion discussed earlier. The screen grid experiences minimal erosion for both congurations due to low incident CEX ion energies, which is conrmed by experimental observations [1]. C. XIPS Accel Voltage Sensitivity From the above results, we see that the barrel erosion rate for the XIPS low-power condition (0.104 mg/khr) is noticeably higher than the rate for XIPS at TH15 (0.070 mg/khr). Since the two conditions use similar upstream plasma conditions, it is apparent that the lower voltage of the XIPS accel grid for the low-power case (300 V compared to 180 V for TH15) is likely the primary reason for the increased erosion rates, since it will result in increased impact energy of the CEX ions. The 300-V accel grid voltage used by XIPS is conservative since it is much lower than the electron backstreaming limit for the XIPS grid system [10]. Therefore, we can examine the change in erosion rate at higher accel voltages. To analyze the effect of accel voltage on barrel erosion rate for the XIPS thruster, we ran the model for accel voltages 300, 240, and 180 V. The results from this analysis, shown in Fig. 8, suggest that higher accel voltages will noticeably decrease barrel erosion for the XIPS accel grid. However, for the high-power case, an accel voltage of 180 V increases the upstream erosion due to direct impingement caused by under focusing of the beam. From these results, we see that the optimal accel voltage from an erosion standpoint is higher than the standard 300 V XIPS operating condition; however, one must consider electron backstreaming and changes in thruster geometry due to erosion to determine the most desirable accel grid voltage for long-term mission performance. III. D ISCUSSIONS AND C ONCLUSION Analysis of the XIPS grids with JPLs CEX3D grid erosion model shows that, for BOL conditions, the decel grid signicantly reduces the total erosion rate of the accel by effectively eliminating the pits and grooves erosion of the downstream accel surface commonly observed for two-grid systems. For all operating conditions and geometries examined herein, the decel grid reduced the downstream erosion of the accel grid by over an order of magnitude. Examining the results, it is apparent that most of the CEX ions created in the downstream region of the beamlet are accelerated away from beamlet axis and toward the pits and grooves region of the accel grid. For a two-grid geometry, these CEX ions are accelerated through a potential of at least 200 V, due to the low potential of the accel grid (300 to 180 V), before impacting the upstream surface of

Fig. 5. Comparison of downstream erosion of the accel grid for XIPS twogrid and three-grid geometry at high power. The ordinate axis for the pits and grooves erosion is an order of magnitude greater than the bridge erosion plot (see Fig. 2 for orientation of plots Location axes).

Referring to Table II, results for the XIPS low- and highpower conditions show that the barrel and upstream surface erosion rates are much less at low power. Reduced erosion for the low-power condition for these surfaces is due in part to lower beamlet current as well as improved beam focusing. Improved beamlet focusing (also known as operating closer to the optimal perveance fraction [5]) results in lower hole wall erosion since, upstream of the accel grid, only the CEX ions created near the edge of the beamlet contribute to barrel erosion, while the CEX ions created near the center of the beamlet tend to get accelerated into the beam [5]. The erosion results in Table II suggest that the improved focusing realized at the lowpower condition also reduces the upstream erosion rate. For the high-power condition, the barrel erosion for the twogrid geometry is about 33% greater than the three-grid geometry (0.750 and 0.562 mg/khr, respectively). This difference appears largely due to the near-axis electric eld just downstream of the accel grid; for the two-grid geometry, the potential

WIRZ et al.: DECEL GRID EFFECTS ON ION THRUSTER GRID EROSION

2127

Fig. 6.

Potential () proles for XIPS two- and three-grid geometries at high power.

Fig. 7.

Potential () proles for XIPS two- and three-grid geometries at low power.

Fig. 8. Barrel erosion versus accel grid voltage for XIPS grid set. Lower accel grid voltage reduces barrel erosion but can increase erosion at high-power condition due to the onset of direct impingement on the upstream accel grid surface.

the accel grid at energies sufcient to cause the signicant pits and grooves erosion seen in many experiments. For a three-grid geometry with the decel grid at only 0 V, the majority of these pits and grooves ions see only a 25 V potential drop, thus attaining energies that lead to sputter yields that are three orders of magnitude less than yields for the 200 V ions for the twogrid case. Therefore, the decel grid protects the accel grid from pits and grooves erosion while sustaining minimal erosion of its own upstream surface. These observations are schematically summarized in Fig. 9 using approximate trajectories inferred from the information in Figs. 3, 4, 6, and 7 for CEX ions born at

relatively slow velocities. Per the recommendation of a reviewer of this paper, in future efforts, we will use the model to show simulated CEX ion trajectories. JPLs CEX3D accurately predicts the erosion patterns for the accel grid, although it appears to overpredict the pits and grooves erosion rates due mainly to uncertainties in incident energies and angles for sputtering ions and lack of knowledge of redeposition of sputtered material [2]. Since the model accurately simulates the erosion pattern but overpredicts the erosion rates for both the two- and three-grid sets, this comparative analysis is sufcient to show how the decel grid signicantly

2128

IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 36, NO. 5, OCTOBER 2008

Fig. 9. CEX ions created in the downstream region of the beamlet attain signicant impact energies in the two-grid conguration, leading to the pits and grooves erosion seen during NSTAR testing. In a three-grid geometry, the decel grid effectively protects the accel grid while sustaining minimal erosion due to much smaller impact energies. Trajectories shown are approximate.

suppresses the erosion of the downstream surface of the accel grid as observed in experimental tests [10], [12]. Erosion rates for the aperture wall (barrel) were similar for the cases examined; the most signicant change was seen when the decel grid was removed from the XIPS thruster at high power, resulting in a 33% increase in barrel erosion for the hypothetical two-grid XIPS geometry. As expected, decreasing the accel grid voltage of the XIPS reduces barrel erosion of the accel grid. For BOL, the results show that an accel voltage as low as 180 V will decrease erosion of the accel grid for the low-power case. From an erosion standpoint, an optimal voltage for the XIPS high-power condition lies between 300 and 180 V. The signicantly higher erosion for the XIPS high-power case is due to the fact that the beamlet is far from the optimal perveance fraction. Essentially, at low power and TH15 the beamlets are much better focused (closer to the optimal perveance fraction compared with the high-power case) and a larger fraction of CEX ions are created near the beamlet axis upstream of the grid plane; from the near-axis location the CEX ions are more likely to be accelerated into the beam instead of into the grids where they can cause erosion [12]. For the high-power case, the beamlet occupies more of the aperture region away from the beamlet axis; CEX ions created on the periphery of these large beamlets are likely to cause erosion of the accel grid.

In future analyses, we will use additional experiments and computational modeling to determine the optimal accel grid voltage for long term thruster life and performance to meet NASA mission needs. These efforts will include consideration of the electron backstreaming, double ion effects, detailed erosion estimates for all grid surfaces, and the inuence of the temporal morphology of the grids on long-term erosion and electron backstreaming limits. Future efforts will also include detailed comparison with XIPS grid erosion once this information is available. We will also improve the codes by including redeposition of the sputtered grid material and angular dependence of sputtering rates. R EFERENCES
[1] J. E. Polk, J. R. Anderson, and J. R. Brophy, An overview of the results from an 8200 hour wear test of the NSTAR ion thruster, presented at the 35th Joint Propulsion Conf.Paper AIAA 99-2446, Los Angeles, CA, Jun. 1999, AIAA 99-2446. [2] J. R. Anderson, I. Katz, and D. Goebel, Numerical simulations of two-grid ion optics using a 3D code, presented at the 40th AIAA/ ASME/SAE/ASEE Joint Propulsion Conf. Exhibit, Fort Lauderdale, FL, Jul. 2004, Paper AIAA 2004-3782. [3] V. Rawlin and C. Hawkins, Increased capabilities of the 30-cm diameter Hg Ion thruster, in Proc. Conf. Adv. Technol. Future Space Syst., Hampton, VA, May 1979. [4] R. Brophy, L. Pless, and C. Garner, Ion engine endurance testing at high background pressures, presented at the AIAA/SAE/ASME/ASEE 28th Joint Propulsion Conf. Exhibit, Nashville, TN, Jul. 1992, Paper AIAA1992-3205.

WIRZ et al.: DECEL GRID EFFECTS ON ION THRUSTER GRID EROSION

2129

[5] J. R. Brophy, I. Katz, J. Polk, and J. R. Anderson, Numerical simulations of ion thruster accelerator grid erosion, presented at the 38th AIAA/ASME/SAE/ASEE Joint Propulsion Conf. Exhibit, Indianapolis, IN, Jul. 2002, Paper AIAA 2002-4261. [6] J. Wang, J. Polk, J. Brophy, and I. Katz, Three-dimensional particle simulations of ion-optics plasma ow and grid erosion, J. Propuls. Power, vol. 19, no. 6, pp. 11921199, 2003. 0748-4658. [7] R. Wirz, Discharge plasma processes of ring-cusp ion thrusters, Ph.D. dissertation, Aeronautics, Caltech, Pasadena, CA, 2005. [Online]. Available: http://etd.caltech.edu/etd/available/etd-05232005-162628/ [8] R. Wirz and I. Katz, Plasma processes of DC ion thruster discharge chambers, presented at the 41st AIAA/ASME/SAE/ASEE Joint Propulsion Conf., Tucson, AZ, Jul. 2005, Paper AIAA-2005-3690. [9] E. M. Diaz and G. C. Soulas, Grid Gap Measurement for an NSTAR Ion Thruster. IEPC-2005-244. [10] W. G. Tighe et al., Performance evaluation of the XIPS 25-cm thruster for application to NASA discovery missions, presented at the 42nd AIAA/ASME/SAE/ASEE Joint Propulsion Conf. Exhibit, Sacramento, CA, Jul. 2006, Paper AIAA 2006-4666. [11] R. P. Doerner, D. G. Whyte, and D. M. Goebel, Sputtering yield measurements during low energy xenon plasma bombardment, J. Appl. Phys., vol. 93, no. 9, pp. 58165823, May 1, 2003. [12] J. R. Beattie, J. N. Matossian, and R. R. Robson, Status of xenon ion propulsion technology, J. Propuls. Power, vol. 6, no. 2, pp. 145150, Mar./Apr. 1990.

John R. Anderson, photograph and biography not available at the time of the publication.

Richard E. Wirz received the B.S. degree in aerospace engineering and the B.S. degree in ocean engineering from Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, in 1992 and 1993, respectively, and the M.S. and Ph.D. degrees in aeronautics and applied sciences from California Institute of Technology (Caltech), Pasadena, in 2001 and 2005, respectively. He is a Senior Engineer with the Electric Propulsion Group, Jet Propulsion Laboratory, Pasadena, CA, specializing in modeling, experimental testing, and mission integration of electric thrusters. He designed and developed the worlds rst noble gas miniature ion thruster and is a recognized expert in miniature and precision plasma thruster development, plasma modeling for electric thrusters, and advanced propulsion concepts. He has three patents pending in this area and has authored over 30 publications. Previously, he was the Manager of Renewable Energy Technologies at Gibbs & Cox, Inc. and Technical Lead at SeaSun Power Systems, Inc., both in Alexandria, VA, where he developed alternative energy technologies.

Dan M. Goebel (M93SM96F99) received the B.S. degree in physics, the M.S. degree in electrical engineering, and the Ph.D. degree in applied plasma physics from the University of California, Los Angeles, in 1977, 1978, and 1981, respectively. He is a Senior Research Scientist with the Jet Propulsion Laboratory, Pasadena, CA, where he is responsible for the development of high-efciency ion thrusters, advanced long-life components such as cathodes and grids, and thruster life model validation for deep space missions. Previously, he was a Research Scientist with Hughes Research Laboratories (HRL), Malibu, CA, and Principal Scientist with Boeing Electron Dynamic Devices, Inc., Torrance, CA, where he was the Supervisor of the Advanced Technology Group for microwave tube development and the Lead Scientist of the Xenon Ion Propulsion System ion thruster program for commercial satellite station keeping. He is a recognized expert in advanced plasma and ion sources, microwave sources, high-voltage engineering and pulsed-power switches. He is the author of over 100 technical papers, one book entitled Fundamentals of Electric Propulsion: Ion and Hall Thrusters to be published this year, and is the holder of 42 patents. Dr. Goebel is the Chair of the American Institute of Aeronautics and Astronautics Electric Propulsion Technical Committee, Chair of the IEEE Electron Devices Society (EDS) Technical Committee on Vacuum Devices, member of the IEEE EDS Publications committee, and Life Member of the American Physical Society and Sigma Xi. In 2004, he received the William Dunbar High Voltage Research Achievement Award.

Ira Katz received the B.S. degree in physical chemistry from Case Institute of Technology, Columbus, OH, in 1967, and the Ph.D. degree in chemical physics from the University of Chicago, Chicago, IL, in 1971. He is the Supervisor of the Electric Propulsion Group; Jet Propulsion Laboratory (JPL), Pasadena, CA, which is responsible for electric propulsion systems for JPLs space science missions. Previously, he was a Senior Vice President of Maxwell Technologies S-Cubed Division, where he led investigations in spacecraftplasma interactions and electric propulsion generated plasmas. He is a recognized leader in computer models of ion thruster physics and spacecraft charging, and has authored over 70 peer-reviewed articles and almost 100 conference publications.