Investigation of a Microwave Plasma for a Novel Ion Beam Thruster

Investigation of a Microwave Plasma for a Novel Ion Beam Thruster IEPC-2024-108 Presented at the 38th International Electric Propulsion Conference, Toulouse, France June 23-28, 2024 Alexander Spethmann∗, Thomas Trottenberg†and Holger Kersten‡ Institute of Experimental and Applied Physics, University of Kiel, Germany This article presents ongoing research on a novel microwave plasma thruster, which aims to generate its radio-frequency power using semiconductor technology. An unconventional magnetic field geometry is utilized to enable electron cyclotron resonance. The magnetic field is provided using longitudinally magnetized ring-shaped permanent magnets. A coaxial electrode arrangement inside the magnets creates the electric field perpendicular to the magnetic field. It is shown that promising ion currents can be extracted from the plasma. A method for ion acceleration is proposed that uses a tubular ring electrode. Nomenclature P= electric microwave power Q= gas flow B= magnetic flux density Icp = ion current onto the collector plate Ucp = potential at the collector plate Isat = ion saturation current (Faraday cup) A= area of the aperture of the Faraday cup Iloss = ion loss current to the tubular ring Uacc = acceleration voltage at the tubular ring j= ion current density ωce = electron cyclotron frequency e= elementary charge me= electron mass ∗Research Associate, Group Plasma Technology, [email protected]. †Research and Teaching Associate, Group Plasma Technology, [email protected]. ‡Professor, Head of Group Plasma Technology, [email protected] 1 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved. I. Introduction Microwave technology has rapidly developed in recent decades. While klystrons, magnetrons, and traveling wave tubes, along with devices with lower efficiencies like Gunn diodes, were predominantly used in the past to generate microwaves, semiconductor technology is increasingly becoming a viable alternative. For instance, the Japanese Hayabusa mission1,2 successfully utilized traveling wave tubes to generate microwaves for its gridded ion engines. Traveling wave tubes attain high efficiencies, but their design is complex, which implies higher costs. For industrial applications of gridded ion sources, magnetrons are often used to generate a microwave plasma.3Magnetrons, like those used in microwave ovens, have a very simple design, are robust, and provide a high power output. However, the required cooling, which is easily accomplished in an industrial setup, might be a problem for space applications. Today, small transmitters and receivers are suitable for the mass market in telecommunications. The rapid development in the field of mobile communications has significantly contributed to advancements in microwave semiconductor technology, particularly with respect to miniaturization and cost reduction. It is therefore reasonable to use microwaves for plasma generation with the help of modern semiconductor technology, similar to how a frequency generator in the megahertz range is required for a Radio Frequency Ion Thruster (RIT).4 This article reports on our ongoing efforts to develop a new type of microwave thruster based on semiconductor technology. We aim for a division similar to that in gridded thrusters, meaning that microwaves should, firstly, generate a plasma from which the ions are then, secondly, extracted and accelerated using an additional acceleration voltage. Thus, a higher complexity (generator for plasma generation and power supply for acceleration) is accepted compared to direct current thrusters such as Hall thrusters5and HEMPT.6,7 It is likely that a neutralizer will also be required. With our approach, we do not intend to make acceleration voltage power supplies and the neutralizer dispensable using a magnetic nozzle, as is currently being attempted elsewhere.8,9 We describe the experimental thruster setup for plasma generation and ion extraction and present first measurement results. Microwave power, gas flow, and pressure are varied. Beam diagnostics are performed using a Faraday cup and current measurements at a collector plate. II. Development of a plasma source for the thruster Plasma generation using microwaves can be particularly efficient, even at lower gas densities, when magnetic fields, especially those enabling electron cyclotron resonance (ECR), are utilized. The magnetic field should ideally be perpendicular to the oscillating electric field that heats the plasma via the electrons. The desired orientation of the fields can be accomplished in a cylindrical geometry by a longitudinal, homogeneous magnetic field and a radial electric field.10,11 A longitudinal magnetic field can be generated using various methods. For instance, a long coil would produce such a field within its interior. We opted for a solution utilizing permanent ferrite magnets. The magnets are four stacked rings with magnetization in longitudinal direction, see Fig. 1. The finite element calculation shown in Fig. 2 indicates that the electric field strength Eis quite homogeneous in the inner near the midplane. Since the magnetization of the ferrite is not precisely known, we measured the flux density at several externally accessible points and then adjusted the magnetization to match these measurements. Accordingly, the magnetization should be 300 kA/m. Based on this calculation, the magnetic field strength in the center is somewhat higher than B= 87 mT when four rings are stacked. At this magnetic field strength, the electron resonance occurs at a frequency of ωce =e me B , (1) which is ωce/2π= 2.44 GHz, where emeans the elementary charge and methe mass of an electron. The stack of four magnets can be seen in the center of the photograph. The radial electric field is generated by coaxial electrodes. The electrode carrying the RF voltage is the inner conductor coming from the microwave generator, which extends a short distance out of the outer conductor. The counter electrode is a grounded cylinder that encloses the discharge volume. In this manner, the electric field lines run radially from the inner to the outer electrode, i.e. perpendicular to the magnetic field lines. A ceramic covers the protruding part of the inner conductor of the waveguide, see Fig. 1. In the next steps of the development, a ceramic will also cover the outer conductor and the interior of the grounded 2 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved. counter electrode in order to insulate them completely from the plasma. This will be necessary, similar to a gridded ion thruster, to raise the potential of the plasma using an anode for the purpose of ion acceleration. However, for the tests we are reporting on here, we have not yet implemented the galvanic insulation of the microwave waveguide. At this stage of development, the ceramic serves primarily to channel the gas to the right location. The ceramic has fine holes, indicated by the interruptions in the yellow line, allowing the gas to escape. This design aims to achieve the highest gas density where the electric field strength is greatest, which is crucial for an efficient use of the propellant. The challenge is to achieve sufficient gas densities without inhibiting the ECR process. When operating with microwaves in the range from 2.4 to 2.5 GHz, electron cyclotron resonance can occur in the inner cylindrical region of the magnetic stack where the gas is released. (a) (b) Figure 1. Plasma source. (a) Three-dimensional view along the axis into the interior of the permanent magnet stack. (b) Longitudinal section, highlighting the ceramic (yellow), through which the gas is directed to the location where the ECR condition is fulfilled. The magnetic field created by the stack of ring magnets has another advantage. The numerical field simulation shown in Fig. 2 reveals that near the end planes of the stack, the field lines acquire an increasing radial component and even reverse direction. The field lines coming from the interior of the stack end on the end faces of the magnets, primarily at smaller radii. The field lines therefore describe an arc outside the stack and have a strong radial component, which efficiently impedes the movement of electrons in the axial direction. In addition, the flux density increases significantly in front of the end faces of the magnets due to converging field lines. This creates a magnetic cusp, and due to the magnetic mirror effect it further weakens plasma recombination at the surface where the field lines end.12 The outwardly curved, returning, and focused field lines thus act like a screen grid in gridded thrusters: they confine the electrons and enable the application of an accelerating field that would otherwise be screened by the plasma. The magnetic field predominantly affects the electrons due to their much smaller mass and gyration radii. In contrast, ions have gyration radii that are large compared to the system length, and therefore ions are not confined by the magnetic field. Figure 3(a) shows a photograph of the demonstration model of the plasma source and Fig. 3(b) shows the plasma source in operation. 3 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved. Figure 2. Finite element simulation of the magnetic field in a longitudinal section through the permanent magnet stack in cylindrical coordinates. (a) Contour lines for 30, 50, 75, and 87 mT. (b) Magnetic field lines, displaying only lines from the interior of the magnet stack. The calculation is based on a magnetization of 300 kA/m for the ferrite. Figure 3. (a) Plasma source with the stack of four magnets. (b) Plasma leaving the magnet stack at an Ar flow of 5 sccm and a power of 5 W. III. Test of the plasma source Figure 4. Schematic of the experimental setup with collector plate. 4 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved. Figure 5. Ion extraction from the plasma source. Exemplary current-voltage characteristics at gas flows (a) Q= 1 sccm and (b) Q= 3 sccm for microwave powers P= (5 −50) W. Crucial for the utilization of the microwave plasma source in a future thruster is the question of how much ion current can actually be extracted from the plasma. For this purpose, a biasable collector plate was placed at a short distance behind the opening of the plasma source. The square surface area measures 10×10 cm2, significantly larger than the opening, which has a diameter (also the inner diameter of the magnets) of 32 mm, see the sketch in Fig. 4. This setup ensures that, with an attractive potential, the measured current should at least provide a lower limit for the extractable current. The vacuum chamber was pumped down to a base pressure below 10−3Pa. During the operation of the discharge, the chamber pressure was several 10−2Pa, depending on the chosen gas flow rate. It was possible to maintain a discharge at a very small gas flow of Q= 1 sccm. Certainly, the collector plate contributed to achieving the required gas density in the plasma source through reflections of gas atoms on its surface. Then, the plasma source was operated at various combinations of gas flows Qand electrical microwave powers P. We tried to operate the source for all combinations of Q= (1,2,3,4,5) sccm and P= (5,10,20,30,40,50) W. Figure 5(a) shows the achieved currents at Q= 1 sccm for the different powers Pand voltages Ucp applied to the collector plate. The negative voltages were in the range 0> Ucp >−200 V, and currents Icp in the order of magnitude of mA could be measured. As the potential becomes increasingly attractive, the currents initially increase more and then less and less, so that saturation can be expected somewhere above the applied voltages. Saturation at higher (more negative) extraction voltages indeed occurs when the flow is increased to Q= 3 sccm, see Fig. 5(b). This observation suggests that in this saturation regime nearly all ions usable for propulsion are extracted, with the exception of an unavoidable portion of ions lost due to recombination at the surfaces in the plasma source. These measurements at small gas flows show that a significant current of ions can be extracted from a stable microwave discharge. It is promising that we can achieve currents in the milliampere range with low gas flow and low power. IV. Plasma source supplemented by an ion acceleration chamber After the plasma generation stage has proven to be a suitable source for the ions to be accelerated, an ion acceleration stage is now to be added. For this purpose, a dielectric funnel-shaped chamber has been constructed, which is attached to the previously described assembly, see Fig. 6. This chamber is intended to mount the accelerating electrode and to confine the unionized gas escaping from the plasma generation zone, in order to at least partially return it to the ionization zone (see Sec. III). The walls of this acceleration stage in our experimental setup are made of glass to allow visual observation from the outside. At the rear end of the conical glass body, acceleration electrodes of various geometries can be attached. The main requirement for an acceleration electrode should be high transparency. We made tests with 5 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved. Figure 6. (a) Plasma source with connected ion acceleration chamber. (b) Observation of the plasma flowing through the glass acceleration chamber and exiting at the rear. different grids. The tested grids included one that consisted of a few crossed wires, but also plates with many holes drilled into them, similar to a gridded ion engine. V. Tubular ring as an acceleration electrode The most promising geometry so far was a tubular ring with an outer diameter of 80 mm, an inner diameter of 78 mm, and a thickness of 4 mm, see Fig. 7. It is mounted concentric to the glass cone in its exit plane. The transparency for ions can be qualitatively assessed by measuring the current density behind the exit plane with a Faraday cup and simultaneously measuring the absolute current losses to the electrode itself. The losses should be minimized, while the current density of the extracted ions is to be maximized. The following plots show the current density measured with the Faraday cup for increasingly negative collector voltages (collector of the Faraday cup). Instead of measuring at a fixed collector voltage, we preferred to record the entire characteristic to be sure that all electrons coming from the plasma are repelled. Finally, the ion saturation current can be used to calculate the ion flux density j=Isat/A using the area A= 25.5×10−6m2of the aperture of the Faraday cup. The plots show two cases of particular importance. Fig. 8(a) was taken at the lowest possible argon flow at which the plasma could be operated, while Fig. 8(b) represents the operating conditions for the highest achievable ion current (higher acceleration voltages caused sparking). The loss currents flowing onto the tubular ring were Iloss = 0.61 mA and Iloss = 0.88 mA, respectively. The voltages at the tubular ring (extraction electrode) were Uacc =−260 V and Uacc =−270 V, respectively. Table 1 shows the current densities obtained from the saturation currents and the acceleration voltages for the two cases shown in the above figures and the measurements at all other operating conditions as well for which measurements were made. 6 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved. Figure 7. (a) Tubular ring used as acceleration electrode attached to the acceleration chamber. The cylindrical device next to the acceleration assembly is the Faraday cup. (b) Light emission from the plasma during operation. Figure 8. Measurements conducted on the setup with tubular ring acceleration electrode. Faraday cup currentvoltage characteristics (a) at the lowest possible Ar flow of 2 sccm and 10 W microwave power, (b) at the highest possible current of 80 µA at an Ar flow of 5 sccm and 20 W microwave power. The circular aperture of the Faraday cup has a diameter of 5.7mm. 1 sccm 2 sccm 3 sccm 4 sccm 5 sccm 10 sccm 5 W no ignition Uacc: 270 V j: 0.47 A/m2 Uacc: 265 V j: 0.86 A/m2 Uacc: 265 V j: 0.82 A/m2 Uacc: 265 V j: 0.74 A/m2 Uacc: 255 V j: 0.82 A/m2 10 W no ignition Uacc: 260 V j: 0.78 A/m2 Uacc: 265 V j: 1.02 A/m2 Uacc: 265 V j: 1.14 A/m2 Uacc: 265 V j: 1.14 A/m2 Uacc: 260 V j: n.a. 20 W no ignition Uacc: 265 V j: 0.94 A/m2 Uacc: 265 V j: 1.61 A/m2 Uacc: 265 V j: 2.16 A/m2 Uacc: 270 V j: 3.14 A/m2 Uacc: 275 V j: 2.04 A/m2 50 W briefly Table 1. Table of ion currents densities jmeasured by the Faraday cup for different power levels Pand flow rates Q. The Uacc values indicate the applied voltage at the tubular ring. 7 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved. VI. Conclusion and next steps A novel type of plasma thruster has been proposed and first laboratory experiments have been carried out. The concept uses coaxial electrodes and a magnetic field aiming at electron cyclotron resonance for microwave frequencies. The electrical power shall be in the range of semiconductor microwave generators that can be operated in space. The magnetic field is generated by permanent magnets in the form of longitudinally magnetized ferrite rings. The work has shown that a suitable plasma can be generated in a space-like environment with the proposed plasma source. Considerable ion currents were extracted from the plasma. The plasma source was complemented by an acceleration chamber with an acceleration electrode. A promising geometry for the acceleration electrode was identified. Measurements of the extracted ion current densities were presented. Our next steps involve the galvanic separation of the microwave electrodes from the plasma generation chamber and the introduction of an anchor electrode in the plasma generation chamber. The currently negatively biased acceleration electrode will then be grounded. Furthermore, we plan to extend our tests to higher acceleration voltages. Our diagnostic techniques will be enhanced with ion energy measurements utilizing a retarding potential analyzer, and thrust measurements with a force probe.13,14 Acknowledgments This work is supported by the German Aerospace Center DLR, Project No. 50 RS 2204. References 1H. Kuninaka, K. Nishiyama, Y. Shimizu, I. Funaki, H. Koizumi, S. Hosoda, and D. Nakata, “Hayabusa Asteroid Explorer Powered by Ion Engines on the way to Earth” in 31st International Electric Propulsion Conference, Ann-Arbor, Michigan, USA, IEPC–2009–267, 2009. 2K. Nishiyama, S. Hosoda, K. Ueno, R. Tsukizaki, and H. Kuninaka, “Development and Testing of the Hayabusa2 Ion Engine System” in 34th International Electric Propulsion Conference, Kobe, Japan, IEPC–2015–333, 2015. 3M. Zeuner, F. Scholze, H. Neumann, T. Chassé, G. Otto, D. Roth, A. Hellmich, and B. Ocker, “A unique ECR broad beam source for thin film processing”, Surf. Coat. Technol., vol. 142–144, pp. 11–20, 2001. 4K. Holste, P. Dietz, S. Scharmann, K. Keil, T. Henning, D. Zschätzsch, M. Reitemeyer, B. Nauschütt, F. Kiefer, F. Kunze, J. Zorn, C. Heiliger, N. Joshi, U. Probst, R. Thüringer, C. Volkmar, D. Packan, S. Peterschmitt, K.-T. Brinkmann, H.-G. Zaunick, M. H. Thoma, M. Kretschmer, H. J. Leiter, S. Schippers, K. Hannemann, and P. J. Klar, “Ion thrusters for electric propulsion: Scientific issues developing a niche technology into a game changer”, Rev. Sci. Instrum., vol. 91, p. 061101, 2020. 5J.-P. Boeuf, “Tutorial: Physics and modeling of Hall thrusters”, J. Appl. Phys., vol. 121, p. 011101, 2017. 6G. Kornfeld, N. Koch, and H.-P. Harmann, “Physics and Evolution of HEMP-Thrusters”, IEPC–2007–108, 2007. 7T. Trottenberg, F. Bansemer, S. Böttcher, D. Feili, H. Henkel, M. Hesse, H. Kersten, T. Krüger, J. Laube, A. Lazurenko, D. Sailer, B. Schuster, L. Seimetz, A. Spethmann, S. Weis, and R. F. Wimmer-Schweingruber, “An in-flight plasma diagnostic package for spacecraft with electric propulsion”, EPJ Techn. Instrum., vol. 8, p. 16, 2021. 8A. E. Vinci, S. Mazouffre, V. Gómez, P. Fajardo, and J. Navarro-Cavallé, “Laser-induced fluorescence spectroscopy on xenon atoms and ions in the magnetic nozzle of a helicon plasma thruster”, Plasma Sources Sci. Technol., vol. 31, p. 095007, 2022. 9Y.-A. Chan and G. Herdrich, “Experimental Verification of an Electromagnetic Nozzle on Plasma Confinement and Acceleration” in 37th International Electric Propulsion Conference, Cambridge, Massachusetts, USA, IEPC–2022–428, 2022. 10A. Spethmann, T. Trottenberg, and H. Kersten, “Mikrowellen-Zyklotron-Resonanz-Plasma-Triebwerk und zugehöriges Betriebsverfahren sowie Verwendung”, Patent DE102022112292B3, 2023. 11A. Spethmann, T. Trottenberg, and H. Kersten, “Microwave-Cyclotron-Resonance Plasma Thruster and associated Operating Method and Use”, International Patent Application WO002023222155A1, 2023. 12M. G. Haines, “Plasma containment in cusp-shaped magnetic fields”, Nucl. Fusion, vol. 17, pp. 811–858, 1977. 13T. Trottenberg, A. Spethmann, and H. Kersten, “An interferometric force probe for beam diagnostics and the study of sputtering”, EPJ Techn. Instrum., vol. 5, p. 3, 2018. 14D. Bock, A. Spethmann, T. Trottenberg, H. Kersten, and M. Tajmar, “In-plume thrust measurement of NanoFEEP thruster with a force measuring probe using laser interferometry” in 35th International Electric Propulsion Conference, Georgia, Atlanta, USA, IEPC–2017–391, 2017. 8 The 38th International Electric Propulsion Conference, P. Baudis Convention Center, Toulouse, France, June 23-28, 2024 Copyright 2024 by the Electric Rocket Propulsion Society. All rights reserved.