Skip to main content
Skip to article control options
Open AccessTechnical Notes

Ion Thruster Operation with Carbon Nanotube Field Emission Cathode

Published Online:

I. Introduction

Ion engines have been used in a number of space missions, including Deep Space 1 [1], Gravity Field and Steady-State Ocean Circulation Explorer (GOCE) [2], Hayabusa [3] and Hayabusa 2 [4], Boeing’s All Electric Satellite [5], and Hodoyoshi-4 [6]. These engines contain electron sources, such as hollow cathodes [79], radiofrequency discharge neutralizers [10], and microwave discharge neutralizers [11], for the neutralization of the extracted ion beam from the ion source. Unfortunately, the electron sources consume propellant gas, and this leads inevitably to degradation of total specific impulse.

Field emission cathodes (FEC) [1215] offer a promising alternative, as they have lower power consumption than hot cathodes and no consumption of propellant, though the current density is small, on the order of several milli-amperes per square centimeter. Research and development on carbon nanotube FECs for space applications has been the subject of considerable efforts in this decade because of their well-balanced performance in terms of electron extraction capabilities, simplicity, and robustness. FECs emit electrons by applying a strong electric field on the emitter, approximately 109  V/m, by the so-called “tunneling effect.” Carbon nanotubes (CNT) are a good emitter material, because the diameter of the tip of the carbon nanotube is nanoscale, and so the electric field is strongly focused on the tip of the carbon nanotube. Applied voltage is therefore relatively low, and this leads to reduced power consumption. In addition, the density of the tube is high, which leads to large emission current. The disadvantage of the carbon nanotube FEC is low tolerance to atomic oxygen (AO), but this could be overcome by coating the surface of the carbon nanotube with AO resistant material, or avoiding direct exposure to AO [16].

The use of FECs with carbon nanotubes in space was demonstrated in The Space Technology 7 Disturbance Reduction System [17] and Kounotori Integrated Tether Experiments (KITE) [18]. In KITE, the FEC produced 2 mA electron emission with a 40×80  mm2 surface area in a space environment (altitude 370 km). This is enough electron emission current for neutralization of a miniature ion thruster [19,20]. To the best of our knowledge, the present study represents the first demonstration of the capabilities of the FEC as neutralizer.

II. Experimental Setup

All of the experiments described here were performed at the Space Science chamber at the Institute of Space and Astronautical Science (ISAS) at the Japan Aerospace Exploration Agency (JAXA). The chamber is 2.5 m in diameter and 5 m in length, and has two cryopumps (each with a pumping speed of 23,000 l/s for argon gas).

An FEC with a CNT emitter was used, as shown in Fig. 1. The size of the cathode is 88×88  mm2, and the maximum emission current achieved is 20 mA at potential difference of 500 V between gate grid and emitter. Single-walled CNT attached to molybdenum substrate was used as the emitter material. The cathode has two grids: a gate grid, to apply high electric field to the carbon nanotube, and a mask grid, which covers the emitter (CNT) to reduce direct electric collision with the gate grid [21]. The open area is rectangular in shape; the open ratio is about 80%. Both electrodes were made of thin molybdenum plate. The gate voltage against emitter potential can be varied from 0 to 500 V. The electron emission current to the outside of the FEC is calculated by subtracting the drain current to the gate electrode from the current emitted from the CNT emitter. These currents were measured using IV circuits (uncertainty of 5%, which based on specification of this circuit).

Fig. 1
Fig. 1

Field emission cathode using carbon nanotube.

A 30 W class microwave discharge ion thruster [22] was used for the ion source. Three flat grids were used to extract the ion beam, the screen grid, acceleration grid, and deceleration grid, with grid parameters optimized using a numerical analysis code developed by Nakano et al. [23]. The grid parameters are shown in Table 1. The screen grid and acceleration grid are made of molybdenum and the deceleration grid is made of pyrolytic carbon. The maximum extracted ion beam current of this ion thruster is 21.4 mA, at a xenon mass flow rate of 0.036 mg/s, beam voltage of 1500 V, and incident microwave power of 16 W [22]. Microwave power at 2.45 GHz was fed through a coaxial line and into a star-shaped antenna [24]. A DC block, which cannot pass a direct current component but will pass microwaves, with a loss of 0.43 dB at 2.45 GHz, was inserted to protect the microwave amplifier. The extracted ion beam current was estimated by subtracting the sum of the current from the acceleration grid and deceleration grid from the current through the screen power supply. The current was measured using a current probe (accuracy of 3% at 1 mA). Pure (99.995%) xenon gas was used as propellant, and at mass flow rate kept constant at 0.010 mg/s using a thermal mass flow controller. Pressure in the vicinity of the thruster and CNT-FEC was measured using an ionization gauge located 210 mm from the ion thruster and located 50 mm behind the FEC. A mesh shield prevents plasma from reaching the gauge. The pressure at 0.010 mg/s xenon mass flow rate is 1×104  Pa (7.5×107  torr).

The FEC with carbon nanotube was located 200 mm from the ion thruster, as shown in Fig. 2. The entire ion engine system was electrically isolated from the vacuum chamber (i.e., ground) in order to evaluate the neutralization of the system as shown in Fig. 3; the cathode is electrically floating in the chamber, which has the same potential as the deceleration grid. For safety, a 47 V varistor is installed between cathode and ground. The current through the varistor was 0.8 mA when 45 V voltage was applied between the two electrodes of the varistor. The potential difference between chamber and cathode was monitored using a differential probe with accuracy of ±2%. The ion thruster head has an electrical shield, whose potential was set as the chamber potential.

Fig. 2
Fig. 2

Setup of ion thruster head and FEC.

Fig. 3
Fig. 3

Electrical circuit for coupling operation.

III. Results

Conventional FECs are usually operated in high vacuum conditions, to avoid damage to the emitter material. For neutralizing the ion beam from the ion thruster, the carbon nanotube FEC must function in the presence of xenon atoms. Emission currents as a function of gate voltage were measured with/without xenon gas feed, as shown in Fig. 4. To form the electrical current loop via the cathode and the vacuum chamber, a potential of 100  V was applied to the FEC emitter relative to the grounded vacuum chamber, with cathode common tied to the grounded chamber and serving as the current return path. (The voltage simulates the operation of the ion thruster.)

Fig. 4
Fig. 4

Emission current from FEC with/without xenon gas flow. The uncertainty of the emission current is 5% and that of gate voltage is 2%.

Xenon gas was supplied from the ion source at a mass flow rate of 0.01 mg/s. The ion gauge pressure in the vicinity of the ion thruster and FEC was 1×104  Pa with xenon gas and 5×105  Pa without xenon gas. The electron emission current under xenon cold flow conditions is almost the same as under no-gas conditions, reaching about 3 mA at a gate voltage of 400 V; the current from the emitter is 5.1 mA and the current into the gate is 1.7 mA under vacuum conditions, and 4.8 and 1.5 mA under cold flow conditions, respectively. That is, emission degradation due to the presence of neutral atoms was not observed.

After demonstrating that the FEC works under cold-flow conditions, we moved on to a demonstration of ion beam neutralization by FEC. Figure 5 shows the emission current from the FEC and the floating voltage of the system. Here, the gate voltage was kept constant at 400 V and the ion beam current was increased by increasing the microwave power of the ion source. Figure 5 indicates that the electron emission current from FEC and the ion beam current were roughly balanced throughout the operation range, and the floating voltage was changed self-consistently corresponding to the ion beam current levels. Because the whole system was electrically isolated from the vacuum chamber as mentioned in the previous section, this consistency means that the ion beam current was neutralized by the FEC.

Fig. 5
Fig. 5

Electron emission current and floating voltage vs ion beam current. The uncertainty of the emission current is 5%, that of ion beam current is 3%, and that of floating voltage is 2%.

Although the emission currents from FEC were matched to within at most 16% of the thruster beam current, a closer examination of the data shows unbalanced operational conditions at some points. For example, the electron emission current was 2.25 mA with ion beam current of 2.49 mA, and electron emission current was 1.0 mA with ion beam current of 0.87 mA. This is probably due to uncertainty (3% and 5%) in the current measuring system, which is based on specifications for the current probe and IV circuit used. Furthermore, ion collision with the thruster shield and/or the chamber wall and the current loop could take place through the varistor and/or differential probe; other unintended connections are also possible. Nevertheless, the floating potential of the cathode was constant at 34  V (leak current is less than 50  μA), which means that the neutralization succeeded. The electron emission cost, defined as total power consumption over extracted electron current, was 360 W/A. This could be reduced by adding a bias power supply; when 45  V was applied between the emitter and the ion thruster base (deceleration grid), the current into gate was decreased to 1.5 mA and the cost was improved to 320 W/A. The total specific impulse of this system is 1.25 times higher than in a conventional ion engine system with our microwave discharge neutralizer [25].

Figure 6 shows the current from the emitter and current into the gate electrode of the FEC versus the ion beam extraction current under the same conditions as Fig. 5. As the ion beam current increases, the emission current from emitter increases slightly and the current into the gate electrode decreases. This result shows that the excess electron current from the FEC goes back to the gate electrode. The gate voltage of 400 V is high enough to draw back the excess electrons. The FEC has a shield, but it has a hole where electrons go through and the diameter of the hole is 7 mm, which is larger than the estimated Debye length, 0.6 mm.

Fig. 6
Fig. 6

Current vs ion beam current. The uncertainty of the current is 5% and that of ion beam current is 3%.

The floating potential varied with the electron emission capability (such as change of gate voltage) and with distance between cathode and ion thruster head, among other factors. Figure 7 shows the floating potential versus gate voltage under constant ion beam current of 1.1–1.2 mA (see Fig. 4). With an increase in gate voltage, the floating voltage was increased from 45   to 20  V, and it saturated at 20  V for gate voltages larger than 380 V. The saturation of the floating voltage would be due to the space charge limitation [20]. The emission current from the FEC was in the range of 1.2–1.3 mA and therefore balanced, and so the neutralization was maintained, except at gate voltage of 360 V. With gate voltage of 360 V, the emission current from FEC was 1.0 mA and the ion beam current was 1.1 mA; the unbalance would be due to the current through the varistor. At this condition, the leak current through the varistor was larger than 0.2 mA (the voltage between the varistor electrodes was 42 V).

Fig. 7
Fig. 7

Floating potential vs gate voltage. The uncertainty of the floating potential is 2% and that of gate voltage is 2%.

IV. Conclusions

The neutralization of a miniature ion thruster using xenon as a propellant with field emission neutralizer was successfully demonstrated using an field emission cathodes (FEC) with carbon nanotube emitter for, to our knowledge, the first time. The emission current from the FEC is balanced with the ion beam current from the ion thruster head, by changing the potential of the neutralizer relative to the chamber (space plasma potential). The electrically isolated potential of the cathode is kept constant and is changed by changing the ion beam current and the gate voltage. The electron emission cost was 360 W/A. This is higher than that of a conventional hollow cathode (less than 30 W/A) [26,27], but considering the current range and the lack of propellant consumption, this is a competitive cost. A remaining challenge is demonstration of the 30 W class miniature ion thruster with CNT-FEC and the life expectancy of the FEC with carbon nanotube.

J. BlandinoAssociate Editor


The results were obtained at the Space Plasma Laboratory of ISAS, JAXA. The work was supported by the 2016-2018 JAXA Strategic Development Program and JSPS KAKENHI Grants JP16H04595 and JP16K14506.


  • [1] Sovey J. S., Rawlin V. K. and Patterson M. J., “Ion Propulsion Development Projects in U.S.: Space Electric Rocket Test I to Deep Space I,” Journal of Propulsion and Power, Vol. 17, No. 3, 2001, pp. 517–526. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [2] Wallace N. C. and Fehringer M., “The ESA GOCE Mission and the T5 Ion Propulsion Assembly,” IEPC Paper 2009–269, Sept. 2009. Google Scholar

  • [3] Kuninaka H., Nishiyama K., Funaki I., Yamada T., Shimizu Y. and Kawaguchi J., “Powered Flight of Electron Cyclotron Resonance Ion Engines on Hayabusa Explorer,” Journal of Propulsion and Power, Vol. 23, No. 3, 2007, pp. 544–551. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [4] Nishiyama K., Hosoda S., Ueno K., Tsukizaki R. and Kuninaka H., “Development and Testing of the Hayabusa2 Ion Engine System,” Transactions of the Japan Society for Aeronautical and Space Sciences, Aerospace Technology Japan, Vol. 14, No. ists30, 2016, pp. Pb_131–Pb_140. doi: CrossrefGoogle Scholar

  • [5] Feuerborn S. A., Neary D.A. and Perkins J.M., “Finding a Way: Boeing’s ‘All Electric Propulsion Satellite’,” AIAA Paper 2013-4126, July 2013. Google Scholar

  • [6] Koizumi H., Komurasaki K., Aoyama J. and Yamaguchi K., “Development and Flight Operation of a Miniature Ion Propulsion System,” Journal of Propulsion and Power, Vol. 34, No. 4, 2018, pp. 960–968. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [7] Goebel D. M. and Chu E., “High-Current Lanthanum Hexaboride Hollow Cathode for High-Power Hall Thrusters,” Journal of Propulsion and Power, Vol. 30, No. 1, 2014, pp. 35–40. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [8] Friedly V. J. and Wilbur p. J., “High Current Hollow Cathode Phenomena,” Journal of Propulsion and Power, Vol. 8, No. 3, 1992, pp. 635–643. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [9] Becatti G., Goebel D. M., Polk J. E. and Guerrero P., “Life Evaluation of a Lanthanum Hexaboride Hollow Cathode for High-Power Hall Thruster,” Journal of Propulsion and Power, Vol. 34, No. 4, 2018, pp. 1–8. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [10] Watanabe H., Takegahara H. and Aoyagi J., “Dependence of Cathode Configuration on Performance in an Inductively Coupled Plasma Cathode,” Journal of Propulsion and Power, Vol. 34, No. 3, 2018, pp. 668–678. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [11] Ohmichi W. and Kuninaka H., “Performance Degradation of a Spacecraft Electron Cyclotron Resonance Neutralizer and Its Mitigation,” Journal of Propulsion and Power, Vol. 30, No. 5, 2014, pp. 1368–1372. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [12] Ohkawa Y., Matsumoto K., Kawamoto S. and Kitamura S., “Performance Improvement of a Carbon Nanotube Field Emission Cathode,” 63rd International Astronautical Congress, IAC Paper 12-C4.4.11, 2012. Google Scholar

  • [13] Gasdaska C. J., Falkos P., Robin M., Hruby V., Demmons N. and McCormick R., “Testing of Carbon Nanotube Field Emission Cathodes,” AIAA Paper 2004-3427, July 2004. LinkGoogle Scholar

  • [14] Makela J. M., Washeleski R. L. and King L. B., “Regenerable Field Emission Cathode for Spacecraft Neutralization,” Journal of Propulsion and Power, Vol. 25, No. 4, 2009, pp. 970–975. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [15] Williams L. T., Walker M. L. R., Kumsomboone V. S. and Jud Ready W., “Experimental Characterization of Carbon Nanotube Field Emission Cathode Lifetime,” AIAA Paper 2009-5003, Aug. 2009. LinkGoogle Scholar

  • [16] Shimada A., Tanaka Y., Ohkawa Y., Matsumoto K., Tagawa M., Matsui M. and Yamagiwa Y., “Effect of Atomic Oxygen Irradiation on Field Emission Cathodes in Low Earth Orbit,” Transactions of the Japan Society for Aeronautical and Space Sciences, Aerospace Technology Japan, Vol. 12, No. ists29, 2014, pp. Pb_59–Pb_64. doi: CrossrefGoogle Scholar

  • [17] Ziemer J., Marrese-Reading C., Dunn C., Romero-Wolf A., Cutler C., Javidnia S., Le T., Li I., Franklin G., Barela P. and et al., “Colloid Microthruster Flight Performance Results from Space Technology 7 Disturbance Reduction System,” IEPC Paper 2017-578, Oct. 2017. Google Scholar

  • [18] Ohkawa Y., Okumura T., Horikawa Y., Miura Y., Kawamoto S. and Inoue K., “Field Emission Cathodes for an Electrodynamic Tether Experiment on the H-II Transfer Vehicle,” Transactions of the Japan Society for Aeronautical and Space Sciences, Aerospace Technology Japan, Vol. 16, No. 1, 2018, pp. 63–68. doi: CrossrefGoogle Scholar

  • [19] Marrese C., Polk J., Jensen K. L., Gallimore A. D., Spindt C. A., Fink R. L. and Palmer W. D., “Performance of Field Emission Cathodes in Xenon Electric Propulsion System Environments,” Micropropulsion for Small Spacecraft, AIAA, Reston, VA, 2000, pp. 271–302. doi: Google Scholar

  • [20] Marrese C. M., Wang J. J., Gallimore A. D. and Goodfellow K. D., “Space-Charge-Limited Emission from Field Emission Cathodes for Electric Propulsion and Tether Applications,” Micropropulsion for Small Spacecraft, AIAA, Reston, VA, 2000, pp. 423–447. doi: Google Scholar

  • [21] Ohkawa Y., Izawa A., Yamagiwa Y., Kawamoto S., Nishida S. and Kitamura S., “Research and Development of Carbon Nanotube Cathodes for Electric Propulsion,” Transactions of the Japan Society for Aeronautical and Space Sciences, Aerospace Technology Japan, Vol. 8, No. ists27, 2010, pp. Pb_27–Pb_32. doi: CrossrefGoogle Scholar

  • [22] Yamamoto N., Kondo S., Chikaoka T., Masui H. and Nakashima H., “Effects of Magnetic Field Configuration on Thrust Performance in A Miniature Microwave Discharge Ion Thruster,” Journal of Applied Physics, Vol. 102, Dec. 2007, Paper 123304. doi: JAPIAU 0021-8979 CrossrefGoogle Scholar

  • [23] Nakano M., Tachibana T. and Arakawa Y., “Scaling Law of the Life Estimation of the Three-Grid Optics for an Ion Engine,” Transactions of the Japan Society for Aeronautical and Space Sciences, Vol. 45, No. 149, 2002, pp. 154–161. doi: TJASAM 0549-3811 CrossrefGoogle Scholar

  • [24] Yamamoto N., Masui H., Kataharada H., Nakashima H. and Takao Y., “Antenna Configuration Effects on Thrust Performance of Miniature Microwave Discharge Ion Engine,” Journal of Propulsion and Power, Vol. 22, No. 4, 2006, pp. 925–928. doi: JPPOEL 0748-4658 LinkGoogle Scholar

  • [25] Yamamoto N., Hiraoka Y. and Nakashima H., “Development of a Miniature Microwave Discharge Neutralizer for Miniature Ion Engines,” Journal of the Japan Society for Aeronautical and Space Sciences, Vol. 62, No. 4, 2014, pp. 123–128 (in Japanese). doi: NKGAB8 0021-4663 CrossrefGoogle Scholar

  • [26] Wirz R., Goebel D., Marrese C. and Mueller K., “Development of Cathode Technologies for a Miniature Ion Thruster,” AIAA Paper 2003-4722, July 2003. LinkGoogle Scholar

  • [27] Pedrini D., Misuri T., Paganucci F. and Andrenucci M., “Development of Hollow Cathodes for Space Electric Propulsion at Sitael,” Aerospace, Vol. 4, No. 2, 2017, p. 26. doi: CrossrefGoogle Scholar


Table 1 Grid parameters

Hole diameter, mm1.200.701.20 
Potential, V5001000 
Thickness, mm0.300.300.30 
Hole pitch, mm1.501.501.50 
Grid gap, mm0.251.0 
Number of holes919191