Performance Characteristics of a DME Propellant Arcjet Thruster

Trans. JSASS Aerospace Tech. Japan Vol. 10, No. ists28, pp. Pb_1-Pb_6, 2012 Original Paper

1) 1) 1) 1) By Akira KAKAMI , Shinji BEEPPU , Muneyuki MAIGUMA and Takeshi TACHIBANA

1) Department of Mechanical Engineering, Kyushu Institute of Technology, Kitakyushu, Japan (Received June 20th, 2011)

This paper describes the influence of cathode configuration on performance of an arcjet thruster using dimethyl ether (DME) propellant. DME, an ether compound, has suitable characteristics for a space propulsion system; DME is storable in a liquid state without being kept under a high pressure, and requires no sophisticated temperature management such as a cryogenic device. DME can be gasified and liquefied simply by adjusting temperature whereas hydrazine, a conventional propellant, requires an iridium-based particulate catalyst for its gasification. In this study, thrust of a 1-kW class DME arcjet thruster is measured at a discharge current of 13 A, DME mass flow rates ranging 15 to 60 mg/s under three cathode configurations: flat-tip rods of 2 and 4 mm in diam. and 4-mm-diam. rod having a cavity of 2 mm in diameter. Thrust measurements show that thrust is increased with propellant mass flow rate. Among the tested cathodes, the flat-tip rod of 4 mm in diam. with 55 mg/s DME flow rate yielded the highest performance: specific impulse of 330 s, thrust of 0.18 N, discharge power of 1400 W and specific power of 25 MJ/kg.

Key Words: Electric Propulsion, Arcjet Thruster, Dimethyl Ether, Cathode, Thrust Measurement

Nomenclature utilized for arcjet thrusters. Hydrogen not only requires complex storage systems such as cryogenic device but also : outer diameter of cathode embrittles materials including metals and rubbers, although the : discharge current lowest atomic weight can yield the highest specific impulse. ݀ୡ : plenum chamber pressure Ammonia, a hydrogen-rich molecule, can yield a ܫௗ : thrust comparatively high specific impulse. On the other hand, it has a ܲ௖ : discharge voltage pungent odor and is reactive to most metals because of its ܶ alkalinity. ௗ 1. Introductionܸ The authors propose to use dimethyl ether (DME) as a propellant for arcjet thrusters7). DME, an ether compound, has Among electric propulsion devices, arcjet thrusters have the molecular formula CH3-O-CH3; it has a structure where two higher thrust-to-power ratios and specific impulses than carbon atoms are bonded through an oxygen atom, permitting chemical thrusters. Arcjets have been applied in various low soot production in chemical reactions. Its freezing and 1-3) boiling points are -154 and -54 °C at 1 atm, respectively, and a missions including North-South station keeping ; Telstar-401 8) with 1.8 kW class arcjets launched in 19934); Kodama, a vapor pressure is 6 atm at room temperature . These features Japanese data-transmitting satellite launched in 2002, had enable not only gasification simply by adjusting temperature arcjet thrusters5). Currently, hydrazine is mainly used as without a catalyst but also liquid-form storage in satellites propellant because it can be shared with chemical thrusters. without complex temperature management. The propellant can be decomposed into gases such as ammonia, DME, which has relatively low toxicity and reactivity, nitrogen, and hydrogen using iridium-based particulate catalyst. requires no exhaust gas treatment system for ground testing, and allows us to make tanks, storage and delivery devices with Because the decomposed gas has a relatively low molecular ® weight, a hydrazine arcjet thruster provides a comparatively relatively low-cost materials such as stainless steel or Teflon . high specific impulse. MR-510, a thruster of this type The 6-atm vapor pressure, enabling self-pressurization in tanks, manufactured by Aerojet, yielded a specific impulse of 580 s eliminates not only a pressurant such as nitrogen or helium, but with a discharge power of 2.2 kW6). Hydrazine can be stored in also its storage devices. At present, DME is more readily a liquid state without cryogenic device. available because many studies and development activities On the other hand, hydrazine has a freezing point of 1 °C and related to its synthesis, storage, and transportation have hence requires temperature management. Materials for tanks enhanced the efficiency of its production and distribution, and tubes should be compatible with this propellant due to reducing its cost. This ether is now applied to industrial uses as strong reactivity. Gas treatment systems are required for diesel fuel and coolant. From this viewpoint, DME has various ground tests because of its strong toxicity. Increased characteristics preferred for an arcjet thruster propellant. temperature would deteriorate iridium-based particulate Our research shows that a designed 1-kW class DME arcjet catalysts. thruster successfully produced arc plasmas at mass flow rates Diverse propellants such as ammonia, and hydrogen were and discharge currents that are comparable to those of

Pb_1 Trans. JSASS Aerospace Tech. Japan Vol. 10, No. ists28 (2012)

conventional arcjet thrusters7). Discharge voltage, power, and DME port Ceramic insulator plenum chamber pressure for DME are higher than those for nitrogen at each mass flow rate or discharge current. The DME arcjet thruster also exhibits both high- and low-voltage modes. Thrust was measured with a vertical thrust stand using a torsional spring to evaluate discharge power, thrust, specific impulse, and thruster efficiency9,10). In this study, a prototype 1-kW class DME arcjet thruster was tested to clarify the correlation between performance and cathode configuration at Anode nozzle a discharge current of 13 A. Feeder (a) Flat 2-type cathode. 2. Experimental Apparatus Cathode

2.1. Designed thruster Figure 1 and Table 1 show a schematic diagram of a prototype 1-kW class arcjet thruster and its configuration, respectively. Gaseous propellant, supplied from a DME port, goes through a gap between a feeder and a cylindrical ceramic insulator into a plenum chamber of 6 mm in diameter. Then, the propellant gas enters a tungsten nozzle having convergent and (b) Flat 4-type cathode. divergent angles of 15 and 45 deg with an expansion ratio of 100. The gap between the flat-tip cathode and the anode was Cathode kept at 1.0 mm in the present study. Afterwards, the plasma is expanded and expelled from the thruster. 2.2. Cathode configuration Three types of cathode were tested in the current study. Figure 1 and Table 2 present cathode schematics and configuration, respectively. All types of cathode had a flat tip, whereas most arcjet thrusters have a conical cathode to enhance resistive heating in the neighboring arc plasma. Our previous (c) Cavitary-type cathode. study shows that conical cathodes reduce the reproducibility of Fig. 1. Schematics of designed thrusters. experimental results for DME propellant7); in some tests, stable arc plasmas were successfully produced for more than 60 s; Table 1. Configuration of designed arcjet thruster. whereas, in other tests plenum chamber pressure rose Constrictor diameter， mm 1.0 repeatedly and dropped even under the same experimental Constrictor length， mm 1.0 conditions. In the worst cases, arc discharge was autonomously Half angle of convergent section in constrictor, interrupted within 5 s following ignition because of a sudden 45 deg increment of plenum chamber pressure. In contrast, a flat tip Half angle of divergent section in constrictor, deg 15 cathode exhibited better reproducibility of experimental results Expansion ratio 100 compared to conical tip cathodes. Hence, in the study, flat tip Electrode region， mm 1.0 cathodes were used to sustain stable arc discharge. Plenum chamber diameter， mm 6.0 Flat 2- and flat 4-type cathodes have flat-tip thoriated tungsten rods of 2 and 4 mm in diameter. In a past study, a thruster using a flat 2-type cathode was tested to evaluate Table 2. Cathode geometry. 9,10) performance . Flat4-type cathode was also used to Cathode type Geometry dc Schematics investigate dependence on cathode diameter. A cavitary-type Flat 2 Flat tip rod 2 mm Fig.1 (a) cathode, with a 4-mm outer diameter thoriated tungsten rod Flat 4 Flat tip rod 4 mm Fig.1 (b) having a 5-mm-depth 2-mm-diam. hole, was designed to Flat-end rod with a hole 5 Cavitary 4 mm Fig.1 (c) prevent the constrictor from being choked by soot. Such a mm in depth hollow-type cathode using a methane propellant was tested to reduce soot accumulation in the inter-electrode region11). 2.3. Experimental setup The designed arcjet thruster was tested in a cubic vacuum Table 3. Experimental conditions. chamber with a 300 mm side. The vacuum chamber had Propellant DME 15, 20, 25, 30, 35, vertical and horizontal cylindrical extensions. The vertical Mass flow rate, mg/s cylindrical extension holding a thrust stand increased the length 40, 45, 50, 55, 60 of a thrust stand arm to improve measurement resolution and Discharge current, A 13 sensitivity. More detailed information on the thrust stand is given in the next section.

Pb_2 A. KAKAMI et al.: Performance Characteristics of a DME Propellant Arcjet Thruster

0.2 Pc 0.2 , MPa , MPa c

T c P

T P 0.1 0.1 , V , V

, A P , A d

c d d d I I V V T , N Start propellant feeding T , N 150 0 150 0 Start propellant feeding Thrust

Ignition Thrust Vd 100 Vd 100

50 Ignition 50 Discharge current Discharge current Discharge voltage Discharge voltage Plenum chamber pressure Id Id Plenum chamber pressure 0 0 20 40 60 80 100 0 0 20 40 60 80 100 Time t, s Time t, s Fig. 2. Time history of discharge current, voltage, plenum chamber Fig. 4. Time history of discharge current, voltage, plenum chamber pressure, and thrust for the flat 2-type cathode at a target discharge current pressure, and thrust for the cavitary-type cathode at a target discharge of 13 A at mass flow rate of 55 mg/s. current of 13 A at mass flow rate of 55 mg/s. supply provided a train of 10-kV peak voltage pulses for arc discharge ignition. T 0.2 An RTAI/Linux personal computer with an analog interface

P , MPa board (Interface, PCI-3521) stored the displacement-sensor c c 0.1 P output for the thrust chamber arm, discharge current, voltage, plenum chamber pressure, and back pressure of the vacuum , V , A d d I V Start propellant feeding T , N chamber at a sampling rate of 100 Hz. All the sensor outputs 150 0 were smoothed using a resistive-capacitive low-pass filter with Vd

Ignition Thrust a cut-off frequency of 1000 Hz. 100 3. Results and Discussion 50 Discharge current Discharge voltage Id Plenum chamber pressure 3.1. Time history 0 0 20 40 60 80 100 Figures 2 to 4 show time histories of discharge current, Time t, s voltage, plenum chamber pressure, and thrust for flat2-, flat4-, Fig. 3. Time history of discharge current, voltage, plenum chamber and cavitary-type cathodes. Time origin t=0 represents the pressure, and thrust for the flat 4-type cathode at a target discharge current moment when arc discharge is initiated. Focusing on Fig. 2, at of 13 A at mass flow rate of 55 mg/s. t=-9 s, plenum chamber pressure is increased soon after DME The horizontal cylindrical extension was used to reduce supply started. At t=0 s, an arc discharge is initiated by a train thrust-measurement errors originating from the intense heat of high voltage pulses. Shortly after ignition, the thruster yields provided by the arcjet thruster. The high-temperature plume relatively intense fluctuations of discharge voltage, which from the thruster negatively affects thrust measurement subsequently become small. Whereas thrust and plenum accuracy because additional heat causes thermal deformation chamber pressure exhibit variations accompanied by discharge of the vacuum chamber and the thrust stand. As a result, it gives voltage fluctuations, the arc discharge is successfully a bias to the displacement of the thrust stand arm. During tests, maintained without autonomous extinguishment. At t=40 s, the the horizontal extension cylinder, cooled with water, absorbed plenum chamber pressure rises due to the adhesion of plume heat. agglomerated carbon to the electrodes. At t=62 s, thrust and The vacuum chamber was evacuated using a rotary pump plenum chamber pressure are decayed by interrupting the with an exhaust rate of 240-L/min. In the tests, back pressures discharge current. of the vacuum chamber were kept below 2 kPa. In the case of the flat 4-type cathode, the thruster exhibits a DME was stored in a pressure vessel in a vapor-liquid similar time history with an exception, as depicted in Fig. 3; for equilibrium state. Gaseous DME was supplied to a pressure the first 5 s following ignition, the thruster showed a regulator, suppressing its pressure to 4 atm, and then was fed to low-voltage mode, thereafter it maintained a high-voltage a mass flow controller (Kofloc, 3665) to maintain the required mode. Compared to the flat 2-type, the flat 4-type yields mass flow rates. Mass flow rate tested in the experiments are large-amplitude low-frequency ripples of discharge voltage. summarized in Table 3. Thrust and plenum chamber pressure also vary with time in Discharge current was supplied with a current-stabilized accordance with the variation of discharge voltage. power supply at a rated voltage and current of 120 V and 35 A. As illustrated in Fig. 4, the cavitary-type cathode, showing a In the current study, discharge current was fixed at 13 A. low-voltage mode for the initial five seconds of arc discharge, Discharge voltage and current were measured using a resistive finally yielded a high-voltage mode after gradually increasing voltage divider and a Hall effect current sensor. The power discharge voltage. Although the high-voltage mode is

Pb_3 Trans. JSASS Aerospace Tech. Japan Vol. 10, No. ists28 (2012)

2000 400 Flat2 Flat2 Flat4 Flat4 Cavitary 1500 Cavitary 300

1000 200 500 100 Specific impulse, s 120 0 Discharge power, W 0.2 0 90

60 0.1

30 Thrust, N Discharge voltage, V 0 0 10 20 30 40 50 60 70 0 DME mass flow rate, mg/s 0 10 20 30 40 50 Specific power, MJ/kg Fig. 5. Dependence of discharge voltage and power on mass flow rate. Fig. 7. Dependence of thrust and specific impulse on specific power.

0.27 400 Flat2 Flat2 Flat4 Flat4 300 Cavitary Cavitary 200 0.18 100 0.2 0 Specific impulse, s

0.09 Thruster efficiency 0.1 Thrust, N

0 0 10 20 30 40 50 0 0 10 20 30 40 50 60 70 Specific power, MJ/kg DME mass flow rate, mg/s Fig. 6. Dependence of thrust and specific impulse on mass flow rate. Fig. 8. Dependence of thruster efficiency on specific power. maintained until discharge current interruption, the discharge due to plasma confinement inside the thruster. voltage has ripples with smaller amplitudes and higher At a given mass flow rate, the flat 4-type cathode exhibited frequencies than those of the flat 4-type cathode. The time the highest discharge voltage and resultant power among the variations of thrust and plenum chamber pressure are tested cathodes, and the flat 2 and cavitary-types showed accompanied with variations of discharge voltage. almost the same values. The constrictor was choked for all the cathodes tested. 3.3. Dependence of thrust and specific impulse on mass Accordingly, the measured thrust chamber pressure exceeded flow rate 0.2 MPa. As depicted in Fig. 6, mass flow rate influences thrust and 3.2. Dependence of discharge voltage and power on mass specific impulse. Thrust and specific impulse also spike at 30 flow rate mg/s, then increase with flow rate in the same way as discharge Figure 5 illustrates the dependence of discharge voltage and voltage and power. The flat 4-type cathode showed maximum power on mass flow rate. Plotted symbols express the average thrust and specific impulse among the tested cathodes, while values, and error bars exhibit the standard deviations. The the flat 2 and cavitary-types produced almost the same values. values are evaluated in the period between the start of high In the current study, the flat 4-type cathode yielded a specific voltage mode and discharge-current interruption. For both the impulse of 330 s, a thrust of 0.18 N at a discharge power of flat 2- and flat 4-type cathodes, discharge voltage was 1400 W, and a specific power of 25 MJ/kg. discontinuously increased with mass flow rate in the vicinity of The dependence of thrust and specific impulse on cathode 30 mg/s. At flow rates of 30 mg/s or more, the thruster showed configuration is partially ascribed to augmented discharge higher discharge voltage and yielded the plume downstream voltage and resultant power for the flat 4-type cathode at a from the nozzle exit. Below 30 mg/s, it exhibited a low given mass flow rate. As mentioned in the following sections, discharge voltage and generated no plume outside. Hence, the the flat4-type yielded the highest specific impulse at a given flat4-type cathode also shows high- and low-voltage modes, specific power. which are dependent on flow rate. The cavitary-type cathode 3.4. Dependence of thruster efficiency on specific power would also have the low-voltage mode in a low-flow rate range, Figures 7 and 8 illustrate the dependence of thrust, specific although it was not tested below 25 mg/s to prevent the low impulse, and thrust efficiency on specific power. At a given voltage mode operation, which can sometimes melt the cathode mass flow rate, the flat 4-type cathode shows a maximum

Pb_4 A. KAKAMI et al.: Performance Characteristics of a DME Propellant Arcjet Thruster

Cavitary-type cathode Flat2-type cathode Concave soot rod

Concave soot rod

Fig. 9. Soot agglomeration on flat 2-type cathode. Fig. 10. Soot agglomeration on cavitary-type cathode. specific impulse of 330 s at 55 mg/s and a thruster efficiency of DME dissociation is not completed in the current experiments, 0.23 at 60 mg/s among the tested cathodes. The flat2 and and thermal decomposition helps reduce the molecular weight cavitary types yielded almost the same specific impulse and of the exhaust gas. thrust efficiency at a given specific power. 3.5. Soot adhesion to cathode 4. Summary Under all of the cathode configurations tested, soot accumulated on the tips of cathodes and convergent sections of We proposed applying DME as an arcjet thruster propellant, anode nozzles. The soot was a brittle cratered rod with a and evaluated the performance of a DME arcjet thruster to relatively sharp edge, as shown in Figs. 9 and 10, and was clarify the influence of cathode configuration at DME mass easily removed from the cathode by hand. flow rates ranging 15 to 60 mg/s. The following is a summary The soot for the flat 4-type had almost the same shape with of the current paper. the exception of its diameter. The concave soot rod has 1. Three types of cathode were tested at a discharge current diameters of 2, 1, and 0.6 mm for the flat 4-, flat 2-, and of 13 A: the flat 2 and 4 types are rods having a flat tip of cavitary-type cathode, respectively. With regard to the 2 and 4 mm in diameter, respectively; and the cavitary-type cathode, the concave soot rod on the cathode tip cavitary-type cathode is a 4-mm circular cathode having a extended its length especially toward the anode nozzle rather 2-mm-diam. cavity. than toward the cavity. Hence, the cavitary-type cathode had a 2. At high mass flow rates, all cathode types tested showed a slight influence on the period for which the accumulated soot high-voltage mode where thrusters yield higher chocked the constrictor. In all the cathodes tested, the top of the performance. concave soot rod was closer to the anode nozzle than the 3. At a given flow rate, the flat 4 type yielded the highest cathode tips; hence, the concave soot rod would behave like a thrust and specific impulse among the cathodes tested. virtual cathode during arc discharge. The other types of cathode showed almost identical 3.6. Influence of cathode configuration performance. At a specific power, the flat 4 type exhibited As shown in Figs. 4 to 9, thrust measurements showed that the best specific impulse and thruster efficiency among the flat 4-type cathode had advantages in terms of performance the tested cathodes. among the tested cathode configurations both at a given mass 4. For all cathodes tested here, soot adhered to the cathode flow rate and a specific power. This is ascribed to enhanced tip, forming rods having a crater with a sharp edge. The heat exchange between arc plasma and its surrounding DME concave soot rod had outer diameters of 2, 1, and 0.6 mm gas. As mentioned in the previous section, soot accumulated for the flat 4, flat 2, and cavitary-types, respectively. more on the flat 4-type cathode than the other cathodes. With this cathode, heat from the arc plasma is most intensely Acknowledgments exchanged between the arc plasma and the gaseous neutral DME. This work was partially supported by KAKENHI, On the other hand, the flat 2-type and cavitary-type had Grants-in-Aid for Scientific Research (B) (22360356). almost the same diameter of accumulated soot. Accordingly, the heat exchange intensity is identical for these cathode types. References Enhanced heat exchange contributes not only to accelerating exhaust velocity but also to producing lighter molecules in 1) Kuriki, K. and Arakawa, Y.: Introduction to Electric Propulsion exhaust gas. In a high-temperature environment, DME is Rockets, University of Tokyo Press, 2003. (In Japanese) readily decomposed into lighter weight molecules such as 2) Jahn, R.G.: Physics of Electric Propulsion, Dover Publications, methane, hydrogen, and carbon monoxide. Moreover, DME INC., New York, 1996. requires 67 MJ/kg for completely dissociating into atoms, 3) Sutton, G.P. and Biblarz, O.: Rocket Propulsion Elements 7th edition. John Wiley & Sons, Inc., New York, 2001. whereas specific power ranged from 10 to 40 MJ/kg. Hence,

Pb_5 Trans. JSASS Aerospace Tech. Japan Vol. 10, No. ists28 (2012)

4) Stephenson, R. R., Electric Propulsion Development and 8) Japan DME Forum: DME Handbook, Ohmsha, 2007. (In Application in the United States, IEPC-95-01, Proceedings of Japanese) International Electric Propulsion Conference, Moscow, Russia, 9) Kakami, K., Beppu, S., Maiguma, M. and Tachibana T.: Thrust Sept. 1995. measurement of dimethyl ether arcjet thruster. Acta Astronautica, 5) Fujiwara, Y., Sudo, Y., Nagano, H. and Kanamori Y.: Japan’s 68 (2011), pp. 1228–1233. First Data Relay Test Satellite (DTRS), Proceedings of 54th 10) Kakami, K., Beppu, S., Maiguma, M. and Tachibana T.: Design International Astronautical Congress of the International and Thrust Evaluation of Arcjet Thruster using Dimethyl Ether Academy of Astronautics, and the International Institute of Space Propellant. Journal of the Japan Society for Aeronautical and Law, IAC-03-M.1.03, Sept. 29-Oct. 3, 2003. Space Sciences, 3 (2010), pp. 74-78 (in Japanese). 6) Sackheim, R. L.: Overview of United States Space Propulsion 11) Hruby, V., Kolencik, J., Martinez-Sanchez, M., Dvore, D., Annen, Technology and Associated Space Transportation Systems, K.D. and Brown, R.C.: Methane Arcjet Development, Proceedings Journal of Propulsion and Power, 22 (2006), pp. 1310-1333. of International Electric Propulsion Conference, IEPC-97-014, 7) Kakami, A., Ebara I., Yokote, J. and Tachibana, T.: Application of pp.104-111. dimethyl ether to arcjet thruster as propellant. Vacuum 8(2008), pp. 77-81.

Pb_6