hollow cathodes development at sitael - sitael … · sp2016_3124925 1 space propulsion 2016,...

SP2016_3124925

1

SPACE PROPULSION 2016, MARRIOTT PARK HOTEL, ROME, ITALY / 2–6 MAY 2016

HOLLOW CATHODES DEVELOPMENT AT SITAEL

D. Pedrini(1)

, F. Cannelli(2)

, C. Ducci(3)

, T. Misuri(4)

, F. Paganucci(5)

, M. Andrenucci(6)

(1) (2) (3) (4) (6) Sitael S.p.A., Via A. Gherardesca 5, 56121 Pisa, Italy

[email protected] [email protected]

[email protected] [email protected]

[email protected]

(5) Department of Civil and Industrial Engineering, University of Pisa, Italy

[email protected]

KEYWORDS: Hollow cathodes, Hall effect thrusters, electric propulsion.

ABSTRACT:

Hollow cathodes are electron sources used for the gas ionization and the beam neutralization in Hall effect thrusters. Several hollow cathodes have been developed and tested at Sitael (former Alta), each intended for a specific power class of thrusters. An in-house numerical model has been used during the cathode design to define the relevant geometric parameters, in accordance with the thruster unit specifications in terms of discharge current, mass flow rate and lifetime. LaB6-based cathodes were successfully developed for Hall effect thrusters with discharge power ranging from 100 W to 5 kW. The design of a cathode for a 20 kW Hall thruster was also carried out. Experimental campaigns were performed in both stand-alone and coupled configurations. The comparison between experimental results and model predictions are here presented offering a sound theoretical framework to drive the design of future hollow cathodes.

1. INTRODUCTION

Hollow cathodes are sources of electrons to ionize the propellant and neutralize the ion beam exhausted by ion and Hall effect thrusters. A complete understanding of the operation of hollow cathodes is hindered by the complexity of their driving physical processes along with the difficult plasma diagnostics due to the small size (typically a few millimeters in diameter) and high operating temperatures (above 1000 K). Nevertheless, a deeper study of hollow cathodes is important to improve the entire propulsion system, whose

performance and lifetime are both affected by the hollow cathode operation [1]. A theoretical model has been developed and refined at Sitael, to be used as a quick numerical tool for the design of hollow cathodes. A reduced-order approach appeared to be the most practical compromise between including details and keeping the model flexible and time saving. Several hollow cathodes have been developed following the suggestions from the numerical results, each suited for a power class of Hall thrusters, ranging from 100 W to 20 kW. LaB6 was selected as the emitter material, taking into account the considerable space heritage of this technology. LaB6 has been used as the electron emitter in hollow cathodes since 1970s: by the time being, over 200 SPT Hall thrusters have flown in telecommunications satellite applications using LaB6 cathodes [2]. The major reason for using a LaB6 cathode is its robustness, high current density and long life, with respect to conventional refractory metal or impregnated-dispenser cathodes [3]. The experimental data collected so far at Sitael during the testing of cathodes for the different power levels explored are in good accordance with the numerical model.

Figure 1. Schematic of an orificed hollow cathode.

2

2. CATHODE DESIGN

The general architecture of a hollow cathode consists of a thin refractory metal tube, housing the active electron emitter, which provides the electrons via field-enhanced thermionic effect (Fig. 1). The gas-receiving tube ends with an orifice to increase the internal pressure, improving the gas ionization. The cathode can be provided with a heater and heat shields to ease the discharge initiation, and a keeper electrode, positively biased with respect to the inner tube, is used at ignition. Since the main tube must handle the high thermal fluxes coming from the emitter, the tube thickness is selected as the minimum value according to the fabrication technology, whereas the tube length is suggested by the thermal studies to prevent overheating of the support base. Aside these considerations, the critical dimensions of the cathode are selected on the basis of the results of the model hereafter briefly described.

2.1. Model Description

From a theoretical point of view, the cathode design relies upon a reduced-order model previously developed at Sitael [4], which describes the cathode performance as a function of the geometry and the operating conditions. The model results are analyzed to select the cathode geometry, in terms of the emitter dimensions, i.e. its length, inner and outer diameter, and the orifice diameter and length. The theoretical model has been refined and improved, as soon as experimental results have become available. A more comprehensive thermal model has been included to estimate the heat exchange mechanisms occurring in the cathode, considering the plasma in the cathode-to-keeper gap in addition to the emitter and orifice regions. As a matter of fact, a simplified sub-model of the plasma region in the inter-electrode space has been included, in order to consider the voltage drop at the keeper sheath. The resulting electrical characteristics reflect the correct dependence of the total voltage drop as a function of the cathode mass flow rate, which was previously in contrast with the experimental data [5]. As a further step in the theoretical study, the model was adapted to krypton propellant by including its peculiar properties (e.g. the viscosity model, the ionization energy, and collision cross sections). A lifetime estimation based on the emitter evaporation is included in the model. The plasma parameters, such as electron temperature, plasma density and neutral density, are evaluated along with the heat exchanges and the temperature profiles by combining different modules as hereafter summarized.

Plasma Model

The plasma formed inside the hollow cathode is ideally divided in three coupled regions (Fig. 2): the emitter, the orifice, and the cathode-to-keeper gap. Equations expressing particle, momentum, and energy balances are numerically solved to compute the plasma properties in each region. Current density equations are introduced to estimate the voltage drop at the sheaths formed at the emitter and keeper surfaces, for a given current demand. The model assumes the formation of double sheaths at the boundary between the different regions of the cathode, being characterized by different plasma properties [4].

Figure 2. Illustration of the plasma divided in three regions for the numerical computation.

The pressure model is based on a Poiseuille flow in the orifice region [6], with a correction term for the transitional regime as a function of the Knudsen number. The ionization model has been revisited to include the contribution of the step-wise mechanism [7].

Thermal Model

A dedicated thermal model was developed and coupled with the plasma models to estimate the power exchanges and the temperature profile along the cathode, by means of power balance equations. The cathode was schematized as in Fig. 3, nevertheless the heater and thermal shields could be included or not according to the real cathode configuration considered.

Figure 3. Cathode schematization for the thermal model.

3

Effective Emission Length

The prediction of the cathode parameters is strongly affected by the plasma penetration depth, which impacts the current emission process and, ultimately, the cathode power consumption. The plasma penetration depth identifies the so-called effective emission length, experimentally shown to be dependent on the operating conditions, in terms of current level and mass flow rate [8,9]. The fraction of the emitter actively involved in the thermionic emission of electrons is not always coincident with the emitter length. In particular, the effective emission length decreases if the cathode internal pressure increases (due to higher mass flow rates and/or smaller orifice diameters), whereas the dependence on the current is less evident and behaves contrastingly according to works by different authors [7,10]. A method to predict the effective emission length has been introduced in the theoretical model, which is based on the principle of minimum entropy production rate proposed by Prigogine [11], according to which the plasma penetrates the insert in such a

way to minimize the entropy production rate, �̇�. For a hollow cathode, this quantity can be computed as follows:

�̇� =

(𝑉𝑐 + 𝑉𝑝)𝐼𝑑

𝑇𝑐

(1)

being 𝑉𝑐 the voltage drop at the emitter sheath, 𝑉𝑝

the resistive potential drop of the plasma, 𝐼𝑑 the

discharge current, and 𝑇𝑐 the emitter surface temperature. The hollow cathode model has been

then used to evaluate �̇� for fixed geometry and operating conditions, while parametrically varying the effective emission length as a progressive percentage of the total insert extent. The selected effective emission length corresponds to the

minimum of �̇�.

Lifetime Evaluation

The cathode model includes a lifetime estimate, based on the emitter evaporation at the operating surface temperature [4]. This computation is useful for a comparison between theoretical results, since other mechanisms such as orifice clogging and ion bombardment damage are expected to cause a performance degradation shortening the cathode lifetime.

Solution Procedure

The plasma and thermal sub-models are combined together to iteratively compute the plasma and wall parameters in the cathode assembly. The model requires the cathode dimensions, the gas characteristics and the material properties, as well

Figure 4. Flow chart of the cathode numerical code.

as the operating conditions defined by the current and the mass flow rate. A flow chart of the numerical code is shown in Fig. 4, along with the iteration variables 𝑉𝑐, 𝑇𝑒 (electron temperature in

the emitter region), 𝑇𝑐 (emitter surface

temperature), 𝑇𝑜 (orifice surface temperature),

𝑇𝑘 (keeper surface temperature).

The following sections report the experimental results collected during the test campaigns of the designed cathodes. In some of the figures, the measurement error-bar is smaller than the marker size.

3. HC1

A hollow cathode for the 100 W-class Hall thrusters (HC1) was developed to provide 0.3-1 A discharge current. The cathode operates at mass flow rates in the range 0.08-0.5 mg/s Xe, with an expected lifetime higher than 4000 hours.

3.1. Cathode Description

The HC1 cathode features a LaB6 emitter, placed in a tantalum tube ending with a sub-millimetric orifice. A titanium-alloy keeper encloses the inner components. A tungsten spring along with an alumina spacer is used to hold the emitter in place. The cathode has been conceived to be tested both with and without a heater, which could be easily included in the cathode assembly. The cathode mass (without cables) is lower than 50 grams.

4

3.2. Experimental Setup

The heaterless configuration of HC1 was tested in the Sitael IV1 vacuum chamber, which was provided with an anode plate for the cathode characterization. The IV1 facility consists of a cylindrical stainless steel body 600 mm in diameter and 1600 mm in length. The test bench is equipped with a primary scroll pump, a 300 l/s turbo molecular pump and two 900 l/s cryogenic pumps. The facility is able to reach a background pressure of 1×10

−6 Pa when the combined

pumping system is activated. The pressure level within the chamber is continuously monitored by two Leybold-Inficom IT90 Pirani/Bayard-Alpert sensors and recorded via LabVIEW DAQ. A current-limited Huttinger PFG5000 (1000 V, 6 A) power supply controlled the cathode-to-keeper voltage during discharge initiation and the current during operation. An external anode plate was located about 15 mm downstream of the cathode. The cathode-to-anode current was controlled using a Sorensen DLM 300-3.5E (300 V, 3.5 A) power supply. Both power supplies were connected to a common negative reference and the set-up was electrically floating with respect to ground. Grade N48 xenon was used during the test campaigns. K-type thermocouples were installed on the mounting flange constituting the mechanical interface, and on the backside of the anode.

3.3. Cathode Performance

The cathode was characterized in diode mode with the keeper, in triode mode with both the keeper and the anode, and with the anode electrode only. While carrying out the characterization test, the cathode was started 10 times by flowing 1 mg/s Xe, applying 700 V to the keeper, and regulating the keeper power supply to 1 A. The cathode totally cumulated more than 200 hours of operation. After this period, 20 further ignitions were repeated, showing a decrease in the keeper voltage required to start the cathode, which was between 300 and 470 V, with a mass flow rate of 1 mg/s.

Figure 5. HC1 in diode mode with the keeper (0.1 mg/s Xe, 1 A discharge current).

Figure 6. HC1 in triode mode with keeper and anode (keeper current 1 A, anode current 0.9 A, mass flow rate

1 mg/s Xe).

Figure 7. HC1 in triode mode with keeper and anode (keeper current 1 A, anode current 0.3 A, mass flow rate

0.3 mg/s Xe).

Figure 8. HC1 in diode mode with the anode (0.3 mg/s Xe, 0.9 A discharge current).

HC1 is shown in Fig. 5 operating in diode mode with the keeper, in Figs. 6-7 operating in triode mode with anode and keeper, in Fig. 8 operating in diode mode with the anode.

The cathode internal pressure was measured at room conditions, flowing a xenon mass flow rate between 0.1 and 1.5 mg/s. The collected data are reported in Fig. 9 together with the theoretical predictions, which show a lower slope as compared to the experimental curve, nevertheless being in a fairly good agreement. The pressure comparison for the cathode operated in diode mode with the keeper is shown in Fig. 10, for a current equal to 1 A. The parameter Th/Tw represents the ratio of the heavy particles

5

temperature, Th, to the cathode wall temperature, Tw. As expected [12], a value of Th/Tw comprised between 1 and 4 gives a good estimate of the cathode internal pressure. The electrical characteristics collected in diode mode with the keeper are shown in Fig. 11, at different mass flow rates between 0.08 and 0.5 mg/s. The discharge voltage was found to decrease from about 30 to 15 V, when increasing the current from 0.3 to 1.5 A. The power consumption of HC1 is in the range from 9 to 20 W, depending on the operating conditions. The results are also shown in Fig. 12 as a function of mass flow rate, for different values of the discharge current. The discharge voltage showed a negligible dependence on the mass flow rate. The results of the model, in terms of electrical characteristics, are in good agreement with the experimental data (20% of discrepancy for the worst prediction), as shown in Fig. 13 for two different mass flow rates.

Figure 9. HC1 internal pressure as a function of Xe mass flow rate at room temperature.

Figure 10. HC1 internal pressure as a function of Xe mass flow rate in diode discharge with the keeper at 1 A.

Figure 11. HC1 electrical characteristics in diode mode with the keeper, at different Xe mass flow rates.

Figure 12. HC1 discharge voltage in diode mode with the keeper, as a function of Xe mass flow rate.

Figure 13. Comparison between experimental and theoretical electrical characteristics for HC1, at two Xe

mass flow rates.

0 0.5 1 1.50

20

40

60

80

100

120

Mass Flow Rate, mg/s

Pre

ssure

, hP

a

Model

Experiment

0 0.2 0.4 0.60

50

100

150

200


Pre

ssure

, hP

a

Th/T

w = 1

Th/T

w = 4

Experiment

0 0.5 1 1.5 20

5

10

15

20

25

30

35

Discharge Current, A

Dis

charg

e V

olta

ge

, V

0.08 mg/s

0.1 mg/s

0.2 mg/s

0.3 mg/s

0.4 mg/s

0.5 mg/s

9 W

20 W

0 0.2 0.4 0.60

5

10

15

20

25

30

35


Dis

charg

e V

olta

ge

, V

0.3 A

0.5 A

0.7 A

0.9 A

1 A

1.5 A

0 0.5 1 1.5 20

5

10

15

20

25

30

35


Dis

charg

e V

olta

ge

, V

0.08 mg/s, Model

0.08 mg/s, Experiment

0.5 mg/s, Model

0.5 mg/s, Experiment

6

4. HC3

A hollow cathode for the range 1-3 A (HC3) was developed for the low-power (<500 W) ion and Hall effect thrusters. The cathode operates at mass flow rates between 0.08 and 1 mg/s, with an expected lifetime higher than 10000 hours.


The cathode features a LaB6 emitter enclosed in graphite sleeves (acting as a barrier against potential chemical reactions with the refractory metal), and placed inside a tantalum tube. A tungsten-alloy spring is used together with graphite spacers to hold the emitter in place. The keeper is made of titanium alloy. The cathode is provided with a heater made of a W-3Re wire immersed in an alumina potting surrounding the cathode tube. The cathode mass (without cables) is lower than 100 grams. The design allowed for an easy replacement of the orifice plate to assess the change of performance as a function of the orifice dimensions.

4.2. Experimental Setup

The HC3 cathode was tested in a stand-alone experimental campaign in the Sitael IV1 facility (§ 3.2), and in the IV4 vacuum chamber both in stand-alone mode and coupled with the Sitael HT-100D Hall thruster. The IV4 facility consists of two different bodies made of AISI 316L stainless steel with low magnetic relative permeability (μr<1.06). The Auxiliary Chamber - AC - is the main vessel (2 m dia., 3.2 m length) and the Small Chamber - SC - is the service chamber (1 m dia., 1 m length). The two bodies are connected through a 1 m dia. gate valve. The small chamber was used to accommodate the cathode and thruster setup, with electrical and gas-feeding systems, while the AC (directly connected to the main pumping system) allowed for a free expansion of the plasma plume. At the far end of the AC a bi-conical, water cooled, Grafoil-lined target is installed to dump the beam energy down. The chamber pumping system is capable of maintaining a back pressure in the range of 10

-5 Pa by using a primary stage located

in the AC and a secondary stage located in the SC. The combined pumping speed of the system is approximately 130,000 l/s for xenon. The pressure level within the chamber is continuously monitored by three Leybold-Inficom IT90 Pirani/Bayard-Alpert sensors and recorded via LabVIEW. A current-limited Huttinger PFG-5000 (1000V-6A) DC power supply controlled the cathode-to-keeper voltage. The gas feeding system consisted of two independent lines, each equipped with dedicated mass flow controllers (Bronkhorst F-201C-FAC-22-V and Bronkhorst F-201C-FAC-88-V for cathode

and anode lines, respectively), connecting the xenon tank to the test item. The tests were performed using grade 4.5 xenon. The electrical parameters were measured by using current (LEM LA25-NP) and voltage probes (LEM LV25-P), whereas the cathode pressure was recorded by means of a Kulite (HKM-375-25A) pressure transducer located along the feeding line at a distance of about 20 cm from the insert. All the sensing probes were calibrated before the test and connected to the DAQ system controlled via LabVIEW software.


The HC3 cathode was tested in diode (Fig. 14) and triode mode, both with an anode plate and with the HT-100D Hall thruster (Fig. 15). HT-100D is a thruster operating in the 120-400 W range to give a thrust between 6 and 18 mN and a specific impulse between 1000 and 1600 s [13].

The feasibility of a heaterless ignition was explored during the test campaign, demonstrating the need of higher keeper voltage and mass flow rate in the absence of a heating phase. As a matter of fact, whereas the ignition occurred at 300 V keeper voltage and 0.4 mg/s with a heater power of about 50 W, the heaterless starting required keeper voltages as high as 950 V combined with an anode voltage between 250 and 500 V and a cathode mass flow rate in the 1-2 mg/s range. More than 60 ignitions were accumulated by the cathode, along with about 120 hours of continuous operation.

Figure 14. HC3 in diode mode with the keeper (1 mg/s Xe, 1.5 A discharge current).

Figure 15. HC3 operating with the HT-100D Hall thruster.

7

The cathode inlet pressure was monitored during the test campaign, showing a linear trend as a function of mass flow rate. At room temperature conditions, a pressure from about 5 to about 50 hPa was recorded during a mass flow rate sweep from 0.2 to 1.5 mg/s. During the coupling test with the HT-100D Hall thruster, the cathode inlet pressure was found to vary from about 8 to 170 hPa at increasing mass flow rate from 0.1 to 1 mg/s. The higher values compared to the room condition pressure are tied to the gas heating effect due to the current drawn from the cathode (which was comprised between 2 and 2.5 A). The comparison of the cathode pressure measured at room conditions and the corresponding theoretical values shows an excellent agreement between the two curves (Fig. 16). The pressure as a function of mass flow rate in the presence of the discharge at 3 A is compared to the numerical results in Fig. 17, for three different values of the parameter Th/Tw

(the ratio of the ions and neutrals temperature to the cathode wall temperature). The comparison suggests that the heavy particles temperature is around twice the wall temperature of the cathode.

Figure 16. HC3 pressure comparison, room temperature.

Figure 17. HC3 pressure comparison at 3 A discharge current (diode mode).

The comparison of the HC3 electrical characteristics with the model results is shown in Figs. 18-21, at four different mass flow rates in diode mode with the keeper. The discharge voltage is also shown in Figs. 22-24 as a function of the mass flow rate, for three values of current. The results of the updated model are more accurate for variations in the mass flow rate, since the new plasma module of the cathode-to-keeper gap, albeit simplified, allows for a correct prediction of the trends in the discharge voltage [5]. As a matter of fact, consistently with data available in the literature, a general trend of a lower voltage with higher mass flow rates was observed during the experiments.

The cathode power consumption was found to be comprised between about 15 and 60 W. The discharge voltage decreased when increasing the discharge current, settling in a range from 14 to 35 V in the various operating conditions.

Figure 18. Comparison of the HC3 electrical characteristics at 0.08 mg/s Xe.


0 0.5 1 1.50

10

20

30

40

50


Pre

ssure

, hP

a

Model

Experiment

0 0.2 0.4 0.6 0.8 10

50

100

150

200


Pre

ssure

, hP

a

Th/T

w = 1

Th/T

w = 2

Th/T

w = 3

Experiment

0.5 1 1.5 2 2.5 3 3.50

5

10

15

20

25

30

35


Dis

charg

e V

oltage, V

Model

Experiment

0.5 1 1.5 2 2.5 3 3.50

5

10

15

20

25

30

35


Dis

charg

e V

oltage, V

Model

Experiment

8


Figure 21. Comparison of the HC3 electrical characteristics at 1 mg/s Xe.

Figure 22. Comparison of the HC3 electrical characteristics at 1 A.



5. HC20

HC20 is a hollow cathode conceived for the 8-20 A range (1-4 mg/s mass flow rates), specifically intended for the 5 kW-class Hall effect thrusters, with an expected lifetime higher than 10000 hours.


This cathode includes a LaB6 emitter placed inside a molybdenum-alloy tube. A graphite keeper is used, to limit the sputtering damage due to ion bombardment on this electrode. The cathode mass is about 300 grams, without cables and connections. HC20 is a heaterless cathode: to reduce the complexity of the device as well as to increase the system reliability, the heater has been removed since it represents a single point of failure for the cathode. The cathode was tested both in a stand-alone configuration (Fig. 25) and coupled with Sitael HT5k Hall thruster (Fig. 26). HT5k

0.5 1 1.5 2 2.5 3 3.50

5

10

15

20

25

30

35


Dis

charg

e V

oltage, V

Model

Experiment

0.5 1 1.5 2 2.5 3 3.50

5

10

15

20

25

30


Dis

charg

e V

oltage, V

Model

Experiment

0 0.2 0.4 0.6 0.8 1 1.20

10

20

30

40


Dis

charg

e V

oltage, V

Model

Experiment

0 0.2 0.4 0.6 0.8 1 1.20

5

10

15

20

25


Dis

charg

e V

oltage, V

Model

Experiment

0 0.2 0.4 0.6 0.8 1 1.20

5

10

15

20


Dis

charg

e V

oltage, V

Model

Experiment

9

operates in the 2.5-7.5 kW range to produce a thrust comprised between 150 and 350 mN and a specific impulse in the 1700-2800 s range [14].


An endurance test was performed during the stand-alone cathode experimental campaign in the Sitael IV1 facility (§ 3.2). The cathode was operated at 16 A discharge current with the anode (keeper left floating) at 1 mg/s Xe for 300 hours. After the endurance test the cathode was also characterized with krypton propellant.

The cathode was tested in diode mode with the keeper electrode, selecting the current range from 1 to 5 A. The corresponding comparison between the pre-endurance experimental data and the results of the updated theoretical model are shown in Figs. 27-29, at 2 mg/s, 1.5 mg/s, and 1 mg/s, respectively. The numerical data correctly predict the measured ones at 5 A, deviating from the experimental curve when decreasing the current. Nevertheless, the maximum discrepancy is lower than 20%. Figs. 30-32 refer to the HC20 operated with the anode plate and with floating keeper. The comparison with the theoretical results is shown at three mass flow rates, with the anode collecting up to 18 A of current. The predictions at 2 mg/s Xe (Fig. 30) are more accurate at the higher level of current, whereas at 7 A the discrepancy raises to about 35%. The comparison at 1.5 mg/s Xe (Fig. 31) shows a discrepancy of about 25% at 2 A, and a better agreement is found when increasing the current. The experimental data follow a non-monotonic trend, which could be tied to instabilities due to the cathode-to-anode coupling. The comparison at 1 mg/s Xe (Fig. 32) shows an offset of about 5 V between numerical and experimental data.

After the 300-hour endurance test, the HC20 cathode was characterized with krypton propellant. The related electrical characteristics are reported in Figs. 33-35, along with the theoretical results, in triode mode with 2 A keeper current. A maximum discrepancy of about 5 V is observed at the higher current levels investigated with 4 mg/s Kr (Fig. 35), whereas the other operating points are in excellent agreement. A higher power consumption up to 60 W with krypton as compared to the corresponding operating points with xenon was observed, which is ascribed to a less efficient ion production tied to the higher ionization energy of krypton with respect to xenon.

Temperature measurements were collected during the endurance test. The highest value of 1168 ± 3 °C was recorded by a B-type thermocouple located on the main tube outer surface in the vicinity of the

thermionic emission zone. The temperature of the metallic flanges did not exceed 80 °C throughout the endurance test.

Figure 25. HC20 ready for the stand-alone test campaign.

Figure 26. HC20 operating with the 5 kW Hall thruster HT5k.

Figure 27. Comparison of the HC20 electrical characteristics at 2 mg/s Xe, diode mode with keeper.

0 1 2 3 4 5 60

10

20

30

40

50

60


Dis

charg

e V

oltage, V

Model

Experiment

10

Figure 28. Comparison of the HC20 electrical characteristics at 1.5 mg/s Xe, diode mode with keeper.

Figure 29. Comparison of the HC20 electrical characteristics at 1 mg/s Xe, diode mode with keeper.

Figure 30. Comparison of the HC20 electrical characteristics at 2 mg/s Xe, with anode, floating keeper.

Figure 31. Comparison of the HC20 electrical characteristics at 1.5 mg/s Xe, floating keeper.

Figure 32. Comparison of the HC20 electrical characteristics at 1 mg/s Xe, floating keeper.

Figure 33. Comparison of the HC20 electrical characteristics at 1 mg/s Kr, triode mode with 2 A keeper

current.

0 1 2 3 4 50

10

20

30

40

50

60


Dis

charg

e V

oltage, V

Model

Experiment

2.5 3 3.5 4 4.5 5 5.50

5

10

15

20

25

30


Dis

charg

e V

oltage, V

Model

Experiment

5 10 15 200

5

10

15

20


Dis

charg

e V

olta

ge

, V

Model

Experiment

0 5 10 15 200

10

20

30

40


Dis

charg

e V

olta

ge

, V

Model

Experiment

10 12 14 16 18 200

5

10

15

20


Dis

charg

e V

olta

ge

, V

Model

Experiment

0 5 10 15 200

10

20

30

40


Dis

charg

e V

olta

ge

, V

Model

Experiment

11


current.


current.

The HC20 cathode cumulated more than 100 cold ignitions, and about 1000 hours of operation, including 500 hours of continuous coupling with the HT5k thruster fed with a mixture of xenon (25%) and krypton (75%).

Given the heaterless configuration, keeper voltages as high as 800 V were required to start the cathode, with a mass flow rate of 5 mg/s. The mass flow rate was decreased as soon as the discharge stability was obtained.

6. HC60

A hollow cathode has been developed to be used for the propellant ionization and the ion beam neutralization of the Sitael HT20k. HT20k is a thruster designed to operate at a nominal discharge power of 20 kW to produce a thrust greater than 1 N and a specific impulse higher than 2500 s [15]. The HC60 cathode has been designed

to operate at mass flow rates between 2 and 6 mg/s, at discharge currents between 30 and 60 A. The cathode predicted lifetime should be higher than 10000 hours.


The design features a LaB6 cylindrical insert in a molybdenum-rhenium tube with a titanium-alloy keeper. The LaB6 insert is protected from direct contact with the refractory metal tube by a thin graphite sleeve, to avoid chemical compatibility problems at high temperatures. The insert is held in place with a tungsten spring placed inside the tube. The tube is sufficiently long and thin to minimize heat conduction from the insert to the base plate according to the results of the thermal simulations. A heater was included to reduce the emitter thermal stress during the cathode ignition and to ease the discharge initiation. The estimated overall mass of the cathode is about 500 g without harnesses. The maximum dimensions are 160 mm length and 80 mm diameter. A CAD drawing of the hollow cathode is shown in Fig. 36.

6.2. Cathode Predicted Performance

The cathode electrical characteristic computed with the theoretical model is shown in Fig. 37, considering 5 mg/s xenon mass flow rate. For the operating point of 60 A discharge current and 5 mg/s Xe mass flow rate, a discharge voltage of 12.2 V was numerically computed, along with an emitter temperature of about 1700 K, an orifice temperature of about 1920 K, and a total power consumption of 730 W. The electron temperature is estimated to be about 1.18 eV in the emitter region, and 1.88 eV in the orifice. The theoretical model has been used to predict the cathode operation with krypton propellant. The results showed a discharge voltage up to about 3 V higher using krypton with respect to xenon, along with higher temperatures, i.e. a difference of about 30 K for the emitter temperature, and up to 70 K for the orifice temperature, considering krypton in place of xenon.

Figure 36. CAD drawing of the HC60 cathode.

0 5 10 15 20 250

5

10

15

20

25

30

35


Dis

charg

e V

olta

ge

, V

Model

Experiment

0 5 10 15 20 250

5

10

15

20

25

30

35


Dis

charg

e V

olta

ge

, V

Model

Experiment

12

Figure 37. Theoretical electrical characteristic of the HC60 (5 mg/s Xe).

The thermal analysis of the cathode showed that the cathode structure is able to withstand the high temperature gradients tied to the thermionic electron emission. The predicted thermal deformation is expected not to compromise the cathode operation, since the displacement is mainly directed along the cathode axis.

7. CONCLUSION

Hollow cathodes for a wide range of power levels have been developed and tested at Sitael. The results are promising and in good accordance with both experimental data available in the literature and the theoretical predictions coming from the in-house model. Future developments will include a deeper investigation of the HC1 low-current cathode, the test campaigns of the HC60 cathode, and a study of the influence of the heater power on the LaB6 cathodes ignition. A more extensive comparison between theoretical and experimental data will be carried out, in order to deeply validate the numerical model.

8. ACKNOWLEDGEMENTS

The authors wish to express their gratitude to Carlo Tellini, Ugo Cesari, Nicola Giusti, and Luca Pieri, for their valuable assistance in preparing and performing the experimental campaigns.

9. REFERENCES

1. Domonkos, M. T., Patterson, M. J. & Gallimore, A.D., (2003). Low-Current, Xenon Orificed Hollow Cathode Performance for in-Space Applications. J. Propuls. Power 19(3), 438–443. 2. Kozubskii, K. N., Murashko, V. M., Rylov, Yu. P., Trifonov, Yu. V., Khodnenko, V. P., Kim, V., Popov, G. A. & Obukhov, V. A. (2003). Stationary Plasma Thrusters Operate in Space. Plasma Phys. Rep. 29(3), 251-266.

3. Frollani, D., Coletti, M. & Gabriel, S. B. (2015). Flexible LaB6 Hollow Cathode for Lab Tests. IEPC-2015-46, 34

th International Electric Propulsion

Conference, Hyogo-Kobe, Japan. 4. Pedrini, D., Albertoni, R., Paganucci, F. & Andrenucci, M. (2014). Theoretical Model of a Lanthanum Hexaboride Hollow Cathode. IEEE Trans. on Plasma Sci., 43(1). 5. Pedrini, D., Albertoni, R., Paganucci, F. & Andrenucci, M. (2015). Modeling of LaB6 Hollow Cathode Performance and Lifetime. Acta Astron., 106(2015), 170–178. 6. Domonkos, M. T. (1999). Evaluation of Low-Current Orificed Hollow Cathodes, Ph.D. dissertation, Dept. Aerosp. Eng., Univ. Michigan, Ann Arbor, MI, USA. 7. Downey, R. T. (2008). Theoretical and Experimental Investigation into High Current Hollow Cathode Arc Attachment, Ph.D. dissertation, Univ. Southern California, CA, USA. 8. Ferreira C. M. & Delcroix J. L. (1978). Theory of

the Hollow Cathode Arc, J. Appl. Phys., 49(4),

2380-2395.

9. Goebel, D. M. & Chu, E., High Current Lanthanum Hexaboride Hollow Cathodes for High Power Hall Thrusters (2011). IEPC-2011-053, 32

nd

International Electric Propulsion Conference, Wiesbaden, Germany. 10. Krishnan M., Physical Processes in Hollow Cathodes (1977). J. Propuls. Power 15(9), 1217-1223. 11. Christen, T. (2009). Modeling Electric

Discharges with Entropy Production Rate

Principles. Entropy, 11, 1042-1054.

12. Goebel, D.M. & Katz, I. (2008). Fundamentals of Electric Propulsion: Ion and Hall Thrusters, JPL Space Science and Technology Series. 13. Ducci, C., Albertoni, R. & Andrenucci, M. (2013). HT100D Performance Evaluation and Endurance Test Results. IEPC-2013-140, 33

rd

International Electric Propulsion Conference, Washington, D.C., USA. 14. Ducci, C., Andreussi, T., Arkhipov, A., Passaro, A., Andrenucci, M., Bulit, A. & Edwards, C. (2015). Investigation of a 5 kW Class Hall-Effect Thruster Operating with Different Xenon-Krypton Mixtures. IEPC-2015-126, 34

th International

Electric Propulsion Conference, Hyogo-Kobe, Japan. 15. Leporini, A., Giannetti, V., Andreussi, T., Pedrini, D., Rossodivita, A., Piragino, A., Andrenucci, M. & Estublier, D. (2016). Development of a 20 kW-Class Hall Effect Thruster. SP2016_3125196, Space Propulsion 2016, Rome, Italy.

20 30 40 50 60 70 800

5

10

15

20


Dis

charg

e V

oltage, V

hollow cathodes development at sitael - sitael … · sp2016_3124925 1 space propulsion 2016,...

Documents