In spacecraft propulsion
, a Hall effect thruster
(HET) is a type of ion thruster
in which the propellant
is accelerated by an electric field
. Hall effect thrusters wrap electrons in a magnetic field
and then use the electrons to ionize propellant, efficiently accelerate the ions
to produce thrust
, and neutralize the ions in the plume. Hall effect
thrusters are sometimes referred to as Hall thrusters
or Hall current thrusters
. Hall thrusters are often regarded as a moderate specific impulse
(1,600 s) space propulsion
technology. The Hall effect thruster has benefited from considerable theoretical and experimental research since the 1960s.
Hall thrusters operate on a variety of propellants, the most common being xenon
. Other propellants of interest include krypton
, and zinc
. Hall thrusters are able to accelerate their exhaust to speeds
between 10–80 km/s (1,000–8,000 s specific impulse
), with most models operating between 15–30 km/s (1,500–3,000 s specific impulse). The thrust produced by a Hall thruster varies depending on the power level. Devices operating at 1.35 kW produce about 83 mN of thrust. High power models have demonstrated up to 3 N in the laboratory. Power levels up to 100 kW have been demonstrated by xenon Hall thrusters. As of 2009 , Hall effect thrusters ranged in input power
levels from 1.35–10 kilowatts, and had exhaust velocities
of 10–50 kilometers per second, with thrust
of 40–600 millinewtons
in the range of 45–60 percent.
The applications of Hall effect thrusters include control of the orientation and position of orbiting satellites
and use as a main propulsion engine for medium-size robotic space vehicles.
Hall thrusters were studied independently in the United States
and the Soviet Union
. They were first described publicly in the US in the early 1960s.
However, the Hall thruster was first developed into an efficient propulsion device in the Soviet Union. In the US, scientists focused instead on developing gridded ion thrusters
. Two types of Hall thrusters were developed in the Soviet Union:
Soviet and Russian SPT thrusters
The SPT design was largely the work of A. I. Morozov.
The first SPT to operate in space, an SPT-50 aboard a Soviet Meteor spacecraft
, was launched December 1971. They were mainly used for satellite stabilization in North-South and in East-West directions. Since then until the late 1990s 118 SPT engines completed their mission and some 50 continued to be operated. Thrust of the first generation of SPT engines, SPT-50 and SPT-60 was 20 and 30 mN respectively. In 1982, SPT-70 and SPT-100 were introduced, their thrusts being 40 and 83 mN, respectively. In the post-Soviet Russia
high-power (a few kilowatts
) SPT-140, SPT-160, SPT-200, T-160 and low-power (less than 500 W) SPT-35 were introduced.
Soviet and Russian TAL-type thrusters include the D-38, D-55, D-80, and D-100.
Soviet-built thrusters were introduced to the West in 1992 after a team of electric propulsion specialists from NASA’s Jet Propulsion Laboratory
, Glenn Research Center
, and the Air Force Research Laboratory
, under the support of the Ballistic Missile Defense Organization
, visited Russian laboratories and experimentally evaluated the SPT-100 (i.e., a 100 mm diameter SPT thruster). Over 200 Hall thrusters have been flown on Soviet/Russian satellites in the past thirty years. No failures have ever occurred on orbit. Hall thruster continue to be used on Russian spacecraft and have also flown on European and American spacecraft. Space Systems/Loral
, an American commercial satellite manufacturer, now flies Fakel SPT-100’s on their GEO communications spacecraft. Since their introduction to the west in the early 1990s, Hall thrusters have been the subject of a large number of research efforts throughout the United States, France, Italy, Japan, and Russia (with many smaller efforts scattered in various countries across the globe). Hall thruster research in the US is conducted at several government laboratories, universities and private companies. Government and government funded centers include NASA’s Jet Propulsion Laboratory
, NASA’s Glenn Research Center
, the Air Force Research Laboratory
(Edwards AFB, CA), and The Aerospace Corporation
. Universities include the US Air Force Institute of Technology
, University of Michigan
, Stanford University
, The Massachusetts Institute of Technology
, Princeton University
,Michigan Technological University
, and Georgia Tech
. A considerable amount of development is being conducted in industry, such as Aerojet
in the USA, SNECMA
in France and in Italy.
The first use of Hall thrusters on lunar orbit was the European Space Agency (ESA) lunar mission SMART-1
in 2003. On a western satellite Hall thrusters were first demonstrated on the Naval Research Laboratory (NRL) STEX spacecraft, which flew the Russian D-55. The first American Hall thruster to fly in space was the Busek
BHT-200 on TacSat-2
technology demonstration spacecraft. The first flight of an American Hall thruster on an operational mission, was the Aerojet
BPT-4000, which launched August 2010 on the military Advanced Extremely High Frequency
GEO communications satellite. At 4.5 kW, the BPT-4000 is also the highest power Hall thruster ever flown in space. Besides the usual stationkeeping tasks, the BPT-4000 is also providing orbit raising capability to the spacecraft. Several countries worldwide continue efforts to qualify Hall thruster technology for commercial uses.
The essential working principle of the Hall thruster is that it uses an electrostatic potential
to accelerate ions up to high speeds. In a Hall thruster the attractive negative charge is provided by an electron plasma at the open end of the thruster instead of a grid. A radial magnetic field of a hundred gauss
(about 100–300 G, 0.01–0.03 T) is used to confine the electrons, where the combination of the radial magnetic field and axial electric field cause the electrons to drift azimuthally, forming the Hall current from which the device gets its name.
Hall Thruster. Hall thrusters are largely axially symmetric. This is a cross-section containing that axis.
A schematic of a Hall thruster is shown in the image to the right. An electric potential
between 150 and 800 volts is applied between the anode
. The central spike forms one pole of an electromagnet
and is surrounded by an annular space and around that is the other pole of the electromagnet, with a radial magnetic field in between. The propellant, such as xenon
gas, is fed through the anode, which has numerous small holes in it to act as a gas distributor. Xenon propellant is used because of its high atomic weight
and low ionization potential
. As the neutral xenon atoms diffuse into the channel of the thruster, they are ionized by collisions with high energy circulating electrons (typically 10–40 eV, or about 10% of the discharge voltage). Once ionized, the xenon ions typically have a charge of +1, though a small fraction (~20%) are +2. The xenon ions are then accelerated by the electric field
between the anode and the cathode. For discharge voltages of 300 V, the ions reach speeds of around 15 km/s for a specific impulse of 1,500 seconds (15 kN·s/kg). Upon exiting, however, the ions pull an equal number of electrons with them, creating a plasma
plume with no net charge. The radial magnetic field is designed to be strong enough to substantially deflect the low-mass electrons, but not the high-mass ions which have a much larger gyroradius
and are hardly impeded. The majority of electrons are thus stuck orbiting in the region of high radial magnetic field near the thruster exit plane, trapped in E×B
(axial electric field and radial magnetic field). This orbital rotation of the electrons is a circulating Hall current
, and it is from this that the Hall thruster gets its name. Collisions with other particles and walls, as well as plasma instabilities, allow some of the electrons to be freed from the magnetic field, and they drift towards the anode. About 20–30% of the discharge current is an electron current, which does not produce thrust, so limits the energetic efficiency of the thruster; the other 70–80% of the current is in the ions. Because the majority of electrons are trapped in the Hall current, they have a long residence time inside the thruster and are able to ionize almost all of the xenon propellant, allowing for mass utilizations of 90–99%. The mass utilization efficiency of the thruster is thus around 90%, while the discharge current efficiency is around 70% for a combined thruster efficiency of around 63% (= 90% × 70%). Modern Hall thrusters have achieved efficiencies as high as 75% through advanced designs. Compared to chemical rockets, the thrust is very small, on the order of 83 mN for a typical thruster operating at 300 V, 1.5 kW. For comparison, the weight of a coin like the U.S. quarter
or a 20-cent Euro coin
is approximately 60 mN. As with all forms ofelectrically powered spacecraft propulsion
, thrust is limited by available power, efficiency, and specific impulse
. However, Hall thrusters operate at the high specific impulses
that is typical of electric propulsion. One particular advantage of Hall thrusters, as compared to a gridded ion thruster, is that the generation and acceleration of the ions takes place in a quasi-neutral plasma and so there is no Child-Langmuir charge (space charge)saturated current
limitation on the thrust density. This allows for much smaller thrusters compared to gridded ion thrusters. Another advantage is that these thrusters can use a wider variety of propellants supplied to the anode, even oxygen, although something easily ionized is needed at the cathode.
Cylindrical Hall thrusters
Although conventional (annular) Hall thrusters are efficient in the kilowatt
power regime, they become inefficient when scaled to small sizes. This is due to the difficulties associated with holding the performance scaling parameters constant while decreasing the channel size and increasing the applied magnetic field
strength. This led to the design of the cylindrical Hall Thruster. The cylindrical Hall thruster can be more readily scaled to smaller sizes due to its nonconventional discharge-chamber geometry and associated magnetic field
The cylindrical Hall thruster more readily lends itself to miniaturization and low-power operation than a conventional (annular) Hall thruster. The primary reason for cylindrical Hall thrusters is that it is difficult to achieve a regular Hall thruster that operates over a broad envelope from ~1 kW down to ~100 W while maintaining an efficiency of 45-55%.
Hall thrusters have been flying in space since December 1971 when the Soviets launched an SPT-50 on a Meteor satellite. Over 240 thrusters have flown in space since that time with a 100% success rate.
Hall thrusters are now routinely flown on commercial GEO communications satellites where they are used for orbital insertion and stationkeeping
. The first Hall thruster to fly on a western satellite was a Russian D-55 built by TsNIIMASH, on the NRO’s STEX
spacecraft, launched on October 3, 1998.
The solar electric propulsion
system of the European Space Agency
spacecraft used a Snecma PPS-1350
-G Hall thruster.
SMART-1 was a technology demonstration mission that orbited the Moon. This use of the PPS-1350-G, starting on September 28, 2003, was the first use of a Hall thruster outside geosynchronous earth orbit
(GEO). Unlike most Hall thruster propulsion systems used in commercial applications, the Hall thruster on SMART-1 could be throttled over a range of power, specific impulse, and thrust.
- Discharge power: 0.46–1.19 kW
- Specific impulse: 1,100–1,600 s
- Thrust: 30–70 mN
Evaluating NASA’s Futuristic EM Drive
A group at NASA’s Johnson Space Center has successfully tested an electromagnetic (EM) propulsion drive in a vacuum – a major breakthrough for a multi-year international effort comprising several competing research teams. Thrust measurements of the EM Drive defy classical physics’ expectations that such a closed (microwave) cavity should be unusable for space propulsion because of the law of conservation of momentum. EM Drive:
Last summer, NASA Eagleworks – an advanced propulsion research group led by Dr. Harold “Sonny” White at the Johnson Space Center (JSC) – made waves throughout the scientific and technical communities when the group presented their test results on July 28-30, 2014, at the 50th AIAA/ASME/SAE/ASEE Joint Propulsion Conference in Cleveland, Ohio. Those results related to experimental testing of an EM Drive – a concept that originated around 2001 when a small UK company, Satellite Propulsion Research Ltd (SPR), under Roger J. Shawyer, started a Research and Development (R&D) program. The concept of an EM Drive as put forth by SPR was that electromagnetic microwave cavities might provide for the direct conversion of electrical energy to thrust without the need to expel any propellant.
This lack of expulsion of propellant from the drive was met with initial skepticism within the scientific community because this lack of propellant expulsion would leave nothing to balance the change in the spacecraft’s momentum if it were able to accelerate. However, in 2010, Prof. Juan Yang in China began publishing about her research into EM Drive technology, culminating in her 2012 paper reporting higher input power (2.5kW) and tested thrust (720mN) levels of an EM Drive. In 2014, Prof. Yang’s papers reported extensive tests involving internal temperature measurements with embedded thermocouples. It was reported (in SPR Ltd.’s website) that if the Chinese EM Drive were to be installed in the International Space Station (ISS) and work as reported, it could provide the necessary delta-V (change in velocity needed to perform an on-orbit maneuver) to compensate for the Station’s orbital decay and thus eliminate the requirement of re-boosts from visiting vehicles. Despite these reports, Prof. Yang offered no scientifically-accepted explanation as to how the EM Drive can produce propulsion in space.
Dr. White proposed that the EM Drive’s thrust was due to the Quantum Vacuum (the quantum state with the lowest possible energy) behaving like propellant ions behave in a MagnetoHydroDynamics drive (a method electrifying propellant and then directing it with magnetic fields to push a spacecraft in the opposite direction) for spacecraft propulsion. In Dr. White’s model, the propellant ions of the MagnetoHydroDynamics drive are replaced as the fuel source by the virtual particles of the Quantum Vacuum, eliminating the need to carry propellant. This model was also met with criticism in the scientific community because the Quantum Vacuum cannot be ionized and is understood to be “frame-less” – meaning you cannot “push” against it, as required for momentum. The tests reported by Dr. White’s team in July 2014 were not conducted in a vacuum, and none of the tests reported by Prof. Yang in China or Mr. Shawyer in the UK were conducted in a vacuum either. The scientific community met these NASA tests with skepticism and a number of physicists proposed that the measured thrust force in the US, UK, and China tests was more likely due to (external to the EM Drive cavity) natural thermal convection currents arising from microwave heating (internal to the EM Drive cavity). However, Paul March, an engineer at NASA Eagleworks, recently reported in NASASpaceFlight.com’s forum (on a thread now over 500,000 views)
that NASA has successfully tested their EM Drive in a hard vacuum – the first time any organization has reported such a successful test. To this end, NASA Eagleworks has now nullified the prevailing hypothesis that thrust measurements were due to thermal convection.
A community of enthusiasts, engineers, and scientists on several continents joined forces on the NASASpaceflight.com EM Drive forum
to thoroughly examine the experiments and discuss theories of operation of the EM Drive. The quality of forum discussions attracted the attention of EagleWorks team member Paul March at NASA, who has shared testing and background information with the group in order to fill in information gaps and further the dialogue. This synergy between NASASpaceflight.com contributors and NASA has resulted in several contributions to the body of knowledge about the EM Drive. The NASASpaceflight.com group has given consideration to whether the experimental measurements of thrust force were the result of an artifact. Despite considerable effort within the NASASpaceflight.com forum to dismiss the reported thrust as an artifact, the EM Drive results have yet to be falsified. After consistent reports of thrust measurements from EM Drive experiments in the US, UK, and China – at thrust levels several thousand times in excess of a photon rocket, and now under hard vacuum conditions – the question of where the thrust is coming from deserves serious inquiry. Applications:
The applications of such a propulsion drive are multi-fold, ranging from low Earth orbit (LEO) operations, to transit missions tothe Moon
, and the outer solar system
, to multi-generation spaceships for interstellar travel. Under these application considerations, the closest-to-home potential use of EM Drive technology would be for LEO space stations – such as the International Space Station.
In terms of the Station, propellant-less propulsion could amount to significant savings by drastically reducing fuel resupply missions to the Station and eliminate the need for visiting-vehicle re-boost maneuvers.
The elimination of these currently necessary re-boost maneuvers would potentially reduce stress on the Station’s structure and allow for a pro-longed operational period for the ISS and future LEO space stations.
Likewise, EM drive technology could also be applied to geostationary orbit (GEO) satellites around Earth. For a typical geostationary communications satellite with a 6kW (kilowatt) solar power capacity, replacing the conventional apogee engine, attitude thrusters, and propellant volume with an EM Drive would result in a reduction of the launch mass from 3 tons to 1.3 tons. The satellite would be launched into LEO, where its solar arrays and antennas would be deployed. The EM-drive would then propel the satellite in a spiral trajectory up to GEO in 36 days.
Moving out from LEO, Mr. March, from NASA EagleWorks, noted that a spacecraft equipped with EM drive technology could surpass the performance expectations of the WarpStar-I concept vehicle. If such a similar vehicle were equipped with an EM Drive, it could enable travel from the surface of Earth to the surface of the moon within four hours. Such a vehicle would be capable of carrying two to six passengers and luggage and would be able to return to Earth in the same four-hour interval using one load of hydrogen and oxygen for fuel cell-derived electrical power, assuming a 500 to 1,000 Newton/kW efficiency EM Drive system. While the current maximum reported efficiency is close to only 1 Newton/kW (Prof. Yang’s experiments in China), Mr. March noted that such an increase in efficiency is most likely achievable within the next 50 years provided that current EM Drive propulsion conjectures are close to accurate. Far more ambitious applications for the EM Drive were presented by Dr. White and include crewed missions to Mars as well as to the outer planets.
Specifically, these two proposed missions (to Mars and the outer planets) would use a 2 MegaWatt Nuclear Electric Propulsion spacecraft equipped with an EM Drive with a thrust/powerInput of 0.4 Newton/kW. With this design, a mission to Mars would result in a 70-day transit from Earth to the red planet, a 90-day stay at Mars, and then another 70-day return transit to Earth.
According to Dr. White, “A 90 metric ton, 2 MegaWatt nuclear electric propulsion mission to Mars [would have] considerable reduction in transit times due to having a thrust-to-mass ratio greater than the gravitational acceleration of the Sun (0.6 milli-g’s at 1 Astronomical Unit).” Furthermore, this type of mission would have the added benefit of requiring only a “single heavy lift launch vehicle” as compared to “a current conjunction-class Mars mission using chemical propulsion systems, which would require multiple heavy lift launch vehicles.” Presenting at the “Human Outer Solar System Exploration via Q-Thruster Technology” panel at IEEE, 2014, Mr. Joosten and Dr. White explained that “only 12 days would be utilized spiraling up from a 400 km low Earth orbit to achieve escape velocity and only 5 days spiraling down to a 400 km low Mars orbit.” While these spiral trajectories around Earth would have to be carefully designed to avoid or minimize time in the most problematic regions of the Van Allen radiation belts that could expose crewmembers to undesirable levels of radiation, Mr. Joosten and Dr. White note that “These relatively rapid transits would argue for mission strategies where the ‘Q-Ship’ (EM Drive ship) operates between the lowest orbits possible to minimize the launch requirements of crew and supplies from Earth and lander complexity at Mars.” Moreover, this type of EM Drive-enabled mission could negate the need to bring along, for the duration of the mission, a high-speed reentry vehicle to return a Mars crew back to the Earth’s surface because “By quickly spiraling into Earth orbit at the end of the mission, the crew could readily be retrieved via a ‘ground-up’ launch. “While the fast Mars transits that Q-Thruster technology [EM drive] could enable would be revolutionary, the independence from the limitations of departure and arrival windows may ultimately be more so,” added Mr. Joosten and Dr. White. This means that an EM drive ship mission could be designed without consideration of the every-two-year interplanetary conjunction launch windows that currently govern Earth-Mars transit missions and could help stabilize and provide more routine Mars crew rotation timetables. This same elimination of inter-planetary conjunction-enabled launch windows would be applied to crewed missions to the outer planets as well. For such a mission, such as a crewed flight to the outer planets – specifically, a Titan/Enceladus mission at Saturn – an EM Drive would allow for a 9-month transit period from Earth to Saturn, a 6-month in-situ mission at Titan, another 6-month in-situ mission at Enceladus, and a 9-month return trip to Earth. This would result in a total mission duration of just 32 months. However, EM drive applications are not limited to Mars or outer solar system targets. Applications of this technology in deep space missions have already received conceptual outlines. In particular, the Alpha Centauri system, the closest star system to our solar system at just 4.3 lights year’s distance, received specific mention as a potential mission destination. Mr. Joosten and Dr. White stated that “a one-way, non-decelerating trip to Alpha Centauri under a constant one milli-g acceleration” from an EM drive would result in an arrival speed of 9.4 percent the speed of light and result in a total transit time from Earth to Alpha Centauri of just 92 years. However, if the intentions of such a mission were to perform in-situ observations and experiments in the Alpha Centauri system, then deceleration would be needed. This added component would result in a 130-year transit time from Earth to Alpha Centauri – which is still a significant improvement over the multi-thousand year timetable such a mission would take using current chemical propulsion technology. The speeds discussed in the Alpha Centauri mission proposal are sufficiently low that relativity effects are negligible. Bringing EM Drives to reality:
While such mission proposals are important to consider, equally as important are the considerations toward development of the needed technology and procurement long-lead items necessary to make this power technology a reality. Specifically, a useful EM Drive for space travel would need a nuclear power plant of 1.0 MWe (Megawatts-electric) to 100 MWe.
While that sounds significant, the U.S. Navy currently builds 220 MW-thermal reactors for its “Boomer” Ohio class ICBM vehicles. Thus, the technology to build such reactors is available, and the technology needed to build such a device for space-based operations has been around since the 1980s. The limiting factors for further testing and development of this potentially revolutionary space exploration technology are funding to verify and characterize its operations, and the political will to develop nuclear power for space applications. Progress Update:
On April 5, 2015, Paul March reported at NASAspaceflight.com’s Forum that Dr. White and Dr. Jerry Vera at NASA Eagleworks have just created a new computational code that models the EM Drive’s thrust as a three-dimensional magnetohydrodynamic flow of electron-positron virtual particles. These simulations explain why in NASA’s experiments it was necessary to insert a high density polyethylene (HDPE) dielectric into the EM Drive, while the experiments in the UK and China were able to measure thrust without a dielectric insert. The code shows two reasons for this: 1) the experiments in the UK and China used (unlike the ones in the US) a magnetron to generate the microwaves and 2) the experiments in the UK and China were performed with much higher input power: up to 2.5 kiloWatts, compared to less than 100 Watts in the US experiments. In the US tests, microwave frequency generation was controlled via a voltage-controlled oscillator whose signal was passed to a variable voltage attenuator. The tests performed in the UK and China used, instead, magnetron microwave sources (as used in home-use microwave ovens) for their experiments. The magnetron generates amplitude, frequency and phase modulation of the carrier wave (FM modulation bandwidth on the order of +/-20 MHz, at tested natural frequencies of ~2.5 GHz). Dr. White’s computer simulation shows that the modulation generated by the magnetron results in greater thrust force.
Dr. White’s computer analysis also shows that increasing the input power focuses the virtual particle flow from near omnidirectional at the low powers used in the NASA experiments, to a much more focused jet like beam at the higher power (kilowatts as compared to less than 100 Watts) used in the UK and China experiments. The simulation for the 100 Watts input power (as used in the latest tests at NASA) predicted only ~50 microNewtons (in agreement with the experiments) using the HDPE dielectric insert, while the 10 kiloWatts simulation (without a dielectric) predicted a thrust level of ~6.0 Newtons. At 100 kiloWatts the prediction is ~1300 Newton thrust. The computer code also shows that the efficiency, as measured by the thrust to input power ratio, decreases at input powers exceeding 50 kiloWatts. A note of caution is that Dr. White’s simulations do not assume that the Quantum Vacuum is indestructible and immutable. The mainstream physics community assumes the Quantum Vacuum is indestructible and immutable because of the experimental observation that a fundamental particle like an electron (or a positron) has the same properties (e.g. mass, charge or spin), regardless of when or where the particle was created, whether now or in the early universe, through astrophysical processes or in a laboratory. Another reason is that the Quantum Vacuum is assumed to be the lowest possible (time-averaged) energy that a quantum physical system may have, and therefore it should not be possible to extract momentum or energy from the Quantum Vacuum. Due to these predictions by Dr. White’s computer simulations NASA Eagleworks has started to build a 100 Watt to 1,200 Watt waveguide magnetron microwave power system that will drive an aluminum EM Drive shaped like a truncated cone. Initially a teeter-totter balance system will be used in ambient conditions to see if similar thrust levels (0.016 to 0.3 Newton) as reported in the US and China can be reproduced at NASA with this approach. For the last three years, Dr. White’s team has been conducting experiments to find out whether it is possible to measure, with an interferometer, a distortion of spacetime produced by time-varying electromagnetic fields. The ultimate goal is to find out whether it is possible for a spacecraft traveling at conventional speeds to achieve effective superluminal speed by contracting space in front of it and expanding space behind it. The experimental results so far had been inconclusive. During the first two weeks of April of this year, NASA Eagleworks may have finally obtained conclusive results. This time they used a short, cylindrical, aluminum resonant cavity excited at a natural frequency of 1.48 GHz with an input power of 30 Watts.
This is essentially a pill-box shaped EM Drive, with much higher electric-field intensity, aligned in the axial direction. The interferometer’s laser light goes through small holes in the EM Drive. Over 27,000 cycles of data (each 1.5 sec cycle energizing the system for 0.75 sec and de-energizing it for 0.75 sec) were averaged to obtain a power spectrum that revealed a signal frequency of 0.65 Hz with amplitude clearly above system noise. Four additional tests were successfully conducted that demonstrated repeatability. One possible explanation for the optical path length change is that it is due to refraction of the air. The NASA team examined this possibility and concluded that it is not likely that the measured change is due to transient air heating because the experiment’s visibility threshold is forty times larger than the calculated effect from air considering atmospheric heating. Encouraged by these results, NASA Eagleworks plans to next conduct these interferometer tests in a vacuum. *Click Here For larger images of those used in the article
- Jump up ^ Hofer, Richard R. “Development and Characterization of High-Efficiency, High-Specific Impulse Xenon Hall Thrusters”. NASA/CR—2004-21309. NASA STI Program. Retrieved 17 October 2011.
- ^ Jump up to: a b Choueiri, Edgar Y. (2009). “New Dawn for Electric Rockets”. Scientific American 300: 58–65.doi:10.1038/scientificamerican0209-58.
- Jump up ^ Janes, G.; Dotson, J.; Wilson, T. (1962). Momentum transfer through magnetic fields. Proceedings of third symposium on advanced propulsion concepts 2. Cincinnati, OH, USA. pp. 153–175.
- Jump up ^ Meyerand, RG. (1962). Momentum Transfer Through the Electric Fields. Proceedings of Third Symposium on Advanced Propulsion Concepts 1. Cincinnati, OH, USA. pp. 177–190.
- Jump up ^ Seikel, GR. (1962). Generation of Thrust – Electromagnetic Thrusters. Proceedings of the NASA-University Conference on the Science and Technology of Space Exploration 2. Chicago, IL, USA. pp. 171–176.
- Jump up ^ “Hall thrusters”. 2004-01-14.
- Jump up ^ Morozov, A.I. (March 2003). “The conceptual development of stationary plasma thrusters”. Plasma Physics Reports (Nauka/Interperiodica) 29 (3): 235–250. Bibcode:2003PlPhR..29..235M. doi:10.1134/1.1561119.
- ^ Jump up to: a b “Native Electric Propulsion Engines Today” (in Russian) (7). Novosti Kosmonavtiki. 1999. Archived from the original on 6 June 2011.
- Jump up ^ ALTA
- Jump up ^ “Hall-Effect Stationary Plasma thrusters”. Electric Propulsion for Inter-Orbital Vehicles. Retrieved 2014-06-16. 
- Jump up ^ Y. Raitses and N. J. Fisch. “Parametric Investigations of a Nonconventional Hall Thruster” (PDF). Physics of Plasmas, 8, 2579 (2001).
- Jump up ^ A. Smirnov, Y. Raitses, and N.J. Fisch. “Experimental and theoretical studies of cylindrical Hall thrusters” (PDF). Physics of Plasmas 14, 057106 (2007).
- Jump up ^ Polzin, K. A.; Raitses, Y.; Gayoso, J. C.; Fisch, N. J. “Comparisons in Performance of Electromagnet and Permanent-Magnet Cylindrical Hall-Effect Thrusters”. NASA Technical Reports Server. Marchall Space Flight Center. Retrieved 17 October 2011.
- Jump up ^ Polzin, K. A.; Raitses, Y.; Merino, E.; Fisch, N. J. “Preliminary Results of Performance Measurements on a Cylindrical Hall-Effect Thruster with Magnetic Field Generated by Permanent Magnets”. NASA Technical Reports Server. Princeton Plasma Physics Laboratory. Retrieved 17 October 2011.
- Jump up ^ This article incorporates public domain material from the National Aeronautics and Space Administrationdocument “”In-space propulsion systems roadmap.” (April 2012).” by Meyer, Mike, et al.
- Jump up ^ “National Reconnaissance Office Satellite Successfully Launched” (PDF). Naval Research Laboratory (Press Release). October 3, 1998.
- Jump up ^ Cornu, Nicolas; Marchandise, Frédéric; Darnon, Franck; Estublier, Denis (2007). PPS®1350 Qualification Demonstration: 10500 hrs on the Ground and 5000 hrs in Flight. 43rd AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit. Cincinnati, OH, USA. doi:10.2514/6.2007-5197.
- Jump up ^ “Ion engine gets SMART-1 to the Moon: Electric Propulsion Subsystem”. ESA. August 31, 2006. Retrieved 2011-07-25.