There are three basic types of electric propulsion systems, categorised according to the method used to accelerate the propellant as electrothermal, electrostatic, and electromagnetic. Practical propulsion systems frequently make simultaneous use of two or even all three of these methods.

Electrothermal Propulsion

Electrothermal propulsion systems accelerate the propellant using heating. There are three sub-types: resistojets, arcjets and inductively or radiatively heated systems.

Resistojets

Resistojets operate by passing gaseous propellant through an electric heater and then expanding it through a conventional nozzle to create thrust. Resistojets are normally operated as enhanced chemical propulsion systems where electric heating is used to further expand and accelerate propellant that has already undergone a chemical reaction.

The most successful application of this technique to date has been the superheating of catalytically decomposed hydrazine, which offers the advantage of fuel commonality with a frequently used monopropellant chemical propulsion system.

The specific impulse that can be achieved with hydrazine resistojets is limited because the molecular mass of the gases used is relatively high and because the maximum heating surface temperature that can be sustained with available materials is limited to around 3000 K. This results in an achievable exhaust velocity of about 3500 m s^-1 (I_sp = 350 s) some 40% better than without superheating, with an efficiency of up to 80%.

Resistojets were first used experimentally in space during the mid nineteen-sixties. Their first operational use was for north-south station keeping on the Intelsat-V series of geostationary communication satellites in the 1980s. They were also used for orbit insertion, attitude control and de-orbiting of the Iridium satellite constellation.

Arcjets

For electrothermal propulsion systems to achieve exhaust velocities much higher than 10 000 m s^-1, portions of the propellant flow must attain temperatures in excess of 10 000 K while being kept out of contact with the engine component walls. Arcjets accomplish this by passing the propellant through an electric arc that heats it before it expands through a nozzle.

Core arc temperatures of 10 000 to 20 000 K mean that exhaust velocities of 5000 - 6000 m s^-1 (I_sp = 500 - 600 s) at efficiencies of around 40% are possible using catalytically decomposed hydrazine.

Arcjet thrusters entered commercial use for north-south station keeping on the Telstar-4 series of geostationary communication satellites in 1993. Higher power arcjets providing sufficient thrust for orbit transfer or primary propulsion manoeuvres have been demonstrated on test flights but problems with electrode erosion and availability of sufficient electric power have delayed their introduction for operational missions.

Inductively or Radiatively Heated Systems

Inductively or radiatively heated systems heat the propellant stream with some form of electrodeless discharge. The free electron component of the propellant gas is heated by the application of an oscillating electromagnetic field. Frequencies from low radio frequency up to microwave have been tested on the ground. No examples of this technology have yet been flown.

Electromagnetic Propulsion

Electromagnetic propulsion uses orthogonal electric and magnetic fields to apply a Lorentz body force to ionised propellant atoms, accelerating them out of the plane of the crossed fields. The electromagnetic propulsion techniques currently in use or being investigated include pulsed plasma thrusters and magnetoplasmadynamic thrusters.

Pulsed Plasma Thruster

The pulsed plasma thruster (PPT) with solid propellant is used for low power propulsion systems; typically with an average power less than thirty watts. Capacitor stored electrical energy is used to create a pulsed arc discharge across the face of a block of propellant, teflon in most implementations to date. The combination of thermal flux, particle bombardment and surface reactions ablates and ionises a small amount of the solid material (~ 1.5 µg J^-1). The peak discharge current is high, in the region of tens of kA. The self-induced magnetic field acts on the ions moving in the electric field creating the discharge to create a Lorentz body force that accelerates the plasma.

Teflon PPTs have enjoyed limited use in east-west station keeping and sun pointing applications. PPTs do not produce sufficient thrust to be considered as candidates to provide primary propulsion of future out of earth orbit missions and were not considered for SMART-1.

Magnetoplasmadynamic Thruster

Magnetoplasmadynamic (MPD) thrusters pass a large current radially outwards through a neutral plasma, from a central cathode to an annular anode. The radial current induces an azimuthal, circular magnetic field. The Lorentz body force acting on ions moving in the discharge current accelerates the plasma along the axis of the electrode structure. MPD thrusters operate at high power levels, kilowatts to megawatts, and provide high thrust with moderate specific impulse. A variant of this design uses an externally applied magnetic field to enable operation at lower power levels.

MPD thrusters have been tested in the laboratory and have flown on test and demonstration missions. MPD thrusters are an insufficiently mature technology and consume too much power to be considered as candidates to provide primary propulsion of future out of earth orbit missions and were not considered for SMART-1.

Electrostatic Propulsion

Electrostatic propulsion systems accelerate the ionised propellant by means of an electric field. The principal techniques are field effect electrostatic propulsion (FEEP), colloidal thrusters, and gridded ion accelerators.

Field Effect Electrostatic Propulsion

Field Effect Electrostatic Propulsion (FEEP) is based on the ability of a strong electric field (~ 10¹⁰ V m^-1) to extract individual ions from an easily ionisable metal, such as one of the alkali metals. The technique has been demonstrated in the laboratory using liquid caesium as the propellant. With an extraction/acceleration voltage of 10 kV, an exhaust velocity of 100 000 m s^-1 is possible (I_sp = 10 000 s). Caesium FEEP thrusters offer high efficiency (approaching 100%) but the available thrust is very small (1 µN to 5 mN tested on ground) and the thrust per unit power is also low (~ 15 µN W^-1). The low thrust to power ratio and the contamination and surface attack problems caused by the use of caesium have so far prevented the use of FEEP in practical operational applications.

Colloidal Thruster

Colloidal thrusters work by electrostatically accelerating charged, sub-micron diameter droplets of a conducting, non-metallic liquid. Early efforts were frustrated by an inability to obtain a high enough charge to mass ratio for the extracted droplets, leading to a requirement for very high acceleration voltages (> 100 kV), and non-uniformity of charge to mass ratio, giving rise to high beam divergence. These problems have been partially solved and colloidal thrusters with exhaust velocities of 10 000 m s^-1 (I_sp = 1000 s) have been demonstrated on the ground.

Colloidal thrusters are an insufficiently mature technology to be considered as candidates to provide primary propulsion for future out of earth orbit missions and were not considered for SMART-1.

Ion Engine

In gridded electrostatic ion accelerators, also known as ion engines, ions are produced in a magnetically contained ionisation chamber by a direct current discharge, radio frequency energy or tuned electron cyclotron resonance. The exit from the ionisation chamber is covered by a double-grid structure with a space between the grids of around half to one millimetre, across which the ion acceleration potential is applied. Ions that move near to the inner (screen) grid are extracted from the chamber and accelerated by the field between the grids. The ion optics are arranged so as to minimise collisions with the outer, accelerating grid. Electrons are extracted from the chamber by an anode and pumped by the power supply to an external cathode/neutraliser held slightly above the potential of the accelerating grid. The electrons from the cathode combine with the exiting ion stream to neutralise it. Neutralisation of the ion stream is necessary because ejecting charged particles from a spacecraft causes the vehicle itself to acquire a charge, which affects the operation of other spacecraft systems and may cause permanent damage. Also, without neutralisation the emerging ion beam would stall on its own internal potential profile.

Ion thrusters achieve exhaust velocities in the region of 30 000 m s^-1 (I_sp = 3000 s). ESA's EURECA spacecraft demonstrated the operation of RITA, an ion thruster employing radio frequency ionisation, in 1992. Ion thrusters have been in operational use since the mid nineteen-nineties for station keeping on geostationary satellites. In 1998, NASA's Deep Space 1 became the first interplanetary mission to use ion propulsion.

Ion thrusters suffer from low thrust density (available thrust per unit exhaust area) because the maximum ion current density that can be sustained is limited by space-charge distortions of the applied electric field. One advantage of the Hall effect thruster selected for SMART-1 compared to an electrostatic ion engine is that, as the plasma in the Hall effect thruster remains substantially neutral due to the presence of the electrons that constitute the Hall current, they are able to sustain higher ion current densities and hence offer greater thrust densities.

	Power for Electric Propulsion
	Hall Effect Thrusters