Before delving into what electric propulsion (EP) is, we should consider why it is even of interest. Space missions are characterized by the kinetic energy (velocity) a vehicle must achieve to reach a destination. For example, an object must reach a velocity of roughly 8 km/s (18,000 mph) to reach Earth orbit. If an object is to escape from the Earth's gravitational influence and enter interplanetary space (e.g., for a trip to Mars), the object must be accelerated to a velocity of at least 11 km/s (25,000 mph). Therefore, when NASA sends probes to deep space, these vehicles must be accelerated to velocities in excess of 11 km/s. Today, large rockets are used to propel probes to these speeds. However, the faster the desired speed of the probe (e.g., to reach a distant planet such as Pluto in a "reasonable" amount of time), the smaller the probe must be. To understand the reason behind this, one must consider the so-called Rocket Equation (equation 1), perhaps first developed by Ziolkovsky in Russia:

	Note
	Equation 1 assumes impulsive engine operation, where the "engine-on" time is very small in comparison to the trip time. This approximation allows one to neglect the effect of gravity losses, which depending on the mission, may be due to the Earth, the Sun or both. Electric propulsion systems, with their long operating times, violate this assumption. To account for gravity losses, the true exhaust velocity (41 km/s) was multiplied by 0.8. This 20 percent penalty is reflected in the figure below.

The Rocket Equation relates the vehicle mass prior to engine operation (initial mass,M_i ), when the vehicle is full of the propellant it will use to accelerate to its finalvelocity (ΔV), the final mass of the vehicle (M_f ), and the exhaustvelocity of the propulsion system (Ue). The equation shows that the mass ratio(M_f  /M_i ) decreases exponentially with the ratio of ΔV overthe propellant exhaust velocity, which means that for ΔVs larger than Ue, most of the initial mass of the vehicle is propellant.

Consider Arthur C. Clarke's 1968 science fiction classic 2001: A Space Odyssey, wherein a team of American astronauts are sent on a nine-month, half-billion-mile mission to Jupiter to find the origin of the alien culture that planted a monolith on the Moon millions of years ago (when dug up in about 1999, the monolith sent a powerful radio transmission to Jupiter). The giant spacecraft featured in this story, Discovery 1, would need to achieve a ΔV of at least 50 km/s to reach Jupiter in nine months. Clarke, trained as an engineer, postulated that Discovery 1 would have to use a propulsion system much more advanced than chemical rocket motors. In his book Lost Worlds of 2001, Clarke stated that he conceived the giant craft as being nuclear powered with some form of electric propulsion. To understand why Clarke arrived at this conclusion, we will revisit equation 1 and apply it to Clarke's voyage to Jupiter.

Mass ratio vs. ΔV for propulsion exhaust velocities of 4.5 and 41 km/s.

This figure plots equation 1 for propellant exhaust velocities of 4.5 km/s (chemical) and 41 km/s (electric) through a ΔV up to 50 km/s. In order for the chemically propelled spacecraft to achieve a ΔV of just 20 km/s, more than 99 percent of the initial mass of the vehicle must be propellant. For a 50 km/s ΔV, the ratio of the delivered mass, which includes the tank mass, vehicle structure and engines, as well as the actual payload, to the initial mass is less than one part in 150 thousand (<1/150,000) for the vehicle using chemical propulsion (CP). This means that for every kg of delivered mass, the equivalent of 150,000 kg (a medium-sized airliner) is needed at the start of the mission. Clearly such a mission is impractical (if not impossible) with chemical propulsion. In comparison, the mass ratio for the electrically propelled craft is over 20 percent (>1/5) for a ΔV of 50 km/s. The propellant savings in switching from CP to EP in the case of the 50 km/s ΔV is truly mission enabling.

What the early pioneers of rocketry had to say about EP
ften associated with his seminal work on liquid rocket propulsion, and rightfully so. However, it is often forgotten that Professor Goddard was a physicist and in fact conducted experiments with electrical gas discharges. The earliest known technical discussion on the concept of EP is attributed to passages in Goddard's lab notebook written while he was conducting experiments with gas discharge tubes in 1906. Goddard noted that charged particles were being accelerated to great velocities by the electric fields within the tube and yet the tube walls remained relatively cool. In contrast, no known material could tolerate the temperatures needed to propel gas particles at similar speeds through heating; that is, in a chemical rocket motor. Goddard went on to further conclude that particles accelerated through electrostatic means (i.e., with electric fields) could be the basis of a high-exhaust velocity propulsion system.

Over the course of six years, Goddard frequently revisited the notion of using electrostatically accelerated particles for spacecraft propulsion. He even went so far as to postulate that high-velocity streams of negative [electrons] and positive [ions] particles could be "energized" by solar-electric power supplies to provide thrust for an interplanetary spacecraft. This spacecraft would use CP to reach Earth escape velocity but then use EP both to accelerate further and then to decelerate once its target was reached. Goddard postulated that the source of the ions could come from exposing alkaline atoms, such as mercury or cesium, to hot tungsten surfaces. The prophetic nature of Goddard's postulations is almost frightening. Some 82 years after Goddard's notebook entry, DS1, a solar-powered interplanetary spacecraft that uses electrostatically accelerated xenon ions to provide primary propulsion (via a device called an ion thruster), was launched to Earth escape velocity by a chemically propelled Delta II rocket. Moreover, earlier versions of ion thrusters created ions by either shooting electrons toward mercury ions (the electrons emerged from tungsten filaments) or by having cesium atoms come in contact with a heated tungsten wall.

By 1916 Goddard and his students were conducting perhaps the world's first electric propulsion experiments with ion sources. Four years later Goddard devoted passages of his technical reports to his EP experiments.

While little is known about Ziolkovsky's thoughts on electric propulsion, it is quite apparent that Oberth, like Goddard, was a big fan. In fact, Oberth dedicated an entire chapter of his seminal 1929 book on space travel and rocketry, The Road to Space, to EP. He wrote on EP again in his 1957 book Man into Space. Like Goddard, Oberth felt that the electrostatic acceleration of ions was the key to achieving high exhaust velocity propulsion in space. Oberth describes a possible ion propulsion system in Man into Space as porous plates electrically biased to high voltage that provide a finely distributed "spray" of high-speed charged particles. He states that such an engine could use almost anything as propellant, even the refuse from the spacecraft crew! He notes that the thrust would likely be small (in comparison to the vehicle mass) but steady, and that the engine would operate for a long period of time.

As in the case of Goddard, Oberth's vision of the future was truly uncanny. If we consider DS1, which we will discuss in greater detail in Session 4, the vehicle's ion engine used two dished plates, each with over 15,000 holes, to accelerate the xenon ions. The two plates (called grids) were roughly 30 cm in diameter and spaced less than a millimeter apart. Approximately 1,300 V was placed across the two grids to accelerate the xenon particles to over 30 km/s. As far as using "crew refuse" as propellant, early designs of the International Space Station called for drag make-up propulsion to be provided by electrically heating human waste (mostly water) to high temperature and expanding the vapor through a nozzle. This process would be accomplished in an electric propulsion device called a resistojet. Since the early 1980s, communication satellites have used resistojets (but not with human waste as propellant!) to maintain orbital position.

While Oberth and Goddard recognized the potential payoff electric propulsion could have to interplanetary flight, it was Wernher von Braun who sanctioned the first serious study on EP. In 1947, at Fort Bliss, von Braun assigned a young engineer named Ernst Stuhlinger the task of giving Professor Oberth's early concepts of electric spacecraft propulsion "some further study." Von Braun went on to tell Stuhlinger, "Professor Oberth has been right with so many of his early proposals; I wouldn't be a bit surprised if one day we flew to Mars electrically!" Though reluctant, the dutiful Stuhlinger proceeded to investigate the feasibility of using electric propulsion to enhance space travel. Some 15 years later, Stuhlinger published a book entitled Ion Propulsion for Space Flight and directed NASA Marshall Space Flight Center's work on arcjet and ion propulsion systems. Clearly Stuhlinger became a believer in EP. While Stuhlinger was laying the technical groundwork for ion thruster development between 1947 and 1960, a number of engineers and scientists were writing on the subject.

Two prominent American rocket engineers of the late 1940s, L. R. Shepherd and A. V. Cleaver, performed a rigorous analysis on the virtues of electric propulsion for space travel. They considered using a nuclear reactor to heat a propellant such as hydrogen or to provide electric power to an ion thruster. These propulsion approaches are now called nuclear-thermal propulsion (NTP) and nuclear-electric propulsion (NEP), respectively. They concluded that NEP generated much higher exhaust velocities than NTP systems and with considerably less waste heat to reject from the craft. They predicted that exhaust velocities up to 100 km/s would be possible with an NEP system (vs. 10 km/s with a hydrogen-propellant NTP engine). They also suggested configuring the ion exhaust as a beam of parallel streams (as opposed to Oberth's spray concept), concluded that an electron source would be needed to prevent the spacecraft from charging up as positive ions are expelled by the engine, and considered heavy particles to be preferential to light particles to provide high thrust at reduced beam current.

One question that propelled EP naysayers--or at least caused people to question EP's true utility in space travel--was the low inherent thrust-to-weight ratios of electric engines. While chemical propulsion systems have thrust-to-weight ratios that often exceed unity (indeed this is needed to get a vehicle off the ground from a vertical launch position), EP systems are expected to havethrust-to-weight values thousands of times smaller than this. In 1953 H. S. Tsien designed trajectories and thrust alignment procedures for low-thrust, EP-propelled spacecraft. Tsien showed that thrust-to-weight ratios as low as 10^-5 (1 in 100,000) are sufficient to change the trajectory of a space vehicle over a realistic period of time.

In 1954 Stuhlinger presented the first comprehensive study of the major components associated with an electrically propelled spacecraft. Stuhlinger was the first to show the relationship between the optimum exhaust velocity, the desired ΔV, and the specific mass of the power source rigorously. The last quantity, often called α (power supply mass [kg] divided by power source output power [kW]), is a key parameter both for space vehicle design and for identifying the necessary operating characteristics of the EP engine. Stuhlinger also noted that propellants with large mass-to-charge ratios, such as mercury (or xenon) are desirable to minimize the size of the ion engine for a given thrust level.

Over the course of the next several years, a number of people looked at multiple aspects of EP space travel, including the notion of varying exhaust velocity to optimize the trajectory (J. H. Irving) and deriving an expression for the optimum exhaust velocity as a function of α and ΔV (D. Langmuir). Project Snooper was started in 1957 by M. I. Willinksky and E. Orr to design a nuclear-electric space vehicle that would carry scientific instruments to deep space. The Snooper vehicle would use chemical propulsion to achieve escape velocity but NEP thereafter. That year also marked a huge transition in the field--from concentrating on overall vehicle attributes to identifying specific technical problems that needed to be addressed. In other words, the year Sputnik I was launched was also the year the scientific community accepted EP as a viable propulsion technology and transitioned from proving EP's worth to solving technical challenges that impeded EP's implementation. These technical challenges included ion beam generation and neutralization, developing "light weight" (low α) power sources, and intense arc stabilization.

	Relevant Links
	Air Force Office of Scientific Research (AFOSR) (www.afosr.af.mil) Office of Naval Research (www.onr.navy.mil) Defense Advanced Research Projects Agency (DARPA) (www.darpa.mil)

To this end, the Air Force Office of Scientific Research (AFOSR), which is still one of the largest sponsors of EP research in the nation, issued the first electric propulsion research grants in 1957. It was also at this time (1957-58) that engineers began to consider other mechanisms--beyond ion acceleration--for electric propulsion. Besides ion propulsion, researchers began conceptualizing systems that use electric arcs to heat gases to high temperature and thus provide thrust by expanding and accelerating the hot gases through a nozzle (as is the case of a chemical rocket), and that use electromagnetic forces to accelerate a highly ionized gas (plasma). By 1959 the Office of Naval Research (ONR), the Army Ballistic Missile Agency (ABMA) and the Advanced Research Projects Agency (ARPA) had all issued research grants in EP. By 1960 almost every large rocket and aircraft firm had an active EP program.

Some of the better-known companies from that time that are still active in the aerospace field today include Lockheed (builder of launch vehicles and communication satellites and the first US company to use EP on a commercial spacecraft, now Lockheed-Martin), Rockwell (builder of the Saturn V launch vehicle and the space shuttle, now part of Boeing), Thiokol Chemical Corporation (builder of the Space Shuttle Solid Rocket Booster, now Morton-Thiokol), General Electric, Aerojet-General (builder of chemical rocket and electric propulsion engines for launch vehicles, space probes and satellites, now called simply Aerojet), and Hughes Aircraft (builder of communication satellites and the first to use ion propulsion in space, now Boeing). In 1958 the National Advisory Committee for Aeronautics (NACA), the predecessor of NASA, established an EP program at the Lewis Flight Laboratory (now the John Glenn Research Center) outside of Cleveland, Ohio. By 1960 EP programs had sprouted up at the Jet Propulsion Laboratory in Pasadena, California, and at the Marshall Space Flight Center in Huntsville, Alabama. By early 1962 the NASA Lewis Research Center was designated as the lead center for Electric Propulsion Research within the agency, a distinction that exists today, some 40 years later! By 1964 NASA Lewis Research Center developed two ion thrusters for testing in space under the SERT I (Space Electric Rocket Test I) program. The thrusters were to run off of batteries mounted on a small capsule that was launched into a 50-minute ballistic flight by a Scout missile. Although one of the thrusters failed to operate, the other was able to answer the question SERT I was developed for: Do ion thrusters operate in space as well as they do in vacuum chambers on the ground? In particular, ion beam neutralization from an auxiliary electron stream in space versus on the ground was of great interest. Over the next few years, a number of space missions would confirm that the performance of EP devices could be measured accurately on the ground.

In 1970 NASA flew SERT II. While SERT I was developed to address ion beam neutralization concerns and whether ion thrusters operate as efficiently in space as they do on the ground, SERT II was designed to demonstrate long-term ion thruster operation in space with a solar power source. On February 3, 1970, the SERT II spacecraft was launched into polar orbit, which permitted the vehicle to receive continuous sunlight (and hence power) during its entire mission. SERT II carried two mercury-propellant ion thrusters, one of which operated for more than five months and the other for nearly three months. Each engine was 15 cm in diameter, consumed 850 W of power and produced 28 mN (~0.006 pounds) of thrust. Extended ion thruster restarts were conducted from 1973 to 1981. The engines were started more than 200 times in space and confirmed that ion thruster performance could be both measured on the ground in vacuum chambers and predicted by monitoring grid voltages and beam current. Subsystems of the SERT II spacecraft such as solar arrays were tested through 1991, making the 21-year mission of SERT II one of the longest in history.

Between 1970 and 1990, EP research in the United States focused on mercury- and later xenon-propellant ion thrusters, hydrazine-propellant resistojets and arcjets, hydrogen-propellant arcjets and magnetoplasmadynamic (MPD) thrusters, Teflon-propellant pulsed plasma thrusters (PPTs) and argon-propellant MPD thrusters and Hall thrusters. The 1990s marked a number of significant events for electric propulsion development worldwide. These events included:

• Lockheed-Martin launching a communication satellite that used hydrazine-propellant arcjets for stationkeeping (i.e., keeping the satellite in its proper orbit and orientation with respect to Earth);
  
  
• Hughes launching the world's first commercial ion engine on the communications satellite, PanAmSat 5;
  
  
• DS1, NASA's first deep-space mission to use ion propulsion; and
  
  
• The collapse of the Soviet Union, which resulted in an influx of technical information on their EP development activities.

The principal EP product to come westward from the Soviet Union was the Hall thruster. Because of its performance characteristics (e.g., exhaust velocity and efficiency) that are ideal for commercial applications in Earth orbit, the Hall thruster has in effect supplanted the hydrazine arcjet (and perhaps to some degree in the future, the ion thruster) as the propulsion system of choice for stationkeeping and orbit raising of large commercial space vehicles. We will now delve into the operating principles of electric propulsion systems in the next session and use missions or applications (e.g., DS1) to illustrate the state-of-the-art of a particular EP technology.

Editor's Note: You have just completed Part I (sessions 1-3) of this seminar. Part II (sessions 4-6) discusses the specific scientific operations of different electric propulsion systems and may be advanced for some learners. We recommend that the learners continuing on to part II have a basic knowledge of and interest in physics and mathematics.