The Little Rocket Engine That Could


CubeSats are space age hitchhikers, miniature spacecraft that fly into orbit aboard rockets whose primary payloads are full-size satellites. Measuring about 10 centimeters on each side and weighing less than 1.5 kilograms, CubeSats often ride for free. And, like their human counterparts on Earth, they are often dropped off short of where they want to go.

Because beggars can’t be choosers, CubeSats—which could perform tasks as varied as monitoring disasters and repairing orbiting structures—must use their own propulsion systems to get to their destinations; to make the adjustments needed to stay there or go to another location; or even to reach escape velocity and travel to interplanetary space.

Paulo Lozano and his team at MIT’s Space Propulsion Lab have developed a unique kind of rocket engine for these microsatellites. Dubbed the ion electrospray propulsion system, the electric engine fires tiny streams of ions that push these mini-spacecraft into desired orbits and keep them there. 

Using semiconductor manufacturing technology, Lozano’s team creates chip-sized thruster modules that measure only 10 x 10 x 2.5 mm and could comfortably fit on a dime. An engine that controls yaw or pitch might use four modules, while a main propulsion engine would house many more, depending on the amount of thrust required.

The thruster modules themselves consist of an array of hundreds of small volcano-like cones called emitters. Instead of burning chemical fuels, they accelerate ions out of openings so small, their diameters are measured in nanometers and their thrust in nanonewtons.

“If you take a mosquito, cut off its antenna and divide it into maybe 50 sections—one of those sections would be a nanonewton in weight,” Lozano said.

Because a pound of thrust translates into 4.8 billion nanonewtons, it would take about a million billion ion engines to produce the thrust of just one of the five rocket engines on the first stage of the Saturn V moon rocket. 

Yet, given enough modules firing over a long enough time, an ion engine will vault a CubeSat from a low earth orbit (2,000 km or lower) into a 36,000 km geosynchronous orbit, or even beyond the clutches of Earth’s gravitation and onto the moon or other planets. It can do it with only 150 g of fuel and still leave 70-90 percent of the CubeSat free for critical sensors and electronics.

No other propulsion system comes close.

The Anti-Rocket

For conventional space rockets to escape Earth’s gravity, they have to generate a lot of thrust and burn a lot of fuel. A rocket that relies on chemical combustion must carry 20 to 40 times more fuel than the weight of the payload. Starting from a launchpad, it will burn through that fuel in minutes to reach escape velocity, just over 40,000 km per hour.

Lozano developed his electrospray engines for spacecraft that are already in orbit and that no longer have to do that type of heavy lifting. They can travel and maneuver with far less thrust.

In fact, ion engines use fuel so parsimoniously that they can fire for prolonged periods, shut down, and then fire repeatedly without fully depleting their reserves. The thrusters accelerate ions to many times the velocity of a chemical rocket’s exhaust, producing more thrust than might be expected from such a small stream of ions. As long as time is not an object, firing long bursts of high-speed ions provides all the thrust needed to accelerate CubeSats into higher orbits and beyond.

While rocket scientists typically describe engines as “firing,” this is misleading. “Firing” is to ion propulsion what “filming” is to digital videos.  Electrospray engines do not ignite. They have neither the bell-shaped nozzle that characterizes rocket engines, nor the valves, regulators, and pumps.

Instead, ion engines use passive capillary action to wick propellant—an ionic liquid such as a salt solution—from a plastic holding tank through a porous substrate and up to the cone emitters. There are no moving parts. Since ionic liquids already contain ions, there is no need for a reaction chamber to ionize particles.

The cone-shaped emitters accelerate ions through an electrical field generated by the CubeSat’s batteries, which are recharged by solar panels. It takes a mere 5 W of electricity—at voltages up 1,000 V—to produce a field at the top of the thruster module’s emitters, where the propellant rushes to meet the vacuum of space.

The trick to building a successful ion electrospray propulsion system, Lozano explained, is to increase thrust density by jamming together as many emitters as possible. A single 1x1 cm module may contain 400 or more emitters. Add enough modules to a 10 cm x 10 cm CubeSat surface and total thrust approaches the millinewton range.

“They produce little force, but because they can fire for a long time, you accelerate the spacecraft to a velocity that would be impossible to get with a chemical engine,” Lozano says. “That is a big value.”

Electrospray engines also differ greatly from another form of ion propulsion, plasma ion, which also eschew chemical combustion for the efficiency of the electron. Plasma engines have repositioned Boeing satellites already in geosynchronous orbit since the 1990s. Last year, for the first time, they pushed two Boeing satellites from low Earth to geosynchronous orbits.

Like ion electrospray engines, plasma engines fire for long periods of time. While they take up less room than chemical rockets, which are 90 percent propellant by weight, plasma engines are still bulky. To start with, they require fuel tanks of compressed fuel, usually an inert gas. Xenon is a popular choice because it compresses well for storage, ionizes easily, and has high mass that produces more thrust than would a lighter atom when accelerated.

A regulator metes the gas through pipes and into a reaction chamber. There, a discharge cathode injects high energy electrons into the gas, producing a plasma. An electromagnetic field contains the plasma, which is then accelerated through an electrically charged grid at very high speeds.

It takes power to operate the discharge cathode, magnetic containment, and acceleration grid. Boeing’s electric satellites operate at 3 to 9 kW, compared to 5 W for Lozano’s ion electrospray system.

“You have thrust that is about an order of magnitude higher than what we have now. If you wanted to substitute that thruster for ours, you would need an area about 10 times larger,” Lozano said. 

At first, Lozano tried to miniaturize plasma engines to fit CubeSats, but they were too complex to fit into their 1,000 cubic centimeter volume.  He turned to ion electrospray engines as an alternative. Without valves, pipes, pumps, and pressurized tanks, they are to conventional and plasma electric rockets what thin LED displays are to cathode ray tube TVs.


In some ways, Lozano’s transition to ion engines mirrored his own journey from gunpowder to electricity and then space travel as a child in Mexico City.

“It was kind of normal, during the holidays, for people to set off fireworks all the time,” he said. “We would get their gunpowder for our own explosives. We blew up a lot of stuff.”

But as Lozano, now a MIT associate professor of aeronautics and astronautics, grew older, he was drawn more towards electricity than explosions, spending time in bookstores because public libraries in Mexico City were scarce. While electromagnetism might seem safer, it too had its hazards.  “I got shocked many times,” he recalled.

He also fell in love with space travel, thanks in large part to Star Trek, Star Wars, and Battlestar Galactica.

“In all the science-fiction movies and television programs, for me the most exciting part was always the engines—Star Wars’s TIE and X-Wing fighters—and how they were able to move around so quickly,” he said. 

It’s not surprising that Lozano merged his love for electricity with his interest in rocket engines. After all, the “TIE” in TIE fighters stood for “Twin Ion Engines.” Building the ion drive, Lozano said, “has fulfilled one of my dreams.”

While no one is going to confuse a CubeSat with a TIE fighter, the tiny satellites are surprisingly versatile. This is because ion engines can drive and reposition CubeSats during long missions and still leave lots of room for electronics and even mechanical devices.

Developers hope to take advantage of this flexibility. They are contemplating missions that range from removing debris and nonfunctioning satellites from orbit to nudging existing satellites onto new flight paths. CubeSats, working alone or in groups, could become the maintenance staff of space, inspecting, docking, assembling, and repairing orbiting structures. They could even be used to explore interplanetary space.

Last year Lozano and his team of a dozen post-docs, graduate students, and undergraduates, sent three of their engines to NASA for evaluation. Engineers at NASA Glenn Research Center in Cleveland, Ohio, are now putting the engines through their paces to understand their behavior. 

“Electrospray thrusters are very new. But more and more people are realizing that they may be the way to go in electric propulsion, because they have so many advantages,” Lozano said.

A big advantage, when asking for permission to hitchhike a ride into orbit, is that ion electrospray propulsion engines cannot explode and destroy a rocket’s primary payload.

“By definition, chemical thrusters can blow up and people don’t want to put a $500 million satellite in danger,” Lozano says.

Even plasma engines are vulnerable because they store gaseous fuel under high pressure. Ion engines, on the other hand, do not contain combustible materials, pressurized containers, or even moving parts. Unlike the firecrackers of Lozano’s childhood, they are very, very safe. 

“If you can demonstrate that what you have cannot blow up, then people are happy with you,” he said.

Scaling Up

A second advantage of ion engines is their modularity and, consequently, scalability. Need more thrust?  Just add more modules, Lozano said.  A small CubeSat, for example, might host four thrusters to handle attitude control and main propulsion.  A larger CubeSat might need 16 for these functions.  The only limitation is the surface area on which the modules can be mounted.

Even then, engineers could design satellites to provide more area, if needed. They could, perhaps, add thrusters to pop-out structures that deploy like solar panels once in orbit. 

Another approach, which Lozano and his team are investigating, is to find better propellants.

“There are hundreds and thousands of possible propellants described in the literature, so it is very unlikely that we are using the optimal one,” he said. 

While CubeSats are a near-term opportunity, Lozano aspires to greater things. Ultimately, he hopes to make ion engines powerful enough to propel full sized satellites that weigh thousands of kilograms.

“We are about an order of magnitude lower in thrust density compared to the plasma thrusters that put the Boeing satellites into their orbits,” Lozano said.

A big obstacle to powering such large objects is doing the one thing that ion electrospray engines cannot do: Get satellites moving quickly.  Conventional chemical engines can boost navigational and communication satellites from low earth to geosynchronous orbits in a few days of repeat burns.  Lozano’s ion thrusters would need weeks, even months, to make the transition.

Still, that might not be such a bad thing, considering the weight savings possible with ion engines and the long lifespan of these spacecraft, Lozano said.

“After all, the satellites are going to survive for 15 years or so. Their owners really aren’t going to care if they can’t invoice their use for the first three months,” he said.

Lozano is working with a spinoff company, Accion Systems (Accion is short for “accelerated ion”), to commercialize ion engines. The company’s executives are mostly youthful space propulsion specialists. In addition to Lozano, the advisory board includes Steve Isakowitz, the president of Virgin Galactic, and Bill Swanson, the former chairman and CEO of Raytheon.

In 2015, Accion signed a $3 million Department of Defense research contract, cashed its first commercial checks, and won a Fortune magazine contest by convincing judges it could eventually grow into a $1 billion company.

More to the point, Accion is readying hardware that is scheduled to fly one mission this year and a second in 2017. Next year’s mission features a larger ion engine that could take a CubeSat beyond the pull of Earth’s gravity.

While Lozano is notably reticent about mission details and what entity is underwriting them, he says that the flight, if successful, will demonstrate the ability of ion engines to send CubeSats “on escape trajectories and interplanetary maneuvers.” He also confirms that the target of the 2017 mission is the moon, noting that “we’re not going try to land on it—but we might hit it.”

Which is certainly a long way for a hitchhiker to travel.

Greg Freiherr is a Wisconsin-based technology writer.

As long as time is not an object, firing long bursts of high-speed ions provides all the thrust needed to accelerate CubeSats into higher orbits and beyond.


August 2016

by Greg Freiherr