Rockets For The Red Planet

Engineers rethink how to get to Mars and back

12 min read
Franklin R. Chang Díaz
Photo: Randi Silberman

This is part of IEEE Spectrum’s Special Report: Why Mars? Why Now?

Franklin R. Chang Di\u0301azRocket Star: Franklin R. Chang Díaz, a former astronaut, is building a rocket that might get people to mars.Photo: Randi Silberman

From his corner office at Ad Astra Rocket headquarters near Houston, Franklin R. Chang Díaz hatches big plans. He’s tucked away behind a strip mall on a bland suburban street, but his mind is wandering the cosmos. He envisions multibillion-dollar mining operations extracting iron, cobalt, and platinum from asteroids for use in cities on the moon and Mars. He dreams of space infrastructures so evolved that astronauts freely roam the moons of Jupiter and Saturn. He sees parallel societies grown teeming and rich, and Earth gradually transformed into a grand nature preserve.

But first, he confides, he hopes to trade his comfy landing pad in Houston for an office on the moon.

If anyone can help launch a spacefaring society, it’ll be Chang Díaz. The former astronaut has spent more than two months in space during seven space-shuttle missions. Three times he has gazed down through his helmet’s mirrored faceplate at his white-swirled, blue-green ball of a home. Now he’s building the rocket engine that might make some of those galactic fantasies come true.

Decades ago, Chang Díaz, who holds a Ph.D. in applied plasma physics, concluded that chemical rockets were a dead end, owing to their modest performance specs and huge appetite for fuel. Voyaging in a chemical rocket is the celestial analogue of drifting around the world on a yacht that got its one burst of speed by charging out of port like an angry elephant. It’s heavy, it’s inflexible, and it breaks all the rules of sensible travel. So in the late 1970s he began developing an alternative technology he calls VASIMR, for “variable specific impulse magnetoplasma rocket.” In its most ambitious form, VASIMR would be a nuclear-electric rocket engine—a fission reactor with a plasma thruster that could potentially push people to Mars and back using a fraction of the propellant and time needed for a chemical rocket.

With a power plant similar to the ones on nuclear submarines, the plasma rocket could carry several people from Earth to Mars in 39 days, as opposed to what would be at least a 180-day journey on a chemical rocket, Chang Díaz says. The savings in food, water, air, tedium, and cosmic-ray exposure would be immense. In 2012, Ad Astra plans to test a prototype—using solar power rather than nuclear—on the International Space Station. An astronaut will spacewalk out to attach the 200-kilowatt engine, and if all goes well, it will bump the ISS into a more attractive orbit with about 5 newtons of thrust. The tests will begin to indicate whether VASIMR can figure in NASA’s grand plan to shuttle people and cargo to the moon and perhaps Mars over the next couple of decades. In particular, engineers will analyze two things: how efficiently the engine uses its electricity to produce plasma and how fast its radiator can siphon away excess heat.

On a hot, cloudless February day in Costa Rica, that radiator is undergoing intense scrutiny. Its home is a sleek white warehouse that hulks in a meadow of feathery grasses, an awkward edifice that looks like it dropped from the sky in a space-age remake of The Wizard of Oz . Next to the building, six cars are parked with sun shields propped against their front windshields.

A bumpy dirt road links the warehouse, with its zippy Ad Astra logo, to an unnamed highway, one of two thoroughfares that connect the city of Liberia to the rest of Costa Rica. Rental-car agencies hawking rugged vehicles line the highway. In the window of one agency, a poster advertises a “Race to Space.” Silhouetted runners glide across the surface of some space orb, with Earth hovering behind them on a field of luminous blue. In the foreground, Chang Díaz, who was born in Costa Rica, smiles benevolently in a bright orange space suit. The footrace is to raise money to build roads, but Chang Díaz’s engine may reach Mars before Liberia gets good roads.

Ad Astra’s warehouse lab in this Central American burg is the world’s foremost—and only—dedicated center for heat management in plasma rockets. With an average age of 28, the engineers make up a team as remarkable as it is improbable. The story begins in 2004, when Chang Díaz tapped his younger brother, Ronald, then running a construction company in the city of San José, to start up an Ad Astra office in Liberia. At age 42, Ronald embarked on a real-life Costa Rican version of Rocket Boys . He skimmed the best and brightest from local tech companies and Costa Rica’s universities. He added others as people showed up whose drive and aptitude appealed to him.

One electrical engineering grad appeared at the facility’s ribbon-cutting ceremony and refused to leave. Ronald hired him. The 21-year-old master of the machine shop, an immigrant from Nicaragua, was plucked from a local gas station, where he was an attendant. He’s now also the electrical technician, and he dabbles in computer-aided design.

On this sunny February day, the dozen engineers in the warehouse are scattered around a 50-kW version of the engine. They must design a lightweight thermal jacket for the thruster. The challenge lies in choosing a material that conducts heat well but electricity poorly, perhaps a ceramic made from silicon nitride.

The jacket would collect the heat between the magnetic fields and the thruster’s walls, radiating some of it back to the plasma and some of it into space. The engineers’ task is formidable: Many experts suspect that the combined weight of VASIMR’s power plant and radiator will bog it down too much.

To test their radiator, the engineers prepare to fire the thruster. They settle into chairs at a row of desks facing a vacuum chamber the size of a school bus. Attached to one side is the business end of the apparatus: permanent magnets, a radio-frequency generator, a tank of argon gas, and the tube where they will generate the plasma before venting it into the vacuum chamber. The argon is flowing, and the magnets are powered up.

“Cinco, cuatro, tres,” Jorge Oguilve-Araya, a lead engineer, chants into a walkie-talkie. “Dos, uno. Pulso!” The RF generator switches on and releases a torrent of RF waves into the argon stream. The gas heats up and ionizes, turning into a plasma of about 50 000 kelvin. Magnetic fields generated by the permanent magnets hold and channel the viciously hot material, protecting the thruster walls from melting on contact. A purplish light fills the vacuum chamber before fading to black.

There’s a similar setup in Houston, but with one more stage. Another antenna generates an electric field to heat the plasma to a million kelvin. When the ions’ rotation frequency matches the frequency of the field, the potential energy in the electric field changes into kinetic energy for the ions, accelerating them in a direction perpendicular to the magnetic field lines. This configuration forms a magnetic beach—waves on which the particles then surf their way out of the rocket.

Because it can modulate how the power is distributed between the two heating stages, this rocket has one unique and extremely desirable feature, one that explains much of the effort and expense: It can vary its specific impulse.

A rocket’s specific impulse reflects the efficiency with which it consumes its propellant, which depends heavily on the rocket’s mass. “Impulse” refers to a change in momentum, and it can become “specific” when divided by a mass. Dividing the rocket’s thrust by the amount of exhaust it produces per unit time results in a value whose unit is given in seconds.

Rocket engineers love the idea of variable specific impulse, because it allows a spacecraft to behave more like a race car, adjusting its acceleration at each turn around a track. Chemical rockets are fixed at a relatively low specific impulse of around 450 seconds. They need lots of propellant and can produce lots of thrust. Heading off to Mars, a chemical rocket would thrust for half an hour to escape Earth’s gravitational well and then coast the rest of the way. VASIMR, on the other hand, can run at specific impulses between 5000 and 15 000 seconds using deuterium or as low as 4000 with argon. For wandering the interplanetary voids, high specific impulse—or low thrust—is good: With highly efficient propulsion, the engine can keep firing until it reaches a high velocity, generating minimum thrust near the middle of the trip.

But first, the thruster must escape terra firma. For that to happen, the Costa Ricans must master the heat problem, and the Houstonians must resolve the closely tied power problem. The Houston thruster’s 200-kW rating is impressive when compared with other electric thrusters, but it’s not enough. The engine will need 5 to 10 megawatts—and possibly as much as 200 MW—to send people from Earth to Mars.

There’s no easy way out. Using existing technology—chemical rockets—a human trip would cost hundreds of billions of dollars. Previous missions have shown that a spaceship can deposit a package the size of a refrigerator on the Martian surface. Sporty rovers can explore their environs with minimal solar energy and a 19-minute communications delay. But hoisting fragile, needy humans to Mars and then returning them complicates the mission by an order of magnitude, if not two.

Chemical rockets move by virtue of a contained explosion. In one form, liquid hydrogen and oxygen are pumped from separate tanks into a combustion chamber. They react to produce water vapor and lots of energy, which blasts the vapor out through a nozzle and pushes the spacecraft in the opposite direction. Those tanks account for about 90 percent of a mission’s initial mass.

A space agency could send people on a round-trip jaunt to Mars using only chemical rockets. But that approach undervalues the basic metric of celestial shipping—that a trip’s cost often boils down to a spacecraft’s mass. Even if a crew traveled without fuel for the return trip, the passengers would be facing the most massive camping trip ever undertaken. Aside from Robert Zubrin’s spartan visions [see “How to Go to Mars—Right Now!” in this issue], most chemical-rocket scenarios are leviathan operations with numerous heavy-lift launches to heave all the components into orbit. Space tankers carrying only propellant would depart early to sit in Mars orbit, ready for the return flight. So would all the other necessities: a landing vehicle, some kind of power plant, rovers to let the astronauts explore, and a preassembled habitat—the Martian equivalent of a welcome mat and logs crackling in the fireplace.

Robert Braun, an aerospace engineering professor at Georgia Tech, estimates that the total mass in low Earth orbit would add up to almost eight International Space Stations, or about 1.8 million kilograms. To put that in perspective, the Saturn V rocket, which launched men to the moon and back and remains the biggest thing to leave Earth’s surface, could deliver 119 000 kg per launch. That works out to 15 Saturn Vs to lift the propellant, engines, and payload to start the outbound voyage.

By one common estimate, it costs US $20 000 to place 1 kg in low Earth orbit using a standard launch vehicle. So getting everything floating in the thermosphere would have a starting price tag of about $36 billion, or double this year’s budget for NASA, the world’s largest space agency. After getting to orbit, as many as 400 million kilometers of travel would remain.

Dozens of mission architects have drafted their own flight plans, but each one faces the same trade-off: The less beefy the propulsion system and its fuel, the less preposterous the mission starts to look. For human space transport, there’s only one plausible alternative to chemical reactions, and that’s nuclear power [see slideshow, "From Here to Eternity"]. In an electric rocket like VASIMR, the reactor’s heat would be converted into electricity. A competing configuration, called nuclear thermal propulsion, is more basic: A nuclear reactor heats a gas and blasts it directly out a nozzle. It doesn’t dabble in antennas or magnets or variable specific impulse. When official committees assess future rocket technologies, it always gets a nod. But nuclear rockets have one troubling feature: their radioactive exhaust.

NASA has always wanted a nuclear rocket. Almost immediately after the agency was formed in 1958, it began working on nuclear reactors for space, under a program known as Rover/NERVA, which stands for “nuclear engine for rocket vehicle application.” In the spring of 1969, just before Neil Armstrong planted his boot in the Sea of Tranquility, the NERVA team finished ground testing its first complete mock-up of a nuclear reactor, the NRX-XE. The reactor went through 28 start-and-shutdown cycles at the Nevada Test Site, where the United States tested nuclear bombs.

During the 13 years of its existence, the program’s engineers built and tested 20 reactors and nearly produced a flight-qualified propulsion system. They measured thrust and vented radioactive exhaust at an isolated spot known as Jackass Flats, bordered by mountains and mesas. They demonstrated systems with half the mass of a chemical rocket and a specific impulse of about 845 seconds. They tested engines that could get a crew to Mars and back in 80 days. But before the reactors could fly, the program ended. It was the 1970s, and political pressures were marginalizing space science.

The Soviet Union kept nuclear reactors in play a bit longer. Between 1965 and 1988, it launched a series of naval satellites with small reactors on board. At least two of them failed, releasing radioactive materials and spooking politicians worldwide.

To use nuclear reactors for a trip to Mars safely, a launch vehicle would deliver a spacecraft with three inactive nuclear engines to low Earth orbit. Around 220 nautical miles up, at roughly the altitude of the International Space Station, the reactors would start up. They’d run for no more than 45 minutes, producing about 330 000 newtons of thrust and kicking the ship beyond gravity’s grip. Like a chemical rocket, the vehicle would coast most of the way to Mars and then fire its engines briefly to decelerate. The vehicle would ease into orbit and be greeted by a lander vehicle. The lander would ferry the astronauts to the surface, where the real mission would begin.

At the peak of nuclear rocket research, engineers were reaching thrust levels of almost a million newtons, well beyond what they’d need. “We’re the only propulsion technology that I think is scaling down in size,” says Stan Borowski, an engineer pursuing nuclear thermal propulsion at NASA’s Glenn Research Center, in Sandusky, Ohio. In a typical nuclear rocket design, the fuel consists of graphite pellets mixed with particles of uranium-235 and bundled into fuel rods. Channels perforate the bundle, enabling hydrogen or helium coolant, which is also the propellant, to flow through. The nuclear reaction heats the rods and the propellant, which blasts out into space.

There is a problem, though. Tiny cracks can form in the core, releasing some of the uranium into the propellant. In space, radioactive sputter isn’t a big deal. But without ground testing, the reactors won’t ever get to space. “If we’re trying to sell the public on a Mars mission and the image is of us leaking a radioactive gas, that’s going to be a problem,” notes Steven Howe, the director of the Center for Space Nuclear Research at the Idaho National Laboratory, in Idaho Falls.

Because human exploration of Mars has yet to receive its mandate, none of the work-arounds that could save nuclear propulsion is getting more than theoretical consideration. No space agency is building big enough reactors, and there’s a solid chance they will live on just as a testament to the scientific exuberance of the 1960s. Any dream of humans harnessing the great beyond would be mothballed, with an unbuilt Martian settlement languishing next to a missing philosopher’s stone and a nonexistent flying car.

That’s not to say humans will never make it. A chemical rocket may indeed deliver a small crew to Mars [see sidebar, "Exotic Options for Chemical Rockets" ]. But it wouldn’t be for much more than bragging rights: Humans went all the way to Mars and they didn’t even get the lousy T-shirt or the reinvention of life science—and they certainly didn’t get the space colony.

Ad Astra’s vasimr shares some of the technical and political problems of nuclear thermal propulsion. Only a nuclear reactor can deliver the megawatts needed for a Mars mission. But given a reactor, Chang Díaz is confident he can easily convert at least 60 percent of its electrical power into rocket power. For now, he plans to build his business around closer targets, such as solar-powered moon visits and trips to investigate near-Earth objects.

The ventures would be scaled-up versions of the activities now performed by Hall thrusters and other ion engines, whose related technologies propel space probes and nudge satellites. Ion thrusters have taken European Space Agency and Japanese probes to the moon and to an asteroid named 25143 Itokawa, and one spacecraft is now hurtling toward another asteroid and the dwarf planet Ceres. Solar panels power the thrusters, which rarely use more than a couple of kilowatts.

Size is everything. Says Brent Sherwood, a space architect at NASA’s Jet Propulsion Laboratory, in Pasadena, Calif.: “The real question is, how do you scale up from a thing that’s got a blue glow in a lab to a thing that sends half a dozen people to Mars and back?” To test higher-powered VASIMRs, Ad Astra will need a vacuum chamber even bigger than the aluminum monolith it has in Houston. In fact, it would be so big and expensive that Chang Díaz figures he might as well test the rocket in space.

Under this model, Ad Astra employees would blast off on assignment to the moon for a few months at a time, touching down near a facility surrounded by vast arrays of solar panels. Working within the moon’s peculiar schedule—two weeks of light followed by two weeks of night—the lunar operatives would fire their engine, accumulating performance data in preparation for an eventual flight to Mars.

Then will come the hard part. No amount of testing can mimic humanity’s first flight to Mars. That knowledge gap could be Ad Astra’s greatest challenge. Les Johnson, deputy manager for the Advanced Concepts Office at NASA’s Marshall Space Flight Center, in Huntsville, Ala., puts it this way: “Imagine getting in a Winnebago with your four best friends and saying, ‘We’re not going to leave this Winnebago for three years.’ And between us and complete death is a thin aluminum skin, and the lowest bidder is going to send us out.” Then he pauses. “I would be a bit more mundane,” he says. That is, he would rather see the advanced technology invested in better life-support systems, perhaps in the form of artificial gravity on board a familiar chemical rocket.

In rocket science as in life, differences of opinion are often cultural. On a table next to Chang Díaz’s desk in Houston lies a DVD of a television show he grew up watching in the 1950s. It’s about an eccentric scientist named Captain Video who defended law and order by jumping into his spaceship from his mountain retreat. Before each episode, a voice booms across the mountaintops: “Captaaaaaaiinn Video! And his Videoooo Rangers!” Then antics ensue. Chang Díaz’s name may not echo across the suburbs of Texas, but in the world of rockets, his presence is booming.

For more articles, go to Special Report: Why Mars? Why Now?

To Probe Further

For more, see “Costa Rica’s Radical Rocket.”

For more on propulsion, see “Warp Speed, Mr. Sulu.”

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions