In early October, a dead Soviet satellite and the abandoned upper stage of a Chinese rocket narrowly avoided a collision in low Earth orbit. If the objects had crashed, the impact would have blown them to bits and created thousands of new pieces of dangerous space debris. Only a few days prior, the European Space Agency had published its annual space environment report, which highlighted abandoned rocket bodies as one of the biggest threats to spacecraft. The best way to mitigate this risk is for launch providers to deorbit their rockets after they’ve delivered their payload. But if you ask Jeffrey Manber, that’s a waste of a perfectly good giant metal tube.
Manber is the CEO of Nanoracks, a space logistics company best known for hosting private payloads on the International Space Station, and for the past few years he has been working on a plan to turn the upper stages of spent rockets into miniature space stations. It’s not a new idea, but Manber feels its time has come. “NASA has looked at the idea of refurbishing fuel tanks several times,” he says. “But it was always abandoned, usually because the technology wasn’t there.” All of NASA’s previous plans depended on astronauts doing a lot of the manufacturing and assembly work, which made the projects expensive, slow, and hazardous. Manber’s vision is to create an extraterrestrial chop shop where astronauts are replaced by autonomous robots that cut, bend, and weld the bodies of spent rockets until they’re fit to be used as laboratories, fuel depots, or warehouses.
The Nanoracks program, known as Outpost, will modify rockets after they’re done with their mission to give them a second life. The first Outposts will be uncrewed stations made from the upper stages of new rockets, but Manber says it’s possible that future stations could host people or be built from rocket stages already in orbit. In the beginning, Nanoracks won’t use the interior of the rocket and will mount experiment payloads, power supply modules, and small propulsion units to the outside of the fuselage. Once company engineers have that figured out, they can focus on developing the inside of the rocket as a pressurized laboratory.
Rockets headed to orbit are launched with at least two stages, each equipped with its own propellant tanks and engine. The large first stage boosts the rocket to the edge of space before decoupling and falling back to Earth—or, in SpaceX’s case, landing on autonomous drone ships in the ocean. The smaller second stage brings the payload up to orbital speed before releasing it. At that point, the upper stage typically has just enough fuel left to fire its engine so that it plummets back to Earth. If the upper stage doesn’t do a deorbit burn, it will keep circling the planet as an uncontrolled satellite.
The Nanoracks team is targeting these upper stages for development because they already have many of the qualities that are needed for a space station. A rocket’s fuel tanks are designed to hold pressure, and they’re made out of incredibly durable material to withstand the rigors of launch. They’re also roomy. The upper stage of SpaceX’s Falcon 9 is 12 feet in diameter and around 30 feet tall, which is enough space to make a New York apartment dweller jealous.
But these tanks need a little sprucing up before they can host experiments or astronauts. The first step is to vent any remaining fuel to prevent an explosion. Then, the robots take over. These automatons will attach necessary components like solar panels, surface-mounted connectors, or small propulsion units. Nate Bishop, the Outpost project manager at Nanoracks, says the company will do several small in-space demos before attempting to convert a full upper stage into a functioning space station. “Right now, we’re not really modifying anything,” says Bishop. “We’re focused on showing we can control the upper stage with attachments. But in the future, just imagine a bunch of little robots going up and down the stage to add more connectors and stuff like that.”
There’s just one problem—no one has ever demonstrated the core metalworking and fabrication techniques needed to convert a space station in orbit before. Next May, Nanoracks will change that during its first Outpost demonstration mission. The company has developed a small chamber that will be deployed with several other payloads as part of a SpaceX ride-share mission. Inside the chamber, a small robotic arm tipped with a rapidly spinning drill bit will cut three small pieces of metal made from the same materials used in rocket fuel tanks. If the experiment goes well, the tool should be able to make a precise cut without generating any debris. It will be the first time that metal was ever cut in the vacuum of space.
The fundamental challenge of converting rockets in orbit is understanding how materials react to the space environment. For example, the temperature of a material can differ by hundreds of degrees if one side is facing the sun and the other side is facing away. Without going to space to try it, it can be difficult to predict how that material will react to standard manufacturing techniques like cutting or welding. Other techniques, like making thin film materials for solar panels, require an ultra-pure environment to prevent imperfections. Although space is a vacuum, it still contains a substantial amount of dust and radiation that could interfere with conventional manufacturing processes exported from Earth.
“It’s remarkable how little we still know about manufacturing in space after 70 years,” says Manber. “There’s a lot we need to learn if you really go into reuse in space hardware. These sorts of things seem mundane, but we just have to do it step by step.”
Mission extension programs like Outpost are new to the space industry. Ever since Sputnik, the stuff that was put into orbit was either intentionally deorbited or abandoned and left to fall back to Earth. There simply wasn’t the technology to move a satellite once it ran out of fuel or to commandeer an abandoned rocket hull. And that meant there weren’t any regulations on how to do it safely—or any consensus on whether it was legal to do it at all.
But things are starting to change. Last year, a Northrop Grumman satellite successfully latched onto another satellite that had depleted its fuel supplies and moved it to a new orbit. This maneuver will extend the satellite’s lifetime by at least five years, and it officially ushered in the era of space mission extensions. During a talk at the International Astronautical Congress this year, Joseph Anderson, vice president of the Northrop Grumman subsidiary Space Logistics, described how the company had to work with several different US agencies to modify licensing requirements so that it could launch the historic mission. “It simply didn’t fit the licensing structure that the US government had established,” Anderson said. “Ultimately, we landed on a solution in which the FCC acts as our primary oversight agency.” (That’s the Federal Communications Commission, which also regulates things like radio, television and broadband systems.)
If Nanoracks wants to turn rockets into space stations, it will also have to forge new licensing policies to make it happen. Northrop Grumman’s mission may have laid the foundation for extending the lifespan of new rockets heading to orbit, but what is less clear is whether a company can refurbish rockets that have been abandoned in orbit without the permission of the country or company that launched them.
This is an issue that James Dunstan, the principal attorney at the space law firm Mobius Legal Group, has been grappling with for years. On Earth, international maritime law allows sailors to salvage wreckage they find at sea, but Dunstan says that under the Outer Space Treaty, an international agreement signed in 1967, spent rockets remain the property of whoever launched them. Under this law, if a company or country were to take over an abandoned rocket stage without permission, they would be trespassing on the property of the launching state. But Dunstan describes this interpretation of the law as a fallacy, because, he says, “neither the launching states nor launching companies really care about the spent stages. They’d love for them to go away.”
For now, though, Dunstan says “the legal risk would be significant” for any company that commandeered a rocket stage without asking. He’s spent more than a decade advocating that “find and salvage” maritime laws should be applied to orbital debris like rocket bodies, but he says regulators at agencies like the FCC and Federal Aviation Administration have been slow to act. “It really is going to take a test case to move the needle on the issue of salvage,” Dunstan says. And Nanoracks may very well be the company to do it.
Manber sees recycling rockets as the next logical step to increase orbital commerce and expand humanity’s reach in the solar system. Launching stuff into space is expensive, but developing the techniques to take advantage of resources that are already there could drastically lower the cost of living and working beyond Earth. “When I look 15 or 20 years ahead, there will be scout missions looking for good things to salvage,” Manber says. “You’re going to have prospectors looking for parts and using them for in-space assembly. It’s going to be one of the big markets of the future.”
Manber’s vision has been a long time coming. Over the past 50 years, engineers at NASA have explored several different methods for converting old rockets into habitats. The agency’s first space station, Skylab, was originally meant to be built out of the upper stage of a Saturn V, the massive launcher that carried Apollo astronauts to the moon. This concept, known as a wet workstation, was fairly developed before the engineers on the project decided it would be easier to just launch a bespoke space station instead. But the dream of recycling rockets didn’t die.
Bill Stone is an extreme caver who has been to some of the deepest places on Earth, and he is the CEO of Stone Aerospace, a company he founded to build robots for exploring the oceans on the icy moons of Jupiter and Saturn. Before that, he spent a decade at the National Institute of Standards and Technology working to turn a space shuttle’s external tank into an orbital habitat. At the time, NASA was just beginning to explore engineering designs for Freedom, a space station concept that would eventually morph into the International Space Station. The leadership at NIST tasked Stone and his colleagues with assessing all the details of NASA’s plans to look for ways they could be improved.
“One of the things that kept popping up was the fact that the space shuttle was not 100 percent reusable,” says Stone. Although NASA could land the shuttle orbiter and occasionally recover the solid boosters from the ocean, the biggest element on the rocket—the external tank—was lost on every launch. For Stone and his team, this was a massive waste of resources. By the time the external tank was jettisoned from the shuttle, it had reached 98 percent of the velocity needed to achieve orbit. It wouldn’t take much of an extra boost to keep it in space where it could later be converted into an industrial laboratory.
The shuttle external tank was actually two separate tanks—a small one for liquid oxygen and a much larger one for liquid hydrogen—that are connected by an intertank ring to create one massive structure. The NIST team’s plan was to use the intertank section as a temporary pressurized habitat for crew as they prepared one of the larger tanks for occupation. This would have required several modifications to the tank, such as a hatch to allow astronauts inside and a small motor attached to the bottom of the external tank so it could orient itself in orbit. But the payoff would have been a tremendous amount of space to use as a warehouse or research lab. The smaller liquid oxygen tank would have provided 25 percent more habitable volume than is currently available on the ISS. If the entire external tank was used, it would have had six times more volume than the space station.
“There was 65,000 pounds of aluminum and other aerospace-grade components capable of being pressurized for human habitation that was thrown away on every mission,” says Stone. “Even looking at the best rates that SpaceX will give you for a boost to low Earth orbit today, that’s pushing hundreds of billions of assets that were tossed away.”
As NIST’s plans came together in the 1980s, a consortium of 57 universities took a majority stake in a private venture called the External Tank Corporation that would convert spent shuttle tanks for NASA. As Randolph Ware, the company’s president, told The Los Angeles Times in 1987, the program wasn’t meant to compete with the agency’s plans for space station Freedom. “We are not a substitute for the space station, we are a warehouse on the edge of an industrial park,” Ware said. As the External Tanks Corporations led efforts to commercialize the project, Stone and his colleagues at NIST ran digital and physical simulations of their recycled space station. By the late ’80s, they had even built a mock-up of a shuttle tank in the pool at NASA’s Marshall Space Flight Center so astronauts could practice getting in and out of it. The plan was to use two astronauts during the first demo mission—and Stone was going to be one of them.
NIST wasn’t the only organization that had designs on the space shuttle’s external tank. A study led by an engineer at Martin Marietta Aerospace, one half of what would become Lockheed Martin, floated the idea of using the tank as the basis for a larger space station, and a separate Air Force proposal suggested using the tanks as scrap metal for building structures in orbit. Around the same time, a joint research project between Boeing and the Defense Advanced Research Projects Agency suggested converting the external tank into a large-diameter telescope. Even Hilton Hotels had plans for building orbital hotels called Space Islands out of shuttle boosters, although it seems the project never made it beyond a conceptual stage. (Hilton representatives did not respond to WIRED’s request for comment.)
The dream of turning spent shuttle boosters into a space station collapsed in 1993 when the Clinton administration gave a stamp of approval to the International Space Station. Stone and his team at NIST had recently submitted a proposal to turn shuttle boosters into space stations, which had worked its way up through the highest levels at NASA and into the White House. But as the Clinton administration prepared to move ahead with the ISS, Stone recalls, the director of NIST called him into his office to deliver the bad news: NASA had spiked the program. “The space station had become a national jobs program, and the project was viewed as a threat to the space station,” says Stone. “It was a tragic mistake that NASA didn’t store those external tanks, because they would have established the orbital depots that you need to implement an Earth-moon economy.”
For the next two decades, the idea of living and working in old rockets faded from memory as NASA engineers concentrated their efforts on the ISS. It wasn’t until 2013 that the idea made a modest comeback when Brand Griffin, a NASA contractor from Jacobs Engineering, led a study for the agency on how to turn a fuel tank from its next generation Space Launch System rocket into a habitat for deep space exploration. He called his reclaimed space station Skylab II.
Like its namesake, Skylab II would be launched in a single piece in the upper stage of NASA’s SLS, the rocket that the agency will use to send humans back to the moon. The crew compartment would be made from an unused hydrogen fuel tank that would be launched as a payload in the upper stage of the rocket. This is similar to the design of Skylab, which was built from the third stage of a Saturn rocket that had been modified on the ground, rather than converted from a spent upper stage in orbit. All the components needed to turn the tank into a viable habitat—solar panels, antennas, robotic arms—would be integrated before it was launched. Much like the Nanoracks Outpost idea, there would be no need for astronauts to assemble the station. The converted hydrogen tank would have enough space to host up to four astronauts and their provisions for a multiyear journey around the moon or Mars. Once Skylab II was in orbit, the crew would be delivered on a subsequent launch via the Orion crew vehicle, which could dock with the habitat and provide propulsion for the mission.
Griffin says the Skylab II study was motivated by the need to lower the cost of deep space exploration. Building the ISS was expensive, and it took dozens of launches to get all the components into orbit. A similar modular station around the moon or Mars would be more expensive still. But Skylab had demonstrated it was possible to launch a capable space station in one shot. “We wanted to bring that economy to a cislunar habitat,” says Griffin. After the study, Griffin and his team built a full-scale mock-up of a Skylab II station at NASA’s Marshall Space Flight Center.
But despite some enthusiasm for the project from NASA officials, the idea was shelved and the agency proceeded with Gateway, its new plan for a lunar space station. Unlike Skylab II, the Gateway is modular and more closely resembles a scaled-down version of the ISS. “There are lots of reasons why people don’t accept change,” says Griffin. “Sometimes people get an idea of where the solution is going to go and have invested too much already. It needed more pressure, but it wasn’t like people were against it.”
Manber and Bishop are well aware of the long history of failed attempts at turning space junk into space stations. But they believe that they can succeed where others have failed. Today, robots are able to carry out some of the tasks that, during the shuttle era, would have required a team of astronauts. A burgeoning space economy is driving demand for more orbital R&D platforms. And NASA’s lunar ambitions will require the agency to rethink the deep-space supply chain. Nanoracks still has to demo many fundamental technologies before the company can recycle a rocket, but for the first time in decades it seems plausible that future astronauts will be living in a secondhand space station.