When Elon Musk and his team at SpaceX were looking to make their Falcon 9 rocket even more powerful, they came up with a creative idea — keep the propellant at super-cold temperatures to shrink its size, allowing them to pack more of it into the tanks.
But the approach comes with a major risk, according to some safety experts. At those extreme temperatures, the propellant would need to be loaded just before takeoff — while astronauts are aboard. An accident, or a spark, during this manoeuvre, known as "load-and-go," could set off an explosion.
The proposal has raised alarms for members of Congress and Nasa safety advisers as the agency and SpaceX prepare to launch humans into orbit as early as this year. One watchdog group labelled load-and-go a "potential safety risk." A Nasa advisory group warned in a letter that the method was "contrary to booster safety criteria that has been in place for over 50 years."
Concerns at Nasa over the astronauts' safety hit a high point when, in September 2016, a SpaceX Falcon 9 rocket blew up while it was being fuelled ahead of an engine test. No one was hurt, but the payload, a multimillion-dollar satellite, was lost. The question on many people's minds at Nasa instantly became: What if astronauts were on board?
The fuelling issue is emerging as a point of tension between the safety-obsessed space agency and the maverick company run by Musk, a tech entrepreneur who is well known for his flair for the dramatic and for pushing boundaries of rocket science.
In this culture clash, SpaceX is the daring, Silicon Valley-style outfit led by a man who literally sells flamethrowers on the Internet and wholeheartedly embraces risk. Musk is reigniting interest in space with acrobatic rocket-booster landings and eye-popping stunts, such as launching a Tesla convertible toward Mars.
His sensibilities have collided with a bureaucratic system at Nasa that has been accused of being overly conservative in the wake of two shuttle disasters that killed 14 astronauts.
The concerns from some at Nasa are shared by others. John Mulholland, who oversees Boeing's contract to fly astronauts to the International Space Station and once worked on the space shuttle, said load-and-go fuelling was rejected by Nasa in the past because "we never could get comfortable with the safety risks that you would take with that approach. When you're loading densified propellants, it is not an inherently stable situation."
SpaceX supporters say tradition and old ways of thinking can be the enemy of innovation and thwart efforts to open the frontier of space.
Greg Autry, a business professor at the University of Southern California, said the load-and-go procedures were a heated issue when he served on Trump's Nasa transition team.
"Nasa is supposed to be a risk-taking organisation," he said. "But every time we would mention accepting risk in human spaceflight, the Nasa people would say, 'But, oh, you have to remember the scar tissue'— and they were talking about the two shuttle disasters. They seemed to have become victims of the past and unwilling to try anything new, because of that scar tissue."
In a recent speech, Robert Lightfoot, the former acting Nasa administrator, lamented in candid terms how the agency, with society as a whole, has become too risk-averse. He charged the agency with recapturing some of the youthful swagger that sent men to the moon during the Apollo era.
"I worry, to be perfectly honest, if we would have ever launched Apollo in our environment here today," he said during a speech at the Space Symposium last month, "if Buzz [Aldrin] and Neil [Armstrong] would have ever been able to go to the moon in the risk environment we have today."
Nasa is requiring SpaceX and Boeing to meet a requirement that involves some complicated calculations: The chance of death can be no greater than 1 in every 270 flights.
One way to ensure that, as Lightfoot said during his speech, is to never fly: "The safest place to be is on the ground."
Still, the scar tissue runs deep.
Nasa lost 14 astronauts in two space-shuttle disasters, the result of deep systematic problems of a once young and swashbuckling agency that many said had grown sclerotic.
In the investigation into the 2003 disaster, the Columbia Accident Investigation Board blasted Nasa for failing to learn "the bitter lessons" from the Challenger explosion in 1986. Columbia was lost as much by a "broken safety culture" as much as the chunk of foam that broke off and damaged the shuttle's heat shield. That second disaster helped lead to the retirement of the shuttle in 2011, leaving Nasa in the position of being unable to fly astronauts from US soil.
Instead, Nasa pays Russia to ferry its astronauts to the International Space Station, an arrangement that costs the agency millions. In 2006, Russia charged $21.3 million a seat. That jumped to $81.9 million by 2015.
To end the dependence on Russia, Nasa has turned to the private sector, outsourcing the responsibility of flying astronauts to the space station to two companies — SpaceX and Boeing — that have been awarded $6.8 billion in contracts combined. Other private companies eventually could compete for other government launch contracts — including Blue Origin, which was founded by Washington Post owner Jeffrey P Bezos — but none are expected to send people to the space station anytime soon.
The pivot to private companies is enabling Nasa to focus on deep space. But SpaceX and Boeing have both faced challenges and delays. Now, as the drought in human spaceflight extends into its seventh year, Nasa is facing the prospect of even more delays — and questions about whether the contractors it plans to rely on will have a better track record than the agency that put men on the moon.
"It really is a very, very difficult problem to do human spaceflight," said Phil McAlister, the director of Nasa's commercial spaceflight development division. "You've got thousands of pounds of really highly energetic propellants on board. You've got mini controlled explosions going off. You've got to survive the rigours of space, which is not very friendly for the human body. And then you've got to re-enter the atmosphere, and the spacecraft gets heated up to thousands of degrees."
SpaceX pulled off 18 successful launches last year, a record, and is aiming for more this year. But it has also lost two of its Falcon 9 rockets in explosions, and amid all its triumphs, it has never attempted flying humans.
The first failure happened in 2015, when a rocket blew up a couple of minutes after lift-off as it was flying cargo and supplies to the space station. No one was on board, and no one was injured. Then, just over a year later, another rocket exploded, this time on the launch pad while being fuelled ahead of an engine test.
At the time, Musk declared that if crews had been aboard they would have been safely ferried away by the rocket's abort system. Still, that mishap is forcing the company to redesign bottles of pressurised helium that sit inside the rocket's fuel tanks.
Now SpaceX is getting ready to fly astronauts on an upgraded version of the same rocket. And its decision to add propellant to the rocket with astronauts on board is attracting scrutiny.
To get more power out of its rocket, SpaceX brings its propellants — liquid oxygen and refined kerosene — to unusually low temperatures. That causes them to become dense, meaning SpaceX can pack more fuel into its rockets.
To SpaceX, the approach is another example of how it is breaking the mould. The densified propellant "provides greater propellant margin for increased reliability," the company said in a statement. In other words, should something go wrong on the mission, the rocket would have more propellant to adjust to emergencies. SpaceX's dramatic booster landings also require additional propellant.
But to others it is an unnecessary risk. At a Capitol Hill hearing earlier this year, members of Congress pressed Hans Koenigsmann, SpaceX's vice president for build and flight reliability, about the safety of the load-and-go procedure.
Koenigsmann said that the fuelling takes only about a half-hour, a "relatively quick procedure, and we believe that this exposure time is the shortest and therefore the safest approach."
And the company points out that if anything goes wrong during fuelling, the rocket's launch abort system would allow the astronauts to escape safely. It also conducts a "static fire," a quick test firing of the engines in the days leading up to the launch to make sure the rocket is operating properly.
And since its rockets and its Dragon spacecraft are reusable, the company gets to inspect them after each flight, giving it an in-depth understanding of how the vehicles perform.
"As with all hazard analyses across the entire system and operations, controls against those hazards have been identified, and will be implemented and carefully verified prior to certification," the company said in a statement.
But in a 2015 letter to Nasa, Thomas Stafford, a retired Air Force lieutenant general and then chairman of the agency's space-station advisory committee, wrote that "there is a unanimous, and strong, feeling by the committee that scheduling the crew to be on board the Dragon spacecraft prior to loading oxidiser into the rocket is contrary to booster safety criteria that has been in place for over 50 years, both in this country and internationally."
At the hearing this year, William Gerstenmaier, Nasa's associate administrator for human exploration and operations, said the agency had not decided whether it would allow SpaceX to load crews before loading the fuel, but he did not rule it out.
He vowed that the agency would "make sure that we're really, really safe to go fly, and the system is ready for crew before we put them on board."
In an interview, Lightfoot, the former acting Nasa administrator, said the agency is in deep discussions with SpaceX about the safest way to go. The agency has a long history with SpaceX, first hiring it to fly cargo to the station and now looking for it to send humans into space.
"It's a matter of having a good risk discussion so that we understand that," he said. "I would just say that instead of working it in the press, we work in the engineering review boards."
'You have to humanise it'
For all its push-the-envelope swagger, SpaceX says it is serious about flying people safely and is going to great lengths to study every aspect of the vehicle, down to individual valves, so that it will meet and surpass the 1-in-270 chance-of-death metric, said Benji Reed, the director of SpaceX's commercial crew programme.
When Reed was down at Cape Canaveral, Florida, on a recent trip, he came across a room on a special tour where the astronauts' families from the shuttle programme used to wait ahead of the rocket launch.
They were stunned to see that a whiteboard with drawings made by the children of the crew lost in the 2003 Columbia disaster was still there, preserved.
"That really drives it home," Reed said. "This isn't just the people that we're flying — these are all of their families. So we take this extremely seriously, and we understand that our job is to fly people safely and bring them back safely. To do that you have to humanise it. You have to see them as your friends and as your colleagues."
But even with some of the best engineering minds at Nasa, calculating risk is an imperfect science. There are too many unknowns in systems that are inherently dangerous and complex.
"Even identifying all of the risks is impossible," Gerstenmaier said during a speech last year. "Also, risk cannot be boiled down to a single statistic."
Before the very first shuttle flight, Nasa estimated that the chance of death was between 1 in 500 and 1 in 5,000. Later, after the agency had compiled data from shuttle flights, it went back and came up with a very different number.
The chance of death was actually 1 in 12.