Currently a high school senior in California, Lewis-Weber has just published a paper in the journal New Space with what he thinks could be the solution to the upcoming energy crisis: putting self-replicating solar panels in space. These solar panels would to build copies of themselves, autonomously, on the surface of the moon. Then they would enter Earth's orbit, collect the sun's energy, and wirelessly beam it to the ground.
That may sound like a crazy idea, but the notion of space-based solar power actually dates back several decades, to the 1970s oil crisis. It was set aside after oil prices went back down, but since then, two things have happened: One, the world has become a lot more desperate to solve climate change; and two, technological innovations have brought this crazy idea out of the realm of science fiction. The idea is gaining attention, and with some big investments, it's possible that space-based solar power could become a reality within a few decades.
Why Put Solar Panels In Space?
Of all the energy sources available to us Earthlings, the sun is pretty much as good as it gets, says Lewis-Weber. "As Elon Musk has stated, we have a fusion reactor in our sky."
The problem with regular solar power is that the sun isn't always up. We have nights, and cloudy days. The panels also take up a lot of land.
"To power entire world with solar power, we would need to cover an area that's 92 percent the size of Nevada in solar panels, and that's not even counting the batteries," says Lewis-Weber. "To me, that's just not feasible."
But if you were to put those same solar panels in space, high above the atmosphere, there would be no weather to contend with, and the panels would experience nearly constant sunlight. Plus, the light that reaches them would be 27 percent brighter, since it wouldn't need to filter through the atmosphere.
There's no such thing as a cloudy day in orbit
These solar panels would use microwaves to beam power back to receivers on Earth. And in case you're worried, those microwaves wouldn't fry us.
"The system would be designed not to exceed safe power densities," Paul Jaffe, who works on space-based solar panels at the U.S. Naval Research Lab, told us. (Jaffe has proposed a plan to launch solar panels into space from Earth--but they wouldn't be self-replicating.) "It's sort of like how you don't have to worry about breaking the land speed record by riding your bike. The system couldn't be weaponized."
Nor would the large array cast a shadow over the Earth. Sunlight would diffuse around the structure, like it does for other satellites in orbit. Even when the moon eclipses the sun, it only casts a shadow over a small part of the world for a short period of time. The shadow from an array the size of Nevada would be "not even a millionth of that," says Jaffe.
There's another advantage to space-based solar power. These arrays could beam power down across vast portions of the globe, wherever the receivers are set up. That opens up the possibility of sending electricity to villages in developing countries, or to disaster-stricken areas. The receiving equipment, Jaffe says, could fit into a couple of shipping containers.
Plus, since the sunlight would be essentially continuous, space-based solar power doesn't require the development of large batteries to store the power--something that holds back ground-based solar power.
There's Just One Big Problem
It's going to take a LOT of solar panels to power the world, and launching all of those up into space will not be cheap.
Just one SpaceX launch costs about $60 million--and that's much cheaper than the competition. In the paper, Lewis-Weber calculates that it could cost tens of trillions of dollars to send up a meaningful number of solar power satellites.
"Definitely the launch cost is one of the most influential factors in determining the cost of space solar," Jaffe agrees. "Without that cost coming down, or using some alternate means to put the spacecraft in place, it's not going to compete [with fossil fuels] on price."
A Possible Solution?
What if, instead of sending thousands of solar panels into orbit, we could just send up one that's programmed to make copies of itself? And then each machine it makes would make copies of itself, and so on. Like multiplying rabbits, the population of solar panel satellites would grow exponentially, covering the size of Nevada in a few months or years.
Earth orbit doesn't have a whole lot of resources for building all those robots, so instead we could send the self-replicating machine to the moon, Lewis-Weber suggests. There, it could mine the soft lunar regolith for aluminum, iron, and silicon, to turn into parts for its solar satellite babies.
Building self-replicating robots won't be easy, but Lewis-Weber has a plan. The first step would be to simplify the solar panels' design as much as possible. "Instead of having 1,000 different types of screws," he says, "let's have five. Instead of having different molds for different parts, let's have a 3D printer."
Like mutiplying rabbits, the population of solar panel satellites would grow expenentially
With about 18 different "species" of factory machine, each one performing a simple task, such as producing screws or solar cells, it is theoretically possible to turn the moon into a self-sufficient solar cell factory.
Once the solar panels are ready, they could launch back to Earth--a process that's a lot easer than launching from Earth, since the moon's gravity is only about a sixth as strong as our planet's--and take up residence in orbit.
The process requires a lot of new technology, but none of it is particularly distant. There's no need for warp drives or matter transporters (though the latter would certainly be useful). Lewis-Weber thinks he could make it happen for about $10 billion. Most of that would be spent on research and development. After the technology is developed and launched, each solar panel it makes is essentially free.
For that same $10 billion, building and launching the solar panels from Earth would provide an array that's large enough to power 150,000 homes. Not bad, but not as great as being able to power the entire world with the same amount of money.
Even if Lewis-Weber's R&D ends up costing $100 billion, the cost of the electricity it generates ($0.00042 per kilowatt-hour) would be several orders of magnitude lower than that of fossil fuels. Not only would that allow solar power to outcompete coal, oil, and natural gas, it means companies that invest in this technology could stand to earn a lot of money.
But It's Not Possible Yet
Complex, self-replicating robots don't exist yet, and "it'll be a tough engineering challenge," Lewis-Weber admits. But it seems possible. Scientists are making progress on building simple machines that can "reproduce", and one 3D printer comes close to self-replicating; it can print 73 percent of a working copy of itself.
In terms of digging up lunar dust and refining it into usable parts, that might be a job for Deep Space Industries and Planetary Resources, companies that are developing technologies to mine asteroids.
Other research teams are also trying to make space-based solar power a reality. Jaffe and his colleagues, for example, are trying to develop technologies that would help the individual satellites move into an organized array in orbit. "Whether you build these from pieces you launch from Earth or from the moon, it's likely that there's going to be some assembly required."