Is it possible to build a Dyson Sphere?

Yes. Absolutely. In fact, it’s easier than people think.

But first, it has to be said that the common concept of this usually seen in science fiction is just absurd. Most people think a Dyson sphere is a solid habitat with a roughly 1.0 AU radius and a star at the center. Such a habitat is not at all reasonable from an engineering perspective. This is not what Dyson meant when he first described a “shell” at 1.0 AU from the Sun.

His idea was actually that over thousands of years, the planets (gas giants included) would eventually be repurposed to make habitats and solar collectors. These would be parked in many different orbits around the Sun, creating a swarm that ultimately blocks all the visible light from escaping. These habitats would be around the temperature of liquid water, so they would radiate infrared heat in a specific frequency range — which we would be able to observe from a distance.

It’s important to understand that Dyson’s original version was not optimized for efficiency or ease of building. It was a projection of what a growing civilization using solar power would eventually blindly come up with. So the amount of material he pictured being invested was absurdly high — he simply plugged in the entire mass of all of the planets. Because why not? The trillions of habitats that could be produced (which could be similar to O’Neil cylinders) would cumulatively have much more room than any planet.

More recently, Anders Sandberg and Stuart Armstrong wrote a paper where (again, as a rough ballpark) they plugged in the mass of the metallic content of Mercury, and 0.3 AU as the distance. Such a system would collect the same exact amount of solar energy as Dyson’s original type of sphere, but at higher temperatures. The average thickness would be around double that of household aluminum foil. Most of the surface area would be accounted for using much thinner mirrors, with the concentrated energy being caught and converted to electricity at their focal points.

The purpose of their paper was not to establish a lower bound for the material cost of a Dyson sphere, but to show how quickly a project like disassembling Mercury could be done, even if the sphere mass needed was equal to the materials contained in Mercury. If we explore further, there’s a much lighter type of sphere to consider still: The Dyson bubble.

A Dyson bubble is a swarm, not a solid structure. Each component has a mass that is evenly balanced between the solar light pressure and gravity. There is a constant number (0.78 g/m^2 for a perfect reflector) which holds at any given distance from the Sun. This is because the amount of light and the amount of gravity increase or decrease at the same rate when you get closer to or further from the Sun.

A structure massing 0.78 g/m^2 at 1.0 AU has a mass around that of the second-largest asteroid. At 0.3 AU, to be comparable with the above, it would be about 1/10th of that. There are lots of large asteroids of comparable size distributed throughout the Solar system, and they all have very low gravity.

That thickness — far thinner than a sheet of paper or aluminum foil — is around 1000 times as heavy as a single sheet of graphene (0.77 mg/m^2). So a possible configuration would be a monolayer of graphene that is ‘clad’ with metal, glass, or any other material.

One of the well known ways to make graphene is to take a flat surface of another material (such as nickel) and layer carbon atoms on it. This can be most easily done in a vacuum. So it should be very friendly to space based manufacturing.

We could get started on the project rather soon. The first solar panels produced this way would be used to fuel near-earth industries, because at that point we haven’t fully automated everything. Keeping it near earth for a few decades allows us to employ humans as telerobotics operators while we work out how to truly automate everything. Given advances in machine learning, that could be very soon. The materials would come from near-earth asteroids, at least unless/until we set up lunar launching infrastructure.

Then as the mechanisms become better established, we can send fully automated units to more distant asteroids, and the panels could be lowered into the Sun using solar sailing or efficient ion rockets. One interesting possibility would be to redirect asteroids into controlled collision with Mercury, as a way to harvest volcanic ejecta, increase its rotational speed, etc.

Originally shared on Quora.