# Harvesting hydrogen from a star

I've been thinking about what highly technologically advanced civilisations would do once their energy requirements become comparable to the total output of a star, and how such activity could be detected from Earth.

Of course, one possibility is to build a Dyson sphere of one kind or another, which would be detectable as a star with an unusual spectrum, heavy in the infra-red, with emission lines corresponding to the heavy elements from which the structure is composed.

However, if we're talking about a civilisation that's really cracked fusion power in a big way, it strikes me that another possibility would be to extract hydrogen and/or helium nuclei from the star itself and fuse them directly in reactors. In principle this has a significant advantage over a Dyson sphere in that it creates heavy elements as a waste product; with careful control over the nuclear chemistry, it could be used to create not only energy but also the material needed to build more reactors. Thus, if such technology is possible then an initial investment can pay off exponentially; perhaps the rate of power output from such activity could eventually exceed the rate of power output of the original star.

I have two closely related questions in relation to this idea.

1. Is there a practical way to extract hydrogen from the surface of a star? This is assuming a civilisation that has access to literally astronomical amounts of energy, but which nevertheless doesn't want to waste any of it unnecessarily. For example, if one were able to produce a sufficiently strong electromagnetic field, is there some configuration that could be used to funnel charged hydrogen ions away from the star's surface?

2. If this were to be done on a sufficiently large scale, would it be detectable astronomically, and what would the resulting signature be? This signature might come from the act of extracting the material itself, or from its effect on the star, whose distribution of elements would (presumably) eventually be changed by the process.

Note that this is in many ways the opposite question to "What is the easiest way to stop a star?". In that case I'm assuming a forward-thinking civilisation that wants to prevent the star from wasting hydrogen so that it can burn it more slowly in the future; in this case I'm assuming the goal is simply to produce energy and heavy elements as rapidly as possible, perhaps even faster than the star would do by itself.

• I am struggling to understand how the use of solar power on Earth isn't "extracting fusion power from the hydrogen of a star" - and how the amount of power that would be available wouldn't be more than enough. Maybe I'm failing to get into the spirit of your question. Feb 15, 2015 at 3:57
• @Floris the question is coming from the somewhat dystopian perspective that societies' energy requirements can run out of control. Our society is coming close to wrecking our planet because of exponentially increasing energy requirements; the thought experiment is about what would happen if this exponential economic growth continued to the point where we'd end up wrecking the Sun as well. If this happens elsewhere it should be detectable, so I'm hoping this weird dystopian thought experiment can be turned into a testable hypothesis. Feb 15, 2015 at 4:19
• @Floris e.g. this blog post shows that if you naïvely extrapolate past economic trends you find human energy use will pass the Sun's total output in less than 2000 years. I don't think this figure is particularly plausible, but I do think it's plausible that it could happen eventually. (Not good or bad, just plausible.) Feb 15, 2015 at 4:26
• at what population levels, though? Truth is that biological organisms and societies composed of them reach an equilibrium with the environment and the available resources . Feb 15, 2015 at 4:30
• @Nathaniel - that's exactly the bit I was missing. Feb 15, 2015 at 14:33

So: fusing hydrogen to helium yields 0.7% of the hydrogen mass as energy. The amount of hydrogen required per second if you have a 100% efficient energy capture process and need the power of the Sun is $3.8\times 10^{26}/(0.007c^2) = 6\times10^{11}$ kg/s.

Provided you can figure out how to process this (and that seems much harder than building a Dyson sphere), then you consume as much hydrogen as the Sun does. This alone will make almost no difference to the luminosity of and hence the evolution of the Sun, since the Sun only burns 10% of it's hydrogen over it's lifetime - so has plenty of fuel. As you remove 1% of the solar mass per billion years I guess the luminosity will increase at a slightly lower rate and the Sun's life would be modestly extended.

This assumes that as thegravitational potential energy of the material is a lot smaller than the fusion energy you wouldn't bother fusing in situ. If you did, or you threw the resulting helium back into the Sun, that might accelerate the solar evolution by increasing the mean particle mass. The details here depend on how rapidly any mixing down to the core could take place - an unsolved astrophysical problem. However, if mixing is not efficient (and it probably isn't), the material will just be mixed within the outer convection zone. Changing the mean particle mass of the outer envelope might change the Sun's radius and temperature slightly due to a differing opacity and a changing equation of state. The effects are complex (e.g. Tanner et al. 2014), but if the "waste" contained metals heavier than He it would have more effect. Excess He would make the star smaller and hotter; excess metals would make the star bigger and cooler. Helium abundance anomalies might be most readily detected by asteroseismology. Metal anomalies are easier to pick up by optical spectroscopy, but vast quantities need to be dumped into the star to make an observable difference.

Another question is where does the energy go? Assuming that it isn't 100% efficiently converted into some other form like kinetic energy and taken out of the star system, then there will be a lot of waste heat, presumably at temperatures a lot lower than the stellar photosphere.. Then we're back to a Dyson sphere scenario of looking for a significant IR excess in the spectrum. Excesses of more than say 1% of the stellar luminosity emerging in the IR over that expected from the photosphere would be quite suspicious in an old star. There are however (rare) natural events that might produce this - e.g copious dust from a major collision.

• Thanks, some useful information here. I guess the main signature would be the waste heat in that case. Since this would be of a similar or greater magnitude to the star's total output, it would probably be quite easily detectable I suppose. Feb 15, 2015 at 13:25
• @Nathaniel Yes. Far infrared excesses that are of order $10^{-4}$ of the photospheric luminosity are routinely detected. Many (well about 1%) older stars have them. Anything where this ratio is $<10^{-2}$ is usually attributed to optically thin debris discs. Larger excesses are often seen in very young stars, associated with a primordial, optically thick disc. Feb 15, 2015 at 13:33

I can't think of any practical reason to extract hydrogen from a star because it's so abundant and easier to get from say - Saturn, Uranus, even Titan. Any of the outer planets or moons with an atmosphere should have abundant hydrogen. The inner planets - not so much, but the outer planets have lots.

Something to consider is that Solar energy alone is quite abundant. Solar may not be exciting, but the earth gets hit by about 1,000 times as much energy from the sun as we generate on earth with all our fossil fuel and uranium burning/fission. That's why, it's theoretically possible, while somewhat improbable, to power the earth with solar panels about the size of Texas.

But, if you're talking about lots and lots of energy, like space travel energy, first things first, Deuterium is easier to fuse than hydrogen so hydrogen wouldn't be the first choice. It's actually pretty inefficient in the fusion process cause 2 hydrogens don't fuse easily, and other elements like Deuterium, Helium-3 (not Helium-4) and, surprisingly, Lithium, are easier to fuse than hydrogen. Deuterium (2-H) or Helium3 (3-He) would probobly be more likely elements to mine.

On earth alone, there's so much available Deuterium in our oceans that we're not likely to need another source for a very very very long time - assuming we figure out how to make fusion work that is. We don't have much Helium 3 on earth, and there's talk about a moon mission to mine it, and in the more distant future, Saturn, or maybe Titan, cause Titan has both an atmosphere that probobly contains 3-He and a surface you could build on - that might be the best target.

There's enough of an abundance of materials in the solar system and the universe that I can't imagine solar mining ever being practical or necessary. Saturn weighs 95 times what the earth weighs and it's mostly hydrogen. That's an enormous supply. 95 times the mass of the earth is (roughly speaking), about a million pacific oceans full of liquid-density hydrogen. Saturn alone has more than enough hydrogen than we're likely to ever need, and if Saturn runs out, we can start mining Jupiter, which has 3 times as much hydrogen as Saturn does.

Fun stuff to think about. If we ever do get space mining underway and practical, there's no shortage of stuff out there.

I'm not sure that's quite the answer you were looking for, just my thoughts on this subject.

As far as "detectible" - most things in space would be detectible with a big enough telescope of sorts cause anything that reflects or blocks light should be visible and because stars are fairly predictable. How easy it is to see would depend on how far away it is and how good our detection abilities were.

I'm not a professor or anything. This is just a hobby to me.

• I forgot to mention that in the scenario I'm considering, the reason for using the Sun instead of Jupiter is that Jupiter has been used up already, or soon will be. Of course the amount of energy available from fusing the mass of Jupiter is immense, but under exponential growth it won't necessarily last all that long. Although H nuclei are hard to fuse, my feeling is that a civilisation this advanced would have the technology to do so, e.g. via the CNO cycle. But if it's much harder than I'm imagining I'm willing to be corrected. Feb 16, 2015 at 6:35
• Well, again, I'm far from an expert. I didn't mean to imply hydrogen fusion is impossible, only that it wouldn't be a first choice as there are easier, though less abundant options. en.wikipedia.org/wiki/Fusion_power#Fuels Feb 16, 2015 at 7:40
• The amount of deuterium you would need is the same as the amount of H. So how are you going to get that? No, your fusion process has to mimic the pp chain or CNO cycle, like the Sun. Feb 16, 2015 at 8:55