Is is possible to create a nuclear fusion reactor on a satellite and use it to produce heat

I am not a physicist but I would really like to know what trained physicists think about this problem that came up in conversation the other day.

I have been reading about the search for habitable planets for a little while and about the conditions it takes for a planet to be habitable. I understand that the right amount of heat is important for ideal conditions. Then I caught an interesting documentary on Oppenheimer, Heisenberg and the race for the atomic bomb in world war II. The documentary said something to the effect that a nuclear detonation or any nuclear reaction is very much like a small, very temporary star on earth. It got me thinking that nuclear reactors must also be analogous to mini-stars or suns so I started wondering what it would take to create a nuclear reactor that could make a planet (or moon) more habitable by allowing the heat produced to escape the reactor and warm the planet's surface.

As an example, once the technology exists for a nuclear fusion reactor, could it possible to assemble one outside our atmosphere, send it into orbit around, for example, Jupiter's icy moon Europa in order to heat the surface and create more habitable conditions for life?

• Besides the physical challenge of producing enough energy to be useful, you also have the engineering challenges of building a working fusion reactor (components of which must be kept cool to work properly) in space (where keeping things cool is very hard). – Asher Apr 16 '16 at 22:10
• Thanks Asher. I'll have to research this more but since the focus is projecting the heat toward the surface couldn't we just ventilate the heat in the direction of the surface to keep the reactor cool? – Aaron Pennington Apr 16 '16 at 22:36
• "ventilate" is something you do in an atmosphere. In space, your only option to transfer energy is by radiation, and if the outside of your reactor is hot enough for the radiation to significantly warm the moon's surface from orbit, you'll have to deal with some, ahem... material failure. – Asher Apr 17 '16 at 5:20

1 Answer

Not realistically. Ignoring other considerations, it's worth looking at how much power you would need to produce. The Sun produces about $1400$ watts per square metre at the top of the atmosphere: so let's assume we wanted to produce $1000$. Europa's radius is about $1.56\times 10^6$ metres, so we would need to generate $1000\times\pi\times (1.56\times 10^6)^2$ watts, if our system was completely efficient -- all its power output was directed at Europa. This is about $7.65\times 10^{15}$ watts, or a seven million gigawatts. That's not very practical.

• Thanks for the response tfb. I was thinking about this after I posted the question. I didn't mean to imply that the reactor should be able to heat the entire surface of Europa but perhaps create a warmer zone around its equator. I'm not talking about creating t-shirt weather by any means. It would still be cold. I should've have been a bit more clear. Also, what is the time frame for the sun's production. Is that 1400 watts per second? – Aaron Pennington Apr 16 '16 at 22:31
• @AaronPennington power is a measure of energy per unit time, so 1400 watts doesn't need to be per second. I think if you tried to heat just part of a planet you'd find that the atmosphere would rapidly transport the heat elsewhere, probably in a process involving some really impressive wind speeds: people do simulations of tidally-locked exoplanets which effectively are heated on one side only and I think they have winds which are comfortably supersonic. PS I think this is not a dumb question: it just won't work, I think. – tfb Apr 16 '16 at 22:49
• Thanks @tfb Good point. So if this device we're talking about was pushing heat toward Europa on a much smaller scale than the sun to earth it would have to heat almost every part of Europa's surface to avoid creating huge pressure systems in its atmosphere thus making things worse. So its orbit would have to accommodate that requirement but I could be wrong. Maybe I should look more into these simulations. Can you give me a link to the research you are referring to? Atmospheric circulation of tidally-locked exoplanets by Heng, Menou and Phillipps? – Aaron Pennington Apr 16 '16 at 23:37
• @AaronPennington I'm away from work, but the people I know about are at Exeter: there will be other groups I am sure. They use numerical climate modelling programs for exoplanets, which I think is kind of cool. – tfb Apr 17 '16 at 11:17
• The units of power, like the watt, definitely are associated with a particular unit of time. The solar power output of 1400 W/m^2 means that the energy equivalent of 1400 joules is falling on each square meter of area at the top of the atmosphere each second. 1400 joules is the amount of energy it takes to raise the temperature of 330 grams of water by 1 degree Celsius. – user16622 May 20 '16 at 1:26