The ultimate heat sink for a naval reactor is indeed the ocean, via circulating seawater through heat exchangers.
Exactly. Naval reactors are most definitely not closed systems. While there are several closed loop systems inside the sub (including the primary coolant and secondary steam generation system) that do not physically contact each other, all are thermodynamically interconnected by heat exchangers that ultimately dump waste heat to the ocean. Because naval reactors utilize steam turbines, they require a very large heat sink to condense the steam. And of course, the whole point of a nuclear reactor is to produce heat. This heat must be promptly removed, or bad things happen.
There are some other minor issues, like the fact that naval reactors aren’t exactly lightweight, and that they are designed to operate mostly upright in a normal gravity environment. They also utilize seawater (after distillation and ion-exchange) for makeup water, and to provide oxygen (for breathing) by electrolysis.
How much does a submarine reactor weigh anyway? The most powerful rocket currently in operation is the Delta IV Heavy, and it can only lift 29 tons into low earth orbit. I don’t think that would be enough. On the drawing board are Falcon Heavy (54 tons) SLS Block-2 (130 tons), but I suspect those still won’t be enough?
how would you cool it? I think a lot of people don’t realize that nuclear power is- at its heart- steam power. Only with a fission reaction heating the water instead of a combustible like coal or natural gas. it’s not enough to have a reactor, you need a shit ton of water. Both for the reactor, and for the steam generation/turbine (IIRC all Naval reactors are PWRs.) you also need a cooling source to condense the steam after the turbines into liquid water again. Navy ships have that whole “ocean” thing to provide cooling.
The need to reject heat has been well covered above. Worth putting into quantitative terms. The best possible energy extraction from a heat source limited by the formula:
(T[sub]hot side[/sub] - T[sub]cold side[/sub])/T[sub]hot side[/sub] where T is in degrees absolute (ie Kelvin).
Your problem isn’t making something hot, it is making a differential in temperature.
This doesn’t mean the a fission reactor in space is impossible, but it does mean a Navy reactor isn’t going to be the right answer.
Another problem with any fluid cooled reactor is that often the safe modes (ie operating safely with no external power) depend upon convection to move the cooling fluid. That isn’t going to work in space. No apparent gravity, no convection.
But, in addition to radio-thermal generators, there have been space based fission reactors. But the design constraints are radically different. For one, only run the reactor a long way from any people and worries about shielding and its mass are much less a problem. The design of fuel rods is going to be much different, and probably built around designs that can survive an uncontrolled reentry. (The RTG fuel is so designed.)
There is no problem about using a boiling fluid in space (or just hot gasses) to power a system. A Stirling engine can be used as a heat engine that works quite well with relatively low temperature differences. But you can’t rely on gravity for anything, all fluid movements will involve active elements of the design.
But overall, your big problem is getting rid of the heat. This is just an engineering problem, but it isn’t a simple one.
I’ve occasionally pondered the problem of using nuclear power to generate usable electric power, without going through a heat engine. The best I’ve been able to come up with is to put a beta-decay material in the center of a spherical capacitor, charged up to a voltage just under the beta decay energy (positive terminal in the center). Then bleed charge off of the capacitor at the same rate that it’s replenished by the betas. You could in principle get very efficient this way, but without a chain reaction, it’d be difficult to get very much power.
It’s probably more practical to just settle for heat engines, and try to crank the temperature of the hot reservoir as high as possible.
Entropy is a b**ch. No matter what method you use, the engineering design is a trade off of efficiency with things like material limitations, reliability, operability, safety, environment, cost etc.
Heat engine temperatures are limited by material temperature properties (people talk about ceramics and composites but they are still in infancy) and working fluid characteristics ( the higher pressure and temperature you go with water - the more “pure” it needs to be and increases safety and corrosion concerns ).
The power industry moves very slowly and has not seen breakthroughs like semiconductors.
Important to realise that RTGs are heat engines and their maximum efficiency is identically limited by the same expression ((T[sub]h[/sub]-T[sub]c[/sub])/T[sub]h[/sub])
The fins on an RTG are there for the same purpose as any other heat engine. No heat rejection, no power.
And they’ve had several uncontrolled reentries of reactors, notably Kosmos 954 that landed on a large swath of Canada, but also a couple that landed in the ocean.
According to that article, nuclear-fueled satellites have a mechanism to eject the core and put it into a “safe disposal orbit” which apparently is much higher than their original orbit. It seems to me that this is just kicking the can down the road; unless it’s ejected from Earth orbit altogether, it’s eventually going to come down. Perhaps they think that’ll be far enough in the future that the radioactivity will be reduced enough to be safe.
The moon is in Earth orbit. Really the only risk of something coming down is if it is in an orbit that has either some residual atmospheric drag, or the orbit is unstable and will eventually get back too close. (Technically all the orbits are unstable, but some you need to wait until the sun is already cold before you are likely to see it.) Even out of Earth orbit it is in solar orbit, and might get a revist.
Whether the orbit these Russian systems get the core to is actually stable enough is another matter, but there is nothing intrinsically bad about the idea. I do however doubt there is actually enough delta-v available to do more than kick it into an orbit above the worst of the drag. Getting it to maybe up to 2000km would be good. Needs to be a reasonably circular orbit as well. That isn’t going to be trivial with a little emergency rocket.
I have read of a couple of at least partial theoretical ideas in the fusion arena. One was the magnetic bottle: although primarily a heat engine, some electricity would be created directly by ion leakage out the ends – basically the opposite of a particle accelerator. The other is fundamentally the same concept, but using the H-[sup]11[/sup]B process, which produces no neutrons or gamma radiation but a lot of hot He[sup]-[/sup] (alpha) that can be decelerated to generate electric power directly. Unfortunately, that process happens at several billion degrees, so getting there is the hard part.
Well, any orbit with a semi-major axis less than geosynchronous altitude, is going to come down. And if the eccentricity is extreme enough, even that won’t keep the body up there. I didn’t want to go into that kind of detail, because I figured the emergency rocket would not be powerful enough to get it that high.
For anyone not already familiar with it, the “Project Rho” site is an excellent source for details on the problems and practicalities of various space-based designs (and their history in SF.) Herearelinks to info on heat radiation in space.
I don’t know what T[sub]hot[/sub] and T[sub]cold[/sub] might be for an orbiting RTG, but their real-world efficiency is pretty damn low, less than 10%. So if you’ve got an RTG providing 500 watts of electrical power, it’s trying to dump 4500 watts of heat through those fins. Considering the typical residential space heater is just 1500 watts and cools by convection, you can see what a challenge the engineers have in designing a high-power RTG that only cools by thermal radiation.
The tradeoff is that with zero moving parts, RTGs can be extremely reliable; the biggest challenge to operating life is the reduction in power output over time as the radioactive material decays. The Voyager spacecraft are still transmitting data 40 years after launch, and are expected to continue doing so in some capacity until at least 2025.
In contrast, a nuclear reactor driving a steam turbine is going to have a shitload of moving parts; even assuming the heat rejection problem could be solved, it would be extremely difficult to achieve the kind of long-term reliability provided by RTGs.
Come down eventually? Yes. Come down too soon vs. the ever-declining residual radioactivity? That’s the relevant question.
Particularly if you *don’t *design the reactor core to survive reentry, you’ll eventually be spraying a few kilos of low level radionuclides across thousands of cubic miles of air when the whole thing finally re-enters.
With enough dilution in time and space this becomes a non-issue; there’s already some nonzero number of radioactive molecules in each cubic meter of sea water and of ordinary dirt.
So the question ultimately is one of degree, not of kind. I don’t know the details for any specific space reactor, but they could be computed.
There’s nothing special about geosynchronous height as the dividing line between stable and unstable. Low-Earth orbits aren’t stable over long terms, and geosynchronous are, but there are very few satellites in orbits in between those two heights, and so folks don’t often talk about the stability in there. But you can get stable enough without going all the way up to geosynch.
And about cooling fins: On the Earth, a big chunk of cooling comes from convection, and so all a heat sink needs is a lot of surface area for air to flow past, and so you get designs like a whole bunch of parallel plates attached to one plate perpendicular to them. But in space, all you’ve got available is radiation, and compact heat-sinks like that don’t work any more: Radiation leaving one of those plates is mostly just going to hit another one and get re-absorbed, and the amount of radiation that actually escapes will be the same as for any other object that fits in that same envelope. Radiative heat sinks need to be genuinely big, like a sail spread out over a large area. Fortunately they don’t need to be particularly strong, especially in zero G, so you can get away with mostly thin mylar or the like unfolded on rigid supports, but it’s still a nuisance you’d rather not have to deal with.
As has been rightly said, one of the big problems is cooling in space.
All you’ve got is radiation, so your radiator is going to have to pretty hot to lose significant heat. One solution is to use potassium vapour rather than steam.
Has anyone demonstrated a boiler in space? Seems like a big engineering challenge. Without gravity, the vapor would pool around the heating element rather than bubbling up to the top where it can be collected and sent to a turbine/cylinder. The vapor and liquid need to be separated by centrifugal force (perhaps by creating a circular flow of liquid) or some other mechanical method.