Spaaaaaaaaaaaaaaaaaace!
March 25, 2013 5:17 AM   Subscribe

When the US Department of Energy halted Plutonium 238 production as far back as 1988, things looked grim for the future of space exploration. On Monday, March 18th, NASA's planetary science division head Jim Green announced that production has been restarted, and is currently in the test phases leading up to a restart at full scale.
posted by cthuljew (37 comments total) 18 users marked this as a favorite

 
That's insane. We've got 112 tonnes sitting at the bottom of a swimming pool that they could have had... well, assuming they could convince someone to risk shipping it across the Atlantic.
posted by pipeski at 5:23 AM on March 25, 2013 [2 favorites]


That's mostly pu239 intended as fission fuel.
posted by a robot made out of meat at 5:27 AM on March 25, 2013 [1 favorite]


Well, about half pu239, 25% 240, and the rest heavier. Only 2-5% will be 238. Separating them is a real bear.
posted by a robot made out of meat at 5:32 AM on March 25, 2013 [2 favorites]


That's nice, but the two missions mentioned in the link, Titan Mare Explorer (TiME) and Comet Hopper, lost out to sending another probe to Mars. Because as awesome as putting a probe in Titan's ocean or landing on a comet would be, one can never have too much Mars.
posted by Brandon Blatcher at 5:37 AM on March 25, 2013 [2 favorites]


Well, about half pu239, 25% 240, and the rest heavier. Only 2-5% will be 238. Separating them is a real bear.

And basically it's impossible to make 238Pu out of 239Pu, so even if you have a large quantity of weapons grade material, it's not useful for the reaction.

The usual process is to recover 237Np from spent reactor fuel, irradiate it, where it becomes 238Pu, then extract that from the neptunium.

Basically, NASA's position on this is simple. If we don't have 238Pu, we're not exploring past Mars, period, unless someone comes up with a different power source that is also not dependent on solar flux and can offer enough waste heat to keep the spacecraft functioning.

One way we could do it is fission reactors, rather than RTGs. Good luck getting approval for that launch.
posted by eriko at 6:12 AM on March 25, 2013 [8 favorites]


Heh, they're only going to be making about three pounds of the stuff per year. That's got to be one of the most expensive substances on the planet.

According to the density figures, at least assuming I've done my math right, that'll make a cube that's just a hair over 4cm on a side, or about 1.6 inches. You might be able to close your hand around that and hide it.

Well, okay, you could do that if you didn't want your fingers anymore.
posted by Malor at 6:28 AM on March 25, 2013 [2 favorites]


Isn't uranium used as fission fuel and plutonium used for weapons? Can someone break this down please? I heard as late as last year that Sandia was making plutonium pits. I presume these were weapons grade only.
posted by pashdown at 6:31 AM on March 25, 2013


Well, okay, you could do that if you didn't want your fingers anymore.

And if you didn't want anyone within a few hundred meters to be alive a month on.
posted by cthuljew at 6:33 AM on March 25, 2013 [1 favorite]


Pashdown, Pu238 is useless for making weapons and is not the sort of thing that people usually mean when they speak of plutonium in regard to weapons, waste products, etc. Pu238 generates a tidy amount of decay heat. This lets you generate a thermal difference which lets you use the magic of physics (the Seebeck effect, I think) to generate electricity.
posted by introp at 6:41 AM on March 25, 2013


One way we could do it is fission reactors, rather than RTGs. Good luck getting approval for that launch.

Of course if you launched a reactor using enriched uranium as fuel, you wouldn't have to put a crash proof containment around it (which is good because you'd never be able to do that with a reactor the way you can with an RTG). After all, fresh uranium fuel is only moderately toxic, high altitude dispersion of a hundred kg or so wouldn't be so bad. Of course once the reactor has run for any length of time you'd not want it to have any chance of atmospheric re-entry.

The biggest problem would be dumping the heat, you'd replace large arrays of delicate solar panels with large arrays of cooling fins (which are less tolerant of damage than the solar panels).
posted by atrazine at 6:42 AM on March 25, 2013


Isn't uranium used as fission fuel and plutonium used for weapons? Can someone break this down please? I heard as late as last year that Sandia was making plutonium pits. I presume these were weapons grade only.

They can both be used for either purpose depending on configuration, isotope, and enrichment. Most modern weapons are (probably, this is all secret squirrel stuff) plutonium 239. It has the smallest known

There's quite large stockpiles of pu239, but the stuff they're talking about for space power is pu238.

The properties are quite different, as indeed they must be. The weapons stuff, 239, has a half-life of 24,000 years which is desirable for weapons because it sits there doing nothing until you want it. Pu238 has a half-life of 87 years which is why it gets quite hot, that's great for powering an RTG but not so great for a bomb.
posted by atrazine at 7:10 AM on March 25, 2013


Pu238 decays by alpha radiation, but it has a half life of only 88 years. As a result, it's producing a lot of heat constantly, and in fact the pure stuff glows from its own heat production.

That much alpha radiation isn't benign, but it's not like it is gamma radiation. It isn't a room-clearer.
posted by Chocolate Pickle at 7:15 AM on March 25, 2013


The biggest problem would be dumping the heat, you'd replace large arrays of delicate solar panels with large arrays of cooling fins (which are less tolerant of damage than the solar panels).

That seems counterintuitive. Why is that?
posted by Pre-Taped Call In Show at 7:26 AM on March 25, 2013


Space is a nearly perfect insulator. Your ship would melt.
posted by cthuljew at 7:28 AM on March 25, 2013 [1 favorite]


Right, I get you have to dump the heat, not just to keep the vehicle cool but to maintain a thermal gradient across the thermoelectric pile. But are cooling fins actually more susceptible to damage than solar panels, or is it just that the failure mode is worse (overheat vs undervolt)?
posted by Pre-Taped Call In Show at 7:41 AM on March 25, 2013 [1 favorite]


And if you didn't want anyone within a few hundred meters to be alive a month on.

Yeah, as Chocolate Pickle is saying, it decays by throwing off alpha particles, which are easily absorbed by any solid substance, including skin. I imagine you'd cook your hands if you handled it, but you probably wouldn't badly hurt anyone but yourself.

It was used in pacemakers at one time; according to Wikipedia, 250 plutonium pacemakers were created, and 22 of them were still working 25 years later. They link to this article, which talks about the specific dangers of Pu-238 radiation. (not bad by itself, quite dangerous around other unstable nuclei.)

That's an interesting article -- one of the things they mention is being able to make, for a mere $200,000 (in the 1970s!), a device the size of a D-cell that would put out 50 watts, presumably with the same 87-year half-life as the radioactive component.

It's a shame we're so vulnerable to radiation. Wouldn't it be cool to buy one set of batteries every fifty years? Hopefully, for rather less than $200K. :-)
posted by Malor at 7:45 AM on March 25, 2013 [2 favorites]


One way we could do it is fission reactors, rather than RTGs. Good luck getting approval for that launch.

The Russians launched dozens of reactor-based satellites (using molten sodium-potassium coolant, IIRC), so presumably they could be talked into it. Maybe they even have a few surplus reactors sitting around.

Over the course of the program they only ever had one launch issue, and that wasn't especially serious (fell into the Sea of Japan); the big problem there was, because they were LEO spy satellites, if the reactor core didn't eject into a high-altitude "parking orbit" correctly before the main body of the satellite de-orbited, it would re-enter and spew waste products on the way down. They irradiated a bit of Canada this way.

But if you were leaving Earth orbit anyway it's not an unreasonable solution at all. The reactor doesn't even need to be started (can be sub-critical at launch) until the power is needed, which could be far away from Earth, so there's no chance of nasty decay products ending up back home if something goes wrong.

The US had a similar space-reactor program called SNAP, but abandoned it fairly quickly. It was a NaK-cooled system, which we've never done very well -- the Soviets were much better at the LMKRs to be blunt about it, since they used them heavily in their submarine as well as their space program.
posted by Kadin2048 at 7:51 AM on March 25, 2013 [2 favorites]


Right, I get you have to dump the heat, not just to keep the vehicle cool but to maintain a thermal gradient across the thermoelectric pile. But are cooling fins actually more susceptible to damage than solar panels, or is it just that the failure mode is worse (overheat vs undervolt)?

It's more that a leak in a cooling circuit could easily lose you a significant fraction of your coolant even with clever design and isolation capabilities. Losing a bit of solar PV panel just means losing that bit of panel. If you could cool the reactor in such a way that the only bits that were exposed to damage were solid metal fins then you'd be good to go, because solid metal doesn't have that failure mode. I suspect you'd be quite limited in terms of how powerful a reactor you could do that with though.

The Russians launched dozens of reactor-based satellites (using molten sodium-potassium coolant, IIRC), so presumably they could be talked into it. Maybe they even have a few surplus reactors sitting around.

NASA actually bought a few (six? something like that, I think) of their NaK enriched uranium space reactors, they've never been used for anything other than testing though.
posted by atrazine at 8:09 AM on March 25, 2013


Right, I get you have to dump the heat, not just to keep the vehicle cool but to maintain a thermal gradient across the thermoelectric pile. But are cooling fins actually more susceptible to damage than solar panels, or is it just that the failure mode is worse (overheat vs undervolt)?

Radiative cooling is a function of surface area. For you to have more surface area in a compact volume (since the rocket's cargo space is predetermined) requires structures to be more elaborate and delicate in an attempt to expose more surface area to space. The damage of those structures by dust and micrometeorites will slowly whittle away at the ability to dump heat as well.
posted by Talez at 8:10 AM on March 25, 2013


Very informative comments. This is the type of thread that makes me proud to be a MeFite!
posted by fairmettle at 8:53 AM on March 25, 2013 [2 favorites]


"Basically, NASA's position on this is simple. If we don't have 238Pu, we're not exploring past Mars, period..."

Well, apart from the solar-powered Juno mission, which is currently on its way to Jupiter orbit. Sadly, it's not going to last for very long - 32 orbits - although whether that's due to money or expected radiation damage, or whether there's an extended mission ready to take over if it performs well enough, I don't know.
posted by Devonian at 9:14 AM on March 25, 2013


Radiative cooling is a function of surface area. For you to have more surface area in a compact volume (since the rocket's cargo space is predetermined) requires structures to be more elaborate and delicate in an attempt to expose more surface area to space. The damage of those structures by dust and micrometeorites will slowly whittle away at the ability to dump heat as well.

How about a super-long wire you attach to it that trails behind it, like 5 miles long. You could even shoot it up on a separate mission.
posted by Ironmouth at 10:06 AM on March 25, 2013


Ironmouth: How about a super-long wire you attach to it that trails behind it, like 5 miles long. You could even shoot it up on a separate mission.

Wouldn't that just be super-vulnerable to being severed? Also, wouldn't it prevent you from rotating the spacecraft, or interfere with course corrections?
posted by Mitrovarr at 10:34 AM on March 25, 2013


Wow, I assumed RTG's used some kind of low-grade waste isotope, with matching low output, and they were kind of a power source of last resort. But looking into Pu-238... that's just beautiful - an incredible amount of energy output per gram, an 87-year half-life, and easily shielded. It's a fantastic match for space exploration. Pity it's so difficult to manufacture.
posted by anonymisc at 10:38 AM on March 25, 2013 [1 favorite]


The damage of those structures by dust and micrometeorites will slowly whittle away at the ability to dump heat as well.

As long as the micro damage is whittling them down at a rate of 87-years half-life, you're golden :-)
posted by anonymisc at 10:43 AM on March 25, 2013 [2 favorites]


Can we also talk about nuclear thermal rocket engines and other nuclear based propulsion? NERVA, Orion, and more recently Prometheus, for the ion engine buffs. It's hard to beat Orion for cool weird-ass nuclear age-ness - I mean, a rocket fueled by repeated nuclear explosions? Awesome. Unless you're part of the ground crew at liftoff.
posted by Muddler at 11:42 AM on March 25, 2013


God, I'm still really stuck on a D cell that puts out fifty watts. It would still be putting out 25 watts, almost 90 years later!

Scale that up a little, and you could buy a box the size of a suitcase that would let you live, well past your present standard of power consumption, for the rest of your life. And your kids and grandkids could use it for another 90 years, as it decayed to 25% of the power it had when you bought it. If your family added a new one every other generation, your total power budget would be pretty nuts.

Buying power once every 60-90 years. That would certainly make power companies look rather different, wouldn't it?
posted by Malor at 11:55 AM on March 25, 2013 [1 favorite]


I mean, imagine: buy two. Put one in the house, and stick one in your car, and you could drive continuously for probably a hundred years. If your car wears out, swap the suitcase into a new one, and keep driving. Never buy gas again. No gas tank, no huge internal combustion engine, just some small electric motors. Would give you room for a second trunk up front, or a lot more cabin space. Absolutely unreal torque and performance; dump five or ten thousand watts of 'go' into a car, and it will freaking go.

Radioactivity is awesome! Why the hell do we have to be such horribly fragile creatures?
posted by Malor at 12:58 PM on March 25, 2013


Radioactivity aside, I'm not sure that would work too well. RTGs are good at producing a little electricity forever, not a lot on demand. I doubt they'd make a good power source for a car. Plus, I suspect that smaller RTGs are more effective than larger ones per unit mass, since they have a much more favorable surface area:volume ratio. A small RTG embedded in the body also has the significant advantage of being embedded in a giant circulated fluid heatsink.

Plus, uranium is expensive and there are pretty limited reserves, so those power sources would be cripplingly expensive and the cost wouldn't be able to come down very far, and we'd run out in a couple of generations.
posted by Mitrovarr at 1:15 PM on March 25, 2013


If we don't have 238Pu, we're not exploring past Mars

We'd not even be able to explore Mars effectively, since without RTGs we'd have to stick to lower latitudes, spend most of the day charging rather than exploring, and constantly worry about dust.
posted by RobotVoodooPower at 2:14 PM on March 25, 2013


Solid state RTGs have shit efficiency. Cassini's RTG had 6.7% efficiency converting thermal power to electricity. Add in the manufacturing costs and radiation hazard, and you have one pretty craptacular power source.
posted by ryanrs at 2:45 PM on March 25, 2013


Buying power once every 60-90 years. That would certainly make power companies look rather different, wouldn't it?

Yes, you might say it'd be too cheap to meter.
posted by Kadin2048 at 2:46 PM on March 25, 2013 [1 favorite]


Speaking of Cassini: Venus… from Saturn
posted by homunculus at 5:37 PM on March 25, 2013


It's hard to beat Orion for cool weird-ass nuclear age-ness - I mean, a rocket fueled by repeated nuclear explosions? Awesome. Unless you're part of the ground crew at liftoff.

It only seems weird because we tend to think of nuclear bombs as magic death machines. They produce a predictable amount of energy in a predictable way.

It also seems weird because the idea of atmospheric nuclear detonations is taboo for several very good reasons. I really don't care that fallout from a ground-launched Orion would be an order of magnitude less than annual fallout from above-ground nuclear testing in the sixties -- that's still far too high. I suppose it would be defensible if we needed a practical way of launching a 4000 tonne space battleship to fight invaders from outer space, but not otherwise.

The much less toxic option that was seriously considered in the 1960s avoided atmospheric nuclear detonations. One to three Saturn-5 rockets would be enough to put a baby Orion in orbit. When I say baby I mean a ship capable of sending a crew of eight to Mars in 125 days round trip.

For more, read this book.
posted by justsomebodythatyouusedtoknow at 5:51 PM on March 25, 2013


How about a super-long wire you attach to it that trails behind it, like 5 miles long. You could even shoot it up on a separate mission.

Unless it's a superconductor, it would be useless. The real problem with cooling fins is not radiative cooling into space, it's heat propagation along the fin from the heat source.

One of the best solutions to that is a circulating coolent. That's what the shuttle used. After launch they always opened the cargo bay doors, and left them open until just before reentry. That's because the inside of the cargo doors had big radiators, which were heated by a circulation system. If, for some reason, they were unable to open the doors after they achieved orbit they would have had to reenter within a few hours before the whole system overheated.

But if an impact hits part of the circulation system, the coolent leaks away. In the case of the shuttle that would have meant "Land within 4 hours". For a probe going to Jupiter it means "sayonara".

A long wire would be pretty much useless because only the first couple of meters would be warm enough to radiate away any heat. Everything beyond that would soon go to the ambient radiation temperature and stay there.
posted by Chocolate Pickle at 2:18 PM on March 26, 2013 [1 favorite]


In the case of the shuttle that would have meant "Land within 4 hours". For a probe going to Jupiter it means "sayonara".

Presumably you could have some sort of system of pressure-sensitive valves, or coolant that coagulates in vacuum or something. Also, there's (in theory) liquid droplet radiators which, if struck, would simple lose some liquid but keep going fine (unless the actual arm were hit, but you could just shut them off and switch to heat-sinks if that happens.
posted by cthuljew at 10:05 PM on March 26, 2013




« Older The economics of time travel   |   Bye bye Boris Newer »


This thread has been archived and is closed to new comments