Making Plutonium-238 in large enough quantities is an ongoing challenge | Hacker News Reader

Making Plutonium-238 in large enough quantities is an ongoing challenge

popsci.com

136 pointsjonbaer88 comments

I've been wondering if tritium could be used as an alternative power source for the nuclear batteries used on space missions. Of course tritium has to be "manufactured" just like plutonium-238, but on the other hand it can be made from lithium using a fusion reactor [1].

But wait, you're thinking, fusion reactors based on current technology consume more energy than they can produce. That's true, but also irrelevant if your business model is producing tritium rather than energy. This is because the going rate for tritium (according to Wikipedia) is something like $30,000 per gram [2].

I'm not sure what the licensing requirements are on fusion reactors, but I think they're a lot less onerous than licensing a new fission reactor. I don't think there's any licensing requirement at all for a small fusion reactor like a Farnsworth fusor.

Disclaimer: This is just idle speculation because I haven't done the math.

[1] http://en.wikipedia.org/wiki/Fusion_power#Power_Production

[2] http://en.wikipedia.org/wiki/Tritium#Self-powered_lighting

Nuclear 'batteries' are radioisotope thermoelectric generators (RTGs). They require a pretty substantial amount of heat to produce electricity. I don't know about temperatures, but tritium produces a decay energy of 0.018590 MeV, versus plutonium-238's 5.593 MeV. Assuming you can scale it up directly, the 4.8kg of plutonium-238 in Curiosity would be equivalent to 1444kg of tritium. It's just way, way less energy dense.

Edit: This explanation is way, way off. See response from 'throwaway_yy2Di

There's two other differences you'd need to account for:

* One Pu-238 nucleus weighs 238 amu; a triton only weighs 3 amu

* Tritium decays 7 times faster -- a 12.32 year half-life, vs. 87.7 years for Pu-238

Put together, the specific power (watts/kg) of pure tritium is about twice as high. That's chemically just hydrogen gas, which I'd guess isn't practical. The solids with the highest hydrogen density, like polyethylene ([C2H4]n), would be less power-dense, but not by much.

But I think there's some slack here. The ESA is designing their future RTGs to use Americium-241, which is drastically worse (in terms of these figures), yet still usable.

Ah hah -- great points, thanks! Edited my original post to note that my analysis was woefully incomplete.

Maybe the biggest drawback for tritium is the decay energy of 0.02 MeV compared to 5 MeV of Pu238. On longer missions also half-life may create problems (12 years / 70 years).

E: I'm very slow with my phone...

That's a good point. Of course if you can produce tritium cheap enough it might be feasible to just start each mission with a grossly over-powered nuclear battery.

We may have to reconsider some aspects of non-proliferation for the sake of space exploration. Not just for nuclear batteries, but also for things like Project Orion which tried nuclear propulsion.

Orion is a really cool concept and it's a lot of fun to think about, but realistically any space propulsion system that horribly contaminates the atmosphere and fries every satellite anywhere near Earth is not an option.

You wouldn't use the nukes until you're safely out of earths atmosphere which is easily attainable by having two separate propulsion systems.

You lose much of the advantage of Orion, since Orion is extremely heavy, and you're doing the hardest part of the work without it. You also have to wait until you're pretty far away from Earth, not just out of the atmosphere, before you start blowing up nukes unless you want to destroy all electronics in sight.

Not if it is assembled by nanobots on the moon.

It's not really useful for the type of spacecraft we use today, but if you assemble the spacecraft in orbit and only use the nuclear propulsion once you are far enough away from satellites it looks great for gigantic payloads or very high speeds.

>any space propulsion system that horribly contaminates the atmosphere

Modern nukes are very clean. Even in the 70s, we had nukes that were up to 98% efficient. I can't find the numbers for newer bombs, but they are even cleaner.

High efficiency just means you're not leaving lots of uranium and plutonium behind. You're still leaving lots of fission products, and lots of neutron-activated materials from the non-nuclear components of the bomb.

There's a reason that testing nuclear devices in the atmosphere is frowned upon these days.

Doesn't Plutonium-238 occurs naturally along the Thorium decay chain? I believe that it's fairly easy (relatively speaking) to capture 238 from the Thorium nuclear reactor process.

A Thorium-based nuclear reactor prototype has already been successfully run for years at Oakridge National Laboratory- seeing the value of Plutonium-238, it might justify the cost of building a full grade Pu238-extraction nuclear reactor.

I've consistently heard the following two claims about thorium reactors:

1. They offer a number of advantages in safety, efficiency, input materials, byproducts. No comparative disadvantages.

2. The United States is not actively pursuing any plans to implement thorium reactors.

Is the above really true, or are there some comparative disadvantages to a Thorium reactor vs a traditional nuclear reactor? If the above is true, there must be some conflict of interest (imagine that, a conflict of interest in the domain of energy!) causing #2. Is there a legal issue?

A mix of politics, historical nuclear industry in-fighting, 9/11-terrorism-godzilla paranoia and oil industry lobbying make nuclear development very difficult.

It's also made nuclear power one of the safest and cleanest sources of power in the States.

China and India are very aggressively pursuing Thorium reactors though, which risks putting the West at another energy competitive disadvantage within decades.

From what I've read it's not a legal issue, it's a "huge amount of money needed to get thorium reactors to where they need to be" issue. Nobody is really interest in putting up that money because current modern reactors are "good enough" - all the problems with nuclear plants tend to come from old, bad ones that are decades behind the state of the art.

From what I have read, the best traditional fission plant designs are very safe, efficient, and produce few byproducts. It's the older plants that have problems but we have learned from them and made improvements.

Compared to the most state-of-the-art plant, thorium reactors are not significantly better and we have years of first-hand experience with traditional uranium reactors and very little with liquid fluoride thorium reactors so the incentives to start at square 1 with an new technology are small.

So the good news is if we ever got over our irrational fear of nuclear power, we have time-tested plans for safe, clean and efficient power plants ready to be built.

From what I've read, the reason for #2 is historical. You cannot make bombs with thorium reactors, and when bombs were being made is when most of the advances in nuclear power were made. Since then a number of accidents at nuclear power plants have rendered any advancements in nuclear power taboo in the US so there is no development, especially not in anything new nuclear-wise.

My understanding is that thorium reactors have a bit of a chicken and the egg problem.

We've got plenty of experience building more traditional nuclear designs - considering how conservative and concerned the masses tend to be when it comes to nuclear technology (not entirely unreasonably), do you want to be the one "risking" it on an "unproven" reactor design? It's politically tricky, in an age when several countries are phasing out nuclear power entirely in a knee-jerk reaction to Fukushima.

Even ignoring political risk, I'd imagine there's some budgetary risk for investing in the "cutting edge" of nuclear tech - more room for new mistakes to be made, additional complications to be discovered, resulting in budget overruns.

I believe China is investing in building thorium reactors. It wouldn't surprise me if at some point in the future, the US is left in the position of poaching nuclear talent from China to play catch up in the thorium reactor department.

Wikipedia lists a lot of possible disadvantages:

http://en.wikipedia.org/wiki/Thorium-based_nuclear_power#Pos...

http://en.wikipedia.org/wiki/Liquid_fluoride_thorium_reactor...

No, the naturally abundant isotope of Thorium has 6 fewer nucleons. Why would building a special reactor for a program that has so little funding be justifiable?

Because along with a steady source of Pu238, you also get the benefit of a safer, smaller nuclear reactor using a fuel that is significantly more common and less weaponizable than uranium/plutonium.

Please provide evidence that a Thorium reactor is safer. We have the means to make all the 238Pu we want. NASA just needs to pay commercial reactor operators to irradiate 237Np targets. This is a big problem, because NASA has very little funding to accomplish this, and the target material comes in a purified form suitable for direct use in nuclear weapons.

Is it practical to irradiate the target in units of mass that are smaller than the critical mass?

(The thought being that you can then only allow one unit to be in transit from storage to a reactor at any given time, pretty much entirely mitigating the risk of someone gathering enough of it to make a bomb)

You can make a bomb out of a little of this, and a little of that. A bomb requires much more than a critical mass anyhow.

On the safety: http://en.wikipedia.org/wiki/Liquid_fluoride_thorium_reactor...

Economic feasibility: http://www.thoriumenergyalliance.com/downloads/plutonium-238...

I'm a nuclear safety analyst. I need more than a powerpoint to convince me that an untested technology is safer than a technology that has seen billions of dollars of investment in enhanced technology, engineering, modeling, etc; along with decades of lessons learned. I'm not saying it couldn't be, but I think there are better ways to spend money on nuclear energy technologies right now, in terms of carbon offset and prevention of radioactivity release.

The Nature article that PopSci links to goes way more in depth and is a great read: http://www.nature.com/news/nuclear-power-desperately-seeking.... Also brings up discussion of the need of much more powerful energy sources (potentially nuclear) for proposed human space exploration missions.

If people (politicians, and public) would just get rid of their prejudices and we could just develop a nuclear reactor for outer system exploration and be done with it...

Fuel would not be such a pain to manufacture, there'd be oodles of more power available (more bandwidth for communications, nuclear electric propulsion for faster travel times), and the reactor would be just as safe, if not safer to lift to orbit (unused nuclear fuel is not radioactive, Pu-238 is, although it's so well sealed that it's practically impossible to free it to environment on an accident).

I wonder if for activities near the gas giants the massive Magnetospheres could be tapped for energy as the satellites move through them. However over time you'd be trading orbital elevation for energy.

Another great side effect of RTG's is the "waste" heat keeps your electronics from freezing in the deep of space.

The problem is that the production methodologies for Pu-238 are ridiculous and have a ridiculous number of volatile critical dependencies in both the creation and harvesting process.

"This has been the method of choice at the Department of Energy's Savannah River Site (SRS) production reactors. However, there are several shortcomings with this method of production. First, the production efficiency is quite limited, i.e., to approximately 13% efficiency. This is seen from the fact that the Pu-238 produced in the target after only 2.12 day's half-life decay, itself becomes a target for production of higher isotopes of plutonium, thus reducing the Pu-238 purity by producing Pu-239 and Pu-240. Second, this process produce a hazardous Pu-236 by-product. As noted on the decay chain, above, there is a η→2η or γ→η reaction that results in the production of Uranium-236 (U-236) and Pu-236. These reactions increase with exposure to fast neutron flux. Pu-236 decays to U-232, which has a hazardous gamma-ray energy emitting daughter product. Even a few parts per million U-232 increase the radiation exposure hazard to personnel dramatically. Lastly, Np-237 must be chemically purified before target fabrication. This is seen from the fact that Np-237 decays to Protactinium-233 (Pa-233), which in turn has a strong gamma-ray emission with its beta decay to U-233 (half-life 27 days). Therefore, the Np-237 was stored in solution at SRS and chemically processed immediately before fabricating targets. Solution storage of Np-237 may not be practical at an alternate production site."

Source: http://www.google.com/patents/US6896716

The extent to which this should worry our interstellar concerns is huge. The voyager 1 went several billion kilometers away from the Sun. You know it's crazy, it's so far away that it takes 15-20 hours for the signal to reach us.

The voyager 1 is probably going to go out in like 2020, but that's because of the Plutonium- 238.

There's 2 sites in the USA that can produce it.

Hanford in Washington State and Savannah River Site in South Carolina.

The Russians have Mayak, and thank goodness they kept making the stuff.

Hopefully we can pump some steroidal funding into High Flux Isotope Reactors.

I've been watching this issue since 2006. In 2009, someone finally spoke up about it and published some papers on it.

http://www.nap.edu/openbook.php?record_id=12653

Any NRE's or Particle Accelerator Designers on HN?

238Pu is always going to be problematic because of the fact that the target material is only available in the US in weapons-grade form. USG moved all of it to secure storage at INL, and are very cautious about any program to transport it for 238Pu production. Hopefully they can get ATR moving and keep missions supplied.

Was lucky enough to get to see Dave Lavery (Program Executive for Solar System Exploration at NASA) speak about Curiosity a year or so back; on the Rover's radioisotope thermoelectric generator, he mentioned that these devices come as a complete package from the DoE (so Nasa had little say in the power characteristics, for example, which constrained what they could do with it), and that technically they are 'on loan' to Nasa.

The DoE has no plans, so far as I'm aware, to go and get the one on Curiosity back when Nasa's finished with it.

That's the real mission for those unmanned boeing test spacecraft - they are the precursors and work-in-progress test craft for the classified military project to reach Mars 10 years before current estimates of civilian efforts (including NASA) in order to retrieve the generator from the rovers.

Realistically, any form of power generation could be used in space. It's really only a question of whether it's financially and practically feasible. I've been interested to see if Liquid Flouride Thorium Reactors (LFTRs) ever become a thing.

From what I've read about LFTRs, they can produce a whole lot of power- the most efficient system produces 1000-MW(e) from 700kg of fuel. That lasts approximately 12.5 years, give or take.

Obviously, thats on an Earth-based system and most spacecraft probably don't need anywhere close to 1GW.

Pu-238 has the highest energy density which is why its so useful (both as a weapon and an energy source). Its the cheapest to use because it adds the least amount of weight to the launch vehicle.

238Pu doesn't have anything to do with nuclear weapons. It is a highly unwanted contaminant of weapons-grade Plutonium.

Yes totally correct, its Pu 239 that's weapons grade. I got mixed up.

Well, "any" is a very broad definition

As long as it doesn't have any (constantly) moving parts, doesn't have requirements relating to pressure or radiation exposure, works in a variety of temperatures, it could be used.

Solar panels and RTGs fit the bill today. Maybe the ASRGs have moving parts, given their name?

Solar panels don't fit the bill for missions far from the sun.

> "Realistically, any form of power generation could be used in space. It's really only a question of whether it's financially and practically feasible."

Are you referring only to nuclear power sources? Because a literal reading of your comment ("any form of power generation") is obviously silly (e.g. steam power). And questions of financial and practical feasibility are two of the most important considerations for nuclear power systems, so one shouldn't dismiss those concerns so easily.

Yes, good point. That's what I meant.

Closed loop steam power is actually not that unreasonable in space.

Concentrated solar cells have been considered for a range of space operations. They generate a lot of waste heat, which opens the way for a range of heat engines. Currently, non concentrated solar is a clear winner but serious effort went into studying combined systems.

Long term several space mining concepts involve using concentrated sunlight to heat and refine asteroids. This would tend to make quite a bit of waste heat and steam power is one method to utilize that waste heat.

Thorium reactors may be useful on earth. But the hype is misplaced for space probes. Either they need maintenance (extracting fission products from the fuel) or closed cycle processing which makes them quite bulky (by space mission standards, not by nuclear reactor standards).

RTGs and stirling engines are fairly lightweight and compact. reactors provide more power per weight, but they have a much higher minimum weight.

And molten salts aren't exactly the most hassle-free technology there is. they are highly corrosive, so you have to compromise on the materials you use and it also strongly limits their lifetimes. RTGs on the other hand often operate long beyond their intended lifetime, albeit at reduced capacity.

I think the issue with Stirling Engines is the fact that they have moving parts that could wear and break down.

Most (all?) RTGs are for all intents and purposes, solid state.

Yes, but I was comparing to molten salt reactors. Stirling engines are still much simpler than reactors, even if they are slightly more complicated than RTGs. AIUI they already have research models that are capable of running for decades without maintenance.

I worked for about a year and a half on the ASRG project, specifically doing reliability work for the Stirling engines. They had a bit of a ways to go to prove out the system (its a tricky problem trying to "prove" that your device will last 17 years), but they had made a whole lot of progress. The biggest problem for the project though was that the ASRG wasnt being designed for any one mission, so its performance requirements were a bit of a moving target.

>I think the issue with Stirling Engines is the fact that they have moving parts that could wear and break down.

In NASA's ASRG design there is a stirling generator with only has one moving part: a free piston moving inside a coil, cushioned by the working gas. Wear and tear should be minimal, even for a multi-decade mission.

https://solarsystem.nasa.gov/rps/docs/ASRGfacts2_10rev3_21.p... (page 2)

Another aspect of this that the article fails to mention is that NASA has tried to come up with more efficient thermo-acoustic energy converters than the existing seebeck-effect RTG's for radioactive sources [1], but the projects went terribly over budget and got cancelled.

[1] http://www.planetary.org/blogs/guest-blogs/van-kane/20131208...

So I thought, maybe we could beam energy to satellites?

Using: http://www.pseudonomen.com/lasers/calculators/index.html

If I beam 1 GigaWatt to Jupiter from Earth at closest approach, with the same beam divergence as current moon-shooting laser technology, they could gather at 0.10655 W/m^2.. so terrible. Truly RTG FTW.

The article mentions other batteries that can use the existing Plutonium 238 more efficiently, but is it not possible to design a battery/generator that runs on an alternative nuclear fuel source?

Other fuels are possible, however Plutonium 238 is preferred for these purposes for a number of reasons: http://en.wikipedia.org/wiki/Radioisotope_thermoelectric_gen...

That's a great explanation, thank you.

Linkbait headline. As this article itself (last line) alludes to, production of plutonium-238 has restarted[0]. So we have enough for three more batteries, but are producing more...

[0] http://www.popsci.com/science/article/2013-03/first-time-col...

it's not a linkbait title, it's a really weird one (mixing registers/i.e. level of formality.) It sounds like someone setting their foot down, informally.

Apparently, the rate of production is enough for just one deep-space mission per decade.

Could we not reprocess uranium into plutonium from commercial reactors to recycle the waste and use that for space probes?

238Pu is only efficiently manufactured by activating 237Np.

If any nuclear material could be easily turned into any other nuclear material, life would be so much easier, and nuclear waste would be much less of a problem. ;)

well, they have one to use half a decade from now (2020), two more in reserve, and per your statement, they can make one per decade. Seems they are doing fine!

We changed the title to the last sentence of the article, which seems to make its point more neutrally. If anyone suggests a better (more accurate and neutral) title, we can change it again.

Maybe nasa can ask Iran for help.

It's a pity you're getting downvoted, I thought it was funny.

On a serious note, presuming a significant increase in trust among all parties, wouldn't this idea radically change an ominous situation (Iran having weapons-grade plutonium laying around) into one with peaceful, mutually beneficial, trust-building cooperation between formerly antagonistic nations? Have not stranger things happened?

Doesn't a lot of US nuclear power come from dismantling Russian weapons?

Iran was enriching Uranium, not producing Plutonium. Furthermore, this type of Plutonium is not what you desire for nuclear weapons material.

Iran has a plutonium reactor as well, although its exact capabilities and output are a matter of debate: http://en.wikipedia.org/wiki/IR-40

I'm skeptical that this small reactor rivaled the Uranium program. If your rush 239Pu production, you end up with a product that is very difficult to purify, raises the probability of a fizzle, and generates thermal problems for the pit.

The needed isotope (Pu-238) is only present in small amounts in used reactor fuel.

The concern with Iran is that they might enrich uranium to the point where it is useful in a nuclear weapon (it is useful in a reactor well before this point).

Reply on news.ycombinator.com