Hacker News Reader

Comment by riffic | Hacker News Reader

this reads like a too good to be true pitch, so I have to ask, what's the catch?

edit: these are breeder reactors? aren't there still waste and other concerns here?

On the design page they describe this as a molten salt reactor with dissolved nuclear fuel. One of the principle difficulties designing this type of reactor is the choice of materials used to conduct the liquid fuel, which is very hot, caustic, and radioactive.

Steel and aluminum are out, as they are both chemically reactive at those temperatures. Other materials such as nickel alloys or certain ceramics may be possible, but the radioactivity of the fuel tends to embrittle materials over time as atoms in the containing material are converted by exposure to the fuel's neutron emissions.

The affected components include not just the reactor vessel, but also pipes, and in some MSR designs also pumps and heat exchangers. If those are subject to regular replacement, the used components are also low-grade nuclear waste and must be properly contained indefinitely.

https://en.m.wikipedia.org/wiki/Molten_salt_reactor

The other main impediment is regulatory: since we've never had commercial reactors of this type we don't have a legal framework to license these types of reactors to be run (and eventually decommissioned) safely.

At least, these are my novice understandings from some skeptical background reading. If someone with actual nuclear industry and/or research experience has a more detailed understanding, feel free to weigh in and correct me.

A quibble, pipe embrittlement is not mainly from transmutation, but from atoms knocked out of place in the crystal structure.

One way that Thorcon deals with that is by replacing reactor cores every four years. Each module has two sealed "cans," with only one operating at a time, while the other cools down and then is replaced.

https://thorconpower.com/design/

This seems realistic since Oak Ridge ran their experimental molten salt reactor for five years (though only equivalent to 1.5 years at full power). They used an nickel alloy called Hastelloy-N, didn't see much corruption or neutron embrittlement, and did later work on improving resistance to embrittlement.

https://en.wikipedia.org/wiki/Molten-Salt_Reactor_Experiment

The catch is that no units have actually been built yet and the things that work well in theory may not work well in practice. Take it with the same tempered optimism as a startup pitch for a great new battery or solar cell design.

It may be interesting to note that a similar design was built and operated continuously back in 1964 for 4 years successfully. There's some material science that needs to be ironed out for commercialization, but it's a viable technology for sure, IMHO.

> There's some material science that needs to be ironed out for commercialization

Worth noting that this is no small thing. Materials science is the limiting factor of all molten salt reactors, and has yet to be solved satisfactorily.

I like how Moltex asked "wait, why are we pumping all this fuel around in the first place?"

https://youtu.be/7qJpVClxzVM?t=536

Molten has a good point. Pumping, what, tens or hundreds of tons of highly radioactive, hot, corrosive liquid around puts a crushing demand on the perfection of your plumbing, heat exchanger, and pumps. What do you do when (not if, we know plumbing) it leaks?

You're referring to the Molten Salt Reactor Experiment. That was a promising experiment, but the MSRE had thermal power output of 8 megawatts while the "can" in the Thorcon design is supposed to output 557 MWth -- 70 times more. The MSRE also operated for a total of 11555 hours full-power-output equivalent, about 1.3 years, whereas these reactor cans are supposed to run for 4 years at full power before replacement. There are reasons to believe that the physics are fine, but there are a lot of things other than physics that can impair an energy technology's journey from laboratory to industry. Still, I wish them luck.

https://en.wikipedia.org/wiki/Molten-Salt_Reactor_Experiment...

I think the neat thing about this is that they're starting with a design that, even if something goes a bit screwy, is still safe. They're obviously doing their best to design it well, but also relying on actual tests to see how well it performs in practice, which is something that you can't really do with the nuclear regulatory environment in the USA, if I understand correctly.

And part of this document[1] says "if it breaks, send it back". The design allows for replacing a faulty "can" entirely rather than trying to repair it.

[1] https://thorconpower.com/docs/domsr20180119.pdf

There are definitely waste concerns. There are a number of other problems, though most have to do with getting up and running.

Thorium is a by-product of rare-earth mineral mining. It is still radioactive, and it is also subject to all the regulatory red tape of handling any kind of radioactive material.

Rare-earth mineral mining is extraordinarily destructive to the environment. When you read about rare-earth mineral mining, you should think "strip mining" because that is generally how it's done. China does it because they don't care at all about the environment, they care about cornering the market.

Thorium reactors still include uranium to make them work. It may take eight years to exhaust the fuel, but you still have a pile of spent radioactive material that needs to go somewhere, and no one has solved that problem satisfactorily, which is bigger than the NIMBY issue of where the reactor goes in the first place.

Would it be fair to say that it's not a problem that has really been solved because it's a relatively small and non urgent problem? The spent fuel isn't dangerous while it just sits there, and there isn't that much of it. My understanding is that most spent fuel just gets stored on site, because it's the easiest way. I guess my point is that having to deal with storing solid, static nuclear fuel seems like a pretty good problem to have, compared to having to somehow capture the billions of tons of CO2 that we've put into the atmosphere from burning fossil fuels for power.

And I'm not sure about the relative environmental impact, but it's important to note that this thorium and uranium mining would be competing with coal mining.

Also note that coal power plants release many times more radiation in the environment than nuclear power does.

Yeah, I'm personally a proponent of these kinds of reactors over coal or traditional fission reactors. The post I was responding to was asking for devil's advocate instead of all rosy.

I think the upside tradeoffs for this kind of nuclear power are preferable to going without.

Its incredibly hard to build a nuclear reactor and its even more hard to get anybody to allow you to build one. And arguably even harder for somebody to give you the money to do either.

Its not a breeder, they are going for the simplest possible design as far as I remember and that means its gone a burner.

As far as it being harder to let anyone allow you to build one, and to get the money -- they've already got a thing going with the government of Indonesia. It's been in the works for years, and it seems like things are still progressing[1] fine.

[1] https://www.youtube.com/watch?v=azzp0i0BnQ8 [video]

Yes, I know. I wrote about it my other comment. You can find full presentations that they are giving to different officials.

Ah, ok, I misunderstood. What did you mean by "gone a burner"?

I wanted to say 'its going to be a burner type reactor rather then a breeder'.

In nuclear reactors a "burner" is like a conventional reactor, with uranium fuel and slow neutrons, which mainly just fission the U235 (plus a bit of plutonium that shows up). It means you're fissioning about 1% of your uranium ore.

A "breeder" can be one of two things:

1) Thorium fuel, slow neutrons. A neutron hits thorium, converts it to U233, then that fissions. Take a ton of thorium, fission all of it, get one gigawatt-year.

2) Uranium fuel, fast neutrons. The fast neutrons can fission U235, U238, plutonium, and other transuranics. Take a ton of uranium, fission all of it, get one gigawatt-year.

Compared to burners, either type of breeder only requires 1% as much fuel, produces 1% as much waste, and the waste is mostly fission products with much shorter half-lives. Encase it in glass and bury it, and it'll be back to the radioactivity of the original ore in about 300 years, with most of the radioactivity going away in the first few decades.

You could have a Thorium fast breeder as well I think.

I think a thermal breeder can actually be about 10x more efficient then any uranium fast burner can achieve.

But once you do a fast breeder there is really no point in using Thorium at all. The only reason to use thorium is really thermal breeding.

Also you can have both a thermal and a fast burner I'm pretty sure.

"Well, it turns out the fast breeder reactor made out of sticks was a bit of a fiasco."

Peppa Porcine-Bacon, CEO of Threelittlepigs, LLC in a prepared statement.

Reply on news.ycombinator.com