Hacker News new | past | comments | ask | show | jobs | submit login

In the US the EBR I melted down as did the Enrico Fermi reactor in Detroit. Both of them were put back together and operated after the fuel damage.

EBR II and FFTF operated with no major problems in the US, but the Clinch River project was canceled. However the US has failed to construct a mixed-oxide fuel facility which is the most developed fuel for a fast reactor (e.g. pure metal and nitrides are both options but oxide fuels seem to require a high energy ball mill that can turn harmless silica into deadly nanoparticles that can wreck your lungs, just imagine what it does with Pu)

MONJU burned up in Japan.

The Russians have documented hundreds of sodium fires in the BN-350 and BN-400 but they would say fires happen in industrial facilities all the time and people just put them out. They turned on the BN-800 in 2016 and it is now consuming nuclear weapons plutonium.

Old literature suggests that the capital cost of the LMFBR is inevitably worse than the LWR. I'm not sure that's right. Both the LWR and coal burning power plants have been uncompetitive since 1980 against gas turbines for power generation. A closed-cycle gas turbine would fit in the employee break room of the turbine house of a nuclear reactor, but it is an undeveloped technology. A high temperature reactor with a gas turbine powerset could potentially be an order of magnitude smaller than an LWR, not have high pressures, have no risk of a steam explosion -- I think next generation nuclear is not worth pursuing unless the capital cost can be brought radically down with those factors.

There is a precedent for commercial licensing of the LMFBR in the US. Terrestrial Energy (Bill Gates's company) is making noises about building one but they are holding their cards so close to their chest that it's hard to believe they have a realistic plan to do it.




> However the US has failed to construct a mixed-oxide fuel facility which is the most developed fuel for a fast reactor (e.g. pure metal and nitrides are both options but oxide fuels seem to require a high energy ball mill that can turn harmless silica into deadly nanoparticles that can wreck your lungs, just imagine what it does with Pu)

Oxide fuel technology is of course pretty mature, as that's what current generation LWR's all use. However, it has severe disadvantages in a LMFBR. US thinking was apparently that they would jump straight to metal fuel (EBR-II and the canceled IFR), other who have used it see it as a stopgap until nitride fuels become available (e.g. Russia is experimenting with nitrides in their fast reactors).

As for economics, technically breeding is extremely cool (catnip for physicists!), but with current uranium prices there's no particular economic driver to develop and commercialize the technology. I think it's very hard to compare prices for mature LWR technology and various one-off fast breeder prototypes, let alone paper reactors. It's useful as a backstop for rising uranium prices, however.

> There is a precedent for commercial licensing of the LMFBR in the US. Terrestrial Energy (Bill Gates's company) is making noises about building one but they are holding their cards so close to their chest that it's hard to believe they have a realistic plan to do it.

The Bill Gates backed company is Terrapower. They have some agreement with GE Hitachi to develop and commercialize the technology, called "Natrium". AFAIU the Natrium reactor is heavily inspired by the canceled IFR, using metal fuel. We'll see what happens, interesting tech though.

(Terrestrial Energy is the company behind the "IMSR" molten salt reactor. It's a thermal spectrum uranium burning, non-breeding reactor design, though)


You are right about terrapower.

But uranium oxide vs mixed (plutonium) oxide fuel (used in LMFBR) is a big difference. The later is made by making plutonium nano particles that are potentially deadly for workers in the factory.

France doesn’t get excited about it but the US doesn’t seem to think it can be done safely.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: