The complete lack of any details may fly on Chinese Central Television and Reuters.com, but I want more when I come to hacker news. What's next? "North Korea boasts development of time machine!"
They recently built an R&D-scale PUREX plant on the site of a weapons complex. The most likely "story" is that it started operating.
>The Jiuquan Atomic Energy Complex (Plant 404) is centered on a 400-500MW light water graphite reactor [LWGR] fueled with natural Uranium. The facility includes a pilot plutonium reprocessing plant and a arge-scale reprocessing plant with a capacity of 300-400kg Pu/yr (both use PUREX method). It also includes a Nuclear Fuel Processing Plant for refining plutonium into weapons-usable metals.
>The small reprocessing plant in Gansu was completed about six years ago, but Chinese oversight agencies have not yet permitted it to separate any plutonium—as of last month, just two fuel assemblies have been dissolved in the nitric acid tank at the plant. Full-scale operation is not expected before 2012.
China has a long-term vision (not really making progress) of closed fuel cycles based on fast breeder reactors. This is the state media reference to "60-fold increase in uranium usage", "3,000 year uranium supply". Since a fast reactor can't burn through fuel entirely in one go (it will fall apart), one of the requirements of a closed fuel cycle is reprocessing: fuel has to be removed from a reactor and re-fabricated. Certain waste products need to be removed, and pure plutonium* isolated for new fuel (this is a chemical separation).
*(Not in every reprocessing method. COEX (mentioned in the Carnege Endowment article, where the French are trying to persuade China to adopt it) co-extracts plutonium with uranium. The pyroprocessing method developed at Argonne (US national lab) co-extracts plutonium with every trans-uranic, so that the resulting fuel is essentially unusable as weapons. China is pursuing PUREX.)
Who knows what the North Koreans are planning. They could be intending to use their nuclear program to fuel a train capable of travelling at extreme velocities so as to cause time dilation effects so that their leader can rule for an unnaturally long time.
This kind of claims has been heard in France as far as I can remember it. I don't know how technologies differ between French technology (super-phenix) and chinese ones, but I would be warry about the claims before seeing concrete results: fast breeder super phenix was supposed to solve most issues with civil nuclear usage, and it cost a fortune without producing electricity anywhere near reasonable cost.
Thorium-based designs seem more realistic and useful, I wonder why this is not investigated more: it is more common than uranium and produces less nuclear waste than uranium-based designs. The US has a lot of it, Europe has a lot of it so it makes sense for countries like France to look at it, etc...
I'd be wary of calling sodium fast reactors inferior on the basis of their past performance. All SFRs ever built were essentially prototypes; all their operating experience combined is less than what LWRs had in the 1960's. If thorium reactors had fewer spectacular commercial failures (e.g. [1]), it's largely because they had fewer commercial attempts.
(That's not to concede that SFR failures were entirely commercial -- political obstruction was a large factor too. Both anti-nuclear opposition in general (post-Three Mile Island and post-Chernobyl) as well as specific, extremely misguided opposition to plutonium breeding. Argonne's IFR project was killed for this second reason [2]).
>Thorium-based designs seem more realistic and useful, I wonder why this is not investigated more: it is more common than uranium and produces less nuclear waste than uranium-based designs.
There's no meaningful distinction between FBRs and thorium reactors in fuel supply -- both have theoretically thousands of years of potential, so that it's silly to call this a factor. Nor is there a significant difference in waste production. (Though if I remember right, some minor actinides can only be efficiently destroyed by fast neutrons. This isn't an important difference; MAs are a small component, so you can augment a thermal-reactor fuel cycle with dedicated "waste burners" (e.g. [3] [4]) and still have a mostly-thorium fuel cycle).
I did not call anything inferior - I don't know anything about nuclear technologies, so I could not make any judgement of that nature. My argument was that fast breeder is nothing new in principle, and the claims as stated in the article are the same as the ones we heard in france when speaking about super phenix.
I don't think super phenix can be considered as a prototype: as a target production power of > 1000 MW, it is closer to conventional, commercial reactors (those are theoretical numbers, which were never reached, let alone sustained).
Thorium-based reactors are more mature than FBR as far as I understand, and my argument about thorium was against conventional reactors as stated, not fast breeders. As for supply, besides amount, the location also matters a lot. There is thorium in Europe, but little uranium as far I understand, which matters quite a bit.
It's not the size, it's the maturity. As you correctly criticize, Superphenix was non-functional much of the time: there were major engineering issues which hadn't been solved beforehand. This is the character of a prototype.
>Thorium-based reactors are more mature than FBR as far as I understand
Thorium reactors are even less mature than FBRs. Although that's a very loose category, you should distinguish between the extremely different reactor types which burn thorium, such as high temperature gas-cooled reactors (including pebble beds), which were the only commercial thorium reactors (for a short time); and the theoretical molten salt reactors, which unlike HTGRs are breeder reactors. For added confusion, some commercial (both LWR and HWR) reactors can run partially on thorium, but are not self-sufficient without uranium or plutonium fuel.
You need more than a new reprocessing plant to significantly extend uranium supplies, you also need a new kind of reactor.
The Light Water Reactor (currently in use worldwide) produces a small amount of Plutonium while it operates. With reprocessing of both the U235 and Pu in Uranium fuel, you can get 30% more energy per pound. That's nice, but it's nothing like a fast reactor which can produce more Plutonium than it consumes... With fast reactors, it's possible to burn not only U235, but U238, which gives the 60x fold increase in energy mentioned in the articles.
Note that the nuclear waste that we've currently got cooling in dry cast storage is mostly U238 with a little Pu and U235 in it. Reprocessed, the current stock of spent fuel in the U.S. could power the country for at least a century before we'd need to start mining Uranium again.
Yes, the technology isn't mature, but the price is enormous -- essentially unlimited energy, since uranium can be extracted from seawater at a reasonable price if we can burn the U238.
I get the impression that sodium cooled reactors with water as a working fluid in the turbines are a non-starter today. There's not only the reactivity problem, but also the fact that the steam turbines are H-U-G-E. I think commercially workable sodium cooled reactors will use something like supercritical CO2 as the working fluid, which can produce something that's simultaneously safer and much much cheaper.
Not an engineer, but I don't think the sodium/water heat exchangers are a fatal setback -- for one thing, there's an intermediate sodium loop between the core sodium coolant and the turbine loop, so any sodium/water interaction* isn't near the reactor core. I'm pretty sure all SFRs to date have had steam turbines.
* euphemism
>but also the fact that the steam turbines are H-U-G-E.
That's not enough of a setback to stop most power plants in the world from using the Rankine cycle. You should ask an engineer why this is, I am not one.
They're trying to use more of the uranium in the fuel, by extracting left-over U-235 and bred Pu-239 to use as fuel again. The U-238 can also be used in breeder reactors, once we get those going. It's recycling! Sort of.
17 comments
[ 3.6 ms ] story [ 45.9 ms ] thread>The Jiuquan Atomic Energy Complex (Plant 404) is centered on a 400-500MW light water graphite reactor [LWGR] fueled with natural Uranium. The facility includes a pilot plutonium reprocessing plant and a arge-scale reprocessing plant with a capacity of 300-400kg Pu/yr (both use PUREX method). It also includes a Nuclear Fuel Processing Plant for refining plutonium into weapons-usable metals.
http://www.globalsecurity.org/wmd/world/china/jiuquan.htm
>The small reprocessing plant in Gansu was completed about six years ago, but Chinese oversight agencies have not yet permitted it to separate any plutonium—as of last month, just two fuel assemblies have been dissolved in the nitric acid tank at the plant. Full-scale operation is not expected before 2012.
http://carnegieendowment.org/publications/index.cfm?fa=view&...
More about PUREX, reprocessing:
http://en.wikipedia.org/wiki/Nuclear_reprocessing
http://www.inl.gov/technicalpublications/KeyResults.aspx?key...
China has a long-term vision (not really making progress) of closed fuel cycles based on fast breeder reactors. This is the state media reference to "60-fold increase in uranium usage", "3,000 year uranium supply". Since a fast reactor can't burn through fuel entirely in one go (it will fall apart), one of the requirements of a closed fuel cycle is reprocessing: fuel has to be removed from a reactor and re-fabricated. Certain waste products need to be removed, and pure plutonium* isolated for new fuel (this is a chemical separation).
*(Not in every reprocessing method. COEX (mentioned in the Carnege Endowment article, where the French are trying to persuade China to adopt it) co-extracts plutonium with uranium. The pyroprocessing method developed at Argonne (US national lab) co-extracts plutonium with every trans-uranic, so that the resulting fuel is essentially unusable as weapons. China is pursuing PUREX.)
Thorium-based designs seem more realistic and useful, I wonder why this is not investigated more: it is more common than uranium and produces less nuclear waste than uranium-based designs. The US has a lot of it, Europe has a lot of it so it makes sense for countries like France to look at it, etc...
(That's not to concede that SFR failures were entirely commercial -- political obstruction was a large factor too. Both anti-nuclear opposition in general (post-Three Mile Island and post-Chernobyl) as well as specific, extremely misguided opposition to plutonium breeding. Argonne's IFR project was killed for this second reason [2]).
>Thorium-based designs seem more realistic and useful, I wonder why this is not investigated more: it is more common than uranium and produces less nuclear waste than uranium-based designs.
There's no meaningful distinction between FBRs and thorium reactors in fuel supply -- both have theoretically thousands of years of potential, so that it's silly to call this a factor. Nor is there a significant difference in waste production. (Though if I remember right, some minor actinides can only be efficiently destroyed by fast neutrons. This isn't an important difference; MAs are a small component, so you can augment a thermal-reactor fuel cycle with dedicated "waste burners" (e.g. [3] [4]) and still have a mostly-thorium fuel cycle).
[1] http://en.wikipedia.org/wiki/THTR-300
[2] http://www.pbs.org/wgbh/pages/frontline/shows/reaction/inter...
[3] http://energyfromthorium.com/2009/11/11/lftr-and-msrs-achiev...
[4] http://www.oecd-nea.org/ndd/reports/2002/nea3109.html
I don't think super phenix can be considered as a prototype: as a target production power of > 1000 MW, it is closer to conventional, commercial reactors (those are theoretical numbers, which were never reached, let alone sustained).
Thorium-based reactors are more mature than FBR as far as I understand, and my argument about thorium was against conventional reactors as stated, not fast breeders. As for supply, besides amount, the location also matters a lot. There is thorium in Europe, but little uranium as far I understand, which matters quite a bit.
>Thorium-based reactors are more mature than FBR as far as I understand
Thorium reactors are even less mature than FBRs. Although that's a very loose category, you should distinguish between the extremely different reactor types which burn thorium, such as high temperature gas-cooled reactors (including pebble beds), which were the only commercial thorium reactors (for a short time); and the theoretical molten salt reactors, which unlike HTGRs are breeder reactors. For added confusion, some commercial (both LWR and HWR) reactors can run partially on thorium, but are not self-sufficient without uranium or plutonium fuel.
The Light Water Reactor (currently in use worldwide) produces a small amount of Plutonium while it operates. With reprocessing of both the U235 and Pu in Uranium fuel, you can get 30% more energy per pound. That's nice, but it's nothing like a fast reactor which can produce more Plutonium than it consumes... With fast reactors, it's possible to burn not only U235, but U238, which gives the 60x fold increase in energy mentioned in the articles.
Note that the nuclear waste that we've currently got cooling in dry cast storage is mostly U238 with a little Pu and U235 in it. Reprocessed, the current stock of spent fuel in the U.S. could power the country for at least a century before we'd need to start mining Uranium again.
Yes, the technology isn't mature, but the price is enormous -- essentially unlimited energy, since uranium can be extracted from seawater at a reasonable price if we can burn the U238.
http://www.neimagazine.com/story.asp?sectionCode=132&sto...
I get the impression that sodium cooled reactors with water as a working fluid in the turbines are a non-starter today. There's not only the reactivity problem, but also the fact that the steam turbines are H-U-G-E. I think commercially workable sodium cooled reactors will use something like supercritical CO2 as the working fluid, which can produce something that's simultaneously safer and much much cheaper.
* euphemism
>but also the fact that the steam turbines are H-U-G-E.
That's not enough of a setback to stop most power plants in the world from using the Rankine cycle. You should ask an engineer why this is, I am not one.
Say again? That means you use more uranium...
Another failure to consider that energy demands grow pretty much exponentially.
http://en.wikipedia.org/wiki/File:World_Energy_consumption.p...
And that's ignoring the fact that there were less than 3 billion people in the world in 1965.