2 ms·

> Superphénix was a fiasco. And France has since given up entirely on fast reactors. It was not really a fiasco; it was shut down before being really productiv

> Superphénix was a fiasco. And France has since given up entirely on fast reactors. It was not really a fiasco; it was shut down before being really productive. Sure, there were some wrinkles, but it was a one-off, not the 10th in a fleet so this sort of kinks are expected. It was an experimental reactor, the design was not supposed to be directly useable as a commercial reactor. Nevertheless, it had a whole scientific life ahead, some experiments would have been useful for nuclear fusion as well. There aren't many reactors able to produce fast neutrons with that spectrum. Regardless of the state of SFRs, closing it was a purely political decision, the consequence of horse trading between the socialists and the environmentalists without considering any engineering or scientific aspect. > BTW, you can infer from this that France doesn't think nuclear is going to power the world. France does not really think anything. There is no master plan, all the decisions are done in a certain context without really any long term thinking. What you see is an industry that has lost its political support. Now that some of the stupidity of the whole mess is dawning on some politicians there is a bit of flailing around, but do not mistake that for some kind of strategy. > If it did, breeders would be necessary, as there's not enough cheap enough uranium to do it with burner reactors. That is not entirely true. There are thermal reactors that can work with unenriched uranium, it's just that PWRs are not the most fuel-efficient design. In any case, no, nobody serious is advocating for using only nuclear. It is just one tool in the toolbox, and there are better sources for some uses.

pfdietz 4y ago
> That is not entirely true. There are thermal reactors that can work with unenriched uranium, it's just that PWRs are not the most fuel-efficient design. That's not a rebuttal at all. The problem isn't enrichment, the problem is there's not enough cheap uranium ore. If the world's primary energy demand (18 TW) were provided by burner reactors, the estimated global uranium resource (at cost sufficient to not seriously affect economics of burner reactors) would be exhausted in perhaps five years. This is so tight that nuclear w. burner reactors cannot provide more than a small fraction of the world's energy. Renewables would have to do almost all of it. In that scenario, there's little reason to use nuclear at all, as nuclear and renewables do not mix well on the grid or elsewhere.
- kergonath 4y ago
  > That's not a rebuttal at all. It's not supposed to be one. It's just that there is more nuances to this than 'thermal reactors are going to run out of fuel in 10 years' and 'fast reactors are going to solve the uranium supply'. PWRs and reactors that depend on enriched uranium use much more uranium ore than those that can burn natural uranium, and not all of them are breeders. > he problem isn't enrichment, the problem is there's not enough cheap uranium ore. Which is directly related to enrichment, because to enrich uranium you need one order of magnitude more ore than for unenriched fuel and we end up throwing away a lot of perfectly fine nuclides. Then there is the issue of reprocessing, because you are also throwing away a lot of perfectly fine nuclides if you cannot re-use spent fuel, because you need to reach certain enrichment levels. That is something doable even in PWRs. > If the world's primary energy demand (18 TW) were provided by burner reactors, the estimated global uranium resource (at cost sufficient to not seriously affect economics of burner reactors) would be exhausted in perhaps five years. I don't understand why you think this is not related to the type of reactor. Different designs have different fuel efficiencies, so a given amount of raw materials would go further with more efficient designs. Just like condensing furnaces allow to use less gas for a given heat output. > This is so tight that nuclear w. burner reactors cannot provide more than a small fraction of the world's energy. That is under very conservative assumptions about the types of reactors and state of ore deposits. Realistically, it's about 100 years, and that's likely overestimating needs because nuclear cannot do everything. > Renewables would have to do almost all of it. In that scenario, there's little reason to use nuclear at all, as nuclear and renewables do not mix well on the grid or elsewhere. There is nothing specific preventing nuclear and renewables to be used on the same grid.
  - pfdietz 4y ago
    > PWRs and reactors that depend on enriched uranium use much more uranium ore than those that can burn natural uranium. This has little to do with enrichment, and more to do with the better neutron economy of heavy water as a moderator. CANDU reactors have moved to use enriched fuel due to superior economics. CANDU reactors are still economically inferior to LWRs, though. In any case, the difference is negligible to the point being made: that burner reactors cannot power the world for very long before the uranium runs out. CANDU reactors still leave the vast majority of the 238U unfissioned. > Which is directly related to enrichment, because to enrich uranium you need one order of magnitude more ore than for unenriched fuel and we end up throwing away a lot of perfectly fine nuclides. To first order, burner reactors are burning 235U. It doesn't matter if this 235U has been concentrated or not. The 238U that enrichment tosses out can't be used in burner reactors in any case. (This is not entirely correct, since some 238U is converted to Pu even in burner reactors. But only a small part of it can be, since the breeding ratio is well below 1, enriched fuel or not.) > I don't understand why you think this is not related to the type of reactor. Because it's not. Burner reactors, by definition, run out of fissionables before they've converted most of the 238U to plutonium. > Realistically, it's about 100 years This must be assuming a much larger uranium resource. > There is nothing specific preventing nuclear and renewables to be used on the same grid. Yes there is. They are both inflexible sources, and compete with each other to be supported by sources of flexibility (dispatchable demand, dispatchable generation from e-fuels, storage, hydro). In general, when you look at optimal solutions to providing power for the grid, the optimal solution is either all renewables or all or mostly nuclear. There's no middle ground. Putting it another way: if it makes sense to have a grid that 10% nuclear and 90% renewable, it would make even more sense to have more nuclear. Nuclear either goes big or it goes home.

pfdietz 4y ago

> That is not entirely true. There are thermal reactors that can work with unenriched uranium, it's just that PWRs are not the most fuel-efficient design. That's not a rebuttal at all. The problem isn't enrichment, the problem is there's not enough cheap uranium ore. If the world's primary energy demand (18 TW) were provided by burner reactors, the estimated global uranium resource (at cost sufficient to not seriously affect economics of burner reactors) would be exhausted in perhaps five years. This is so tight that nuclear w. burner reactors cannot provide more than a small fraction of the world's energy. Renewables would have to do almost all of it. In that scenario, there's little reason to use nuclear at all, as nuclear and renewables do not mix well on the grid or elsewhere.

kergonath 4y ago
> That's not a rebuttal at all. It's not supposed to be one. It's just that there is more nuances to this than 'thermal reactors are going to run out of fuel in 10 years' and 'fast reactors are going to solve the uranium supply'. PWRs and reactors that depend on enriched uranium use much more uranium ore than those that can burn natural uranium, and not all of them are breeders. > he problem isn't enrichment, the problem is there's not enough cheap uranium ore. Which is directly related to enrichment, because to enrich uranium you need one order of magnitude more ore than for unenriched fuel and we end up throwing away a lot of perfectly fine nuclides. Then there is the issue of reprocessing, because you are also throwing away a lot of perfectly fine nuclides if you cannot re-use spent fuel, because you need to reach certain enrichment levels. That is something doable even in PWRs. > If the world's primary energy demand (18 TW) were provided by burner reactors, the estimated global uranium resource (at cost sufficient to not seriously affect economics of burner reactors) would be exhausted in perhaps five years. I don't understand why you think this is not related to the type of reactor. Different designs have different fuel efficiencies, so a given amount of raw materials would go further with more efficient designs. Just like condensing furnaces allow to use less gas for a given heat output. > This is so tight that nuclear w. burner reactors cannot provide more than a small fraction of the world's energy. That is under very conservative assumptions about the types of reactors and state of ore deposits. Realistically, it's about 100 years, and that's likely overestimating needs because nuclear cannot do everything. > Renewables would have to do almost all of it. In that scenario, there's little reason to use nuclear at all, as nuclear and renewables do not mix well on the grid or elsewhere. There is nothing specific preventing nuclear and renewables to be used on the same grid.
- pfdietz 4y ago
  > PWRs and reactors that depend on enriched uranium use much more uranium ore than those that can burn natural uranium. This has little to do with enrichment, and more to do with the better neutron economy of heavy water as a moderator. CANDU reactors have moved to use enriched fuel due to superior economics. CANDU reactors are still economically inferior to LWRs, though. In any case, the difference is negligible to the point being made: that burner reactors cannot power the world for very long before the uranium runs out. CANDU reactors still leave the vast majority of the 238U unfissioned. > Which is directly related to enrichment, because to enrich uranium you need one order of magnitude more ore than for unenriched fuel and we end up throwing away a lot of perfectly fine nuclides. To first order, burner reactors are burning 235U. It doesn't matter if this 235U has been concentrated or not. The 238U that enrichment tosses out can't be used in burner reactors in any case. (This is not entirely correct, since some 238U is converted to Pu even in burner reactors. But only a small part of it can be, since the breeding ratio is well below 1, enriched fuel or not.) > I don't understand why you think this is not related to the type of reactor. Because it's not. Burner reactors, by definition, run out of fissionables before they've converted most of the 238U to plutonium. > Realistically, it's about 100 years This must be assuming a much larger uranium resource. > There is nothing specific preventing nuclear and renewables to be used on the same grid. Yes there is. They are both inflexible sources, and compete with each other to be supported by sources of flexibility (dispatchable demand, dispatchable generation from e-fuels, storage, hydro). In general, when you look at optimal solutions to providing power for the grid, the optimal solution is either all renewables or all or mostly nuclear. There's no middle ground. Putting it another way: if it makes sense to have a grid that 10% nuclear and 90% renewable, it would make even more sense to have more nuclear. Nuclear either goes big or it goes home.