6 ms·

Nuclear waste could power the U.S. for 100 years if utilized

20 points by serverlessmom 4y ago

Vladimof 4y ago
Nuclear waste from power plants?
lazyeye 4y ago
What kind of waste do fast reactors produce?
- credit_guy 4y ago
  Reactors in general produce 3 types of waste: - fission products: after a Uranium or Plutonium atom splits, the daughter atoms are virtually always radioactive; the vast majority of these atoms have a half life of less than 100 years - actinides: Uranium and Plutonium absorb a neutron but don't undergo fission. The resulting atoms, heavier than U or Pu, are part of the Actinide group, and they are quite nasty - unburned fissile material: no reactor has a 100% efficiency, a part of the U235 or Pu239 is left unburned - low and intermediate level waste: various materials inside and around the reactor, which either absorb neutrons and become "activated", i.e. radioactive, or simply get mixed up with particles of the first 3 categories. Fast reactors produce very little actinides. Maybe 1-2% of what regular reactors produce. They also burn much more of their fuel, so there's less waster of the third category. The coolant in various fast reactors is sometimes Sodium. Sodium very rarely becomes activated, and when it does, it transmutes into an element with half life of 15 days, which then decays into a stable isotope. So, after a few years, what's left is not radioactive anymore. If however, the coolant gets contaminated with particles from the nuclear fuel, then it becomes more complex. For modern reactors, the Department of Energy envisions using Triso fuel, which they consider to be virtually indestructible [1]. If that is so, then the coolant in sodium-cooled fast reactors can be considered a very law level waste. Another type of waste is the structural steel around the core subjected to the neutron bombardment coming from the core. That steel, when activated, transmutes into some isotopes that have a half-life of decades, so that's a bit nasty. That's not specific to fast reactors, it's something in common with regular reactors. [1] https://www.energy.gov/ne/articles/triso-particles-most-robust-nuclear-fuel-earth https://www.energy.gov/ne/articles/triso-particles-most-robu...
pfdietz 4y ago
Such a shame that it's more expensive than fuel made from fresh uranium.
cyanydeez 4y ago
Until the political system is stable, I wouldn't trust nuclear energy. Just imagine trump deux telling everyone nuclear reactors no long need regulation
- bergenty 4y ago
  Yeah I wouldn’t trust trump/trumpesk with it but we can’t stop that from us doing what’s best for the environment and the country.
  - ncmncm 4y ago
    We are doing what is best by not throwing good $billions after bad down the nuke rathole. Renewables are now by a huge margin the cheapest power generation system ever created. After enough renewables have been built out to supply power and charge up storage, it will be time to start building out storage, which will then serve base load. Storage cost is falling even faster than wind or solar PV ever did, so by that time many kinds of storage will also be very cheap. We will use many kinds of storage because each has different cost structure: Batteries, per MWh stored, MW in or out cheap, so good for peak smoothing. Synthetic ammonia and hydrogen, also liquified nitrogen, per MW in, tankage almost free, and transportable, MW out via existing combined cycle turbines. Pumped hydro, per MW, MWh cheap up to a hard limit. Undersea pumped hydro, per MWh but very cheap if the pump is onshore. Undersea buoyancy, very cheap all around provided the winch is onshore. Underground compressed air, per MW in, MWh stored very cheap, MW out via turbines maybe boosted with hydrogen or ammonia combustion. There is no place in the future for overpriced nukes built with coerced money.
Maursault 4y ago
This has been known since fast breeder reactors were initially designed to be able to do this, idk when 2-3 decades ago? The problem here is the reason only conventional fission reactors have been built is that they are the least expensive designs to build. And the problem today is that even the cheaper fission reactors are too expensive to be profitable, even with all the government subsidies and not burdening investors with decommissioning, and especially storage and security of the nuclear waste, which is a significant cost left to be paid for by tax payers. And the expense is not because of regulations. It may cost $10B-$15B to build and bring a new fission plant online, but regulatory costs would only account for about $10M/yr, thus the effect of regulations on the nuclear industry is insignificant compared to the bulk of the costs, such as uranium, construction, waste storage and security, and decommissioning, and the costs of waste storage and security and decommissioning are never included in the cost of a fission plant. Once you include those costs, the real price of a fission plant is revealed to be $50B and much more. So as it is, this idea is a non-starter, because fast breeder reactors are far more expensive to bring online than conventional fission reactors, and those are already too expensive as it is. But I have an idea. Make cheaper fast breeder reactors! Scale them up or down, idk which, until the economics makes sense enough that they can attract investors and actually be built. Just make them economically feasible without burdening the tax payer, and we can shut the pro-nuke crowd up for good.
- credit_guy 4y ago
  > And the expense is not because of regulations. [...] regulatory costs would only account for about $10M/yr. You are quite funny. NuScale received the NRC approval for their Small Nuclear Reactors (SMR) in 2020. They spent half a billion dollars for that, and 2 million man-hours [1]. They were founded in 2007 and the approval came 13 years later. If they are lucky, they'll have their first SMR working by the end of the decade. Of course, that's a happy ending. Oklo, another nuclear startup was founded in 2013, and got their application rejected by the NRC in 2022 [2]. They were told they can try again, and they probably will, but for the time being they have little to show for 9 years of work. [1] https://www.nuscalepower.com/newsletter/nucleus-fall-2020/program-development-update https://www.nuscalepower.com/newsletter/nucleus-fall-2020/pr... [2] https://www.nrc.gov/reading-rm/doc-collections/news/2022/22-002.pdf https://www.nrc.gov/reading-rm/doc-collections/news/2022/22-...
lesam 4y ago
How does this article not talk about France, which had a long-running commercial breeder reactor shut down more by politics than cost? The real issue is that breeder reactors produce weapons grade plutonium, and so the US doesn’t like to see them built.
- pfdietz 4y ago
  France's long running commercial breeder program was an expensive fiasco. It was shut down because of pointlessness, not because of politics.

20 points by serverlessmom 4y ago

Vladimof 4y ago

Nuclear waste from power plants?

lazyeye 4y ago

What kind of waste do fast reactors produce?

credit_guy 4y ago
Reactors in general produce 3 types of waste: - fission products: after a Uranium or Plutonium atom splits, the daughter atoms are virtually always radioactive; the vast majority of these atoms have a half life of less than 100 years - actinides: Uranium and Plutonium absorb a neutron but don't undergo fission. The resulting atoms, heavier than U or Pu, are part of the Actinide group, and they are quite nasty - unburned fissile material: no reactor has a 100% efficiency, a part of the U235 or Pu239 is left unburned - low and intermediate level waste: various materials inside and around the reactor, which either absorb neutrons and become "activated", i.e. radioactive, or simply get mixed up with particles of the first 3 categories. Fast reactors produce very little actinides. Maybe 1-2% of what regular reactors produce. They also burn much more of their fuel, so there's less waster of the third category. The coolant in various fast reactors is sometimes Sodium. Sodium very rarely becomes activated, and when it does, it transmutes into an element with half life of 15 days, which then decays into a stable isotope. So, after a few years, what's left is not radioactive anymore. If however, the coolant gets contaminated with particles from the nuclear fuel, then it becomes more complex. For modern reactors, the Department of Energy envisions using Triso fuel, which they consider to be virtually indestructible [1]. If that is so, then the coolant in sodium-cooled fast reactors can be considered a very law level waste. Another type of waste is the structural steel around the core subjected to the neutron bombardment coming from the core. That steel, when activated, transmutes into some isotopes that have a half-life of decades, so that's a bit nasty. That's not specific to fast reactors, it's something in common with regular reactors. [1] https://www.energy.gov/ne/articles/triso-particles-most-robust-nuclear-fuel-earth https://www.energy.gov/ne/articles/triso-particles-most-robu...

pfdietz 4y ago

Such a shame that it's more expensive than fuel made from fresh uranium.

cyanydeez 4y ago

Until the political system is stable, I wouldn't trust nuclear energy. Just imagine trump deux telling everyone nuclear reactors no long need regulation

bergenty 4y ago
Yeah I wouldn’t trust trump/trumpesk with it but we can’t stop that from us doing what’s best for the environment and the country.
- ncmncm 4y ago
  We are doing what is best by not throwing good $billions after bad down the nuke rathole. Renewables are now by a huge margin the cheapest power generation system ever created. After enough renewables have been built out to supply power and charge up storage, it will be time to start building out storage, which will then serve base load. Storage cost is falling even faster than wind or solar PV ever did, so by that time many kinds of storage will also be very cheap. We will use many kinds of storage because each has different cost structure: Batteries, per MWh stored, MW in or out cheap, so good for peak smoothing. Synthetic ammonia and hydrogen, also liquified nitrogen, per MW in, tankage almost free, and transportable, MW out via existing combined cycle turbines. Pumped hydro, per MW, MWh cheap up to a hard limit. Undersea pumped hydro, per MWh but very cheap if the pump is onshore. Undersea buoyancy, very cheap all around provided the winch is onshore. Underground compressed air, per MW in, MWh stored very cheap, MW out via turbines maybe boosted with hydrogen or ammonia combustion. There is no place in the future for overpriced nukes built with coerced money.

Maursault 4y ago

This has been known since fast breeder reactors were initially designed to be able to do this, idk when 2-3 decades ago? The problem here is the reason only conventional fission reactors have been built is that they are the least expensive designs to build. And the problem today is that even the cheaper fission reactors are too expensive to be profitable, even with all the government subsidies and not burdening investors with decommissioning, and especially storage and security of the nuclear waste, which is a significant cost left to be paid for by tax payers. And the expense is not because of regulations. It may cost $10B-$15B to build and bring a new fission plant online, but regulatory costs would only account for about $10M/yr, thus the effect of regulations on the nuclear industry is insignificant compared to the bulk of the costs, such as uranium, construction, waste storage and security, and decommissioning, and the costs of waste storage and security and decommissioning are never included in the cost of a fission plant. Once you include those costs, the real price of a fission plant is revealed to be $50B and much more. So as it is, this idea is a non-starter, because fast breeder reactors are far more expensive to bring online than conventional fission reactors, and those are already too expensive as it is. But I have an idea. Make cheaper fast breeder reactors! Scale them up or down, idk which, until the economics makes sense enough that they can attract investors and actually be built. Just make them economically feasible without burdening the tax payer, and we can shut the pro-nuke crowd up for good.

credit_guy 4y ago
> And the expense is not because of regulations. [...] regulatory costs would only account for about $10M/yr. You are quite funny. NuScale received the NRC approval for their Small Nuclear Reactors (SMR) in 2020. They spent half a billion dollars for that, and 2 million man-hours [1]. They were founded in 2007 and the approval came 13 years later. If they are lucky, they'll have their first SMR working by the end of the decade. Of course, that's a happy ending. Oklo, another nuclear startup was founded in 2013, and got their application rejected by the NRC in 2022 [2]. They were told they can try again, and they probably will, but for the time being they have little to show for 9 years of work. [1] https://www.nuscalepower.com/newsletter/nucleus-fall-2020/program-development-update https://www.nuscalepower.com/newsletter/nucleus-fall-2020/pr... [2] https://www.nrc.gov/reading-rm/doc-collections/news/2022/22-002.pdf https://www.nrc.gov/reading-rm/doc-collections/news/2022/22-...

lesam 4y ago

How does this article not talk about France, which had a long-running commercial breeder reactor shut down more by politics than cost? The real issue is that breeder reactors produce weapons grade plutonium, and so the US doesn’t like to see them built.

pfdietz 4y ago
France's long running commercial breeder program was an expensive fiasco. It was shut down because of pointlessness, not because of politics.