3 ms·

There is also ongoing work on LFTR reactors https://en.m.wikipedia.org/wiki/Liquid_fluoride_thorium_reactor https://en.m.wikipedia.org/wiki/Liquid_fluoride_thor

by chr1 7y ago

There is also ongoing work on LFTR reactors https://en.m.wikipedia.org/wiki/Liquid_fluoride_thorium_reactor https://en.m.wikipedia.org/wiki/Liquid_fluoride_thorium_reac... which are much simpler and safer than reactors using solid uranium, because they have only fraction of fuel of solid reactors, do not have any pressurised containers, and have stable rate of reaction. Two experimental reactors of this type have worked in 60s, but technology was abandoned because it didn't produce the isotope of plutonium used in nuclear weapons.

acidburnNSA 7y ago
The "doesn't make bombs" story of the molten salt reactor program's cancellation is largely a myth, albeit one that's extremely common on the internet [1]. Fluid fueled reactors, mostly in the form of Molten Salt Reactors (MSRs), of which the LFTR is one brand name of, do have lots of potentially huge benefits in terms of simplicity, online refueling, and straightforward reactivity effects. We do only have about 5 reactor-years of experience with them, and so we don't fully understand the costs of building, operating, and maintaining such systems in today's regulator regimes. There is some reason to be concerned. When the fuel is dissolved in fluid, the radioactive fission products are dispersed everywhere. Literally half the periodic table of the elements is in there, and lots of it is volatile and mobile. It gets on your pumps, your valves, your heat exchangers, etc. Doing routine maintenance becomes very challenging and will need to be done remotely. A line I heard recently regarding MSR maintenance is: "You'll need robots to do the maintenance of your maintenance robots." Another memorable gem is "If you can make robots that can do that, you should just sell the robots" We should absolutely work on progressing this technology, but in the face of a climate disaster, we should just build more of what we know is already safer than almost every kind of energy system we can deploy, which is regular Gen III LWRs. We need to solve problems on the construction yard so we can build them much more cheaply. [1] https://whatisnuclear.com/thorium-myths.html#myth1 https://whatisnuclear.com/thorium-myths.html#myth1
- philpem 7y ago
  Good points. How do some of the other alternatives like Pebble Bed fit into this? I imagine PBRs would be more challenging to recycle fuel as it's combined with the moderator(?), but are there other issues which make it unrealistic compared to traditional PWR/BWR/LWR/Magnox reactors?
  - acidburnNSA 7y ago
    I'm starting to really like gas-cooled reactors like pebble-bed. They have two fundamental technical challenges. First is that gas coolant (typically helium) isn't a great heat-transfer mechanism so you need to spread your heat generation out in a large volume. This low power density results in pretty large structures (which can get pricey to build) for relatively low power, compared to other coolant configurations like water and especially liquid metal. Related, if there's a coolant leak, all the pressurized gas shoots out and you need backup cooling systems or particularly low power density that supports heat removal via conduction and thermal radiation. This can ding your economics pretty bad too. Second, the fuel particles you mention are very expensive to fabricate, $10k/kg. Estimates of future costs go "as low as" $3k/kg, but some say it could even go up to $30k/kg. In all cases, that's really expensive fuel fabrication. There's a plant in China that can fabricate them in moderate bulk right now, so perhaps with time and experience we can figure out how to bring this process down in cost. Aimed at solving power density and decay-heat cooling issues, there's also the salt-cooled version of the pebble bed reactor which gets the high-temperature fuel benefits and trades high-pressure gas for low-pressure salt. A research group from MIT and UC Berkeley has worked on this for years under Prof Charles Forsberg, and a company in the Bay Area called Kairos is now working on commercializing it. Unfortunately you get a lot of tritium production from the best known salt, which is a FLiBe salt, where the Lithium + neutrons results in crap-tons of supermobile tritium. You can cold-trap most of it but it's still a pain. And the fuel is still $10k/kg to fabricate as far as anyone knows, and the graphite cracks, and high temperature corrosion under irradiation is hard.
    - pfdietz 7y ago
      FLiBe also needs Be. World annual production of Be is 230 tonnes. And new Li isotope separation facilities would be needed, as only very small amounts of 6Li could be tolerated. The existing technology for that is not workable now due to mercury leakage.
- ben509 7y ago
  Thanks for the link, it's hard to find balanced stuff on thorium MSRs.
- achalhp 7y ago
  RE: A line I heard recently regarding MSR maintenance is: "You'll need robots to do the maintenance of your maintenance robots." Another memorable gem is "If you can make robots that can do that, you should just sell the robots" Maintenance of MSRE was done by humans without much downtime or exposure[1]. Ease of maintenance is a function of design. Just design it carefully. [1] https://energyfromthorium.com/pdf/NAT_MSREexperience.pdf https://energyfromthorium.com/pdf/NAT_MSREexperience.pdf
  - acidburnNSA 7y ago
    MSRE had fans blowing on pipes as cooling. It had no real fluid-to-fluid heat exchanger or power conversion cycle, which is where lots of maintenance troubles arise from. Furthermore, they were in the 1960s, before the NRC existed and before we worried about ALARA. Furthermore, MSRE cannot to this day account for about half of their radioiodine inventory. Furthermore, Alvin Weinberg himself says in his 1990s autobiography, The First Nuclear Era (amazing read btw) that they were just piping radioactive gas into the forest nearby back then (or, wait, was that the aqueous homogeneous reactor? I forget, I left all my copies of that book at the office...). A lot has changed since MSRE ran, and MSRE was a fairly simple experiment to prove a concept. It didn't have all the systems necessary to make practical energy, and so it didn't have the maintenance issues you'd expect. Also, it only ran for like 5 years, vs the 60-80 we're hoping for these days from these facilities (inherently requires lots of maintenance). Have you ever done work planning for ALARA at a light water reactor? Those who have raise eyebrows really high when academic reactor designers start saying how easy the maintenance will be. Admiral Rickover's quotes from 1953 just keep on coming back to haunt us [1]. He didn't curse us to death, but he sure warned us that we need to be working really hard on new/effective solutions to these kinds of problems. Most advanced nuclear advocates skip over these points. RE: Design it carefully: This takes practical experience. The number of people today designing new reactors who have this kind of experience is very close to zero. We will have to re-learn. [1] https://en.wikiquote.org/wiki/Hyman_G._Rickover#Paper_Reactors,_Real_Reactors_(1953) https://en.wikiquote.org/wiki/Hyman_G._Rickover#Paper_Reacto...
    - pfdietz 7y ago

by chr1 7y ago

acidburnNSA 7y ago

The "doesn't make bombs" story of the molten salt reactor program's cancellation is largely a myth, albeit one that's extremely common on the internet [1]. Fluid fueled reactors, mostly in the form of Molten Salt Reactors (MSRs), of which the LFTR is one brand name of, do have lots of potentially huge benefits in terms of simplicity, online refueling, and straightforward reactivity effects. We do only have about 5 reactor-years of experience with them, and so we don't fully understand the costs of building, operating, and maintaining such systems in today's regulator regimes. There is some reason to be concerned. When the fuel is dissolved in fluid, the radioactive fission products are dispersed everywhere. Literally half the periodic table of the elements is in there, and lots of it is volatile and mobile. It gets on your pumps, your valves, your heat exchangers, etc. Doing routine maintenance becomes very challenging and will need to be done remotely. A line I heard recently regarding MSR maintenance is: "You'll need robots to do the maintenance of your maintenance robots." Another memorable gem is "If you can make robots that can do that, you should just sell the robots" We should absolutely work on progressing this technology, but in the face of a climate disaster, we should just build more of what we know is already safer than almost every kind of energy system we can deploy, which is regular Gen III LWRs. We need to solve problems on the construction yard so we can build them much more cheaply. [1] https://whatisnuclear.com/thorium-myths.html#myth1 https://whatisnuclear.com/thorium-myths.html#myth1

philpem 7y ago
Good points. How do some of the other alternatives like Pebble Bed fit into this? I imagine PBRs would be more challenging to recycle fuel as it's combined with the moderator(?), but are there other issues which make it unrealistic compared to traditional PWR/BWR/LWR/Magnox reactors?
- acidburnNSA 7y ago
  I'm starting to really like gas-cooled reactors like pebble-bed. They have two fundamental technical challenges. First is that gas coolant (typically helium) isn't a great heat-transfer mechanism so you need to spread your heat generation out in a large volume. This low power density results in pretty large structures (which can get pricey to build) for relatively low power, compared to other coolant configurations like water and especially liquid metal. Related, if there's a coolant leak, all the pressurized gas shoots out and you need backup cooling systems or particularly low power density that supports heat removal via conduction and thermal radiation. This can ding your economics pretty bad too. Second, the fuel particles you mention are very expensive to fabricate, $10k/kg. Estimates of future costs go "as low as" $3k/kg, but some say it could even go up to $30k/kg. In all cases, that's really expensive fuel fabrication. There's a plant in China that can fabricate them in moderate bulk right now, so perhaps with time and experience we can figure out how to bring this process down in cost. Aimed at solving power density and decay-heat cooling issues, there's also the salt-cooled version of the pebble bed reactor which gets the high-temperature fuel benefits and trades high-pressure gas for low-pressure salt. A research group from MIT and UC Berkeley has worked on this for years under Prof Charles Forsberg, and a company in the Bay Area called Kairos is now working on commercializing it. Unfortunately you get a lot of tritium production from the best known salt, which is a FLiBe salt, where the Lithium + neutrons results in crap-tons of supermobile tritium. You can cold-trap most of it but it's still a pain. And the fuel is still $10k/kg to fabricate as far as anyone knows, and the graphite cracks, and high temperature corrosion under irradiation is hard.
  - pfdietz 7y ago
    FLiBe also needs Be. World annual production of Be is 230 tonnes. And new Li isotope separation facilities would be needed, as only very small amounts of 6Li could be tolerated. The existing technology for that is not workable now due to mercury leakage.
ben509 7y ago
Thanks for the link, it's hard to find balanced stuff on thorium MSRs.
achalhp 7y ago
RE: A line I heard recently regarding MSR maintenance is: "You'll need robots to do the maintenance of your maintenance robots." Another memorable gem is "If you can make robots that can do that, you should just sell the robots" Maintenance of MSRE was done by humans without much downtime or exposure[1]. Ease of maintenance is a function of design. Just design it carefully. [1] https://energyfromthorium.com/pdf/NAT_MSREexperience.pdf https://energyfromthorium.com/pdf/NAT_MSREexperience.pdf
- acidburnNSA 7y ago
  MSRE had fans blowing on pipes as cooling. It had no real fluid-to-fluid heat exchanger or power conversion cycle, which is where lots of maintenance troubles arise from. Furthermore, they were in the 1960s, before the NRC existed and before we worried about ALARA. Furthermore, MSRE cannot to this day account for about half of their radioiodine inventory. Furthermore, Alvin Weinberg himself says in his 1990s autobiography, The First Nuclear Era (amazing read btw) that they were just piping radioactive gas into the forest nearby back then (or, wait, was that the aqueous homogeneous reactor? I forget, I left all my copies of that book at the office...). A lot has changed since MSRE ran, and MSRE was a fairly simple experiment to prove a concept. It didn't have all the systems necessary to make practical energy, and so it didn't have the maintenance issues you'd expect. Also, it only ran for like 5 years, vs the 60-80 we're hoping for these days from these facilities (inherently requires lots of maintenance). Have you ever done work planning for ALARA at a light water reactor? Those who have raise eyebrows really high when academic reactor designers start saying how easy the maintenance will be. Admiral Rickover's quotes from 1953 just keep on coming back to haunt us [1]. He didn't curse us to death, but he sure warned us that we need to be working really hard on new/effective solutions to these kinds of problems. Most advanced nuclear advocates skip over these points. RE: Design it carefully: This takes practical experience. The number of people today designing new reactors who have this kind of experience is very close to zero. We will have to re-learn. [1] https://en.wikiquote.org/wiki/Hyman_G._Rickover#Paper_Reactors,_Real_Reactors_(1953) https://en.wikiquote.org/wiki/Hyman_G._Rickover#Paper_Reacto...
  - pfdietz 7y ago