4 ms·

This number is based on existing mines (active and inactive) only. There's a lot more fissile material available than just the 40 Gg of usable fissile material.

This number is based on existing mines (active and inactive) only. There's a lot more fissile material available than just the 40 Gg of usable fissile material. And that's assuming we only use U-235 and Pu-239. We also have viable LWR designs for Th-232 and are in the process of creating the first 5 Th-232 SMR test reactors at grid-scale right now.

Schroedingersat 4y ago
> This number is based on existing mines (active and inactive) only. There's a lot more fissile material available than just the 40 Gg of usable fissile material. And that's assuming we only use U-235 and Pu-239. That's all known and inferred accessible reserves. Not current mines. There are about 8-10 million tonnes of natural Uranium which has 0.5% extractable U235 (a bit more if you're willing to pay 10x as much for enrichment). There might be another big, high yield mine in Canada somewhere, but you probably want to check before putting down all our chips on it. U238 is not fissile. > We also have viable LWR designs for Th-232 and are in the process of creating the first 5 Th-232 SMR test reactors at grid-scale right now. So a technology that hasn't made it to the test bench, has no evidence as to its longevity or economics, probably requires more Beryllium than exists, and requires at least twice as much fissile material as is available for startup to breed fissile material from fertile thorium is your solution? One where the only proxy for how the extraction step might work is MOX reprocessed fuel which is more expensive than renewables on its own and releases more radiation under normal operation than Fukushima and TMI combined? Why are you suggesting diverting funds from a technology that works to build a completely different technology from your solution that destroys precious fissile material it needs to scale quickly then? Even in the most optimistic scenarios it will take decades to breed up a fresh load of U233 to double your fleet, and you will have to throw away all your multi billion dollar PWRs. Why not keep doing the thing that's provably working. That way if you solve the whole breeder thing then it will only take a few generations to breed enough fuel rather than hundreds.
- pfdietz 4y ago
  > That's all known and inferred accessible reserves. Resources, not reserves. Resources are turned into reserves when they are proved.
- hardolaf 4y ago
  > Why not keep doing the thing that's provably working. So nuclear then? Plants take on average 1-year longer than natural gas plants and a single nuclear plant produces thousands of hectares worth of solar panels with a tiny fraction of the resource usage. We have viable and in-use grid-scale U-235, Pu-239, and Th-232 reactors. We also have viable and in-use military operated U-233 breeder reactors in multiple nations that are in active production. Converting from a military design to a grid-scale design isn't really that hard as you loosen a ton of the space and thermal management requirements making manufacturing, operation, and maintenance cheaper. > So a technology that hasn't made it to the test bench The technology is fully proven in test reactors. The first 5 grid-scale reactors being built in the USA are part of a US Department of Energy program looking to create shovel-ready Th-232 SMR designs. There isn't a shortage of any of the isotopes we'd need for nuclear. Even if we used only U-235 and Pu-239 reactors using existing reserves to replace all current and projected global energy needs, we'd have 79 years to find more fuel or build something else. Meanwhile, with solar and wind, we still haven't figure out how to cheaply and safely store the energy to smooth the power supply curve. We could buy ourselves over 79 years to figure this out by building nuclear with only existing grid-scale technologies starting today.
  - pfdietz 4y ago
    Natural gas plants take as little as 28-30 months. https://www.powermag.com/controlling-schedule-quality-and-costs-for-new-gas-fired-plants/ https://www.powermag.com/controlling-schedule-quality-and-co...
  - Schroedingersat 4y ago
    > So nuclear then? Plants take on average 1-year longer than natural gas plants Water moderated reactors provably cannot work. Breeders do not exist. You don't get to start gaslighting about build times until you prove it's possible to make the fuel rods. Here's a few hints on how to tell if something works: What was the largest ever deployment of nuclear generation in a single year? For how many years in a row have renewables exceeded this? How many Joules of wholly unsubsidized, non-state-controlled, self-financed, insured nuclear generation have ever been produced? Now how much does unsubsidized solar or wind generation sell for? > a single nuclear plant produces thousands of hectares worth of solar panels How long did Inkai block 3 Uranium mine take to develop? What is its area in km^2 including the exclusion zones where the ground is too poisonous to live on or grow anything on? How many GW net of solar could be placed there in Kazakhstan's climate? How many times less energy does the Uranium it outputs produce? Now do a Namibian open cut mine with 0.01% concentration (the Uranium might even break even). How much fossil fuel does it take to mine the fifty billion or so tonnes of ore you'll need? If you were to expand production significantly that would be a comparatively high concentration mine. > with a tiny fraction of the resource usage. A solar panel produces >100GJ per kg of sand with roughly 10x the silver investment of a NPP or ~50g/kW and traces of B and P. The power density is around 3-6W/kg for high durability panels depending on how they are mounted. A nuclear reactor produces 5-10W/kg. The solar panel doesn't require indium or chromium or cadmium or all of the exotic materials required for a gas centrifuge. A kg of Uranium ore from Rossing produces about 30-80MJ. You can't just repeat a lie based on 20 year old data. You're making a claim that renewables (which have now surpassed the world nuclear fleet and are adding 20-30% per year) are insufficient compared to Nuclear. Prove it. Show me where the fuel can possibly come from. > Even if we used only U-235 and Pu-239 reactors using existing reserves to replace all current and projected global energy needs, we'd have 79 years to find more fuel or build something else. It's ~79 years at current consumption. Which one of the facts I stated about the available energy are you disputing? Is there not roughly 8-10 million tonnes of Uranium resource? Is it not 0.7% fissile? Does 20-30% not get left in tailings? Does a current generation reactor not require roughly 3.5 tonnes net of fissile material per GW? You can't just repeat the lie. Where is the fissile material hiding and how do you start your breeder reactors once you are done without taking a century to build up the U233? > The first 5 grid-scale reactors being built in the USA are part of a US Department of Energy program looking to create shovel-ready So you have a program to maybe finish building the test bench in 10 years? > The technology is fully proven in test reactors No reactor has ever run start to finish at non-negligible capacity factor with a multi year fuel cycle (whether constant reload or not), created >80TJ/kg of usable steam the whole time and ended with more fuel than it started. It is as proven as a 1000Wh/kg AlS battery that costs a few dollars a kilo or a quad junction 45% efficient paintable PV. 'I kinda tried one of the steps but am ignoring the really hard part of separating fission products or not having it corrode' isn't proven.

Schroedingersat 4y ago

> This number is based on existing mines (active and inactive) only. There's a lot more fissile material available than just the 40 Gg of usable fissile material. And that's assuming we only use U-235 and Pu-239. That's all known and inferred accessible reserves. Not current mines. There are about 8-10 million tonnes of natural Uranium which has 0.5% extractable U235 (a bit more if you're willing to pay 10x as much for enrichment). There might be another big, high yield mine in Canada somewhere, but you probably want to check before putting down all our chips on it. U238 is not fissile. > We also have viable LWR designs for Th-232 and are in the process of creating the first 5 Th-232 SMR test reactors at grid-scale right now. So a technology that hasn't made it to the test bench, has no evidence as to its longevity or economics, probably requires more Beryllium than exists, and requires at least twice as much fissile material as is available for startup to breed fissile material from fertile thorium is your solution? One where the only proxy for how the extraction step might work is MOX reprocessed fuel which is more expensive than renewables on its own and releases more radiation under normal operation than Fukushima and TMI combined? Why are you suggesting diverting funds from a technology that works to build a completely different technology from your solution that destroys precious fissile material it needs to scale quickly then? Even in the most optimistic scenarios it will take decades to breed up a fresh load of U233 to double your fleet, and you will have to throw away all your multi billion dollar PWRs. Why not keep doing the thing that's provably working. That way if you solve the whole breeder thing then it will only take a few generations to breed enough fuel rather than hundreds.

pfdietz 4y ago
> That's all known and inferred accessible reserves. Resources, not reserves. Resources are turned into reserves when they are proved.
hardolaf 4y ago
> Why not keep doing the thing that's provably working. So nuclear then? Plants take on average 1-year longer than natural gas plants and a single nuclear plant produces thousands of hectares worth of solar panels with a tiny fraction of the resource usage. We have viable and in-use grid-scale U-235, Pu-239, and Th-232 reactors. We also have viable and in-use military operated U-233 breeder reactors in multiple nations that are in active production. Converting from a military design to a grid-scale design isn't really that hard as you loosen a ton of the space and thermal management requirements making manufacturing, operation, and maintenance cheaper. > So a technology that hasn't made it to the test bench The technology is fully proven in test reactors. The first 5 grid-scale reactors being built in the USA are part of a US Department of Energy program looking to create shovel-ready Th-232 SMR designs. There isn't a shortage of any of the isotopes we'd need for nuclear. Even if we used only U-235 and Pu-239 reactors using existing reserves to replace all current and projected global energy needs, we'd have 79 years to find more fuel or build something else. Meanwhile, with solar and wind, we still haven't figure out how to cheaply and safely store the energy to smooth the power supply curve. We could buy ourselves over 79 years to figure this out by building nuclear with only existing grid-scale technologies starting today.
- pfdietz 4y ago
  Natural gas plants take as little as 28-30 months. https://www.powermag.com/controlling-schedule-quality-and-costs-for-new-gas-fired-plants/ https://www.powermag.com/controlling-schedule-quality-and-co...
- Schroedingersat 4y ago
  > So nuclear then? Plants take on average 1-year longer than natural gas plants Water moderated reactors provably cannot work. Breeders do not exist. You don't get to start gaslighting about build times until you prove it's possible to make the fuel rods. Here's a few hints on how to tell if something works: What was the largest ever deployment of nuclear generation in a single year? For how many years in a row have renewables exceeded this? How many Joules of wholly unsubsidized, non-state-controlled, self-financed, insured nuclear generation have ever been produced? Now how much does unsubsidized solar or wind generation sell for? > a single nuclear plant produces thousands of hectares worth of solar panels How long did Inkai block 3 Uranium mine take to develop? What is its area in km^2 including the exclusion zones where the ground is too poisonous to live on or grow anything on? How many GW net of solar could be placed there in Kazakhstan's climate? How many times less energy does the Uranium it outputs produce? Now do a Namibian open cut mine with 0.01% concentration (the Uranium might even break even). How much fossil fuel does it take to mine the fifty billion or so tonnes of ore you'll need? If you were to expand production significantly that would be a comparatively high concentration mine. > with a tiny fraction of the resource usage. A solar panel produces >100GJ per kg of sand with roughly 10x the silver investment of a NPP or ~50g/kW and traces of B and P. The power density is around 3-6W/kg for high durability panels depending on how they are mounted. A nuclear reactor produces 5-10W/kg. The solar panel doesn't require indium or chromium or cadmium or all of the exotic materials required for a gas centrifuge. A kg of Uranium ore from Rossing produces about 30-80MJ. You can't just repeat a lie based on 20 year old data. You're making a claim that renewables (which have now surpassed the world nuclear fleet and are adding 20-30% per year) are insufficient compared to Nuclear. Prove it. Show me where the fuel can possibly come from. > Even if we used only U-235 and Pu-239 reactors using existing reserves to replace all current and projected global energy needs, we'd have 79 years to find more fuel or build something else. It's ~79 years at current consumption. Which one of the facts I stated about the available energy are you disputing? Is there not roughly 8-10 million tonnes of Uranium resource? Is it not 0.7% fissile? Does 20-30% not get left in tailings? Does a current generation reactor not require roughly 3.5 tonnes net of fissile material per GW? You can't just repeat the lie. Where is the fissile material hiding and how do you start your breeder reactors once you are done without taking a century to build up the U233? > The first 5 grid-scale reactors being built in the USA are part of a US Department of Energy program looking to create shovel-ready So you have a program to maybe finish building the test bench in 10 years? > The technology is fully proven in test reactors No reactor has ever run start to finish at non-negligible capacity factor with a multi year fuel cycle (whether constant reload or not), created >80TJ/kg of usable steam the whole time and ended with more fuel than it started. It is as proven as a 1000Wh/kg AlS battery that costs a few dollars a kilo or a quad junction 45% efficient paintable PV. 'I kinda tried one of the steps but am ignoring the really hard part of separating fission products or not having it corrode' isn't proven.