5 ms·

EVinci nuclear microreactor moves towards commercialization

34 points by peutetre 2y ago

atomic128 2y ago
Reactors like this (and like Bill Gates' Natrium reactor) use HALEU (high-assay low-enriched uranium) which is above 19% fissile (U-235), compared to the usual 5% or 6% in a large pressurized water reactor (e.g., https://en.wikipedia.org/wiki/AP1000 https://en.wikipedia.org/wiki/AP1000). For comparison, a nuclear bomb or military microreactor (on a submarine, for example) uses fuel that's 90%+ fissile. This difference in fissile concentration is the reason why non-military nuclear reactors can never explode like a nuclear bomb. The bottleneck for nuclear energy is the yellowcake. This is the uranium that comes out of the ground, which is 0.7% fissile and must be enriched. A pound of HALEU requires many pounds of yellowcake. Conversion and enrichment are also in short supply today, but the centrifuges can be built. There is a profound near-term undersupply of this natural uranium due to the fact that the mining companies were allowed to die after Fukushima. The mining companies died due to low uranium prices: $16 per pound at the market bottom. The two big miners that survived (Cameco and Kazatomprom) are together unable to increase supply to match demand. Big new mines, like NexGen's, are years away, and the supply they provide will not close the supply/demand gap. Mining uranium is slow and difficult, even if you have access to good ore. Alternative sources of uranium are even more expensive to exploit (seawater, phosphate deposits, etc.). If you want to benefit (monetarily) from nuclear energy, and you're willing to suffer through the EXTREME volatility of owning a niche commodity, one way is through the Sprott Physical Uranium Trust (ticker: SRUUF) which owns 65 million pounds of uranium oxide and purchases more whenever it trades above Net Asset Value. WARNING: this is not an investment for the faint of heart. YOU HAVE BEEN WARNED. You can watch the spot uranium market (intra-day bid/ask) here: https://numerco.com/NSet/aCNSet.html https://numerco.com/NSet/aCNSet.html You can see the uranium long-term contract price (and spot uranium price) monthly history at Cameco's website here: https://www.cameco.com/invest/markets/uranium-price https://www.cameco.com/invest/markets/uranium-price
killingtime74 2y ago
Given the extensive security requirements for nuclear and the cost of this reactor itself, I struggle to see how it can be cost effective compared to renewables or even a diesel generator. Discussions on hacker News, which is very pro nuclear always tend to emphasise the technical advantages and neglect economic considerations.
- atomic128 2y ago
  Hacker News tends to be pro-solar, pro-wind, and anti-nuclear. Discussions on Hacker News tend to overlook the unreliable nature of intermittent solar and wind power sources. The hidden costs of solar and wind are also overlooked: the cost of land, the cost of transmission infrastructure, the cost of battery storage, the short lifespan of the hardware, the requirement for fossil fuel backup generators, etc. Here's Yann LeCun (Vice-President, Chief AI Scientist at Meta) describing the advantages of nuclear versus solar and wind in a different, larger-scale context: AI datacenters will be built next to energy production sites that can produce gigawatt-scale, low-cost, low-emission electricity continuously. Basically, next to nuclear power plants. The advantage is that there is no need for expensive and wasteful long-distance distribution infrastructure. Note: yes, solar and wind are nice and all, but they require lots of land and massive-scale energy storage systems for when there is too little sun and/or wind. Neither simple nor cheap. Link to LeCun's comment on X here, along with a discussion of why solar and wind are not enough: https://news.ycombinator.com/item?id=41621097 https://news.ycombinator.com/item?id=41621097 The same is true on smaller scales.
  - moralestapia 2y ago
    Also, plenty of locations where cooling is free. This will make AI OpEx almost negligible. In line with OpenAI's premise of making intelligence very cheap. Only issue but not a big one is latency, i.e. Starlink will become extremely valuable in the upcoming years.
- foxyv 2y ago
  Most of the security around nuclear reactors is to prevent disruption of critical power infrastructure and barring access to spent fuel rods. This design would need far less security since it is sealed and not a huge plant that would cut power to half the USA if destroyed. Stealing one of these would be pretty pointless. It would be far easier to get much more dangerous isotopes from one of the thousands of hospitals that use such things or a laboratory. The 20% enrichment is not enough to produce viable weapons either. Uranium enrichment has just gotten way too easy with new centrifuge designs. Applications are mostly stuff like Antarctic research projects and remote oil fields. Also, if it is approved for such uses, it could power a container ship saving huge amounts of money on diesel fuel which is about half their operating cost.
BMc2020 2y ago
Can someone knowledgeable tell me what happens if you blow it up?
moralestapia 2y ago
This is great and I think is the way to go. >produces a self-limiting nuclear reaction that cannot go out of control Anyone knows if this holds true even if someone tampers with the device?
bbojan 2y ago
In my opinion TRISO fuel is bad because it has very low burn-up and is difficult to re-process. These things together mean it has quite low utilization of enriched uranium.
- credit_guy 2y ago
  This particular design has a very low burn-up indeed, but that's not because of the TRISO fuel. The Chinese HTR-PM helium cooled reactor, that went live 2 years ago, has a burnup of 90 GWd/ton of uranium, which is enriched at 8.5% U-235 [1]. That is an exceptionally high burn-up rate. The eVinci reactor has an exceptionally low burn-up rate: about 4 GWd/ton, despite using a much more enriched fuel,at 19.75% ([2], but note that this is just an estimate, Westinghouse did not disclose the actual burnup). Why? That's the price you have to pay to have a micro-reactor. The square-cube law says that for such a reactor the surface of the core is very high compared to its volume, so the neutron economy is extremely poor. The only way to make it work is to use highly enriched uranium. Uranium enriched to more than 20% is considered weapons grade, so commercial reactors need to use fuel below that limit, and 19.75 is basically as high as you can go. TRISO fuel is actually a miracle of science. It addresses many problems of the current generation reactor fuels. Fission results in transmutation. By the very nature of the fission process, you end up with fission products in burned fuel. Some of these products are gases (like Xeon) and they create pressure, and when you have hundreds of thousands of fuel elements some will burst, resulting in fission product discharge in the cooling water. Nasty stuff. The fuel in TRISO is encapsulated in some poppy-seed-sized granules, and it can withstand immense pressures, so this bursting scenario just does not happen. In addition to that, they can withstand immense temperatures as well, and they are surrounded by graphite that has an exceptional heat conductivity, and is also a very good moderator. From the point of view of reactivity control, graphite is actually the best moderator out there, ahead of hydrogen, deuterium and beryllium. > is difficult to re-process That's not a problem. You just don't reprocess it. > it has quite low utilization of enriched uranium This reactor will have low utilization, as discussed, but not because of the TRISO fuel. [1]https://aris.iaea.org/PDF/HTR-PM.pdf https://aris.iaea.org/PDF/HTR-PM.pdf [2] https://www.osti.gov/servlets/purl/1501599 https://www.osti.gov/servlets/purl/1501599
  - 2y ago
Log_out_ 2y ago
So what happens if the alkali elements can not remove heat, for example during a fire?

34 points by peutetre 2y ago

Reactors like this (and like Bill Gates' Natrium reactor) use HALEU (high-assay low-enriched uranium) which is above 19% fissile (U-235), compared to the usual 5% or 6% in a large pressurized water reactor (e.g., https://en.wikipedia.org/wiki/AP1000 https://en.wikipedia.org/wiki/AP1000). For comparison, a nuclear bomb or military microreactor (on a submarine, for example) uses fuel that's 90%+ fissile. This difference in fissile concentration is the reason why non-military nuclear reactors can never explode like a nuclear bomb. The bottleneck for nuclear energy is the yellowcake. This is the uranium that comes out of the ground, which is 0.7% fissile and must be enriched. A pound of HALEU requires many pounds of yellowcake. Conversion and enrichment are also in short supply today, but the centrifuges can be built. There is a profound near-term undersupply of this natural uranium due to the fact that the mining companies were allowed to die after Fukushima. The mining companies died due to low uranium prices: $16 per pound at the market bottom. The two big miners that survived (Cameco and Kazatomprom) are together unable to increase supply to match demand. Big new mines, like NexGen's, are years away, and the supply they provide will not close the supply/demand gap. Mining uranium is slow and difficult, even if you have access to good ore. Alternative sources of uranium are even more expensive to exploit (seawater, phosphate deposits, etc.). If you want to benefit (monetarily) from nuclear energy, and you're willing to suffer through the EXTREME volatility of owning a niche commodity, one way is through the Sprott Physical Uranium Trust (ticker: SRUUF) which owns 65 million pounds of uranium oxide and purchases more whenever it trades above Net Asset Value. WARNING: this is not an investment for the faint of heart. YOU HAVE BEEN WARNED. You can watch the spot uranium market (intra-day bid/ask) here: https://numerco.com/NSet/aCNSet.html https://numerco.com/NSet/aCNSet.html You can see the uranium long-term contract price (and spot uranium price) monthly history at Cameco's website here: https://www.cameco.com/invest/markets/uranium-price https://www.cameco.com/invest/markets/uranium-price

killingtime74 2y ago

Given the extensive security requirements for nuclear and the cost of this reactor itself, I struggle to see how it can be cost effective compared to renewables or even a diesel generator. Discussions on hacker News, which is very pro nuclear always tend to emphasise the technical advantages and neglect economic considerations.

atomic128 2y ago
Hacker News tends to be pro-solar, pro-wind, and anti-nuclear. Discussions on Hacker News tend to overlook the unreliable nature of intermittent solar and wind power sources. The hidden costs of solar and wind are also overlooked: the cost of land, the cost of transmission infrastructure, the cost of battery storage, the short lifespan of the hardware, the requirement for fossil fuel backup generators, etc. Here's Yann LeCun (Vice-President, Chief AI Scientist at Meta) describing the advantages of nuclear versus solar and wind in a different, larger-scale context: AI datacenters will be built next to energy production sites that can produce gigawatt-scale, low-cost, low-emission electricity continuously. Basically, next to nuclear power plants. The advantage is that there is no need for expensive and wasteful long-distance distribution infrastructure. Note: yes, solar and wind are nice and all, but they require lots of land and massive-scale energy storage systems for when there is too little sun and/or wind. Neither simple nor cheap. Link to LeCun's comment on X here, along with a discussion of why solar and wind are not enough: https://news.ycombinator.com/item?id=41621097 https://news.ycombinator.com/item?id=41621097 The same is true on smaller scales.
- moralestapia 2y ago
  Also, plenty of locations where cooling is free. This will make AI OpEx almost negligible. In line with OpenAI's premise of making intelligence very cheap. Only issue but not a big one is latency, i.e. Starlink will become extremely valuable in the upcoming years.
foxyv 2y ago
Most of the security around nuclear reactors is to prevent disruption of critical power infrastructure and barring access to spent fuel rods. This design would need far less security since it is sealed and not a huge plant that would cut power to half the USA if destroyed. Stealing one of these would be pretty pointless. It would be far easier to get much more dangerous isotopes from one of the thousands of hospitals that use such things or a laboratory. The 20% enrichment is not enough to produce viable weapons either. Uranium enrichment has just gotten way too easy with new centrifuge designs. Applications are mostly stuff like Antarctic research projects and remote oil fields. Also, if it is approved for such uses, it could power a container ship saving huge amounts of money on diesel fuel which is about half their operating cost.

BMc2020 2y ago

Can someone knowledgeable tell me what happens if you blow it up?

moralestapia 2y ago

This is great and I think is the way to go. >produces a self-limiting nuclear reaction that cannot go out of control Anyone knows if this holds true even if someone tampers with the device?

bbojan 2y ago

In my opinion TRISO fuel is bad because it has very low burn-up and is difficult to re-process. These things together mean it has quite low utilization of enriched uranium.

credit_guy 2y ago
This particular design has a very low burn-up indeed, but that's not because of the TRISO fuel. The Chinese HTR-PM helium cooled reactor, that went live 2 years ago, has a burnup of 90 GWd/ton of uranium, which is enriched at 8.5% U-235 [1]. That is an exceptionally high burn-up rate. The eVinci reactor has an exceptionally low burn-up rate: about 4 GWd/ton, despite using a much more enriched fuel,at 19.75% ([2], but note that this is just an estimate, Westinghouse did not disclose the actual burnup). Why? That's the price you have to pay to have a micro-reactor. The square-cube law says that for such a reactor the surface of the core is very high compared to its volume, so the neutron economy is extremely poor. The only way to make it work is to use highly enriched uranium. Uranium enriched to more than 20% is considered weapons grade, so commercial reactors need to use fuel below that limit, and 19.75 is basically as high as you can go. TRISO fuel is actually a miracle of science. It addresses many problems of the current generation reactor fuels. Fission results in transmutation. By the very nature of the fission process, you end up with fission products in burned fuel. Some of these products are gases (like Xeon) and they create pressure, and when you have hundreds of thousands of fuel elements some will burst, resulting in fission product discharge in the cooling water. Nasty stuff. The fuel in TRISO is encapsulated in some poppy-seed-sized granules, and it can withstand immense pressures, so this bursting scenario just does not happen. In addition to that, they can withstand immense temperatures as well, and they are surrounded by graphite that has an exceptional heat conductivity, and is also a very good moderator. From the point of view of reactivity control, graphite is actually the best moderator out there, ahead of hydrogen, deuterium and beryllium. > is difficult to re-process That's not a problem. You just don't reprocess it. > it has quite low utilization of enriched uranium This reactor will have low utilization, as discussed, but not because of the TRISO fuel. [1]https://aris.iaea.org/PDF/HTR-PM.pdf https://aris.iaea.org/PDF/HTR-PM.pdf [2] https://www.osti.gov/servlets/purl/1501599 https://www.osti.gov/servlets/purl/1501599
- 2y ago

Log_out_ 2y ago

So what happens if the alkali elements can not remove heat, for example during a fire?