Prismatic (or cylindrical) TRISO also makes sense. There are lots of potential problems using pebble beds (circulation, grinding), whereas doing regular refuelling cycles avoids them, in exchange for down-time to refuel.
From the article:
> We said criticality in 2026, electricity production in 2027, and power to the warfighter in 2028.
Are there any other examples of land-based militaries using nuclear power? Seems kind of like since they can't talk about the energy transition or w/e this has to be a military thing instead.
"Stalin is proud to support the rebirth of Soviet Russia’s nuclear industry and ensuring soviets have access to affordable, reliable and secure energy for generations to come". Always mention the fearless leader.
To add a bit of context there were 11 companies participating in program and only 2 achieved critiality, and the deadline included in "DOE Reactor Pilot Program" was "July 4, 2026", and Aalo Atomics is the only one that might also make it in time.
Sigh, everything is being done for political purposes, now Dear Leader Donny can proclaim on the quartermillenial celebrations that "we achieved a nuclear milestone!". And maybe add that this is possible because of his nuclear physics genius abilities.
I wouldn't even be surprised if the achievements are like "Full-Self Driving", announced hastily with boasts and then slowly revealed to be full of Elon.
Is anyone working in the US on a waste solution that isn’t a big hole with a straight out of cyberpunk sci-fi warning plaque?
The French reprocess and recycle fissile material but that’s kind of a gnarly industrial process. Still they do it and it works.
The long term solution is to create a second kind of reactor that has a higher burn fraction which means a more fuel efficient fast reactor. Those would be, ideally, the big base load plants if we did this rationally.
I don't see the issue with dry cask storage medium term, and deep geological storage long term. Spent fuel isn't really that dangerous once it's been cooled down and for a couple decades before putting it in the ground, to the point that there are far more dangerous natural things you can dig up.
What concerns me is that 250 years of fossil fuel energy continues to store its waste products in my lungs and the water I drink. That's the issue we need to solve with urgency.
Europe has about 60,000 tons of nuclear waste storage[1], so lets say the global nuclear waste quantity is 2-3x or 120,000 to 180,000 tons. That sounds like a lot, however it's less than 2 weeks of coal deliveries to a coal plant (at 1 train of 115 cars each with 116 tons of coal = 13,340 tons delivered per day[2]). To take another approach, the average landfill size is 600 acres[3].
The "eh, just bury it" approach is really not a bad one. Its not even that much stuff to bury
IMO, the long term solution will be to simply launch the waste into space. With low enough launch costs the extra mass needed to armor the waste against accidents becomes tolerable.
Unless you live in Sci Fi future where space travel is magically free, this is a pretty bad deal since sending a pound of spent nuclear fuel out of earth orbit takes 50+ pounds of fuel. Sending all the spent nuclear waste into space would be something like 10 million tons of fuel.
Probably not worthwhile. If you just leave it in orbit you're going to have to track it and worry about debris/micrometeoroid strikes. The ideal would be to stick it in some permanently shadowed crater on the Moon, it'd be a stable environment without wild temperature swings and much lower risk of somehow ending up where people are. But that's a long way to go and a lot more risk to take for now.
Once it's in orbit the cost of moving it farther out needed be very high. Just evaporating it with solar energy in very high orbit will allow the solar wind to carry it out of the solar system.
I don't think that's the case, if launch costs are low enough. The options you describe there are quite expensive. Reprocessing costs somewhere in the neighborhood of $1000/kg. Even fast reactor have trouble disposing of the seven very long lived fission products. A fully reusable launcher might get costs down to $10/kg to LEO. Even if one had to wrap the waste in 10x its mass in launch armor this could be much cheaper than reprocessing.
Why would launch costs be low? $10/kg is fantasy land. UPS won’t ship a kilogram across the country for $10.
And who wants spent nuclear fuel in low earth orbit? This is a far worse location for spent fuel than buried in a bunker. This is a worse location than Times Square.
Why wouldn't they be low? In the limit, if launch becomes operationally similar to air travel, costs will be a few times propellant cost. And propellant is very cheap. LOX is almost free (the second cheapest industrial liquid after water) and liquid methane isn't very expensive either.
Low earth orbit would just be where it's transferred to something to carry it farther out, for example using solar-electric engines.
For all this, remember it isn't done immediately, it's done in (say) 300 years when the short lived fission products are mostly gone.
China does: all of the above, where it makes sense.
Renewables and batteries to keep your AC, workplace EV charger, stove, pool heater and (since recently) green ammonia producer going, nuclear to prevent e.g. aluminium smelters from seizing up.
Also the cheapest way to make renewables work 24/7 is to build HVDC lines - they cost as much as a highway per unit length and even undersea cables would deploy for less and faster than equivalent nuclear.
The total length of HVDC lines just in China is currently more than 40k km, so they've literally deployed enough of them to wrap around the globe.
China is also still building coal and has passed Europe and will (if they don’t change course) soon pass the US and Canada and the other big ones on a per capita emitter basis. They already passed everyone as top emitter in an absolute sense.
Not saying they’re not also building renewables and nuclear, but it seems like the policy is more “build anything and everything to satisfy demand” than a focused effort.
BTW: if you look at US emissions, the data center bubble hasn’t had much if any effect. They are still trending down. There’s reasons to dislike that industry but I’m sick of the mindless echo chamber doom on that issue. They’re not that significant in the grand scheme of things.
Coal is already shrinking in the China (in absolute terms, not just as share of production) [1], and share of wind + solar is already larger than in the US [2], so I doubt China will ever reach US proportions in CO2 emissions per capita.
An additional data point to support that is that emissions intensity per GDP is clearly falling fast for China [3].
According to this graph and assuming both US emissions fall and Chinese grow exponentially at the rate they were over the decade 2014-2024, the figures will cross around 2037.
Personally I doubt that, as US emissions have been going up due to data centers:
And while China, like you said, doesn't seem to have a focused goal, it so happens that renewables are the path of least resistance for just having more energy as fast and cheap as possible.
If your location already has a well-run nuclear energy sector (Finland, Sweden, South Korea): invest in nuclear energy.
If you don't: stick to renewables.
And it also depends on what you mean by "we". As a Dane, I don't think us Danish taxpayers should invest in nuclear energy, but I'm perfectly happy that private Danish investors invest in Seaborg/Saltfoss and Copenhagen Atomics.
Nuclear is not on a trajectory to do more than supply a minor amount of world energy. A (say) 10% nuclear, 90% renewable world is not an easier challenge than a 100% renewable world -- the intermittency/seasonality issues aren't eased by having 10% nuclear running as baseload, and keeping it as backup makes its cost per kWh explode.
Nuclear really has to go big (supply most of the world's energy) or go home. But supplying most of the world's energy means burner reactors are inadequate -- there isn't enough cheap uranium. Burner microreactors have even worse neutron economy, so this argument applies even more so to them.
It's more of an engineering call than something that can be purely determined from inductive reasoning. I think most engineers working in the space would say "both" are needed, but partisans exist on both sides.
That's a political and economic question, not a scientific one. Science can provide input information, but the decision involves weighing all sorts of facts and considerations outside the scope of science.
Nuclear partisans like to call renewables ideological, but I think this is another example of "the accusation is a confession".
The empirical evidence has nuclear being uncompetitively expensive. The current focus on variant reactor designs appears to be something of a Hail Mary attempt to get around this sad state of affairs.
You sometimes see them making an argument about energy density, which goes back to Vaclav Smil. But Smil used this argument to massively mispredict how solar would be go in the market. We don't hear him much anymore.
Nuclear advocates increasingly resort to conspiracy theoretic reasoning to explain away the failure of their technology to compete. This should be a red flag.
> The empirical evidence has nuclear being uncompetitively expensive.
France nuclearized 75% of its grid in the 1980s while the solar folks were faffing around. It's not a cost issue, it's that anti-nuclear folks have choked out the industry.
We need to take the boot off the neck of nuclear. Wind and solar aren't an avenue to moving up the tech tree of civilization, which will involve using vastly more power.
We don't actually know how much that cost France, sine it was mixed in with their military nuclear effort. French auditors threw up their hands trying to figure out the actual costs.
What we do know is their attempt to build more NPPs now has gone spectacularly tits up, with costs completely out of control. This should make one view their earlier efforts with great suspicion. Have they become much worse, or were earlier problems concealed?
Tech development != tech tree, history is nonlinear and ecological. Linear, unlockable "tech trees" are an ahistorical fantasy necessary to make some video games fun and digestible. Applying them to the real world is like trying to win a war by replicating the Battle of Helm's Deep or something.
Right but “upgrading to the next level” is exactly the type of cognitive error that I’m against, it’s forward-looking Whig history. There is no “levels” of civilization, there’s only the arrangements we have now and the ones that will occur in the future. Maybe those arrangements will have different pros and cons than we have now, maybe they’ll have more sophisticated engineering practices, but there’s no objective “development score” that is being maximized it’s just more humanity.
Fusion's main accomplishment will likely be to make fission look cheap in comparison, due to fundamental issues of power density in the nuclear island. Why make a huge complex low power density radioactive thing when you can make a much smaller simpler high power density radioactive thing?
We should be investing in all non carbon emitting sources and we should have been doing it since the 1970s when we figured out pretty conclusively that this would be a problem.
Instead we had right wing fossil fuel shills on one shoulder and unscientific woo woo greens on the other, the net effect being that we kept burning more carbon. We still have them, Trump with “beautiful coal” and greens now opposing even solar power and batteries, but climate change is no longer possible to ignore. Some still manage it but those people are nuts.
If we hadn’t stopped improving nuclear we’d probably have emitted half the CO2 we have. It would have become cheaper and safer and more scalable and then when China industrialized they would have copied that instead of burning so much coal.
France with its nearly zero carbon grid is the existence proof.
It wasn’t until the 2010s that solar and wind became grid scale in a big enough way to matter. That was too slow.
Whether someone is at least open to nuclear power is my litmus test for whether they take climate change seriously.
I do. If we hit 600, 800, 1000 ppm CO2, which is possible if the world keeps developing on the back of fossil fuels, we are entering existential risk territory. Earth has had those CO2 levels before, and higher, but our species was not alive then.
We already passed the FAFO threshold for ppm CO2 and now we will FO. But that’s not X-risk yet. I’m talking about the next threshold, which may start around 600 but really kicks in near 1000. This is where you actually start asphyxiating. You get lowered IQ and impaired judgement to a small degree, but across the globe at a time when we really don’t need it.
> "The Trump administration is proud to support the rebirth of America’s nuclear industry and ensuring Americans have access to affordable, reliable and secure energy for generations to come."
> "The demonstration and the licensing pathway it establishes represent a key step toward deploying electricity-producing microreactors for U.S. military installations by September 30, 2028."
So which is it? Power to the people or power to the military? This microreactor concept doesn't seem very well suited for commercial use.
Why would microreactor concepts not be suitable for commercial use? History is overwhelmed with examples of large, rare and expensive tech being produced in small cheap packages and becoming massive commercial successes that make the old way look primitive.
Because large scale production is generally more scalable and efficient. And you probably don't want dozens of "microreactors" scattered across cities.
A nuclear reactor is generally treated as a high security facility. I don't know how this new reactor works but I thought it was safe to assume something like a terrorist attack on one might be bad. It's also a lot more work to inspect and control them when scattered.
On the other hand you can scale production of reactor themselves. And I don't think the idea is to scatter them around, but to have a power plant with dozens of them in one place (instead of 3-4 regular reactors in a regular nuke power plant).
I think that may be exactly wrong. The small scale may make it easier for a reactor to be “walk away safe” ie shut itself down absent external activity. I know that is a design goal of some of the Chinese micro reactors and those are used for civilian power generation.
Secondly although generating large amounts of power is more efficient in terms of generation, generating power close to the point of use is significantly more efficient in terms of power loss on the grid as I understand it.
Large scale production of commodity goods is generally more efficient. Which is why microreactors don't seem to have any inherent disadvantages. The efficiencies tend to kick in with the raw number of items produced.
> microreactors don't seem to have any inherent disadvantages
They have diseconomies of scale. Some of the costs of a nuclear power plant scale sublinearly with power. Neutron economy is improved in a larger core. Larger turbines are more efficient than smaller turbines. It doesn't take 1000x as many operators to operate a NPP with 1000x the power output.
Is that relevant? The economics of nuclear plants doesn't have anything to do with efficiency as far as I'm aware, the fuel costs are relatively negligible. They can afford to be horribly inefficient if they can get economies of scale producing the plant. So you can use inefficient turbines and have bad neutron economy and it wouldn't change the economics by anything in particular.
You'd also probably find similar issues with diesel generators, but small diesel generators do roaring trade and have great commercial applications.
Cost is not only relevant, it's paramount. Efficiency is only important insofar as it affects overall cost.
Diesel generators have the advantage of being very cheap -- an order of magnitude cheaper than NPPs per unit power output -- and of having much of their total cost being fuel cost, so they can operate at lower capacity factor. But even so, we don't see large power plants composed of arrays of diesel microgenerators.
(The solution for current higher capacity factor diesel users, like say remote operation at mines, would be to supplement them with renewables and storage to reduce fuel costs. This is already happening.)
A significant problem with any small power plant is fixed costs. A 1 MW(e) plant (Antares is said to be between 100 kW(e) and 1 MW(e)) making power at 90% capacity factor and selling at $0.05/kWh will gross about $400K/year. A single full time employee, like a security guard, will cost a good chunk of that.
Microreactors have been tried before by the military, for use at bases, which have guards. They not only didn't make sense to install, they didn't make sense to continue to operate once installed.
Great to see engineering deliver on time. I wonder if Rolls Royce will also have a smooth ride. It's a PWR.
Are there any other examples of land-based militaries using nuclear power? Seems kind of like since they can't talk about the energy transition or w/e this has to be a military thing instead.
That distinction matters because nuclear.
Let that sink in for a moment.
I wouldn't even be surprised if the achievements are like "Full-Self Driving", announced hastily with boasts and then slowly revealed to be full of Elon.
The French reprocess and recycle fissile material but that’s kind of a gnarly industrial process. Still they do it and it works.
The long term solution is to create a second kind of reactor that has a higher burn fraction which means a more fuel efficient fast reactor. Those would be, ideally, the big base load plants if we did this rationally.
What concerns me is that 250 years of fossil fuel energy continues to store its waste products in my lungs and the water I drink. That's the issue we need to solve with urgency.
The "eh, just bury it" approach is really not a bad one. Its not even that much stuff to bury
[1] https://worldnuclearwastereport.org/
[2] https://www.eia.gov/todayinenergy/detail.php?id=16651
[3] https://www.colorado.edu/ecenter/2021/04/15/hidden-damage-la...
And who wants spent nuclear fuel in low earth orbit? This is a far worse location for spent fuel than buried in a bunker. This is a worse location than Times Square.
Low earth orbit would just be where it's transferred to something to carry it farther out, for example using solar-electric engines.
For all this, remember it isn't done immediately, it's done in (say) 300 years when the short lived fission products are mostly gone.
Should we double down on renewable energy and solve its issues with lots of batteries or should we invest in next generation nuclear energy?
Both at the same time?
Does anyone know?
Renewables and batteries to keep your AC, workplace EV charger, stove, pool heater and (since recently) green ammonia producer going, nuclear to prevent e.g. aluminium smelters from seizing up.
Also the cheapest way to make renewables work 24/7 is to build HVDC lines - they cost as much as a highway per unit length and even undersea cables would deploy for less and faster than equivalent nuclear.
The total length of HVDC lines just in China is currently more than 40k km, so they've literally deployed enough of them to wrap around the globe.
https://ourworldindata.org/grapher/co-emissions-per-capita
Not saying they’re not also building renewables and nuclear, but it seems like the policy is more “build anything and everything to satisfy demand” than a focused effort.
BTW: if you look at US emissions, the data center bubble hasn’t had much if any effect. They are still trending down. There’s reasons to dislike that industry but I’m sick of the mindless echo chamber doom on that issue. They’re not that significant in the grand scheme of things.
An additional data point to support that is that emissions intensity per GDP is clearly falling fast for China [3].
[1] https://ember-energy.org/data/electricity-data-explorer/?ent...
[2] https://ember-energy.org/data/electricity-data-explorer/?ent...
[3] Chart 75 from here: https://robbieandrew.github.io/GCB2025/
Personally I doubt that, as US emissions have been going up due to data centers:
https://www.reuters.com/business/energy/us-leads-global-co2-...
And while China, like you said, doesn't seem to have a focused goal, it so happens that renewables are the path of least resistance for just having more energy as fast and cheap as possible.
If you don't: stick to renewables.
And it also depends on what you mean by "we". As a Dane, I don't think us Danish taxpayers should invest in nuclear energy, but I'm perfectly happy that private Danish investors invest in Seaborg/Saltfoss and Copenhagen Atomics.
Do it all.
Nuclear really has to go big (supply most of the world's energy) or go home. But supplying most of the world's energy means burner reactors are inadequate -- there isn't enough cheap uranium. Burner microreactors have even worse neutron economy, so this argument applies even more so to them.
https://unece.org/sites/default/files/2022-04/LCA_3_FINAL%20...
I think a low carbon mix will result in the cheapest, most reliable and cleanest energy grid.
The empirical evidence has nuclear being uncompetitively expensive. The current focus on variant reactor designs appears to be something of a Hail Mary attempt to get around this sad state of affairs.
You sometimes see them making an argument about energy density, which goes back to Vaclav Smil. But Smil used this argument to massively mispredict how solar would be go in the market. We don't hear him much anymore.
Nuclear advocates increasingly resort to conspiracy theoretic reasoning to explain away the failure of their technology to compete. This should be a red flag.
I have a solution, take the subsidies spent on "renewables" and put them into nuclear! Easy peasy!
France nuclearized 75% of its grid in the 1980s while the solar folks were faffing around. It's not a cost issue, it's that anti-nuclear folks have choked out the industry.
We need to take the boot off the neck of nuclear. Wind and solar aren't an avenue to moving up the tech tree of civilization, which will involve using vastly more power.
What we do know is their attempt to build more NPPs now has gone spectacularly tits up, with costs completely out of control. This should make one view their earlier efforts with great suspicion. Have they become much worse, or were earlier problems concealed?
Or, maybe, it really is complicated and needs regulation to keep it sufficiently safe in a world where corners are always cut.
Life isn't Civ 6, there is no tech tree to climb.
We should be investing in all non carbon emitting sources and we should have been doing it since the 1970s when we figured out pretty conclusively that this would be a problem.
Instead we had right wing fossil fuel shills on one shoulder and unscientific woo woo greens on the other, the net effect being that we kept burning more carbon. We still have them, Trump with “beautiful coal” and greens now opposing even solar power and batteries, but climate change is no longer possible to ignore. Some still manage it but those people are nuts.
If we hadn’t stopped improving nuclear we’d probably have emitted half the CO2 we have. It would have become cheaper and safer and more scalable and then when China industrialized they would have copied that instead of burning so much coal.
France with its nearly zero carbon grid is the existence proof.
It wasn’t until the 2010s that solar and wind became grid scale in a big enough way to matter. That was too slow.
Whether someone is at least open to nuclear power is my litmus test for whether they take climate change seriously.
I do. If we hit 600, 800, 1000 ppm CO2, which is possible if the world keeps developing on the back of fossil fuels, we are entering existential risk territory. Earth has had those CO2 levels before, and higher, but our species was not alive then.
We already passed the FAFO threshold for ppm CO2 and now we will FO. But that’s not X-risk yet. I’m talking about the next threshold, which may start around 600 but really kicks in near 1000. This is where you actually start asphyxiating. You get lowered IQ and impaired judgement to a small degree, but across the globe at a time when we really don’t need it.
> "The demonstration and the licensing pathway it establishes represent a key step toward deploying electricity-producing microreactors for U.S. military installations by September 30, 2028."
So which is it? Power to the people or power to the military? This microreactor concept doesn't seem very well suited for commercial use.
Crippling diseconomies of scale.
Rooftop solar is an example of small scale decentralized energy production, maximum efficiency is not the only relevant metric.
> And you probably don't want dozens of "microreactors" scattered across cities
Why not? If they're considered safe and pass all inspections, what's the problem?
Rooftop solar does not have these issues.
Secondly although generating large amounts of power is more efficient in terms of generation, generating power close to the point of use is significantly more efficient in terms of power loss on the grid as I understand it.
They have diseconomies of scale. Some of the costs of a nuclear power plant scale sublinearly with power. Neutron economy is improved in a larger core. Larger turbines are more efficient than smaller turbines. It doesn't take 1000x as many operators to operate a NPP with 1000x the power output.
You'd also probably find similar issues with diesel generators, but small diesel generators do roaring trade and have great commercial applications.
Diesel generators have the advantage of being very cheap -- an order of magnitude cheaper than NPPs per unit power output -- and of having much of their total cost being fuel cost, so they can operate at lower capacity factor. But even so, we don't see large power plants composed of arrays of diesel microgenerators.
(The solution for current higher capacity factor diesel users, like say remote operation at mines, would be to supplement them with renewables and storage to reduce fuel costs. This is already happening.)
A significant problem with any small power plant is fixed costs. A 1 MW(e) plant (Antares is said to be between 100 kW(e) and 1 MW(e)) making power at 90% capacity factor and selling at $0.05/kWh will gross about $400K/year. A single full time employee, like a security guard, will cost a good chunk of that.
Oh sorry, I thought you were talking about efficiency. Ok, what is the cost is for these plants?
> A single full time employee, like a security guard, will cost a good chunk of that.
I dunno, a 1MW nuclear plant could end up being pretty small. It might easily be economic to install them places that already have security guards.