According to Chinese state media, a group of scientists recently managed to refuel a working thorium molten salt reactor without causing a shutdown — a feat never achieved before. The success was announced by the project's chief scientist Xu Hongjie during a closed-door meeting at the Chinese Academy of Sciences on April 8, Chinese news outlet Guangming Daily reported last week.
Such a breakthrough could be transformative to the global energy landscape, as thorium has long been hailed as a far safer and cheaper alternative to uranium in nuclear reactors. While also a radioactive element, thorium produces less waste, and the silver-colored metal, mostly found in monazite, is much more common in the Earth's crust.
According to the International Atomic Energy Agency (IAEA), thorium is three times more abundant in nature than uranium, but historically has found little use in power generation due to the significant economic and technical hurdles.
[...] Compared to uranium, thorium can generate a significantly higher amount of energy via nuclear fission. A Stanford University research estimates that thorium's power generation could be 35 times higher. Thorium molten-salt reactors (TMSRs) are also compact, do not require water cooling, cannot experience a meltdown and produce very little long-lived radioactive waste, according to the IAEA.
When announcing the breakthrough, Xu acknowledged that its project was based on previous research by US researchers who pioneered molten salt reactor technology in the 1950s, but abandoned shortly after to pursue uranium-fueled ones.
Xu — who was tasked with the thorium reactor project in 2009 — told Chinese media that his team spent years dissecting declassified American documents, replicating experiments and innovating beyond them.
China's TMSR-LF1 Molten Salt Thorium Reactor Begins Live Refueling Operations:
Although uranium-235 is the typical fuel for commercial fission reactors on account of it being fissile, it's relatively rare relative to the fertile U-238 and thorium (Th-232). Using either of these fertile isotopes to breed new fuel from is thus an attractive proposition. Despite this, only India and China have a strong focus on using Th-232 for reactors, the former using breeders (Th-232 to U-233) to create fertile uranium fuel. China has demonstrated its approach — including refueling a live reactor — using a fourth-generation molten salt reactor.
The original research comes from US scientists in the 1960s. While there were tests in the MSRE reactor, no follow-up studies were funded. The concept languished until recently, with Terrestrial Energy's Integral MSR and construction on China's 2 MW TMSR-LF1 experimental reactor commencing in 2018 before first criticality in 2023. One major advantage of an MSR with liquid fuel (the -LF part in the name) is that it can filter out contaminants and add fresh fuel while the reactor is running. With this successful demonstration, along with the breeding of uranium fuel from thorium last year, a larger, 10 MW design can now be tested.
Since TMSR doesn't need cooling water, it is perfect for use in arid areas. In addition, China is working on using a TMSR-derived design in nuclear-powered container vessels. With enough thorium around for tens of thousands of years, these low-maintenance MSR designs could soon power much of modern society, along with high-temperature pebble bed reactors, which is another concept that China has recently managed to make work with the HTR-PM design.
(Score: 3, Insightful) by corey on Monday May 05, @08:53AM (9 children)
As above, on the surface (I didn’t read the article yet), sounds like a real engineering barrier to feasibility. One of those things you don’t think of when considering the technology, a bit like how to get heat out of a fusion reactor.
It’s good they openly said they replicated the work done in the 50s by Americans, I respect that honesty, but also it acknowledges they are standing on the shoulders of giants, and taking the science further. Hopefully it’s shared so others can do the same for common good.
(Score: 5, Interesting) by Unixnut on Monday May 05, @09:23AM (7 children)
It will probably be shared, as a general rule scientific publishing tends to be global and public and the Chinese publish a lot of their research. Secondly even if they don't publish, just knowing it is possible (and eventually by studying or reverse engineering Chinese designs) the knowledge and understanding will spread globally.
To me however it is a shame to think about all the potential research paths for nuclear power were abandoned in the 1960s because they would not lead to a nuclear weapon.
The present technology was primarily designed to generate fissile material for nuclear weapons,and its suitability for power generation, risks of meltdown, long term waste storage, etc... were afterthoughts. As we can see with Thorium there were better alternatives to research for power generation in the 1960s, but you could not use them to make a nuclear bomb.
Ironically enough, having researched this technology and made their nuclear bombs, the current nuclear powers spent the rest of their time since then trying to prevent others from making nuclear bombs using their own nuclear power plants, as the design is dual-use.
Plus dealing with all the negatives of this particular design has sullied nuclear power in the eyes of the general public due to nuclear waste, risk of meltdowns, etc... making it harder to even get funding for new nuclear research and/or power plants in the western world. Which is another reason why the East has generally steamed ahead in this area.
(Score: 5, Informative) by Rich on Monday May 05, @10:10AM (6 children)
Hardly. CANDU and RBMK qualify, and the few metal cooled breeders left, but all the light-water stuff is energy-out only. The PWR has its origins in nuclear marine propulsion, so it is of military origins, too, but without any consideration for plutonium production.
PWR and BWR simply won out for civilian production, because it's cheapest to just put a big water kettle up and spin a turbine from the steam. Having a huge mass of molten, corrosive, super radioactive stuff around as an alternative will not ever be economically viable, even if factoring out the disposal costs (and here, someone might enlighten us with the neutron budget available for fission product transmutation...).
However, the prospect of long term continuous operation is very interesting where cost is not a primary issue - again, for military marine propulsion. Exhibit: https://en.wikipedia.org/wiki/Type_004_aircraft_carrier [wikipedia.org] . If they get the continuous feeding and fission product right with the molten salt experiment, they've got a winner and get around rebuilding their ship each 10 years or so.
(Score: 2, Interesting) by slon on Monday May 05, @01:27PM (1 child)
Does anyone know how they take the molten fuel out and replace it?
(Score: 2, Informative) by shrewdsheep on Monday May 05, @04:45PM
The salt processing might be integral part of the salt convection loop (online processing): https://en.wikipedia.org/wiki/Molten-salt_reactor [wikipedia.org]
It is my understanding that the reprocessing uses similar processes otherwise used in high-pressure reactors.
(Score: 5, Informative) by Anonymous Coward on Monday May 05, @02:12PM (1 child)
You speak about neutron budget?
Consider the consequence of enforcing a 'commercial reactor grade' limitation on fissile %, with the large remaining concentration of U-238.
Submarine LWR's acheive high burn up because they have the neutron budget to burn through reactor poisons.
The two things wrong with Liquid Thorium-Fluoride Reactors is that they breed nearly pure U-233 with an unremovable contaminant of U-232: which makes them an impractical choice for bombs. This make the nearly-pure fissile material unusable for bombs. This is the first thing 'wrong' with them.
The second thing is the burn-up achievable in a U233 fueled reactor would be 'too high'. This is 'bad' in the sense that a commercial reactor cannot be allowed to have such a high fuel burn up. Compare with low-enriched 'commercial reactor grade' fuels: Even with physically huge cores (RBMK, CANDU) to maximise neutron economy by size scaling, it still is only single-digits %. The U238 simply sucks away too many neutrons.
High burn-up (read: some non-single-%-of-fissile-quantity-fuel-efficiency) would make nuclear power cheaper by approximately the inverse of the burnup improvement. Ie, from say, 3% to 60% would make it 20x cheaper: Poor coal power would be unable to compete, and the powers that be did not want to allow this.
So we're told that "because it's weapons grade, you can't have it".
Meanwhile 100% fissile enriched fuel is routinely loaded into military small modular reactors (ie, submarines) and achieve such high fuel burn up that refuelling is probably not necessary for the service lifespans of the reactors.
NuScale imploded as soon as someone sat down and ran the sums on using low-enriched fuel in LWR-SMR's. It doesn't work, economically, because the physically smaller reactor necessarily has a much poorer neutron economy, so burnup is even worse.
This can also be shown simply from how very few remaining commercial nuclear ships remain: Just that one Russian one, probably because they *can* get high enriched fuel to run it. All the rest of those built in the 50's operated on low-enriched fuel, and were consequently commercially unviable until refit with Diesel engines.
If low-enriched fuel in reactors was all that good, it would already have displaced coal, and it hasn't. SMR's with low enriched fuel make even less sense, and can't even keep up with coal.
If you think my reasoning is wrong, 'because the cost of the fuel is only a tiny part of the costs of running commercial nuclear reactor', that would be because most of those 'other' costs also scale with how much you have to deal with the fuel, all of which would scale down if you didn't need to 'play' with it so much.
The reason LFTR's can have extremely high burnups, is because it's feasible to actually remove the waste - at least the undesirable reactor-poison parts, whilst the reactor remains online.
All fission reactors 'shit where they eat'. The process is stochastic, and happens distributed throughout the core. So the waste ends up there as well. Liquid Fluoride fuel allows the wastes to be removed without stopping the reactor.
Putting the fuel in a form that can be pumped also allows it to flow to visit the heat exchanger when it's hot: This means that the heat exchanger doesn't need to also be a neutron-oven like the core. Doing fewer things at once allows a cheaper design to work better.
Using a fluid that has such an incredibly high boiling point prevents causing steam explosions also. No need to design for those explosions, because no compressibility of fluids in the core.
On top of that, if you do have a leak, it will soon freeze back into a solid rock again, which is soft enough to be easily broken up and removed piece by piece: Not something that is true of solid-fuel-reactor 'corium'.
And then there's the chemical *stability* of the fluoride fuel against ionizing radiation: Unlike water, which constantly ionizes, fluoride salts are ionically bonded: Yes, with radiation present you'll get some F gas come out, but it's not a big deal because you can design for it, and it doesn't cause containment-building-shattering detonations like Hydrogen does.
So far as 'oh no so much radioactive!!!' That doesn't actually matter. The reason is, distance is the best shielding, and you don't need all that much of it (metres). Oh and you can put shielding up, that lets people work only 10's of metres away. It's all very well understood by now. People can live on nuclear submarines and never receive any more dosage than they would at home as a civilian. Do you worry about being unable to safely hug a blast furnace without it burning you? No? Then why worry about not being allowed so close to a reactor core? At least these reactors don't require people to 'play with it's food' for it, unlike solid-fueled reactors, which actually all do.
The fuel can just flow instead. And melt-downs are not a problem either: There can be a arranged a separate, non-critical configuration catchpot underneath, equipped with passive cooling as well as electric heater elements and a pump: so no physical access is required to later 'melt-up' the reactor and put it back into operation. They're melt-down 'compatible'. With the 'turn off' automatically doing this so reliably, that as soon as you start the process, you can just walk away because it'll be fine. This has already been demonstrated before.
This world first - and it is one - was kind of already done before: Online refueling (albeit with Uranium) was demonstrated
previously for a molten-salt reactor. Turns out you only need to drop the metal into the molten salt to do this, maybe add a little more Fluorine. It dissolves. You can add it in as a fluride also, and then it only melts and disperses. Getting the material to have it to put in - and I remind you: Nearly 100% pure fissile material - that's the challenge. IIRC their reactor design is just a huge pool of the fuel salt.
I don't recall whether they surrounded it with a second 'breeder' salt with the Thorium in it, or just mix that into the one salt.
Adding Thorium works the same way as adding Uranium, except that it *becomes* fissile later after absorbing one neutron and a decay process.
Online refuelling from Thorium-derived U233 taken from a breeder blanket salt is genuinely new however, so this would still be a great achievement, assuming that's what they did.
The whole goal of this low-power research reactor (with *passive* liquid-fuel flow through the heat exchangers: it just uses natural convection!) is really to provide a facility to provide 'used' fluoride fuel so they can develop the means to effect online *waste removal*. That new tech (which is theoretical only right now) would unlock the cheapest, safest and most sustainable (and also most green*!) energy source ever known: It will an epoch-defining event, not unlike the advent of steam power.
* ecologically as well as literally: Uranium fluoride has that color too.
(Score: 4, Informative) by turgid on Tuesday May 06, @11:18AM
If you think my reasoning is wrong, 'because the cost of the fuel is only a tiny part of the costs of running commercial nuclear reactor', that would be because most of those 'other' costs also scale with how much you have to deal with the fuel, all of which would scale down if you didn't need to 'play' with it so much.
Exactly. I worked at a Magnox station which ran on natural (unenriched) uranium. The reactors were enormous (the vessels were 20 metres diameter) and the whole thing only put out about 240MW electrical total from both. The cores had to be constantly refuelled on load to maintain the flux profile. There was the order of 20k fuel elements in each reactor. Handling fuel was a major part of the operation.
A submarine gets loaded with fuel once. Thirty years later, they lift the reactor out with a crane. Job done.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 2) by JoeMerchant on Monday May 05, @07:41PM
> get around rebuilding their ship each 10 years or so.
Not a problem when you've got 11 ships on active duty...
🌻🌻🌻 [google.com]
(Score: 2) by gnuman on Tuesday May 06, @08:13AM
CANDU has been designed to burn Plutonium (and everything else), not breed it. Basically, you could use it in a bind, but there are much more efficient ways of dong this.
(Score: 2) by JoeMerchant on Monday May 05, @11:37AM
I have long wondered if the only problem with Thorium is that it doesn't go BOOM? Uranium and Thorium programs are both fabulously expensive to develop, but when you develop the Uranium/Plutonium side of the house you get BIG BOOM along with electricity, develop Thorium and you just get electricity; and there's not a tremendous overlap between the two development paths. Both have their tricky waste disposal issues, Thorium less than Uranium as I understand it, but still significant on both sides.
So, kudos to China for going so far as building a working demonstrator. Maybe they really are more serious about protecting the environment for future generations, just needed to leverage coal and oil for a big enough push to be truly independent of domination by other world powers that might go mad at any 2 year election cycle?
🌻🌻🌻 [google.com]
(Score: 5, Interesting) by turgid on Monday May 05, @09:53AM
The British gas-cooled reactors (Magnox then AGR) were designed for on-load refuelling. It was never actually done at full power in the AGRs for safety reasons.
The reactors were designed to operate with an equilibrium fuel cycle of continuous refuelling, to maintain an optimal reactivity and flux profile.
The refuelling machines (charge machines) were giant shielded cranes (450 tons) that became part of the gas circuit. They could store several channels of new and used fuel elements. They were expensive to maintain and with hindsight, on-load refuelling was an expensive complication.
Due to the design of AGR cores, the re-entrant cooling and the complicated fuel stringers, a safety case could only be made for refuelling at 30% power. The potential for a blockage and a hotspot were too great.
With these molten salt reactors, I believe they don't run at high pressures therefore the refuelling machinery can be far simpler.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 2) by VLM on Monday May 05, @01:21PM (1 child)
LOL uh huh
(Score: 5, Funny) by krishnoid on Monday May 05, @05:32PM
No, no, it all follows step-by-step:
(Score: 2) by VLM on Monday May 05, @01:37PM (2 children)
I wonder why you'd want to do that. The safety risks are high and the rest of the plant wears out and needs inspection so IIRC from my research there's somewhere between rarely and never a situation where the refit was delayed because of refueling process problems. ALL the time they find unexpected (unexpected as in out of limits) corrosion that needs research or some plumbing failed during final closeup tests or there's some bearing/shaft/turbine thing with a painfully long lead time that unexpectedly needs replacement.
Refueling while powered up would NEVER be allowed on US soil. For generations ALARA has been pretty deeply ingrained. "Well we could shut it down and do a room temp zero pressure refuel, but we thought it would be fun to do it during full power operation" NRC would absolutely shit themselves. To voilate ALARA like that would take something like losing an invasion by space aliens, maybe not even that.
The knee-jerk military claim doesn't work, the rest of the ship needs extensive maint around refueling time anyway. Everything in the propulsion system is worn, as is the rest of the ship, and the whole thing needs a refit stem to stern.
An interesting to think about problem for the military would be whatever the Navy analogy is to "GI Proofing" the refueling equipment and material. It'll all be radioactive and as part of hazing or just general inexperience you know some poor sailor is going to shove the nuclear refueling hose in a F-18, not knowing any better, or pour used radioactive refueling salt on the deck to prevent icing ("uh, it said salt on the barrel label"). This just isn't going to turn out well.
One thing I can think of is some kind of "picking up nickels in front of the steamroller" strategy of trying to extend maint intervals. This will probably lead to an accident, not directly because of refueling but because a lot of stuff gets found during maintenance intervals that now will ... not. It'll get found during the subsequent accident investigation.
Another thing I can think of is CIA cloak and dagger stuff to get around non-proliferation treaties. "Oh that barrel thats glowing blue? Thats not illegal nuclear proliferation trafficing thats merely supplies for our sub thats refueling while underway, you know, as you do, totally not illegal nuclear proliferation in violation of the treaty".
(Score: 3, Touché) by Anonymous Coward on Monday May 05, @01:59PM
The current administration will probably defund the NRC anyway, so there wouldn't be any push back from that organization.
(Score: 3, Informative) by gnuman on Monday May 05, @04:19PM
https://en.wikipedia.org/wiki/CANDU_reactor [wikipedia.org]
So, many reasons... maybe not so important these days as uptime of regular reactors has been brought up. Reddit has some interesting comments here,
https://www.reddit.com/r/NuclearPower/comments/twmixy/candu_vs_us_lwrs/ [reddit.com]
You can also put stuff in the reactor that has nothing to do with fuel. Like making medical grade Colbalt 60 -- half world supply comes from CANDU reactors. Having a high flux neutron source is very beneficial here.
https://www.nuclearfaq.ca/malkoskie_cobalt_paper.pdf [nuclearfaq.ca]
and others,
https://www.world-nuclear-news.org/Articles/Canadian-Candu-produces-cancer-therapy-isotope [world-nuclear-news.org]
(Score: 3, Touché) by gringer on Monday May 05, @09:25PM (1 child)
Fixed that for you:
Ask me about Sequencing DNA in front of Linus Torvalds [youtube.com]
(Score: 2) by driverless on Tuesday May 06, @03:30AM
Also:
The way he runs things, it won't last a hundred.
(Score: 4, Funny) by weirsbaski on Tuesday May 06, @02:11AM
Unfortunately, the reactor used Microsoft lead shielding, that means the scientists will have to subscribe to a monthly fuel plan from Microsoft to continue to hot-load thorium in the future.
(Score: 3, Touché) by gnuman on Tuesday May 06, @08:36AM (3 children)
There is nothing special about Thorium for power generation. It's basically like Uranium. You can very similar short term issues with Thorium as you do with Uranium reactors. You can melt them down (depending on reactor design, of course, *not* fuel dependent here), you can create radioactive crap lasting many many generations. It's just a different fuel. So if you have lots of Thorium around and not much Uranium, then you probably want to have Thorium reactors.
One difference between thorium and uranium is that ALL thorium require some uranium to get going, then you *want* to refuel them online and keep only using thorium input as it breeds uranium in the reactor. Basically, you can already use CANDU reactors to burn Thorium -- you don't need fancy designs here.
https://www.sciencedirect.com/science/article/abs/pii/S0196890405002670 [sciencedirect.com]
As for molten salt reactors, well, those are "meltdown reactors" by definition. The fuel is liquid and quite warm. If you get any leaks, you end up with a massive mess on your hands. That's the main reason no one is using MSRs -- they are a nightmare to cleanup. And if primary vessel needs major maintenance --- yeah, good luck!! There are some theoretical calculations how these reactors can run for 200 years or so, but then who maintains that? The current reactors are much much easier to cleanup and it's already difficult enough. I would say, MSRs are great in theory, terrible in practice.
FTFA (I can't resist):
Which is same for Chernobyl, hahaha. It also "cooled down" when "exposed to air". It's the other issues that happen when your core is "exposed to air" that is the problem.
(Score: 2) by turgid on Tuesday May 06, @11:30AM (2 children)
The Chernobyl problem was quite different. When you have a uranium rector, over time, the U-238 captures neutrons and decays to plutonium-239, which is fissile. This leads to the positive temperature coefficient of reactivity. Because the Pu-239 is fissile (fast neutrons), the hotter the reactor, the more fission it undergoes and the more heat it produces. It's positive feedback.
You can get into a situation in an emergency that if you have shut your reactor down and it over-heats, it comes back to life (goes critical) all on its own due to the high temperature. Uranium reactors are designed very carefully to prevent this. They have "hold down" rods as well as control rods. There are other features such as boron balls, boron dust, nitrogen injection and boric acid (in water cooled reactors).
If you get into the terrible situation where the core is disturbed and the fuel melts, if there is enough of it and if it gets hot enough, you might get a re-crtiticality. That won't last long. It'll blow itself apart. It will be quite unpleasant. I believe PWRs are designed to contain melted fuel. This is what saved the day at Three Mile Island.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 2) by gnuman on Tuesday May 06, @12:34PM (1 child)
I think you meant *positive void coefficient* ... it's a property of reactor *design*, not the fuel.
You can have this with thorium just as you can have this with plain uranium.
Thorium is not better or worse than Uranium or even Plutonium as fuel. It's just a fuel. The difference is like using methane to power your car or gasoline or diesel. They all burn. They all produce CO2. They all can cause same engine problems. etc. etc. Nuclear fuels are exactly the same --- they burn similarly but have slight different characteristics. The arguments is like arguing has methane is sooo much safer than gasoline or diesel. In the end, it doesn't matter -- you can die by accident with any of these fuels.
(Score: 2) by turgid on Tuesday May 06, @01:24PM
I think you meant *positive void coefficient* ... it's a property of reactor *design*, not the fuel.
No, that's something else. It is related, however.
you can die by accident with any of these fuels
True, but the probability of an accident is greater or lesser depending on the fuel and the uses, and so are the consequences of that accident.
I worked in Reactor Physics at a nuclear power station for a few years. I had a bit of training. We looked at certain famous accidents.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].