Back

Chinese pebble-bed nuclear reactor passes "meltdown" test

137 points21 hoursans.org
cyberax19 hours ago

Pebble bed reactors are a bad idea in general.

They will be HUMONGOUS because they need a large surface to radiate away the heat for the passive safety, so they can't be easily put into a containment building.

A core of a PWR plant is _tiny_ for the amount of power it produces (around 3GWt!), just around 5 meters in diameter and 15 meters in height.

The pebble bed reactor in the article (HTR-PM) is around the same size, but it produces a mere 0.25 GWt.

Pebbles themselves are also problematic, they tend to swell, crack, and they can't be reprocessed using the current technologies. They MASSIVELY increase the amount of waste.

credit_guy14 hours ago

This reactor indeed has lower power density than a PWR, but not by a factor of 12. I compared it with NuScale's reactor, which is a PWR SMR. Details about both can be found in [1]. The HTR-PM reactor pressure vessel has a volume of about 640 m3, and yields 105 MWe, while NuScale has a volume of 101 m3 and yields 77 MWe. The power densities come to be 6.1 m3/MWe vs 1.3 m3/Mwe, and the ratio is 4.7x.

Still, this is a good price to pay for getting a meltdown-proof reactor.

> Pebbles themselves are also problematic, they tend to swell, crack, and they can't be reprocessed using the current technologies.

It is simply not true that pebbles tend to swell and crack. Quite the opposite happens: fuel elements in the current generation PWRs tend to swell, crack and burst. This happens because some fission products and decay products are gasses, such as xenon, kripton, radon. They build up in time and create internal pressure. The same happens inside the fuel kernels in the pebbles used in this reactor, but those kernels are specially built to withstand much higher internal pressures.

Here's a relevant quote from [2]:

  > As the pressure vessel size is reduced, the more efficient spherical geometry can be adopted, and the required wall thickness drops dramatically to the point that a 35 μm SiC layer can indefinitely contain gas pressures in excess of 100 MPa. This compares to the main reactor steel pressure vessel which may go as high as 20MPa or the Zircalloy cladding which can have pressures up to 10s of MPa in limited conditions – temperatures far below what the ceramic pressure vessel tolerate
As for the reprocessing part, I think you are jumping the gun. There is no reprocessing done in the US, at all, for any type of fuel. Even where reprocessing happens, as in France, the benefit is quite reduced. One can extract some plutonium and unburned uranium, but in the end that will allow you to extract maybe 10-20% more energy from the original amount of natural uranium. It will not make you extract one hundred, or 10 times, or even just twice as much energy. Reprocessing is simply not a game changer. It is not clear at all if it makes economic sense to build the highly complex facilities that do reprocessing, for the limited benefit.

[1] https://aris.iaea.org/publications/SMR_catalogue_2024.pdf

[2] https://www.usnc.com/triso/

huijzer8 hours ago

Is NuScale’s design a good one to compare against? Isn’t it a theoretical design only?

I held their stock for a while until I realized they don’t exist to make a reactor. They exist to get funding.

Sounds crazy, but look through their actions. All press releases are just talk about what they will research and with whom they talked or made a “memorandum of understanding”. The CEO CV is also interesting since it lists a whole lot of board positions and titles but it’s not clear what he has actually done.

instagraham7 hours ago

interesting comment - made me look up the stock and it just jumped. Propelled by this news perhaps?

huijzer7 hours ago

price != value

Otherwise Buffett wouldn't be so rich.

yeknoda13 hours ago

PWR pressure is on order 2-3x higher than an HTGR, so that’s a 2-3x thicker vessel. So good savings there

cyberax12 hours ago

> This reactor indeed has lower power density than a PWR, but not by a factor of 12.

It's more.

> while NuScale has a volume of 101 m3 and yields 77 MWe

VVER1200 has the inner vessel _diameter_ of 4.2m, height of 11m for the internal volume of 153m^3, and 1200MWe capacity (so around 3GWt).

_THIS_ is what you're comparing it with.

> Quite the opposite happens: fuel elements in the current generation PWRs tend to swell, crack and burst

Nope. A swollen or a ruptured fuel rod in a regular reactor is a reason for SCRAM. The water inside the reactor vessel is constantly monitored for fission products. The individual fuel tablets swell, but they are contained inside zirconium rods.

freeone300018 hours ago

So for a 3GW pebble bed reactor, we’re looking at a core the size of small house instead of a master bedroom? I don’t see a huge difference here; it’s the same amount of everything else (cooling, pumps, turbines, security) since it produces the same amount of heat/power.

cyberax18 hours ago

> So for a 3GW pebble bed reactor, we’re looking at a core the size of small house instead of a master bedroom?

No, we're looking at a core the size of a small residential tower. Probably around 30 meters in height.

JojoFatsani18 hours ago

That’s not too big man

+1
Etheryte15 hours ago
pfdietz16 hours ago

It is, when you consider the problem of dealing with spent pebbles. These are not like the fuel rods of a LWR that occupy a fraction of a much smaller volume. The dry casks to store them would be immense.

lambdasquirrel16 hours ago

Sounds terrible for a relatively dense place like Western Europe or Japan, but I think this would be fine in the US and China.

shepherdjerred18 hours ago

Is GWt a common abbreviation for gigawatt? I first read that as gigawatt-tons which is a… confusing unit

howenterprisey18 hours ago

Gigawatt thermal, as opposed to gigawatt electric. Gigawatt thermal is the heat your power plant makes, whereas gigawatt electric is the electricity that the heat is used to generate. They're not the same because not all the heat can be converted into electricity, and the percent of heat that gets converted varies from power plant to power plant.

shepherdjerred18 hours ago

Oh that makes total sense! Thanks for explaining

retzkek12 hours ago

There is a similar unit used for tracking nuclear fuel “burnup,” which is how much energy it produced: GWd/t (or GWd/MTU).

pfdietz17 hours ago

The other problem is mechanical abrasion of the pebbles, creating radioactive dust.

One of Germany's PBRs had to filled with concrete after it was defueled, they couldn't decontaminate it enough to dismantle it.

Grosvenor18 hours ago

Seems like that would be fine for places where you have < 0.25GWt energy needs, and need a safe power source. Like remote installations/towns. Antarctic research stations, etc.

cyberax18 hours ago

Pebble bed reactors are indeed researched as a source for process heat (e.g. for steel or concrete production). But I really dislike that.

If you just need 250MW of power, then just use electricity sourced from a regular PWR for heating. It'll be cheaper.

Retric13 hours ago

The maximum temperature from nuclear reactors is far below what you need for steel production because otherwise you’d melt the steel used to construct them.

Waste heat can be useful for district heating systems because houses don’t need to reach high temperatures, but few designs give you access to even 300C and nothing currently hits 1,000+C.

cyberax12 hours ago

Sorry, you're right. Pebble bed reactors can provide helium at around 600C (reasonably up to 700C, but that's at the limits of material science).

klysm10 hours ago

I’m not convinced the size is actually a problem - is that the most constrained resource here? What do we gain from decreasing power density?

pcl13 hours ago

PWR evidently means Pressurized Water Reactor.

thebeardisred20 hours ago
kordlessagain12 hours ago

Interesting they are now posed to take advantage of this power for applying AI: https://www.washingtonpost.com/opinions/2024/09/24/ai-power-...

dbcooper19 hours ago

I remember that article. An older friend had worked on fluidised-bed reactors at Tsinghua in the 1990s, and I sent it to him.

Using helium indicates a problem with kinematic viscosity of cooling gases? Nitrogen would be non-reactive enough, I assume?

semi-extrinsic19 hours ago

Helium has excellent heat transfer properties (low Prandtl number) and does not undergo nuclear reactions when subjected to a neutron flux.

Nitrogen will undergo an (n-p) reaction to produce carbon-14 which has a half-life of 5700 years.

cyberax18 hours ago

Yup. That's why regular PWRs take care not to nitrogen to pressurize or flush the primary coolant loop after maintenance.

+1
pfdietz17 hours ago
HPsquared17 hours ago

Nitrogen is a big neutron absorber in gas-cooled reactors. It's actually used as a secondary shutdown mechanism in the UK's AGR reactors. So if it was to leak out (lose pressure) you'd see an increase in power at the same time as a loss of cooling... Not a great idea!

yk17 hours ago

> Other pebble beds: The pebble bed technology and design has previously been used in prototype reactors in China and Germany, but not a larger-scale plant like Shidaowan.

That's wrong, Hamm-Uentrop was a full scale commercial reactor. It did run in total for a week or so between 1985 and 1989 and was then shut down. The fundamental problem is, that the pebbles grind against each other, and being of the same material as pebbles they can grind each other down. (Now if you wonder why this wasn't discovered at the experimental reactor in Juellich, those guys just never mentioned that they lost fuel.)

Animats18 hours ago
consumer45116 hours ago

Thanks for sharing those links. That was a very interesting read.

muditmudit20 hours ago

Tangentially related: https://youtu.be/0gskQJE6lxU?si=nztv5C0Et7pBJeMj

This video explores an incident with a reactor of a similar design, and very rudimentarily explains the way pebbles and the helium gas is used.

dongobongo13 hours ago

This actually happened in August and September of 2023 and it’s great validation for High Temperature Gas-cooled Reactor (HTGR) at larger scales. I hope they have the guts to also do a full loss of coolant test. I’ve also heard the these two reactors have been turned off for quite a while due to issues with the primary heat exchangers, temperature fluctuations, and uneven cooling - characteristically disadvantages of pebble beds.

There’s of course 2 flavors of HTGR (prismatic and pebble bed), and people choose the pebble version for continuous refueling despite all the drawbacks [1]. But there’s a lot of reasons to do prismatic. Can’t wait to see China’s prismatic HTGR.

[1] https://lvenneri.com/blog/pebble-bed-nukegumball

8bitsrule13 hours ago

Hmmm. The Titanic had sixteen watertight compartments.

Did the pebble-bed reactor's "commercial-scale inherent safety" also pass a test with an compromised container, which would admit air that would cause the graphite to burn?

JumpCrisscross11 hours ago

> which would admit air that would cause the graphite to burn?

Rats, then it would cause about as many deaths per GWh as the coal it's replacing.

natmaka10 hours ago

The real underlying subject isn't "coal or nuclear" but "nuclear or renewables", and a potential nuclear plant not inducing any risk of major nuclear accident (nor any annoying waste? nor any dependency towards any fuel?...) would induce quite new way to consider it.

JumpCrisscross10 hours ago

> real underlying subject isn't "coal or nuclear" but "nuclear or renewables"

Not in the real world. We're deploying solar and wind as quickly as we can because they're the cheapest sources of power. The bottlenecks are production and permitting, and there is no indication either of those are seeing a step change in the next decade.

There is a reason America and Europe, where anti-nuclear sentiment runs ripe, are building gas power plants and infrastructure at the fastest clip in history. In the West, the choice is gas or nukes. In the China and India, it's coal or nukes.

+1
natmaka9 hours ago
GeoAtreides5 hours ago

the compartments weren't watertight 100%, they weren't sealed at the top

m3kw911 hours ago

The automatic mechanism could fail right?

paranoidrobot10 hours ago

The safety of the system comes from it's design. It doesn't rely on anyone, or any mechanical process cooling it.

> The pebble-bed reactor is designed so that this effect is relatively strong, inherent to the design, and does not depend on moving parts. This negative feedback creates passive control of the reaction process.

From the Wikipedia article on the PBR that was linked in a neighbouring comment.

It should, at least if built correctly, be impossible for it to get into a run-away state. Even if it lost all power, and everyone walked away, it should not melt down.

[1] https://en.wikipedia.org/wiki/Pebble-bed_reactor#Safety

pineaux20 hours ago

Can someone explain to me how this reactor is meltdown proof?

cosmotic19 hours ago

Wikipedia has an explanation: https://en.wikipedia.org/wiki/Pebble-bed_reactor#Safety

Summary: As the temperature rises, neutron absorption increases, reducing fission and thus temperature.

chickenbig19 hours ago

> As the temperature rises, neutron absorption increases, reducing fission and thus temperature.

Negative fuel temperature coefficient is not an unusual feature.

The real question is whether the heat removal system of the reactor as a whole is sufficient to remove the decay heat to keep the fuel within the limits.

Gibbon118 hours ago

I remember talking to an engineer at the old GE nuclear research facility in San Jose. He said you can design reactors to be cooled by natural convection.

chickenbig3 hours ago

> He said you can design reactors to be cooled by natural convection.

That's the point of the Westinghouse AP1000; the containment (steel liner) and protection from the outside world (concrete wall) are separated, allowing the liner to cool by convection and water dripping from above. Admittedly you need to top up the water tank at the top, but that is less of a task than trying to push water into the containment.

pfdietz17 hours ago

It just makes them larger. And it makes the building containing them larger. And this makes them more expensive.

NuScale's reactor was originally motivated by the desire to make it safer by using natural convection. But it ends up requiring 1/3rd more labor hours to build a NPP using their reactors than it does to build a conventional large reactor power plant.

cyberax18 hours ago

The reactor vessel is humongous, so the natural convective cooling can carry away the decay heat. The pebbles themselves can tolerate extremely high temperatures (literally glowing white-hot) without burning.

DennisL12320 hours ago

It’s meltdown proof in principle since the process goes into an equilibrium rather than into a runaway process.

euroderf19 hours ago

IIRC there is a question about graphite fires.

kevin_thibedeau18 hours ago

It's not, just statistically unlikely assuming no fuel pebbles crack and coalesce their fragments.

AtlasBarfed19 hours ago

Can pebble beds have a cooldown pan similar to a LFTR, where a plug melts and the "pebbles" fall and spread into a pan where they won't stay critical because they are too separated / unconcentrated?

Because the real problem with solid rods is that they ... are solid rods, and if they start "overreacting" you can't split up the rods, unlike a pile of pebbles/spheres.

cyberax18 hours ago

The unique "melt plug" safety story of LFTRs is mostly a fairy tale.

Modern PWRs also have this safety feature, if a core melts down, the molten mass will be contained in a core catcher. Where it'll be mixed with inert material that can provide enough surface area and thermal mass to prevent further fuel mass migration.

The biggest problem in the core catcher design was to make sure that the molten fuel lava spreads out enough for the passive cooling to stop it from melting through concrete.

Pebble bed reactors will have a similar problem. You can "drain" pebble beds somewhere, but then you need to make sure that this "somewhere" can conduct away the decay heat without melting.

petre19 hours ago

They should also test with cracked pebbles.

bargainbot3k15 hours ago

> New testing done at China’s Shidaowan nuclear power plant has confirmed its ability to be naturally cooled down, an industry-first milestone for achieving commercial-scale inherent safety, according to researchers.

Amazing. Well done! How far this country has come in the last few decades is nothing short of breathtaking.

cynicalsecurity17 hours ago

The obsession with "green" energy wouldn't do us any good, would it.