[neutronicus]

    TRWTF is that the design requires active cooling for safe operation.

It's less of a WTF than you think. You have to read the fine print of the supposedly "passively cooled" designs. The Westinghouse AP1000, for instance, is "passively cooled" in the instance of a scram, but only for 72 hours. Although this is about 9 times as long as the Japanese cooling systems ran on batteries, it is significantly less time than the reactors have been without power. The fact is that a lot of decay heat is dumped into that system in a pretty short time, and you're going to need mass flows (read: pumps) if you're going to reject it all safely.

(I suppose a molten salt reactor might allow you to get more creative with mass transport, since you can conceivably move the fuel somewhere else, but those designs are a long way off, and I'm sure they come with their own unique hazards)

[jacques_chester]

Thanks, I didn't know that about the 3+ systems. However my understanding of Gen 4 systems is that passive cooling is the central design element.

I mean, it's inefficient, but when you're talking about nuclear it's not as though source energy is hard to come by.

[neutronicus]

"Passive cooling" are words that people like to hear, so everyone's going to tell you that their favorite Gen IV is "passively cooled", but one has to approach these claims with a certain level of realism.

I mean, the reactor's going to be producing about 1% of rated thermal power for a few days after shutdown, so 36 MW. The heat of vaporization of water is about 3 MJ/kg (heat of vaporization is more important than heat capacity, so we'll just use that). That means that the fuel inside a 1 GWe nuclear reactor, a day or two after shutdown, can boil 12 kg of water in a second.

This is a fundamental property of nuclear reactors - no reactor design is going to make decay heat disappear. So I would be skeptical of designs that claim to be "passively cooled". What they mean is "passively cooled for a while", which is an important engineering achievement that I don't want to trivialize at all - I just think it puts the current designs a little in perspective, since some people apparently believe that you can scram a gen 4 reactor, go on a luxury 7-night cruise, and then deal with at your liesure, which isn't true - you've got to find a way to get water in there (or whatever), you just have a lot more time to do so because of the passive safety systems.

[jacques_chester]

So safer, but not failsafe past 72 hours.

An incremental improvement, not a quantum leap.

[neutronicus]

Right. The number may be longer for Gen IV as opposed to Gen 3.

Rejecting 36 MW of heat for weeks at a time without some intervention is a very difficult, if not impossible engineering problem.

[DennisP]

How about liquid thorium? Reactor gets too hot, a plug melts in the bottom and all the fuel drains into a wide basin.

[iwwr]

How about submerged reactors? They'd use ocean currents to cool passively.

[neutronicus]

Probably yes, although possibly no, since rate of heat transfer depends on flow rate past the heat transfer surface.

Given that the AP1000 manages for 72 hours solely with "hot water rises, cold water sinks" levels of flow, a submerged reactor could probably remain cooled indefinitely, although salt water corrosion would then become a part of your life.

[_djo_]

The South African-designed Pebble Bed Modular Reactor[1] was supposed to be designed to be meltdown-proof through the use of passive cooling.

Rather than using rods for fuel and water as the coolant, the PBMR design used graphite and TRICO-coated LEU pebbles as fuel and Helium as the coolant.

In the event of a total loss of control, the large surface area of the pebble bed would theoretically dissipate the heat at enough of a rate that the highest possible temperature the core could reach would be around 1600˚ celsius, too low to cause the fuel to melt.

So even in a worst-case scenario the reactor core temperature would just rise to a safe 'idle' temperature and sit there indefinitely, allowing more than enough time for the pebbles to be safely removed from the core.

At least, that's the theory as I understand it. But the PBMR project was cancelled as part of the South African government's cost-cutting measures, the project's rising cost and a belief that the commercial future of the project was bleak. There were also concerns from some quarters about the resistance of the graphite coatings on the spheres to fire. The intended pilot plant was not built.

However, I understand that there are other high-temperature reactors like the PBMR in development around the world, so the technology is still being pursued. It's an interesting approach to the problem of trying to get the benefits of nuclear power while minimising the downsides.

[1] http://www.pbmr.com/

[neutronicus]

I did a little digging to see how pebble bed designs approach the decay heat problem. I found a decent publicly available example[1] (MOX fuel notwithstanding). The money shot occurs on page 37:

    Heat from the core is transferred to the reactor heat removal system (RHRS)
    that is surrounding the RPV at a distance of 1 m. These
    water panels are assumed to be at a constant temperature 
    of 70°C.

This is a bit of a shell game - the problem becomes cooling the water panels so that assumption remains valid instead of cooling the core. This is a much _easier_ problem, since you can use whatever water you like in those panels without damaging the reactor, and the water in the panels won't get contaminated.

And that's sort of what I'm getting at - "passively cooled" in the nuclear industry tends to mean that "the core is passively cooled via heat transfer to some heat sink that you're eventually going to have to think about cooling".

[1] http://www.inl.gov/technicalpublications/Documents/4655310.p... [pdf warning!]

[_djo_]

So in order to prove that the PBMR's claim of using passive cooling is incorrect, you dug up some docs for a completely different pebble bed design?

Eskom specifically claimed 'walk-away' safety for the PBMR, due to the fact that when the reactor was in an idle mode the temperature within the pebble bed core could never rise high enough to melt the fuel. Here's one such claim from their website:

The PBMR is walk-away safe. Its safety is a result of the design, the materials used and the physics processes rather than engineered safety systems as in a Koeberg type reactor.

The peak temperature that can be reached in the reactor core (1 600 degrees Celsius under the most severe conditions) is far below any sustained temperature (2 000 degrees Celsius) that will damage the fuel. The reason for this is that the ceramic materials in the fuel such as graphite and silicone carbide - are tougher than diamonds.

Even if a reaction in the core cannot be stopped by small absorbent graphite spheres (that perform the same function as the control rods at Koeberg) or cooled by the helium, the reactor will cool down naturally on its own in a very short time. This is because the increase in temperature makes the chain reaction less efficient and it therefore ceases to generate power. The size of the core is such that it has a high surface area to volume ratio. This means that the heat it loses through its surface (via the same process that allows a standing cup of tea to cool down) is more than the heat generated by the decay fission products in the core. Hence the reactor can never (due to its thermal inertia) reach the temperature at which a meltdown would occur. The plant can never be hot enough for long enough to cause damage to the fuel.

Whether the claims about the PBMR were accurate or not is an open question that won't be answered until somebody else manages to commercialise this technology. But it's at least important to question the actual claims made for the PBMR and similar reactors, rather than dismissing them out of hand because some pebble-bed designs rely on a certain level of active cooling.