Comment by neutronicus | Hacker News Reader

Comment by neutronicus | Hacker News Reader

Like they wouldn't have boron near a nuclear reactor. I'm a PWR man myself, but I am 99% sure that GE BWR designs have a couple of big tanks of borated water specifically for a LOCA. I highly doubt they're having reactivity problems, it's probably just managing the decay heat, and now the radiation leaks, that's the problem. "Just" is a relative word, here, of course.

Boiling water reactors are simpler, cheaper, but generally aren’t made anymore because they are perceived as being less safe. That’s because the exotic coolant in the pressurized water reactor can contain boric acid which absorbs neutrons and can help (or totally) control the nuclear reaction. You can’t use boric acid or any other soluble boron-laced neutron absorbers in a boiling water reactor because doing so would contaminate both the cooling system and the environment.

He's completely wrong about industry adoption of BWRs. There are two BWR's planned to be built in the US (along with 3 or 4 PWRs), and I believe that China has contracted with GE for a few as well (along with 4 Westinghouse PWRs and maybe a few Areva ones too).

PWRs are preferred largely because of their higher power densities (a BWR core that produces the same power must be larger) and simpler nuclear calculations and control strategies (two-phase flow makes calculations much more difficult, and it's harder to calculate correct positions for control blades (whose effects are highly localized) than it is to calculate the correct boron concentration (whose effects are smeared over the whole core)). However, now that computers are faster and us nuclear engineers no longer have the excuse of slow computers to hide behind, PWRs are looking to move away from relying on Boron concentration as the main form of control (the Westinghouse AP1000, specifically, relies much more on rod movement than the AP600), because of the cost of performing regular boron dilutions.

He's right that BWRs are simpler and cheaper - about half the moving parts.

> Like they wouldn't have boron near a nuclear reactor.

They probably do, the question is whether they can carry it to the site faster than the US navy, and whether the boron tanks have not been compromised.

The question is also, and I didn't make it clear enough in my original post:

"Wait, they need boron?"

Boron is a neutron absorber, and is therefore used to control the fission rate. You add boron when you're afraid of the fission rate getting out of control (e.g. Chernobyl). This is a legitimate concern here for two reasons:

1) A BWR is significantly more reactive (conducive to a high fission rate) at cold zero power than at hot full power, because cold water is a better neutron moderator than steam.

2) Xenon-135 and Samarium-149 are by-products of nuclear fission that have an effect on reactivity similar to boron. Their half-lives are on the order of hours, so when you crank down the fission rate, a couple hours later you also crank down the concentrations of Xe-135 and Sm-149, which, if you're not careful, can cause the reactor to go supercritical (and possibly prompt supercritical - a form of criticality in which everything happens approximately a thousand times faster - pretty much the worst-case reactivity excursion scenario) again a few hours after shutdown.

So, basically, you need enough negative reactivity from somewhere, either control rods or borated water, to counteract these two reactivity insertions. The Japanese reactor is almost certainly designed so that inserting all of the control rods into the core will kill any and all reactivity increases after shutdown.

What is most likely occurring is that there is very little nuclear fission inside the reactor right now. All the power inside the reactor is coming from decaying fission products, and it's probably on the order of kilowatts, it's just that the when the flow rate through the core drops from gallons per second to essentially nothing, a few kilowatts per cubic foot will get you pretty damn hot pretty damn quick. At this point some of the fuel rods have probably failed as well (if not melted), so the water in the reactor may be nastier than usual.

Unless the control rods have failed or are in the process of failing, I doubt that boron is even necessary for reactivity control, except as insurance. My educated guess is that their problems are entirely thermal- and containment-related, and that there is no danger of a reactivity accident, since a xenon transient or a condensation transient would have run its course by now, so the control rods probably have enough reactivity worth to keep the reactor subcritical indefinitely.

They don't want to control reactivity anymore, they likely want to kill the reactor core completely in order to not take any risks (not just to keep the core subcritical, but to avoid it simply melting as well, if the pumps did not work they don't have the waterflow they need, so the next option is to end the core fast). And for that, boron is the best tool.

Subcritical is "killed completely" from a nuclear standpoint - lowering the reactivity of an already subcritical reactor (e.g. adding neutron absorbers) does nothing but make the fission rate decrease faster. The fission rate already changes on short time scales, (e.g. orders of magnitude in minutes/hours during a scram). If the fission rate is already essentially zero, which it is in the Japanese reactor by now, it does nothing at all.

The heat inside the core is being produced by the decay of fission products, and there's absolutely nothing you can do to stop that except wait for enough half-lives that the activity slows down a bit.

Your general thesis, namely that as soon as they decide the reactor can't be salvaged, they'll dump in a bunch of boron just to be safe on the nuclear front, is correct, but where you're wrong is in assuming that adding a bunch of boron will help cool the reactor at all - it won't. All it will do is ensure that the reactor never goes critical again.

This is why spent fuel needs to sit in a pool of water for 5 years before anyone even considers moving it. Decay heat is serious shit, and it's not related to neutron physics at all.

EDIT: Probably time to get more specific about the term "reactivity". Reactivity is related to the "multiplication factor", which tells you how much bigger each generation of neutrons is than the last. It's zero when each subsequent generation is the same size - this is the normal operating state of a reactor, positive when each subsequent generation of neutrons is larger, and negative ... you get the idea.

If you've ever touched population dynamics in a differential equations class, you'll realize that this is a recipe for exponential growth and decay. Basically the time scale on which nuclear reactions proceed is "reactivity / mean neutron lifetime", with the caveat that if your reactivity is just above zero, neutron population growth is constrained by the longest-lived neutrons (it's like if every family has 2 kids, except for a couple hundredths of a percent, who have three, but put their third kids in cryostasis for a thousand years). In practice, this is how reactors are operated, because the mean neutron lifetime is _very_ small, so the worst thing that can happen is if your reactivity moves outside of this regime (this happened at Chernobyl). If your reactivity is negative, you are _always_ constrained by the longest-lived neutrons, but that's okay, because even this time scale is pretty short.

The upshot here is that if each successive generation of neutrons is smaller than the last, the number of neutrons (and hence things like fission rate that depend linearly on the number of neutrons) decays exponentially with a pretty short time constant.

The definition of "subcritical" is "having negative reactivity", so if the reactor is subcritical for any length of time, the fission rate will have exponentially decayed down to a tiny number.

Someone's under the impression that the US Navy is capable of doing whatever they want, whenever they want. In actuality, it took a very long time for US-Japanese relations to get to the point where the Navy would be able to station a nuclear-powered aircraft carrier, the USS George Washington, at the fleet activities base in Yokosuka. Prior to that, only the conventional carriers Midway, Independence and Kitty Hawk were allowed, and nuclear-powered carriers making a port of call were sometimes diverted due to tensions with the civilian population. Nuclear weapons themselves were banned and the forward carriers were weaponized at sea, in secret, from other vessels.

So, no, given the state of relations between the countries and the general lack of experience in civilian application, not to mention the improbability of possessing the resources, I'd highly doubt that the US Navy would be instrumental in assisting the qualified and experienced staff at a nuclear power plant located in a highly developed country such as Japan. I think the author was just throwing that in there to add some ominous weight to his argument.

So you are saying that Clinton lied? He was quoting Hillary Clinton, a source that should generally be above all suspicions.

The Bosnian airport shooting saga comes to mind. However in this case, there would seem to be little to be gained by fudging the issue.

Is one of the advantages of the BWR that you don't need a large pressure vessel? I remember news articles about JSW in Osaka being the only source for the huge forgings required for PWR pressure vessels, and only having the capacity for ~4 reactors per year.

Not really. BWRs need larger pressure vessels, PWRs need thicker ones.

I believe that JSW is the only place (also I think there's one in Germany, too) where you can forge a reactor pressure vessel in one piece, but that if you're willing to bolt two pieces together there are a few more options. It may be that this is more feasible for a BWR than a PWR, but I actually think that BWR pressure vessels are usually more expensive.

Where BWRs really save all the money is that you don't need to buy, maintain and replace pressurizers, steam generators, and a ton of piping and pumps that need to be rated for 2500 PSI, as well as a whole bunch of instrumentation for measuring and controlling boron dilution levels. You'll be paying more for engineering services, since all the calculations will take longer, but I can only imagine it's worth it, since, high-margin as engineering services are, they're chump change compared to how much the reactor costs.

Just an FYI, the additional reactors going in in South Texas are ABWR's, not BWR's.

I thought the new designs were classified as ALWRs - advanced light water reactors (light water meaning they use regular water).

However they're "classified", some of them (ABWR, ESBWR) work by boiling water inside the core at atmospheric pressure and using the steam to turn a turbine, and some of them (AP1000, EPR) work by heating pressurized (borated) water inside the core, which then transfers heat to other (pure) water at a steam generator, which (pure) steam is then used to turn a turbine.

Reply on news.ycombinator.com