Comment by kurlberg | Hacker News Reader

Comment by kurlberg | Hacker News Reader

Parent Full thread

A long time ago, as part of a "Industry Study Visits"-course at university, we went to visit Asea-Atom and were told about the "PIUS" reactor ("Process Inherent Ultimate Safety"). Supposedly the construction was "passively safe" (i.e., breakdown of cooling pumps etc would not matter.) The idea was to have the reactor encased in an open pipe contained in a larger (closed) tank with borated water. Using stable stratification, the non-borated water was more or less kept contained in the inner pipe using density locks; if something would go wrong the borated water would get sucked in and and shut the reactor down.

The engineers were very optimistic about selling this even as small municipal thermal power plants (for heating water) very close to city centers. The idea of passive safety is quite beatiful, but it seems no PIUS reactors were produced. Rosatom seems to have had some success.

More info about PIUS can be found at

https://www.iaea.org/sites/default/files/publications/magazi...

https://www.euronuclear.org/e-news/e-news-17/nps-kth.htm

I think the issue with safe reactor designs is economic and regulatory / political, instead of intentional.

Nuclear power was invented. A few nuclear accidents happen. The solution was to put more regulation in place. Regulation increases costs and limits reactor designs (large plants, conservative / pre-existing design).

End result? It's impossible to build "safer" reactors, because they're economically uncompetitive under the regulatory burden designed for older reactors. And the regulations never change because no one builds anything but old-style reactors.

I understand the DoE and other similar organizations have prototype programs, but from results evaluation they seem to have gaps in the path to production.

Essentially, we're letting 1960s nuclear fear of 1960s reactor designs write our regulations instead of science.

The tail risk on nuclear is so bad that innovation which merely reduces the cost of energy isn't nearly compelling enough. Energy costs (either direct, or indirect via carbon taxes or other pricing in of externalities of burning stuff) need to rise significantly before there will be a real driver in the system.

If you want nuclear, you want punitive carbon taxes, or carbon trading with low caps.

> If you want nuclear, you want punitive carbon taxes

I think this is the crux of the issue. Carbon-emitting plants get a free pass on safety and long-term environmental costs, while for nuclear all those things are examined with a fine-tooth comb, then compared to the cost of alternatives without those things.

In any regulatory regime that punished carbon emission the same way it did nuclear waste, nuclear contamination, etc, then coal plants would be unthinkable.

I was going to say "then nuclear would be obvious", but I'll hold judgement there because I don't know how that would compare to other non-carbon-emitting sources (hydro, solar, etc).

Carbon-emitting plants are safe. They cause a predictable number of deaths every year. That's a cost, similar to the environmental costs.

That doesn't mean they are less bad than nuclear plants but the claim that they are unsafe is just wrong.

I'm not in the field, but it seems like the risk calculation itself could be used as a driver.

By shifting the regulatory cost from focusing on lesser risks to catastrophic ones, you could make fundamentally safer reactor designs economically competitive while still retaining the primary political goals (avoidance of major event).

I'm concerned but not running around screaming if a reactor leaks tritium. I'm very much the latter if a reactor goes prompt critical.

It's my understanding governments have considered light-touch regulation of nuclear safety with the policy "any nuclear power plant that can get insurance to cover a $100 billion+ meltdown cost can be built" and the result is nobody can provide such insurance, so no plants get built.

From an actuarial standpoint, I wouldn't go near nuclear with a 10' pole either. Simply because the population size is too small.

In the same way I wouldn't offer SR-71 insurance.

That feels like a dodge on the part of regulators.

There's an actual actuarial analysis online somewhere, though I can't remember if it's in German, and the problem is not population size, but a nearly unbounded risk. Take a look at Fukushima and think about what would have happened in terms of cost had the wind blown in another direction.

Maybe this?

https://www.versicherungsforen.at/portal/media/forschung/stu...

Thanks for the pointer! Will be an interesting read.

The president of Exelon has stated that for new nuclear to compete with natural gas combined cycle plants in the US, the CO2 tax would have to be in the range of $300-400/tonne. This is very high, and many cheaper alternatives would become available first at lower values of that tax rate.

The political will to force through long term waste storage is also missing: https://en.wikipedia.org/wiki/Yucca_Mountain_nuclear_waste_r...

This is fairly cheap, a few hundred million, and in terms of safety it's a no brainer, because the US already has nuclear waste that needs long-term storage.

If this can't be solved, it seems almost pointless to dream of new reactors.

Why solve it? Spent fuel can be stored on the surface, safely and cheaply, in dry casks. And stored on the surface the decay heat dissipates into the atmosphere, rather than building up underground.

I have to wonder why burial was being pushed so hard, so early. Maybe they were worried surface repositories would be targets for ground burst nukes in WW3?

Agree. The only thing I see changing this is possibly reactors developed in another country that solve some of the fundamental problems, and becoming successful there. China comes to mind, but there's so little transparency its hard to tell whats real there.

The Chinese are doing something vaguely similar, the DHR-400, which is a low-pressure pool-type reactor for district heating. https://en.wikipedia.org/wiki/China_National_Nuclear_Corpora...

The usual problem is not shutting down the reactor: both TMI and Fukushima was properly shutdown. Important task is cooling down the core (and SFP) after that.

Left on its own, halted thermal reactor generates hundreds MW of heat after shutdown. That heat destroys cladding of fuel, generates explosive Hydrogen and causes meltdowns of former core.

It is believed that ATF, core catchers and electricity-independent passive cooling will reduce risk of this types of incidents, commonly called LOCA - Loss of coolant

Fukishima unit 1 had an Isolation Condenser, which is the sort of passive safety system you're talking about. It required no electricity, merely that some valves be open. Supposedly after the earthquake and before the tsunami, the operators opted to close at least one of those valves for reasons that are a little hazy, but might have had something to do with not cooling the reactor too fast. After the tsunami, they couldn't electrically control the valves, and for some reason there was ambiguity about the state of the valves, and no one went to physically check them and open them until it was too late and the fuel had melted.

Sounds a bit like a Pythagorean cup, just with the liquid overflowing the “lip” due to heat rather than volume.

Reply on news.ycombinator.com