Nobody ever actually explains what requirements would be reasonable to drop and still maintain safety. Because it turns out a whole lot of them are expensive and necessary.
Skimp on containment building? You get Chernobyl. Skimp on wall and backup power? Fukushima. Skimp on relief valve? You get TMI, which while overall harmless was still expensive.
Thanks to Fukushima it got even more expensive, because now they concluded that backup power must be available quickly in the event of a disaster. Which means the availability of the required parts close enough and the ability to bring them by helicopter, if I recall correctly.
Rather than relax safety requirements, it's better to design reactors to meet those requirements without backup power or human action. If you can throw the "off" switch and take a vacation, then safety gets fundamentally cheaper.
High-temperature reactors based on molten salt or TRISO can generally be air cooled, so a ship-based version 'just' needs to survive the worst-case hurricane or tsunami without sinking.
The proposition here is to throw a lot of time and money into a design that might not even work satisfactorily, and might not ever make any money even if it does.
The first is for practical reasons. I'm sure the theory works. But what if the practice is something like a pipe leaking, a liter of highly material spilling on the floor, and it being too contaminated to touch for a year? Yeah, that won't kill anyone, but won't be good business either. There's lots of possibilities for it working perfectly as promised, but just having various issues that make it non-viable anyway.
And I have a very hard time imagining a design that could compete with renewables. Because no matter how you slice it, nuclear has complex requirements that say, solar can just completely ignore. This is especially so when you start talking about not yet production designs, so maybe the first one will be built in 15 years from now.
Assuming that humanity actually has enough collective sense to stop burning fossil fuels and scrap the infrastructure, maintaining 99.999% grid uptime will be very challenging. There are theoretically possible ways for renewables to meet that challenge, but until that actually happens, I think nuclear can remain competitive.
On the other hand, if we continue to burn fossil fuels without paying for carbon emissions, then nuclear is economically doomed.
Nuclear already isn't competitive. Nobody is building anything new, they only maintain what was built, and even that is going down and not being replaced.
The problem is that nuclear is heavy on capital costs, and cheap renewables dig in very badly into its business model, making it even worse.
Also, nuclear isn't completely reliable either. Plants do go down. Spare capacity is still needed.
IMO the long term future is a mix of renewables supplemented by something like gas, batteries, other storage, overprovisioning and interconnection (eg, Europe).
Because we're still burning fossil fuels. It's hard to compete when your opponent is cheating.
> Also, nuclear isn't completely reliable either. Plants do go down. Spare capacity is still needed.
That's what multiple plants are for, ideally of heterogeneous design.
> something like gas
If gas (as in fossil methane) is part of our long-term future, then we are quite screwed. Maybe it's viable with carbon sequestration, but that's unproven technology. The other options you listed will be required to make renewables 99.999% reliable.
> Because we're still burning fossil fuels. It's hard to compete when your opponent is cheating.
No, the cheapest are renewables. Because say, pumping out solar cells by the billion is what mass manufacturing does best, and nuclear has nothing remotely comparable to that.
> If gas (as in fossil methane) is part of our long-term future, then we are quite screwed. Maybe it's viable with carbon sequestration, but that's unproven technology. The other options you listed will be required to make renewables 99.999% reliable.
You seem to be treating this as a moral issue. I see it as a practical one. Gas is just fine to fill whatever holes we might have. Yes, it's not clean. But using say, 10% fossil fuels rather than 60% would be a vast improvement. We can always improve further.
Renewables are cheap most of the time, so we should build lots of them. The interesting question is what to do when the weather is bad. Currently the answer is to burn fossil fuels. We need to stop doing that.
> You seem to be treating this as a moral issue.
No, it's a physics issue. Every gram of CO2 we put into the air, someone in the future needs to remove at great expense. There is no reward for solving 80% of the problem and cooking our civilization a bit slower.
> There is no reward for solving 80% of the problem and cooking our civilization a bit slower.
Heh. The older generations said "we'll be dead by then so it doesn't matter." Now younger generations are thinking along the lines of "If we can delay it until after I'm dead, that'll do."
Nonsense. Perfection is the enemy of "good enough". There are carbon sinks on the planet, for one. We don't need to convert ourselves into some species of tree elves magically in tune with nature. We just need to be a lot better than we are now, which is a good deal easier.
And yes, delaying the problem has a huge value. Because stuff takes time to build, and theoretical stuff takes a long time to develop. If you think nuclear or fusion have something to offer, then time is a huge boon to you. If we need to convert everything NOW, then there's not even a point of thinking about nuclear, it just can't be built fast enough. Let alone fusion.
Certainly I wouldn't say no to complete carbon neutrality world-wide. But given how things are it's clear that this is an unrealistic goal on a short/medium term. But that doesn't mean there aren't big improvements we can't make.
Fukushima is a great example because the takeaway should have been “these systems are already over-engineered”, instead of “we need more regulations”.
Recall that more than 10k people died from the tsunami, of which on the order of 100 from the reactors when you account for lifetime exposure to radiation. I’d call that safe enough. (Maybe a better analysis around cost of cleanup, I haven’t run those numbers.)
More generally, I think a big part of the cost overruns come from the regulatory environment changing mid-project, requiring construction to rework things, or waiting for inspections.
I think it’s a false dichotomy to say our only choices are strict and ever-increasing regulation, or Chernobyl. 10% less regulation would be positive, and you can bet the engineers working in the industry know which 10% is the most stupid.
> Fukushima is a great example because the takeaway should have been “these systems are already over-engineered”, instead of “we need more regulations”.
Absolutely not. There were a serious number of fuckups that were completely avoidable, had the proper rules been in place, and been followed, and had the right people been available. It would have been far cheap than dealing with the mess.
It was known the seawall was too low.
We have stories with heroic measures like workers collecting and hooking up car batteries to get equipment working. That's clearly an improvised effort, not a serious backup plan!
Apparently TEPCO refused US military generators, tried to send their own, and they got stuck in traffic.
There were multiple explosions, which resulted in broken equipment and evacuated workers.
Point is, it wasn't that some stuff got damaged in an ultimately harmless way, and everyone around just ran around like headless chickens for no good reason. Things indeed were serious, people were working around the clock to keep things together, and the situation was far from ideal. Without hard work it could have gotten considerably worse. So it absolutely wasn't over-engineered. Things held up, with a lot of effort and some good luck. They could have not worked out as well as they did.
And there were plenty instances in which the right measures and the right people could have saved a lot of trouble and made the whole thing go a whole lot smoother.
Skimp on containment building? You get Chernobyl. Skimp on wall and backup power? Fukushima. Skimp on relief valve? You get TMI, which while overall harmless was still expensive.
Thanks to Fukushima it got even more expensive, because now they concluded that backup power must be available quickly in the event of a disaster. Which means the availability of the required parts close enough and the ability to bring them by helicopter, if I recall correctly.