Why has nuclear power been a flop? (2021)

Here in the UK it would be super convenient if cost-effective nuclear power plants could be built.

After all, to decarbonise the economy we need to replace petrol cars with EVs, and gas-fired central heating with heat pumps - which will mean a big increase in demand for electricity. A lot of that heat demand will be in winter, when solar output is at its lowest. And a lot of our existing power plants are past the end of their designed life.

The problem is we're trying to build one nuclear power plant, and it's going pretty badly.

It's an extra reactor on an existing site - that should make things easy, right? And it's being built by the French who operate loads of nuclear power plants - should be experienced specialists building a cookie-cutter replica of a proven design, right? And it's using private-sector finance with the government just guaranteeing an electricity purchase price - should remove the incentive for delays and cost over-runs that plague cost-plus contracts used in things like defence procurement, right?

But it turns out none of these tricks was able to overcome the curse of large infrastructure projects, and it's going to be delayed and expensive. And all the time the cost of the project has been going up, the cost of renewables has been falling.

I can understand the argument for building fewer and larger nuclear power plants - you only have to get one set of locals to support you, do all the impact assessment paperwork once and so on - but I can't help but wonder what would have happened if we'd instead set out to build ten nuclear power plants, each one tenth the size, so there weren't so many things being done for the first time in a generation.

Something I haven’t seen discussed in the nuclear debate is that NPPs become a high risk target in times of war.

Look at what’s happened in Ukraine regarding the Zaporizhzhia power plant. The IAEA has been crapping itself the entire time while both Ukraine and Russia have haphazardly been shooting and bombing around it. If Russia were forced out of their position, they could adopt a scorched earth policy and destroy it, potentially irradiating the area.

Whether Russia would actually do that doesn't even matter - just being a possibility allows them to take the area hostage much more easily.

I am well aware that it would not detonate like a nuclear bomb or like Chernobyl. It doesn’t have to, it could still contaminate a huge area and harm a lot of people.

I don't really think he understands why the safety paradigm is the way it is.

The problem is fear. Fear generated by individual incidents that have terrible local harmful effects. And the (perhaps inaccurate) perception that plants may still have issues. Plus the fairly unsolved waste storage problem.

Fukushima, 3 Mile Island and Chernobyl were existential for the nuclear industry. Like it or lump it, they made nuclear feel less safe.

When will we know that the new nuclear power plants are safe enough? Not for a decade or more.

Pretending that the industry isn't taking a rationale line based on the same tired stats about radiation is pointless. Humans are bad at judging risk, but your job as a power plant operator is to deal with the risk people perceive as well as the risk that actually exists.

(And the fact he links to LessWrong which is a site that overemphasises rational thinking to the extent it ignores the human condition says a lot to me).

> There is a great conflict between two of the most pressing problems of our time: poverty and climate change.

This wasn't true in 2021 and has been getting less true over time.

Not one mention of Solar or Wind in the whole article.

Nuclear Power is a complex thing, as any complex thing it require attention and commitment. Therefore it is more prone to destablization by a small number of individuals.

Russia have been quietly lobbying against nuclear in the EU for a long time (looking at you german Green Party).

The conclusions of the article is very good. I mostly agree with them. But there are two major flaws in the article:

1. “we need more like 25 TW”. That seems too high to me. Is that based on a direct conversion of the primary energy consumption today? The alternatives solutions needed for decarbonisation is generally much more energy efficient than the fossil fuel based solutions we use today. In addition, the energy used per capita to live a European lifestyle is decreasing year over year. And finally, the energy use will drop dramatically as we shift from primarily making stuff out of mined virgin materials to using recycled materials. For instance, commercial EV battery recycling is ramping up rapidly already today. Making virgin concrete is energy and CO2 intensive but that might not be a viable option for the long term anyway, as it requires sand that we’re also running out of. There’s already huge pressure to reduce the use of concrete.

It should also be said that 25TW or more of nuclear thermal power will contribute to thermal forcing of the planet that we probably can’t afford when the planet is already on the precipice of dangerous climate tipping points. The global warming effect of the thermal power plants we have today is already on the same order of magnitude as greenhouse gas emissions from airplanes.

2. I miss a discussion about the labour costs when it comes to nuclear. I think one of the reasons it was cheap to build in the past was access to cheap labour in the countries building the nuclear power plants. Changing demographics is an irreversible trend that will keep these labour costs high. Solar lets you outsource much of the labour costs of the production of the panels, and that production will probably end up being fully automated and then on-shored in the end anyway.

The article is too US-centric to a fault. Basically all project's costs after 1970s balloon to an extent. Japan's cost has risen. French, while only slightly, has risen. Same for India. Only S. Korea have seen construction prices fall, but notice the price bottoming out after the 1990s. It's only maintaining the price, not getting lower.

Regardless, these other countries didn't contribute that much to increase global nuclear power capacity. In both India and China nuclear doesn't even count as the majority in their low-carbon energy mix, which is still dwarfed by the massive deployment of coal. S. Korea is slacking on low-carbon sources in general. Japan has lost trust (rightfully) on nuclear and decided to reset their entire nuclear generation capacity. No country has expanded their energy grid with nuclear. Except France, who only did it because of the oil crisis.

> Or they were—costs are so high now that we don‘t even build plants anymore.

That's factually wrong. France has a program of nuclear building. We don't build a lot of plants in Europe, but we do.

Has it been a flop? They should ask France.

About the trains too...

The 1970s: When you have cheap coal and gas it would be financially stupid to invest in nuclear. Which is why most countries never did.

Today: when a single nuclear powerplant is estimated at 20 billion you calculate how many windmills you can build for that money. Ironically wind power is now the proven technology.

This article is missing the biggest elephant in the room: The unsolved problem of safe nuclear waste disposal. Everyone is in favor of nuclear power until the state decides to dispose the nuclear waste near your house: for a hundred thousand years.

Because of the existential threat they pose. They require a military to defend.

A lot of what is said about nuclear power fails to drill down to the bare essentials: how do you de-risk building nuclear power and how do you get the cost down. A lot of smug mumbo-jumbo written by industry insiders who have their heads too far up their own behinds tends to take attention away from the fact that, so far, we've gone about it in ways that don't make sense. You don't actually need to know anything about nuclear power to understand why the approach of building large nuclear power plants doesn't work. Some basic knowledge of financing and the statistical chances of "black swan events" when you stretch out projects in time suffices.

Or in short: if you don't understand the goals you must navigate towards, it doesn't matter how much of the technical implementation you know.

"How Big Things Get Done" by Flyvbjerg and Gardner zooms out a bit (well, a lot) and draws a relatively clear picture of what approximate shape the solution space has for nuclear. Yes, the book gives a high level perspective. And yes, transforming the industry so it can fit within the constraints outlined in the book will be expensive and time-consuming. But it is worthy of serious consideration.

https://www.goodreads.com/book/show/61327449-how-big-things-...

I tend to point out that the country where I live is uniquely suited to design and build reactors precisely because we don't have a nuclear industry. (We had a research reactor, but that's an entirely different matter and you would NOT want to involve the cowboys who ran that)

too expensive and never built on time, only countries where it's cheap are dictatorships where you can hide problems

Because nuclear competency is a strategic asset that the long march has been very successful in degrading.

Simple, people found out the reactors get retired after 50 years and its an absolute shit show. Totally not worth it.

Then there's the unforgiving accidents.

......

Ok, this is crazy. I stopped reading when the author says evacuating the people from Fukushima was unnecessary. Of course it is necessary. What good is a world where everyone is irradiated and dying of cancer ? We have enough already. This guy is either on Westinghouse’s payroll or some kind of pharmaceutical company about to release a new injectable therapy.

Fukushima proved that the nuclear industry in fact isn’t designing reactors safe ENOUGH and that Mother Nature is very difficult to predict. Everything needs redundancies and even higher levels of safety as a nuclear “incident” large enough could destroy humanity.