All those words. I want to highlight what I think it a theme there that is relevant to the framing of today's article:

Nuclear was undergoing a series of massive technical improvement as humanity discovered an exciting new technology. Then people got scared and strangled all innovation out of the nuclear industry some time in the 70s.

Heavy regulation makes it really, really expensive to experiment with new approaches. If we'd kept pushing on with nuclear, the existing nuclear technologies would probably have been obsoleted by now and we'd be using something different, cheaper and better with much more attractive cost structures. There is no point trying to improve a technology that governments are trying to regulate out of existence.

What's your favorite or most closely watched Gen IV reactor? Figure you'd be one to ask.

I think we should be building the best LWRs right now to decarbonize immediately: ABWR, APR-1400, AP1000. My favorite decarbonize rapidly at scale idea is to build a shipyard-based nuclear reactor gigafactory and mass-produce floating PWRs. (This was actually almost done in the 1970s in Jacksonville Florida. They had a manufacturing license from the NRC, the world's largest gantry crane installed and, everything [1]).

Assuming we do build 1000+ gigawatts soon, then we will need to look into nuclear fuel recycling with breeder reactors. The most popular Gen-IV concepts in this area are sodium-cooled fast reactors (SFR) and thorium molten salt breeder reactors (T-MSRs). We have 450 reactor-years of experience with SFRs and like 5 with MSRs. Many of the SFR years were not great, and the MSR tech performance has never been seen at industrial scale (though China is about to turn on a T-MSR really soon!)

So yeah I'm kind of a "decarbonize now with what we know 100% works" person. Focus on Gen-IV is fine for some people, but I think the industry and fans of the industry are way too overhyped on Gen-IV and not nearly hyped enough about Gen-III+. Japan can build gigawatt-class ABWRs in 36 months.

We had a ABWR licensed and ready to build at South Texas Project but we just let it sit there. What a sad sad thing. That's a billion carbon-free watts we are not using.

[1] https://whatisnuclear.com/blog/2020-01-26-offshore-power-sys...

As far as I know there is, after 70 years of research in many nations, not a single industrial satisfactorily working fast-breeder reactor. Is there one? https://en.wikipedia.org/wiki/Breeder_reactor#Development_an...

The most advanced one is Russia's BN-800 (now burning 60% MOX). Its planned big brother was cancelled "In 2015, after several minor delays, problems at the recently completed BN-800 indicated a redesign was needed. Construction of the BN-1200 was put on "indefinite hold", and Rosenergoatom has stated that no decision to continue will be made before 2019." https://en.wikipedia.org/wiki/BN-1200_reactor

This is not due to a lack of interest or funding because Russia launched a new project towards another architecture and builds a small reactor: https://en.wikipedia.org/wiki/BREST_(reactor)

So if nuclear is significantly more expensive that wind/solar like stated in the original article and faster to build (or are you disputing the numbers?), why would we not use the same money to instead build more renewables to get us there faster?

The classic answer is storage and on-demand production. Massive pumped storage is one partial solution the solar energy but I think the environmental impact is great enough that it would prevent it from ever being implemented in the United States.

I haven't seen the numbers that justify that wind/solar are cheaper than nuclear if you aim to fully de-carbonize (get rid of coal and gas completely). Massive amounts of storage would be needed, which is hard except in areas with a lot of hydro dams available.

Salt water + molten salt sounds like a real headache to maintain.

Existing plants are barely competitive with the worst wind/solar from the LCOE graphs I've seen.

There's NO WAY a new LWR will ever beat LCOE of solar + wind + battery. And that is current day prices that doesn't count forthcoming sodium ion storage, LFP, and solar/wind cost improvements.

And it won't come online for a decade at best, with inevitable massive cost overruns.

LWR/PWR is all the crap with waste, fuel rod reprocessing, only using select isotopes of uranium, and while I'm not an expert at nuclear plant economics and accounting, have tons of unpriced externalities.

I wouldn't support any LWR/PWR unless it had LFTR or other reactors built with it that could "online process" the waste. And if we had those, why bother with the huge shield domes?

I'll read your blog posts though carefully, maybe I'll change my mind. But a cursory look seems like it is caged too much in the baggage of the last century of nuclear, which IMO are just dead ends that won't practically advance nuclear for the next century.

As in, we need a couple decades of wind/solar to wipe clean the current slate of nuclear, from politics to actual installations.

Edit: The old designs are interesting, but they are huge and expensive.

What I look forward to in next gen nuclear is basically all of these:

- breeds (so you can use thorium and reprocess old spent waste to usable stuff)

- meltdown proof (LFTR has the melt plug and cooling tank that will decriticalize the liquid)

- scalable to a bunch of shipping containers or smaller (LFTR allegedly fit in a closet for the demo)

- 99% fuel use (per the docs, no idea if this is true or not)

LFTR promises those, if it can deliver due to containment degradation issues, who knows.

When I look at the gee-whiz LFTR presentations, what sticks out to me is the closed fuel cycle/complete usage, promise of breeding old waste to non-waste or new fuel, and so many other things that fix the errors of nuclear history.

The other thing that is missing is computer simulation. We should be able to develop far more usable designs with modern simulation software. We know the military has a lot of fission simulation software. Materials degradation and so many other things can be calculated far better than was available in the 1960s.

The final thing holding back nuclear is that despite a lot of their idiocy, the fact of the matter is that the "greenies" were correct about nuclear energy. It was poorly designed from a long term perspective, played fast and loose with waste, and many other considerations which probably derived from its military inception. The military only cares about the end result, and giving a 100,000 people cancer from spillage/meltdowns, dealing with the full cycle of waste, or actually maintaining safe operation.

Tepco, a japanese company with all the supposed strict adherence to process, was operating the reactor incompetently. Fukushima wasn't an outlier, it was an indictment of the large reactor design over the long haul.

LFTR design addresses SO MUCH of that. Far better meltdown protection, and full fuel use so there's practically no waste (not by the old solid fuel rod standards).

So if the nuclear industry doesn't reformulate around things that LFTR can do, then it will just fail in the long run again.

Targets to ban internal combustion engines will put an enormous strain on battery costs. Someone mentioned "massive pumped storage" but there's hardly any water where you'd want to build solar plants.

Industry and even the modest of living standards depends on energy density & reliability. So there's really no way to get completely off of fossil fuels without nuclear or literally starving people. That's your choice.

If Fukushima wasn't an outlier, then I wouldn't want to live anywhere close to a traditional reactor. Yet accident statistics & population trends don't really support that conclusion.

Awesome, what a great effort. Thank you.

Bookmarked to read on my next vacation!

This is great. Thank you so much.

What do you think about NuScales SMR reactor?

So the previous poster talked about how nuclear is so much more space efficient and the first picture I see on your waste page is a large field of temporary dry cask storage. So what is the size of the facility (including safety zones), is there a no fly zone around it and if yes how large is it?

Do you have news on Westinghouse? Is had filled bankruptcy, did they survive? how? Are they in maintenance mode or do they pursue evoltution of their models? China has bought their patent and is in the process of upscaling the original Westinghouse design, is westinghouse still collaborating with china? Have they abandoned their own models? It seems according to their website they are joining the small reactor fad..

Here's some recent news: China's announced that they're building 4 more Westinghouse AP1000 PWRs. They'll use the Chinese supply chain, but it's still a big win for W.

https://info.westinghousenuclear.com/news/four-westinghouse-...

I really wish we could get a few more orders in the USA so that we could leverage the learning and supply chain being so painfully built in Georgia for the AP1000s there.

We need to build hundreds of Nuclear Power Plants. Electricity should be so cheap, it’s not worth monitoring usage for homes.

The fact that this hasn’t happened is a testament and an indictment of dysfunction in our society.

I’ll turn into a single issue voter if any candidate for the President is pro-nuclear in a big way.

So do we just wave a magic wand and suddenly nuclear power does not cost anything. The article clearly laid out the cost and why even with deregulation LWRs are not cost competitive (while other technology is just not there yet). Or is your argument the government should even heavier subsidise nuclear to make electricity cost nothing? But why electricity and not e.g. housing?

Realistically the quickest way to get to the stage where you don't need to think about your electricity usage is put a solar installation on your roof + battery. Now you have to invest up front (and it might not even be a good investment), but it shows that nothing comes for free.

> do we just wave a magic wand

We have at least 6 magic wands: 1) if the VVER 1200 truly cost 1.4 billion to construct in 4 years, nuclear economics are saved. 2) building a lot of marine nuclear plants https://whatisnuclear.com/economics.html#economies-of-scale-... They require far less concrete (no need to protect against seisms) and much less redundancy because the cooler (water) will always be there (the ocean won't disappear) hence preventing radiations escalation in a passive way. 3) using disruptively simpler/cheaper models https://www.iaea.org/sites/default/files/publications/magazi... 4) using more complex but significantly more efficient models (I'm mainly talking about the underresearched supracritical reactors) 5) if SMRs are not a scam and can really have a fordism era 6) rationalizing et reconciloating regulations with cost effectiveness.

at current costs of fuel maybe power would be too cheap to meter, but is uranium really so abundant to meet demand? I thought I read there was only a couple hundred years supply, but maybe that was propaganda one way or another... I know it exists in seawater but is it easy to extract?

Yeah uranium can power the entire world for about as long as the sun will run. This requires breeder reactors, which were first proven in 1952 at the EBR-1 in Idaho.

https://whatisnuclear.com/blog/2020-10-28-nuclear-energy-is-...

A couple hundred years of proven reserves, with massive deposits known but not proven. It’s never been worth doing more than scratching the surface.

If uranium was used much more, there are economic amounts that are filterable from seawater even.

Seawater: "pumping the seawater to extract this uranium would need more energy than what could be produced with the recuperated uranium" Source: http://large.stanford.edu/courses/2017/ph241/jones-j2/docs/e...

I’ve usually heard it in reference to an add on process to other reasons we move seawater around - desalination, power plant cooling, etc.

very good point though on end cycle efficiencies!

Indeed, and there AFAIK were some "tandem" (desalination+uranium) projects, however nothing appeared therefore there may be severe practical constraints or technical hurdles. Maybe later(?)

Part of it I think is Uranium is easy to get economically right now, and has a pretty limited set of potential buyers (and the set isn’t growing very quickly).

Most Uranium mines (15000 worked claims in the Midwest in particular) no one even bothers with. Cool or not, if one of the 5 major mines decides to cut prices, you’d better be awfully efficient right now.

Uranium was cheap from the 1960 until now, bar ~3 years around 2007 (a 'Bubble'), however research towards ways to not depend on it (mostly towards breeder reactors) was very intense in many nations from the 1950's to 2000, because of an economic perspective (there is no clear reason for uranium prices to stay low, especially after a nuclear 'Renaissance' and given that it is tied to ore grade, which gets lower and lower), a social perspective (breeders and such reduce the amount of nuclear waste and risk associated to it), and for some nations also a strategic perspective (all existing uranium sellers live under a superpower).

The system needs voltage

Your semantics are strong.

for 4 AP1000 they are building ~20 CAP1000, westinghouse should never have sold they patents for a penny.. Also they are obscoleting westinghouse with their 4 planned CAP1400 https://en.wikipedia.org/wiki/List_of_commercial_nuclear_rea...