Neutron Bytes

We can potentially reprocess used nuclear fuel, but that has so many problems.

The reactors are much more advanced designs, and the US and much of the world has been focused the opposite direction: on very simple, very safe, lower cost designs. Making a fast neutron reactor is just a different class of beast, and it's unclear where the appetite is for it. (Which is a shame.) Japan, China, India, and Russia are the only countries to have started new breeder reactors projects since Franc'es last reactor Superhenix in 1985.

Most breeder reactors use a reprocessing mechanism that is somewhat akin to the enrichment used for weapons. It's goal is to extract the good & enriched stuff & leave the rest of the fission product out (which still have hundreds of years to be dealt with). There's still justifiable hesitancy about developing such technology further. Even if we feel like we can use it safely in X Y and Z countries, once the technology becomes more proven (particularly if it can be shown to be cost effective) it's hard to imagine it won't become more available, and that has a real & scary risk.

Ask yourself how many of the above countries that started breeder reactors do we want making new nuclear weapons? And note how many have kicked out nuclear weapons inspectors (Russia, China).

One of my favorite candidate designs was the Integral Fast Reactor (https://en.wikipedia.org/wiki/Integral_fast_reactor), began in 1984 as a faster fast reactor which was considerably safer by…

Information about the authors is in the report, and on the web.

It's not possible to debunk every bit of nonsense written about nuclear energy. And often better not to dignify it by trying. Here goes anyway:

A reactor is a can filled with fuel. Per unit of energy liberated and modulo every other degree of freedom in the design, the size of the can has nothing to do with the amount of fuel or its disposition.

The size of the can does have something to do with the size of the can, per unit of energy liberated. That's because volume scales as radius cubed while surface area scales as radius squared, and the energy depends only on the volume.

So in that, the authors have a point. But the details they provide have little to do with it, and some are misleading or wrong. And why we should care, or how much we should care, about the amount of cans is not investigated.

The result is that there's nothing in this report anyone could use to inform a decision.

Edit: It has been disputed elsewhere...

https://www.powermag.com/small-modular-reactors-arent-difficult-nuclear-waste-generators/

https://www.anl.gov/article/argonne-releases-small-modular-reactor-waste-analysis-report

https://neutronbytes.com/2022/05/31/stanfords-questionable-study-on-spent-nuclear-fuel-for-smrs/

You're referring to the Green Freedom concept:

http://bioage.typepad.com/greencarcongress/docs/greenfreedom.pdf

Its authors badly underestimated how much new American nuclear reactors would cost.

We performed economic analyses on a partially optimized baseline concept based on a single Gen III PWR to provide power for the process. The analyses estimated a capital cost of $5.0 billion for an 18,400 bbl/day synthetic-gasoline plant and $4.6 billion for a 5,000 tonne/day methanol plant. Nuclear power accounts for more than 50% of the total plant capital investment.

If just over 50% of that cost is supposed to be nuclear plant, it would be $2.5 billion for a Gen III pressurized water reactor. ($2.9 billion now, accounting for a decade of inflation.) The just-abandoned VC Summer project was supposed to build two new Gen III PWRs in South Carolina. They pulled the plug because estimated cost of completion had spiraled up to $25 billion ($12.5 billion per reactor).

https://neutronbytes.com/2017/07/31/utilities-pull-the-plug-on-ap1000s-at-v-c-summer/

Good grief.

The exact opposite of what you write is true. Please inform yourself.

They are currently buying up all the nuclear they can get their hands on, reactivating shut down plants, investing both in startups and in existing players to build as much nuclear as they can:

https://www.forbes.com/sites/rrapier/2026/02/19/why-microsoft-and-amazon-are-turning-to-nuclear-power-for-ai/

https://neutronbytes.com/2025/11/20/terrapower-natrium-reactor-on-the-grid-by-2031/

https://www.world-nuclear-news.org/articles/google-to-fund-elementl-to-prepare-three-nuclear-power-sites

https://www.ans.org/news/2025-12-10/article-7608/vc-summer-update-mou-signed-with-brookfield/

https://www.world-nuclear-news.org/articles/palisades-receives-fuel-ahead-of-restart

https://www.msn.com/en-us/money/companies/westinghouse-and-fortum-to-deploy-advanced-ap1000-reactor-in-nordics/ar-AA1HlVv7

https://highlandscurrent.org/2026/03/07/lawler-announces-plan-to-rebuild-indian-point/

Actually it doesn't. There is no reason we could be burning todays "waste" in Thorium or other reactors. Reactor designs that also coast to a stop if their systems are interrupted vs. melting down like many of the current high pressure designs in use. There are many ways to produce nuclear power and there are plenty of designs that are far more efficient and safe than the currently widely used designs. Indeed China doesn't have to deal as much with irrational hype and is set to pass the rest of the western world: https://neutronbytes.com/2018/01/07/recent-developments-in-advanced-reactors-in-china-russia/

It's a combination of politics and hysteria that has painted nuclear into a corner, not science or reason.

The Japanese have been researching micro-reactors for a long time - hopefully they can make it work. I'd love to have a refrigerator sized thing in my basement that makes my power and removes my dependency on the grid.

Because (1) they are at or near capacity, (2) they are very costly to operate so those 'reasonable rates' are very hard to come by when using nuclear as power source. There are other, cheaper sources for power than nuclear that are more in line with what data center operators are willing to pay. There has been some talk of data centers with their own power plants. But usually that centers on fossil fuels rather than on nuclear.

There are some efforts in that direction though:

https://neutronbytes.com/2025/05/07/google-plans-three-600-mw-nuclear-projects-for-data-centers/

And I'm aware of a much smaller scale project in Europe besides this one and there are probably others.