Hacker News

> What has changed here to make them competitive again?

The article has to be read carefully. "On Actinide's engineering estimates, a single Fortitude machine would provide roughly half the isotope-separation capacity of the U.S. government's current electromagnetic fleet."

The "U.S. government's current electromagnetic fleet" is tiny. Oak Ridge is building a modest plant.[1] Idaho has a benchtop-sized separator. That's what Actinide is comparing against. Not the rows of basketball court sized calutrons from WWII. So the announcement gives the impression of a larger operation than it really is.

> Speaking of which, whatever happened to laser enrichment? That was apparently very promising at some point?

That is a very good question. A company called Silex, and their subsidiary Global Laser Enrichment, has been trying to commercialize this for years.[3][4] Exactly how they do this is classified.[5]

There's another startup in this area, crawling along, underfunded, but building something.[6]

Meanwhile, URENCO continues to operate a centrifuge plant in New Mexico.[7] URENCO is a a European company, and seems to be the leader in centifuge technology. Units in France, Germany, the Netherlands, and the US.

I've been expecting something big to happen in the laser enrichment area since the 1990s, but it never has. This suggest that it either doesn't work very well or is being suppressed because it works too well.

[1] https://www.energy.gov/science/articles/doe-expands-stable-i...

[2] https://inldigitallibrary.inl.gov/content/uploads/50/2026/04...

[3] https://www.silex.com.au/

[4] https://www.gle-us.com/

[5] https://www.nrc.gov/docs/ML2304/ML23045A117.pdf

[6] https://laseristech.com/

[7] https://urencousa.com/

They talk about that in the article:

>"A centrifuge plant does one thing, costs billions, and takes years to stand up. Our machines cost a few hundred thousand dollars, produce material within months, deploy anywhere, and are able to be reconfigured in a matter of days to separate various isotopes as they are needed," said Robert Mendelsohn, co-founder and CTO of Actinide.

> What has changed here to make them competitive again?

Nothing. This is a company that specializes in making medical isotopes, which is something Calutrons are good for - you need high levels of enrichment in a single step, you don't need to process large quantities, and the energy consumption doesn't matter. Any talk of using it for reactor fuel production is pure PR spin.

Take these numbers with a grain of salt because I got them by chatting with AIs, but if you're producing electricity from HALEU, then centrifuges require reinvesting <1% of the output, whereas historical calutrons required 200% (useless) and modern technology could potentially bring that down to 10%.

So calutrons will always be less energy efficient than centrifuges, but if the capital cost and construction time is low enough, calutrons might still be economically viable.

>Under the Atomic Energy Act, all information not specifically declassified is classified as Restricted Data, whether it is privately or publicly held

https://en.wikipedia.org/wiki/Separation_of_isotopes_by_lase...

Well, for one thing, the only existing Yb-176 production facilities are in Russia. That means demand for a domestic supply is pretty high right now.

The other thing is that for a nuclear reactor, you need many kilograms of uranium. But for a Pluvicto patient, you need less than a gram of ytterbium.

Totally agree, just summarized what I found in the press release. Even boring engineering is honest work, and this is not boring, even if not cutting edge in my understanding. Nuclear operations have their special challenges which need special care, also from engineering side. Still I think this is less of an engineering news.

Anyways, I didn't mean to downplay it.

they can absolutely see uranium. It ionizes and thus get separated like many other things https://pmc.ncbi.nlm.nih.gov/articles/PMC7470433/

Calutrons are using the same principle of separating ions by their mass/charge ratio, just in a preparative scale (you want to collect what is separated) rather than analytical (you just want to know how much of what).

sounds like this is similar to doing basic paper chromatography and cutting out the strip with scissors just to get the desired substance, spectrometry as means to produce the material rather than means to analyze

Mass spec is about separating things by mass... exactly what enrichment requires.

A mass spec used as an instrument measures the components of a substance, this is just the same concept for actual separation of components not just to look at them

> It's just that we will drop bombs on anyone who tries.

I don't recall us bombing Pakistan, India, Israel, China, or North Korea.

> Nuclear nonproliferation relies on active enforcement.

The comedy hour at HN has arrived. See above ^^^

Nuclear nonproliferation relies on surveillance/monitoring and statecraft.

There are many, many steps before "drop bombs".

It's not difficult from a conceptual engineering perspective but it is difficult from an industrial scale perspective. It requires a tremendous amount of hardware and energy and precision equipment.

It's the same process either way, and the first few percent are the hardest (as it's super dilute so you need to handle a lot of material). It's much easier to go from 20% to 90%, than it is to get to 20%.

https://en.wikipedia.org/wiki/Separative_work_units

5% enrichment (regular reactor grade) is 2/3 the way to weapons grade already.

https://world-nuclear.org/information-library/nuclear-fuel-c...

Hence there technically being no enrichment cap for any country that signed the NPT and stuff (not to be confused with protection from US bombs or rogue states that never signed the NPT).

hah probably help cool things off too.

The logo. Definitely the logo. It both sucks you in and repels you at the same time. 'eye catching' to say the least, but so are laser pointers.

> it is hard to hide the scale of building and power usage from international watchdogs. Because nothing consumes tons of energy and doesn't output tons of product to sell except for nuclear material refinement for weapons.

Until datacenters