This reminds me about one time when I was buying surplus equipment there was an ion milling machine in mint condition and it went for $100 to the only scientist who had a lab to put it in and wanted to move the very heavy base unit.

I could have sworn I read that Intel was already using depleted silicon more than a decade ago and they made it sound like it was a standard for high frequency chips. Is that wrong?

I’m getting a sense the article is playing fast and loose with the state of the state/art to aggrandize their accomplishments. I distrust the author.

How do they get this beam of pure Si-28 though? Do they just end up having to centrifuge a smaller amount, or can they do some sort of mass spectrometry trick?

I remember learning that germanium can be highly purified, so I looked up its itsotopes. What a mess, 5 isotopes with three prominent ones.

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

How much overlap is there between isotope separation technology? The article mentions using centrifuges to separate silicon-28 from -29. Could the same equipment be used to separate uranium-235 from -238?

Is this a question of difficulty? Of scale? Or does Intel become a nuclear proliferation risk?

    The researchers managed to produce samples with fewer than 3 parts per million of
    ^29Si—some 1/10,000 less ^29Si impurities than what exists in natural silicon.

Is 1/10000th less correct?

That doesn't sound like a worthwhile figure?

This is not scalable at all. It's too fragile, the area purified is tiny and rough. A classic academic solution that can only happen in the lab.