Hacker News Reader

Comment by abraxas | Hacker News Reader

In a way the story of nanofabrication seems like a tale of disappointment. Many decades after Feynman's "Plenty of room at the bottom" or Drexler's "Engines of creation" and we have very little to show in a way of progress. What happened? Why were knowledgeable minds like Feynman too optimistic about our ability to make this happen?

View on HN

Its a scaling problem. You can do a lot of cool things with fib/ebip etc. but the process itself is very slow and requires patience unless you want to destroy your structure via codeposits or other side effects. Kinda like the AI field back in the 80s the hardware still isn't there to really make use of the ideas.

> we have very little to show in a way of progress. What happened?

Our semiconductors have had features below 100 nm for a while (actual features, not just process node names), so that's been wildly successful.

Why nanofabrication hasn't been as commercially successful outside of semiconductors is a much harder question to answer.

Yes, of course the etching on silicon process has been refined to a level nobody thought possible. But this is more like a CNC process at a tiny level. What we don't have is additive manufacturing at nano scale. The nano assembler that Drexler and Feynman thought were possible is not panning out.

I'll have to look up that idea of the "nano assembler" that you're referring, but I'll just mention that there is significant additive manufacturing in a modern semiconductor GAAFET node, with the entire transistor being grown.

This video is a good summary of the main steps involved: https://youtu.be/xaKyDrWfHes

It's certainly not as powerful or universal as 3D printing and it might not even be "additive manufacturing" proper, but it is a lot more than just etching/subtractive manufacturing.

Talk's cheap.

The feats we've archived in miniaturisation of logic and memory is already mind-blowing, in my opinion.

Reply on news.ycombinator.com