Ziggurat
Penultimate Amazing
- Joined
- Jun 19, 2003
- Messages
- 64,058
Yes. He's imagining the funnels are aligned such that molecules striking from the wider side will push the funnel open farther, allowing entry. But molecules striking from the narrow side will not do so. Hence the holes being bigger and smaller. This becomes the pawl from the ratchet
OK, I didn't really pay enough attention to that part. Of course, it still won't work: if the vibrational modes of cone expansion are far lower energy than the thermal energy of the gas, then the cones will essentially be flapping in the breeze, and won't make any difference to gas diffusion. If those modes are much higher energy, then the cones will be frozen stiff, and become irrelevant. And if they're comparable, well, when the cones thermally equilibrate with the gas, then they'll be vibrating on their own with enough energy to actually bounce gas molecules out of the cone that would have otherwise passed through. And the odds that a gas molecule will hit the cone and expand it to get through (ie, create an excitation) will be the same as the odds that a gas molecule will hit the cone and get bounced back (ie, absorb an excitation). And we know without bothering to do detailed analysis that these will have to be the same because if they weren't, there would be a net transfer of heat, meaning that the cones and gas weren't at thermal equilibrium. It's true that you can derive work from thermal gradients, but only as long as that gradient lasts, which requires performing work to maintain. So no free lunch.