[ajb]

That one seems like it should fall foul of thermodynamics, I guess. Just melting everything together probably increases entropy to the extent that it's at best like extracting elements from mining ore. Whereas before you do that, there is organisation and substances are more concentrated. Well, that's a bit hand-wavy - perhaps someone with actual knowledge of thermodynamics will comment.

I think what recyclers do currently is at least break everything into small pieces, some of which might have a decent concentration of something useful

[PaulDavisThe1st]

Nothing falls afoul of thermodynamics. This is not a closed system - you can inject as much energy into as you have available. Entropy and thermodynamics play no role here, but I would imagine that (a) the cost of the energy require (b) containment technology (c) what happens after you extract a given substance are/were all very involved in its failure.

This is already done with crude oil, and is called "cat cracking". You heat the crude oil until every component in it becomes gaseous (but still small-molecule) - the smaller the molecules they higher they rise up the "chimney", so you can siphon off particular components at particular heights.

[ajb]

I wasn't arguing that it is impossible, which as you say, is only true in a closed system. What I was saying was that since you wasted the organisation of the system, sorting it out again is going to take more energy than if you didn't. I think that's a consequence of the fact that you increased the entropy.

[landl0rd]

It'd work fine if you just ionize everything; easy to separate then. Just hilariously inefficient given the amount of energy required.

[Ekaros]

And even if you can collect those ions to relatively high purities. It is not often particularly useful. Most of the mass is probably carbon, silica, oxygen, hydrogen and so on. In the end not that much value there compared to virgin or other sources.