Most of it is hydrocarbons, which a sufficiently thorough decomposition process ought to be able to turn into CO2 and H2O. (And maybe HCl in the case of the chl

Most of it is hydrocarbons, which a sufficiently thorough decomposition process ought to be able to turn into CO2 and H2O. (And maybe HCl in the case of the chlorinated plastics? I guess that's neutralized with a base to make a salt?) I don't know a ton about this, but it looks like plasma gasification plants achieve something like this -- producing syngas (mostly H, CO), metal nodules, and a slag material (which seems to roughly be, "igneous rocks or sand, but man made"). I'm curious about the composition of the output products. But it seems to usually be net-energy-positive. This seems like a good intro: https://science.howstuffworks.com/environmental/energy/plasma-converter.htm https://science.howstuffworks.com/environmental/energy/plasm... Sci-fi: I wonder if there are methods from physics that could be used (even if not now cost effectively) to more thoroughly isolate the various elements rather than having the (it seems not totally understood?) chemical reaction of the various components of the plasma as it cools? Like a big industrial scale mass spectrometer? Like the cyclotron scheme to separate isotopes of uranium in the Manhattan Project? Then you could know, "all the lead is in this beam over here" and "all the mercury is in this beam over here", and so on... That'd be neat. Have some plasma physicists working on recycling instead of on fusion... (At this point I Google more...) Interesting, so here they separate copper from zinc using a plasma centrifuge: https://www.sciencedirect.com/science/article/abs/pii/0168583X92959046 https://www.sciencedirect.com/science/article/abs/pii/016858... And here they're looking (hypothetically) at using plasma separation to isolate the component elements in old NdFeB magnets, to recycle them: https://www.sciencedirect.com/science/article/abs/pii/S095965261830372X https://www.sciencedirect.com/science/article/abs/pii/S09596... Hmm...