3 ms·
Wow this looks cool. I didn't know there was a name for this but have had some crazy idea in my head for a few years now along similar lines. My thinking is, al
by seiferteric 6y ago
Wow this looks cool. I didn't know there was a name for this but have had some crazy idea in my head for a few years now along similar lines. My thinking is, all chemical interactions are electromagnetic in nature right? If you could control electromagnetic fields at a small enough scale and at high enough frequencies (i.e. at rates important in chemistry like infrared-x ray) could you not mediate chemical reactions "synthetically"? I imagine something like a "fabric" of electromagnetic transducers (antenna) and transistors along a surface that could both transmit and sense particle states. Obviously, this would require insane improvements in technology, like switching speeds and computation, but is it theoretically possible? Perhaps it could use a computing model similar to a cellular automata to distribute across the "fabric" in order to react fast enough to mediate the chemical reactions. Perhaps it would need to be superconductive in order to have the necessary sensitivity/efficiency? Ultimately could it be used to do things like act as a "perfect" catalyst for electrolysis for example...? or even build things up from the atomic level a la a replicator :) ?
- pharke 6y agoI wonder if you could do this at a fairly macro scale with lots of reactant. I'm thinking along the lines of an electromagnetic enzyme. The goal wouldn't be to build things up from the atomic level or pick and place atoms but to speed up naturally occurring reactions the way enzymes do. As long as you could line the fields up properly or switch them fast enough you would be able to nudge the reactions along. You'd need some way to pump the product out to the next step but you might be able to select for it with a field as well. Edit: after reading further into the paper I see that this is close to what they have in mind