Electrolytic machining of glass may be feasible, using a sort of micro-scale variant of the chlor-alkali process in which the alkali-producing anode directs a stream of dilute near-boiling aqueous alkali against the glass, locally corroding its surface into soluble sodium silicate which is immediately washed away, carrying the ionic current toward the anode. This would be a very slow process, but may be useful for precisely shaping amorphous silicates.
As I understand it, the usual overall reaction of electrolysis of, for example, sodium bicarbonate works like this. The current is carried by Na+ ions migrating from the anode to the cathode (in industrial practice, through a cation-exchange membrane), while the HCO3- ions left behind at the anode give up their electrons to the anode and produce CO2, O2, and water: 4HCO3- -> 4CO2 + 4OH-; 4OH- -> O2 + 2H2O + 4e-. At the cathode, meanwhile, the Na+ ions aren’t actually taking the electrodes from it; instead, they’re taking hydroxyls from it, which the cathode is producing as 4H2O + 4e- -> 2H2 + 4OH-. So the overall reaction is apparently 4NaHCO3 + 2H2O -> 4NaOH + 4CO2 + O2 + 2H2. This contradicts Simon et al., which says it’s 2NaHCO3 + 2H2O -> 2NaOH + 2CO2 + O2 + 2H2, splitting twice as much water, but I guess you could totally have some arbitrary number of hydroxyl anions additionally traveling in the opposite direction through the cation-exchange membrane and thus totally split twice as much water, or ten times as much. Or maybe I miscalculated something above?
Anyway, so for glass ECM in this context you would have a tiny little “chlor-alkali” sodium bicarbonate cell whose cathode is a metal tube, like a hypodermic, nearly pressed up against the glass. Hot water is being pumped through the tube through a membrane, behind which you have a tiny pressure chamber, made of an insulator, full of ridiculously positively charged water. The water is pumped into it from where it’s bubbling out oxygen and carbon dioxide after the anode destroys bicarbonate ions and gives the water that strong positive charge. The cathode neutralizes this positive charge and produces hydroxyl ions and hydrogen gas, which then come out its tip tens of microns away from the glass. After the alkaline water impinges on the glass it spreads out into a rapid current of some kind of buffer or acid, which neutralizes it and keeps it from corroding the rest of the glass. Maybe the acid is also produced by a similar kind of electrolytic cell, from sodium sulfate or something, so you don’t need an acid reservoir. Maybe even the same electrolytic cell.
Now, how plausible is this? I was originally thinking it would require a totally implausible degree of macroscopic charge separation, with unrealistically large coulombic forces from the highly charged water on nearby objects. But of course the reason water is a good solvent for things like sodium ions is precisely that it solvates them with its high permittivity, screening all but a tiny amount of their electric field. And nickel, at least, can withstand contact with hot caustic solutions. You might need to supplement the needle cathode with a high-surface-area reticulated non-graphitizable carbon electrode, and build the anode in the same way, to get high enough current. You might need to use a high overvoltage and thus get poor current efficiency. But it seems clearly feasible.
What if you ditch all the complicated plumbing and just immerse the glass in the hot sodium bicarbonate solution or whatever, and for your cathode use a solid metal needle insulated except for the tip? What happens if you pump a pulse of electrons into the cathode? Won’t they produce hydroxyls that attract more Na+ to the area and enable it to etch the glass there? Won’t they repel HCO3- ions from the area?
Well, I don’t know. Maybe? It seems like it ought to work.
We might be talking about microscopic effects, though; some researchers ran some experiments on glass recycling in 02010 and it took them about 15 days to dissolve 25% of their glass samples, pulverized to 250-800 microns, in 1-molar NaOH at 70°. KOH was somewhat less effective, 5-M NaOH was slightly more effective, using a solution at only 50° was much less effective, and smaller particles sizes were significantly more effective, but overall we’re talking about rates in the ballpark of 30 microns a day or 400 picometers per second.
Yet I’ve seen demonstrations of dissolving silica gel in 30% aqueous NaOH that converted amorphous silica gel to sodium silicate in a few hours. The silica gel in question didn’t have any alkaline earth elements in it, I suppose, which Kouassi et al. mentioned as a factor that slowed dissolution for them. And the demonstrations I’ve seen where lye dissolved glass in seconds or minutes were with molten anhydrous lye, not aqueous lye, which would be something like “25 molar” and is also at something like 350° instead of 70° or 100°.