As I was washing some dried pancake batter off a bowl, I realized that there was a phase transition between the part of the batter that had congealed into masses (stuck to the surface) and the part that hadn’t and could simply be washed away with water. I think this is the percolation-threshold behavior that also governs the sol-gel phase transition: once the particles of flour are close enough together on average, instead of forming small agglomerations of particles, they form a continuous network of particles. The same thing happens in paints at the critical pigment volume concentration (CPVC), and similarly with latex paints when they’re diluted: if the drops in latex or pigment particles in paint fail to form a continuous network, they fall apart.
Naturally my mind ran to how this could be used for digital fabrication. This critical threshold is a point where a very small change in composition provokes a phase change between a sol and a gel, provoked, in the case of the pancake batter, by water evaporation. If you can provoke this change in the places you want to solidify, while keeping the rest of a solution as a sol, you can then just rinse away the sol to extract a green gel object, which can provide geometry for further processing. Until this point, the sol supports the gel weightlessly, since they have essentially the same density.
The new insight here is that the gelation stimulus can be made arbitrarily small to within the precision with which you can control the state of the near-critical sol.
A wide variety of gelation stimuli can be used.
The conventional approach with stereolithography 3-D printing is to use ultraviolet lasers or spatially modulated ultraviolet illumination through an LCD to produce free radicals, which locally initiate polymerization. Two-photon polymerization, where the number of free radicals produced is proportional to the square of the light intensity rather than directly to the light intensity because two photons are needed to initiate the reaction, is becoming popular. Any of these approaches can be used; the benefit of the near-critical sol is primarily in reducing the needed amount of light.
The spatially controlled addition of small amounts of material is another possibility, either a catalyst or a limiting reagent; this can be done either at the surface of a sol bath, similar to powder-bed printing, or throughout its volume with one or more movable nozzles, similar to FDM printing, though this poses the risk of the gelated material sticking to the nozzle instead of the workpiece. Electrolysis is one appealing way to locally deposit some ions, and electrolysis can also be used to initiate other reactions.
Worth special mention here is the possibility of using inkjet nozzles or gas jets scanned over the surface of a bath in order to precisely deposit the reagent.
Related to electrolysis, but different, is the possibility of plasma activation, where corona discharge around sharp points is used to stimulate some points on the surface of the sol but not others.
Another possibility is locally altering the temperature, either down (with a gas jet) or up (with a gas jet, flame, plasma, or laser or other light). Almost any resin system that can will polymerize on its own in enough time, such as commercial casting polyesters and epoxies, can also be convinced to polymerize much more rapidly with some heat, while a wide variety of bistable soluble gelling materials such as agar-agar will gelate upon cooling to a critical gelation temperature, but remain gelled up to a higher melting temperature.
If the sol has an extremely low vapor pressure, or an e-beam window can be moved very close to it, local gelation with an electron beam is also a possibility. This potentially provides finer spatial resolution than visible light.
In an upside-down printing vat like those used for LCD UV stereolithography resin polymerization printers, another possibility is to electrostimulate the bottom of the resin capacitively, for example by pushbroom-scanning a line of electrodes across the bottom of the vat, outside the protective membrane, while modulating an RF signal onto the electrodes. This would at least locally heat it up, which is enough to have the desired effect, but maybe even resistive heating through the membrane would be enough.
By substituting a cathode-ray tube for the vat-bottom membrane, it’s possible to stimulate the sol with light, X-rays, heat, or electron beams that pass through the glass. Ion beams can be substituted for electron beams, with the usual tradeoffs. Ideally the gel would polymerize a slight distance away from the glass to avoid mechanical forces on the glass; this would be least infeasible with the electron-beam approach.
The sol can be formulated with a wide variety of functional fillers which, among other things, reduce the amount of material that needs to be gelled to form a continuous network.
The above gelation stimuli can also be used without the sol, of course, as they are in conventional stereolithography.