This is just something that happened to occur to me this morning, while brewing a pot of tea, of all things.
Ferric oxalate (Fe2(C2O4)3), familiar to those who have dabbled in platinum/palladium printing, is a photo-sensitive chemical that reduces to ferrous oxalate (FeC2O4) upon exposure to ultraviolet light. If ferrous oxalate were “developed” with gallic acid or a suitable, soluble gallate salt, then the ferrous gallate (which is the key component of iron-gall ink) formed could create a dark, contrasty image.
I should point out that all this is conjecture, and practical experimentation would be needed to prove the theory.
A wee while ago (it must be about two years, now), I became interested in diffraction optics, particularly zone plates (aka: Fresnel plates, after Augustin-Jean Fresnel). After pursuing the technology and its properties for awhile, I filed my notes it away in my “curious_projects.txt” file, in order to concentrate on Serious Business.
Zone plates utilize the optical properties of diffraction and interference to converge (focus) incident light, and can be used to create photographic images. They consist of concentric, alternating opaque and transparent rings (“Fresnel zones”) of equal area. The more rings, the sharper and brighter the image. “Binary” zone plates have an interesting property, in that they create images with soft focus and interesting, luminescent halos (reminiscent the halation of old, thick, glass plates examples). Introducing a smooth, sinusoidal transition from transparent to opaque zones (as opposed to a hard edge) eliminates soft focus, creating a sharper image. A zone-plate objective for your SLR can be as easy to make as a good, sharp pinhole (instructions), if you can acquire a spare body cap and some contrasty film with a nice, clear base.
It occurred to me at the time that one of the big drawbacks of zone plate optics was that chromatic aberration could be pronounced and impossible to correct. I suspected that the solution could lie with multiple, RGB-filtered zone plates focussing different wavelengths. Now coming back to the problem (I have plenty of spare time, just now), it occurs to me that coaxial C-M-Y zone plates could present a more elegant solution. The question now is how to prototype a sandwich of coloured zone plates, precisely focussed for three primary wavelengths. I’m not sure I can even get good slide film, these days.