If not for other purposes, you'd all be useful to bring chemistry light (and delight) to me
RF and microwaves being a mere means of heating, this was the way I believed to understand them. In addition to the low photon energy, the too short free flight time precludes selectivity: with EM wave periods of 10µs to 0.1ns, a shock every ps in a liquid or 0.1ns in air means that energy is redistributed before it accumulates. Consistently, we observe no resonances, and for instance the absorption by water is accurately modelled by the lossy orientation of water molecules plus ionic conduction.
Though, surprises do happen, and for instance the coherer used over a century ago isn't still well explainedhttp://en.wikipedia.org/wiki/Coherer
so if non-thermal effects were observed in chemical reactors, the next task would be to interpret them - physics as well is an experimental science...
A catalyst of palladium or metal wire
would more likely show non-thermal effects in microwaves than Pd(II) complexes.
Hot spots, big volumes: just take a lower frequency to obtain even heating. Call it radiofrequency instead of microwaves if you want. Already used in food processing to dry, cook... Polarization losses are bigger at 2.45GHz but ISM power is produced at any frequency: ~200kHz, 13.56MHz, 434MHz... as you wish.