http://www.ibm.com/smarterplanet/us/en/smarter_computing/article/atomic_scale_memory.html ~ don’t forget these guys. Recent stuff. The trick seems to have been to drill a hole just big enough to hold a hydrogen proton ~ but to do it so precisely that the “memory” atom is not interfered with by the containing atoms, nor does it interfere with them.
Interesting concept. I had sort of been bouncing the concept of a “photolithographic well” nanostructure off the walls of the inside of my head for a few days now. I think the “structure” has to be bigger than a single “H” (or “D”)....at least two and likely more, with means for ingress and egress of fuel and reactant products. A “sheet” of such structures fabricated on a high-thermal conductivity substrate should seriously boost heat transfer efficiency, and minimize size/power output. Probably would enhance controllability also.