I used to claum a killowatt / cm2 at 10 nm spacung between the buckyballs for 10^12 diodes / cm2 but backed off to 100 watts / cm2 / 35 nm spacing / 10^11 diodes / cm2. 35 nm spacing gives the reverse bias depletion zone room to grow. I don't know how big the reverse bias depletion zone will be. A little bit of crowding may pay off because a lot of inefficient diodes beats a lot fewer very efficient diodes. Bringing a hundred watts / cm2 of heat to the diodes is hard enough. Making diodes by the billions by sprinkeling buckyballs seems to me cheap and easy. Nature has no guile so the simple coherence of having all the diodes right side up for consistant algnment and metalized into massive parallel is enough to rectify and aggregate the Johnson Noise.
It probably was Mr. Cox who asked me to convince him in a short statement that my device would work (I can look that up)... My last computer ate all its email...
Dr. Crowe at VDI will look up a diode array patch with a lot of small anodes for me when he returns from vacation ~Aug 1.
Aloha, Charlie
