The vacuum-transistor in this article sounds incredible, if it can overcome some of its basic problems. http://spectrum.ieee.org/semiconduc...he-vacuum-transistor-a-device-made-of-nothing I wonder if you could contain a vacuum inside something other than glass--something less fragile. And, weight is still a problem, isn't it? Water is also said to be a potentially fast candidate for computer-brains, but you would still have the problems of sturdy containers and weight, I think. Hexagonal forms of water molecule may potentially be able to retain data in computers. The hydrogen bonding seems to be the key. http://www.aquatechnology.net/electromagneticeffects.html The key may be using a ruthenium ion. Ruthenium absorbs solar energy and is used in some superconductors. http://witcombe.sbc.edu/water/chemistryelectrolysis.html http://www.brightsurf.com/news/headlines/32014/Nano-layer_of_ruthenium_stabilizes_magnetic_sensors.html
Read. The proposed device does not have to actually use a vacuum due to the very short "mean path." I haven't looked at the sources you linked, so I don't know if water or some other medium might be better than silicon semiconductors. However, the big argument for the "vacuum" design is that diffusion is dramatically reduced. The suggested terahertz sensing applications alone would be world changing.