The bipolar transistor was created in 1947 and has essentially stayed the same in its basic concept for about 62 years. It consists of a sandwich of two similarly doped silicon volumes ( P or N ) or the “base” and “emitter” separated by a “collector” volume that has the opposite doping ( N or P respectively ). As the electronics industry has driven down the size of this most basic of all components into the nanoscale it came to a technical wall. It could no longer control the the amount of dopant ( one or two atoms a the nanoscale ) consistently enough to produce the component to component reproducibility it needed for commercial products. No matter how its technical experts tried the barrier could not be broken using the prevailing paradigm. Until the last few years that was still the case.
By introducing a change in paradigm that barrier appears to have been broken. It turns out at that at the nano scale adhering the “dopant” of the “collector” at the surface of the silicon is just as effective and promises better manufacturing control. If proven manufacturable the approach will allow transistors to shrink below current sizes. This clearly has major implications for the medical device field. You can read about it here. Controllable Molecular Modulation of Conductivity in Silicon-Based Devices
This is what I would call a “learning moment”. I believe that engineers applying new technology or even new science in a particular field need to be knowledgeable about the paths that scientists and engineers involved in innovation in other fields take to move through technical barriers. As engineers we can be tasked with or sometimes willingly take on the task to create new solutions in technical areas which appear to be intractable. I have found that at such a point in the innovation process there simply is no substitute for taking one step back and getting back to the basic principles involved in the science of the technology.