The field effect transistor is the workhorse of the consumer electronics industry. Carved into microchips in the billions, these devices beaver away, more or less unnoticed, in practically every home, office, and laboratory in the developed world.
And yet there is a perennial problem with field effect transistors that keeps chip designers awake at night—how to make them ever smaller and thereby keep up the relentless pace of Moore’s Law.
Field effect transistors are already so tiny that making them any smaller leads to a multitude of challenges that are by no means easy to solve. The components of today’s state-of-the-art field effect transistors are just a few nanometers in length—that’s just a few atomic silicon layers thick.
These silicon layers have to be doped with other atoms—just a handful will do the trick in such small components. And therein lies the problem. Even small random variation in the number of dopant atoms in semiconductor components can have a huge effect on the behavior of the transistor. How to control these variations during manufacture is by no means clear. Then there is the physical problem of making a device with three terminals even smaller.
So chip designers would dearly love to have another device they could rely on to build chips that are more densely packed with ever smaller components.
Today, Jason Marmon at University of North Carolina in Charlotte and a few pals unveil just such a device in the form of a light effect transistor. This is essentially a wire that conducts when it is bathed in light and insulates when it is dark. In other words, it is a switch modulated by light. The team says its new device is simpler than a field effect transistor and does not rely on dopant atoms, so it can be made smaller and thereby continue Moore’s law.
To read more, click here.