A computer engineer believes that graphene will eventually replace silicon in commercial circuits, but not until current production methods reach a dead end.
According to James D. Meindl at the Georgia Institute of Technology, graphene will take centre stage once production processes reach the seven nanometre mark.
It is around this point that it is generally thought that CMOS semiconductor manufacturing hits a brick wall. According to Moore’s Law this is due to happen around 2024, and it is not until this point that graphene is seen as a viable alternative by Meindl.
Since the discovery of graphene by scotch tape wielding researchers at the Manchester University in 2004, there have been many advances in the use of the material and its revelatory properties. While there are some more short term uses for graphene it seems that it could still be some time before it takes a place at the heart of a new breed of processors.
According to EETimes graphene switches are one of the most likely uses in circuitry, Meindl believes, and he is currently working on 15 nanometre-wide ribbons which could give silicon a run for its money. Graphene transistors also offer a boost over conventional cooper interconnects, and would be ideal for manufacturing MEMS.
However, there are many hurdles in bringing graphene to mass production, as Meindl notes.
Although there have been successes in graphene integrated circuits, these have amounted to “less than a handful of transistors so far”. That is some way off the billions of transistors which are currently crammed onto processors.
There are concerns over the lack of ‘energy gap’ in graphene, with doubts raised about the ability to work in a CPU.
This means that a hybrid of graphene and silicon is more likely in the near future at least, with Samsung combining the two materials in memory designs.