Introduction

Computer integrated circuits (ICs), also known as computer “chips”, seem to be getting faster and faster, as components are made smaller and smaller through miniaturisation. In 1964, Gordon Moore discovered the speed of computer chips (i.e. the number of transistors that a given area of IC could hold), had doubled every year since they had been invented [5]. He also then predicted that this trend would continue, with the speed of chips doubling roughly every 18 months. Up until now, this has proved to be true, but for how much longer can we sustain this growth? It is predicted that in the next 10 to 15 years, we will meet a physical barrier to the Moore exponential growth theory. This is because of two major problems:

Resistance

Computer chips consist of silicon and copper connections [1]. Although the resistivity of Copper is constant (3×10^-8 ohm m), the resistance of a given piece will differ according to its size and shape, or more precisely, its length and cross-sectional area:

e.g. a piece of copper, diameter 1 mm, length 1 cm:

BUT a piece of copper, diameter 1 nm, length 10 nm:

With miniaturisation, R increases drastically, so the voltage input required also increases (due to Ohm's law). This leads to high currents which are inefficient and generate a lot of wasted heat.

Manufacturing Process

There is a limit to how small components can be made [4]. Current manufacturing processes mean the smallest components that can be created are 200 nm in size (in comparison, carbon nanotubes are roughly 1 nm in diameter). It would cost billions of pounds to completely redesign manufacturing processes in order to be able to produce smaller conventional components, but with nanotubes, it is much easier to create molecular-sized components.

 

< Previous | Next >