You may have heard of Moore's law, which dictates that, the number of transistors in an integrated circuit doubles every 12 to 24 months. This has held true for about 40 years now, but the current lithographic technology has physical limits when it comes to making things smaller, and the semiconductor industry, which often refers to the collection of these as the "red brick wall", thinks that the wall will be hit in around fifteen years (the best resource for information on these limits is the International Technical Road map on Semiconductors - see http://public.itrs.net/). At that point a new technology will have to take over, and nanotechnology offers a variety of potentially viable options.

The total potential for nanotechnology in electronics has been estimated to be about $300 billion per year within 10 years, and another $300 billion per year for global integrated circuit sales (R. Doering, “Societal Implications of Scaling to Nanoelectronics,” 2001). But it's actually much harder to predict the commercially successful technologies in the world of electronics than in the world of materials.

The assumption that continually increasing processing power will automatically slot into a computer hardware Market that continues to grow at the rate it has done historically, is not necessarily sound. Most of the growth over the last decade has been driven by personal computers and some argue that this market is nearing saturation. Certainly there will be other applications. Increasing the intelligence of computers, and giving them the capability to interact verbally in a sophisticated manner, would certainly bring benefits, but increasing hardware capabilities is only half the story, with the biggest challenges being designing the software. Another area predicted to see major growth, is ubiquitous computing, whereby Processors start to be incorporated in all manner of objects around us, which then communicate with each other and us. However, the requirements here are for relatively simple processors and in many cases there is no need for them to be particularly small either. Cost improvements remain a critical factor and it is the correlation of increasing transistor density with a reduction in cost per transistor that has probably kept Moore's law on track for so long. This relationship need not continue, however and several new Approaches, which aren't even nanoscale, hold promise of creating simple circuits cheaply, such as using arrays of MEMS-based micro mirrors to build custom circuits, or the use of ink jet printers to churn out simple ones (interestingly, nanoparticulates figure in the potential of this probably near-market technology). These approaches also offer the ability to create low runs of circuits, or even one-off bespoke designs, at low prices, whereas photo lithographic approaches need massive production runs to achieve economies of scale. Soft lithography, too, offers cheap, micro scale, circuitry and is being pursued in the creation of flexible displays. These technologies could take a share of the existing semiconductor market and certainly future markets such as electronic tagging of goods or the processors required for ubiquitous computing.

High-production-run electronics will continue to be dominated by photo lithographic approaches for years to come, with the advent of the molecular nanotechnologies that could dramatically improve the power and density of processors while competing with photo lithography on cost still way off the radar for the investment community. The reason for this is the challenge of using such approaches to create the complex structures required for processors, an obstacle that doesn't apply to data storage, as we will see later. Soft lithography and nanoimprinting, however, are showing promise of coming into investment range in the near future. A company has already been formed to develop one flavor of soft lithography, the step and flash approach, for nanoelectronics and Stephen Chou of Princeton recently developed a variant of his nanoimprinting approach (already used to make commercially available sub wavelength optical components) to make nanoscale structures by melting silicon.