John A. Venables
Cambridge University Press (2000)
One of the amusing features, but with serious side effects, is Moore's law itself. This is not a scientific law at all, but a simple historical observation. Over a substantial time period, semiconductor feature sizes have become smaller at a more or less constant rate when plotted on a log size versus linear time plot. Thus the extrapolation of this 'law' predicts a certain crunch time early in this new century, when feature size hits molecular dimensions, prompting the question "what happens then?"
While we are all busily worrying about such issues at specialist conferences, such as the Nanostructure Fabrication 2000 conference I attended in July 2000, we should be clear what is at stake here. We don't have to keep following Moore's 'law', but maybe our jobs do depend on doing so to a certain extent. The unprecedented improvement in device performance, embodied in Moore's 'trend', has built in an assumption with investors and the general public that it will go on for ever on its own accord. We have to use our collective wisdom to put them right gently, since soft landings are better than hard... However, that may not be the best way to look at it, because, as some parts of the electronic economy get tricky, other avenues open up with possibilities one had never thought of before. This makes it an exciting time to be in the field. These conferences have been held every two years since 1976. The next conference in the series will be held from June 13-18, 2008. All previous programs of this Nanofabrication series are available on the Gordon Research Conference site.
Meanwhile, the source of all such information is the International Sematech organization, and the idea of the Roadmap, which gets updated from time to time. By consulting these websites, one gets a strong impression of how many related industries try to march in step, in order to get the benefits of standardization and inter-operability. What is remarkable is that the whole industry does actually use the roadmap as a point of departure. For example, the March 2002 issue of the MRS Bulletin is devoted to Alternative Gate Dielectrics for Microelectronics (Wallace and Wilk, 2002). Their editorial references the roadmap in the first sentence, before introducing a series of articles aimed at the possible replacement of SiO2 as the gate oxide in ever shrinking CMOS devices.
More recent articles which should definitely be added to the list include Hasegawa et al.(1999) and Hasegawa (2000). Hasegawa's group in Tokyo is one of the relatively few which has studied thin film conductivity in these systems, in addition to structure and microstructure. Previous papers include work by a Sussex group (Iraji-Zad and Hardiman 1992), a Polish-German collaboration (e.g. Jalochowski et al. 1996), and the Hannover group (Henzler et al. 1999, Pfennigstorf et al. 2000). It may be that, with the wider range of techniques that are now available, this topic is due for a rebirth - or at least a student project to find out what has, and has not been done so far. Professors Henzler and Hasegawa organized a conference on 2D conductivity in surface states and monolayers from March 5-8, 2001, at the Physik-Zentrum Bad Honnef in Germany, and Hasegawa spoke at the Gordon Research Conference on Thin film and crystal growth mechanisms in July 1-6, 2001, in Williamstown, Massachusetts. We have just had the 2003 meeting in this series at Mount Holyoke College in South Hadley, Massachusetts from June 22-27.
New references for section 9.1
New references for section 9.2
On page 302, I give some general references on sensors, and particularly on gas-sensing materials. This, you understand, was instead of discussing the topic, not in addition to having discussed it. This topic is one of the many aspects of sensors, and can itself be subdivided further: for example, there is a recent review of gas sensors (Kohl 2001), which concentrates on 'homogeneous semiconductor gas sensors'. Most of these sensors rely on semiconducting oxides, where the subsurface oxygen vacancies are in equilibrium with the gas phase. These vacancies affect the resistivity, which is measured and interpreted. This paper gives a feel for the range of skills needed to make new developments: everything from quantum calculations for prediction and interpretation of near-surface and grain boundary reactions, to neural networks for automatic signal processing is discussed, and relevant web-sites are quoted.
New references for section 9.3
Book Contents, Introduction/ordering information, or my Home page.