IMPORTANT LEGAL NOTICE - The information on this site is subject to a disclaimer and a copyright notice.
|
|
|
|
|
 |
  |
||||
 |
 |
||||
 |
Nanotechnology |
||||
Blurring the Boundaries at the Atomic Scale |
|||||
The microelectronics industry's race for miniaturisation is fast approaching the realm of nanotechnology. A new, more multidisciplinary paradigm is required, with major implications for fields as diverse as microelectronics, quantum physics and biology.
Just what is nanotechnology? Pick up a paperback at the next bookshop you visit and you'll read about tiny machines patrolling your blood vessels, constantly repairing the slings and arrows of outrageous fortune. Other books will warn of the `Star Trek scenario', where rampaging nanomachines turn the planet into grey sludge, or summon up a vision of perfectly built structures - vehicles, or perhaps even spacecraft - emerging like butterflies from a soup of tiny assembling machines, programmed using the industrial equivalent of DNA. But what do the scientists think? "Nanotechnology has often been defined as the science of fabricating, characterising and using structures from the atomic scale up to around 100 nanometres," says Dr Marc Van Rossum, head of Advanced Materials and Nanoelectronics research at IMEC, the Belgian microelectronics R&D institute. "But this definition is simply not much use because it embraces so many fields - from electronics and physics, through biology and chemistry and on to mechanical engineering. And who set this 100 nanometre limit?"
What's in a Name? Dr Rossum was speaking at an industrial workshop on nanotechnology organised by PHANTOMS, a network supported by the Esprit programme he chairs. It was held at IBM's Zurich Research Laboratory, where Gerd Binnig and Heinrich Rohrer invented the scanning tunnelling microscope (STM). The STM is to nanotechnology what the telescope was to astronomy, and won the pair the 1986 Nobel Prize for Physics. Almost every speaker at the workshop had a different angle on nanotechnology. For Harold Craighead of Cornell University, it includes the precise control of individual biological molecules, propelling biochemistry into "a new regime where an enormous potential remains largely untapped". Peter Vettiger of IBM-Zurich, on the other hand, demonstrated the `millipede', a prototype of a revolutionary - and purely mechanical - data storage device. The difficulty of pinning down a useful definition for nanotechnology can even be found in PHANTOM's full name - Physics and Technology of Mesoscopic Systems. Mesoscopic? "Mesoscopic fills the gap between the atomic and micrometer scales, where quantum mechanical effects come into play," Van Rossum explains. "Arguing about definitions may seem pedantic, but if you cannot define a science how can you run a research programme?" Nanoelectronics: Driving Force The arguments may continue, but it is certain that for the last decade the driving force in the field has unarguably been nanoelectronics - by any definition the future of the microelectronics industry. This emphasis - widely supported by the Esprit Programme at the European level - was natural, as by then the microelectronics industry had looked ahead and seen serious challenges for its cherished CMOS(1) process. CMOS technology has been refined for over 20 years, driving the `line width' - the width of the smallest feature in an Integrated Circuit (IC) - down from 10 to 0.25 microns. This is the force behind Moore's Law, which predicts that the processing power of ICs will double every 18 months. This cannot continue indefinitely. Early next century, feature sizes will enter the mesoscopic range of under 0.1 microns (100nanometres), where a number of serious problems await. Some of these problems are `merely' technological - it becomes increasingly difficult to manage heat dissipation from circuits as they get closer together, for example, while the cost of semiconductor production facilities is expected to exceed US$5 billion by 2006 (see table). Eventually, however, CMOS may hit a more fundamental barrier - the quantum world. How does one design a circuit which is so small that Heisenberg's Uncertainty Principle becomes a factor? In 1997, therefore, Esprit responded with the 25 million euros Advanced Research Initiative in Microelectronics (MEL-ARI) under its `Long-Term Research' activity. "MEL-ARI aimed to pave the way for a quantum leap to a new generation of computing devices," says Esprit Officer Kostas Glinos. "It is in fact two project clusters - one focusing on optoelectronic interconnections, the other on nanoscale IC production techniques." There are 13 projects in the second cluster. All concern radical advances in chip technology that could impact memories and logic processors early next century, including single-electron electronics, molecular electronic devices, nano-imprinting techniques, quantum ICs and nanoscale interconnects ( "Drawing the finest line"). There are also two associated projects on nanoscale resists and vertical CMOS.
A New Paradigm? Since MEL-ARI was launched, however, a number of more radical ideas have come off the science fiction bookshelves and into serious discussion. "New computing and manufacturing paradigms are being considered, and the field is starting to feel more multidisciplinary," says Glinos. "We're not just talking nano-electronics any more, we are talking about molecular scale processing in general, whether it be electronic, mechanical, biochemical or even quantum in nature. We have to establish a vision of systems, not just devices." This, according to IBM Zurich Nobel laureate Heinrich Rohrer, is about time. "The paradigm of the microelectronics industry is not the way forward," he argues. "We can only miniaturise two more orders of magnitude before we reach the atomic level, and it will cost an absolute fortune. Do we really need pentabit devices? Or would we be better off pursuing higher complexity, instead of smaller transistors?" Rohrer points to biological systems as a better paradigm. "Natural systems do a lot of processing at the periphery, and only transmit useful information - not raw data - to the centre," he observes. "They achieve this through `integrated complexity' - combining physics, chemistry, biology and electronics. That's why we need greater multidisciplinarity in tomorrow's research programmes." The time is right Glinos agrees with Rohrer that more interdisciplinary research is needed. So has the EC's view of nanotechnology evolved? "It is time to get more adventurous," he says firmly. "The nanoscale research has to result in some revolutionary feature, or provide new functionality. The reason is clear. We are unlikely to ever make atomic-scale patterns using today's top-down approach, so we will need bottom-up fabrication techniques, such as self-assembly. But these techniques are not faultless, so we will need to adopt the fault tolerance and fault detection architectures found in biochemical systems." In the new interdisciplinary structure of the Fifth Framework Programme, therefore, nanotechnology will no longer be only nanoelectronics (see box). "This shift in thinking was a matter of timing", says Glinos. "We have had `exotic' projects in the past, but we could never achieve critical mass because the field was too immature, there was not enough demand from science and industry. Today, however, the time is right." (1) CMOS - complementary metal oxide semiconductor, the basic process used in the microelectronics industry. Contact
|
|||||
|
