HALE GORDON MOORE: April 19th (Low) marks the famous prophecy by the famous man that computing power will increase drastically, stimulating the technological age of the late twentieth century.
Electronics and information technology is now touching almost every aspect of life. Beginning with the invention of integrated circuits in 1958, the continuing electronics revolution is, in large part, down to the industry’s faithful adherence to a technology known as Moore’s Law.
In 1965, Moore, a chemist-turned-electronic engineer, observed that in the years following the creation of the first integrated circuit, engineers had managed to nearly double the number of components, such as transistors, on a chip each year. He also predicted that the rate of component shrinkage – which he later revised to double every two years – would continue for at least another decade.
The semiconductor industry has never looked back. It has continued to shrink transistors and produce computer chips that combine increasingly high performance and functionality.
For the first few decades, the semiconductor industry accomplished Moore’s Law primarily through engineering talent and vast advances in manufacturing processes. But the important role of fundamental science is also worth remembering, especially as researchers today are looking for ways to keep up with the rate of progress.
The invention of the transistor at Bell Laboratories in Murray Hill, New Jersey in the 1940s was based strongly on the development of semiconductor band theory.
And scientific breakthroughs played an important role in the subsequent development of the technology. Notably, in 1970, Russian physicist Nikolay Basov and colleagues developed excimer lasers that would later be used to etch tiny circuit patterns on the silicon wafers from which the chips are made.
The 1990s called for further innovation. Until then, as transistors got smaller, their speed and energy efficiency increased. But when the components reached about 100 nanometers across, miniaturization had the opposite effect, deteriorating performance.
Chip makers such as Intel, which Moore co-founded, and IBM again looked to basic science to improve the performance of transistor materials. Big help came from condensed-matter physicists.
They had known for decades that silicon’s ability to conduct electricity is greatly improved when its crystal lattice is stretched—for example, by laying it on another crystal that has a different spacing of atoms. Engineers introduced viscous silicon into chips in the 2000s, and Moore’s Law held true for many more years.
State-of-the-art microprocessors now have transistors that are only 14 nanometers wide, and Moore’s Law is finally approaching ultimate physical limits. Waste heating in particular has become a concern. This has already caused the exponential acceleration of the computer ‘clock speed’ – a form of Moore’s Law – to come to a halt. Power-hungry chips also limit the ability of mobile devices to survive for more than a few hours between charges.
The introduction of advanced materials such as hafnium oxide, which provides insulation even when it is a few atomic layers thick, has managed to keep the chips slightly cooler. Heroic efforts may still yield one or two generations of tiny transistors, perhaps down to the size of 5 nanometers. But further improvements in performance will require fundamentally new physics.
where are we going? Transistors that use quantum tunneling, perhaps? Or those in which currents transport quantum spin rather than electric charge? Laboratories around the world are experimenting with methods and materials that can drastically cut energy consumption.
One opportunity that could be exploited is the inherent stability of the collective ‘topological’ properties of atoms: a modern twist on the ancient practice of tying information together in knots. Some researchers are trying radical ‘neuromorphic’ circuit architectures inspired by the plasticity of the brain’s neuronal networks.
A theory that works well in a physics lab doesn’t necessarily translate into something that can be mass-produced. And inevitably, most of today’s efforts will lead nowhere. However, society must be convinced that somewhere, basic science will provide a way to sustain progress. Moore should be proud that we have yet to find the exception that proves his law.