Chip-shrinking may be nearing its limits
The Associated Press
Published Saturday, December 15, 2007 11:05PM EST
SAN JOSE, Calif. - Sixty years after transistors were invented and nearly five decades since they were first integrated into silicon chips, the tiny on-off switches dubbed the "nerve cells" of the information age are starting to show their age.
The devices -- whose miniaturization over time set in motion the race for faster, smaller and cheaper electronics -- have been shrunk so much that the day is approaching when it will be physically impossible to make them even tinier.
Once chip makers can't squeeze any more into the same-sized slice of silicon, the dramatic performance gains and cost reductions in computing over the years could suddenly slow. And the engine that's driven the digital revolution -- and modern economy -- could grind to a halt.
Even Gordon Moore, the Intel Corp. co-founder who famously predicted in 1965 that the number of transistors on a chip should double every two years, sees that the end is fast approaching -- an outcome the chip industry is scrambling to avoid.
"I can see (it lasting) another decade or so," he said of the axiom now known as Moore's Law. "Beyond that, things look tough. But that's been the case many times in the past."
Preparing for the day they can't add more transistors, chip companies are pouring billions of dollars into plotting new ways to use the existing transistors, instructing them to behave in different and more powerful ways.
Intel, the world's largest semiconductor company, predicts that a number of "highly speculative" alternative technologies, such as quantum computing, optical switches and other methods, will be needed to continue Moore's Law beyond 2020.
"Things are changing much faster now, in this current period, than they did for many decades," said Intel Chief Technology Officer Justin Rattner. "The pace of change is accelerating because we're approaching a number of different physical limits at the same time. We're really working overtime to make sure we can continue to follow Moore's Law."
Transistors work something like light switches, flipping on and off inside a chip to generate the ones and zeros that store and process information inside a computer.
The transistor was invented by scientists William Shockley, John Bardeen and Walter Brattain to amplify voices in telephones for a Bell Labs project, an effort for which they later shared the Nobel Prize in physics.
On Dec. 16, 1947, Bardeen and Brattain created the first transistor. The next month, on Jan. 23, 1948, Shockley, a member of the same research group, invented another type, which went on to become the preferred transistor because it was easier to manufacture.
Transistors' ever-decreasing size and low power consumption made them an ideal candidate to replace the bulky vacuum tubes then used to amplify electrical signals and switch electrical currents. AT&T saw them as a replacement for clattering telephone switches.
Transistors eventually found their way into portable radios and other electronic devices, and are most prominently used today as the building blocks of integrated circuits, another Nobel Prize-winning invention that is the foundation of microprocessors, memory chips and other kinds of semiconductor devices.
Since the invention of the integrated circuit in the late 1950s -- separately by Texas Instruments Inc.'s Jack Kilby and future Intel co-founder Robert Noyce -- the pace of innovation has been scorching.
The number of transistors on microprocessors -- the brains of computers -- has leaped from just several thousand in the 1970s to nearly a billion today, a staggering feat that has unleashed previously unimagined computing power.
"I think (the transistor) is going to be around for a long time," Moore said. "There have been ideas about how people are going to replace it, and it's always dangerous to predict something won't happen, but I don't see anything coming along that would really replace the transistor."
But there have been considerable stumbling blocks in recent years.
One problem has been trying to prevent too much heat from escaping from thinner-and-thinner components. That has led chip companies to look for new materials and other ways to improve performance.
Earlier this year, Intel and IBM Corp. separately announced that they discovered a way to boost transistor efficiency.
The solution involves replacing the silicon dioxide used for more than 40 years as an insulator, but has since been shaved too thin, with various metals in parts called the gate, which turns the transistor on and off, and the gate dielectric, an insulating layer, which helps improve transistor performance and retain energy.
Still more novel ways to prevent electricity leakage -- and other problems -- are being pursued. And nobody has won a bet against maintaining the pace of innovation in technology.
"The only thing that's been predicted more frequently than Moore's Law has been its demise -- everybody's been wrong," said Sun Microsystems Inc. Chief Technology Officer Greg Papadopoulos. "It's a pretty robust set of observations and really it's about techno-economics ... It's a dangerous thing to bet against because of the economic investment cycle that's in there."